Physiognomy in ancient philosophy

Boys STNESL, S George Boys-Stones u: os Ar Physiognomy in Ancient Philosophy Some philosophers in antiquity believed that appearance and character necessarily tracked each other -- or tended to do so. This is a comprehensive study of the theoretical (psychological, metaphysical) underpinnings of this view, covering Socrates, Plato, Aristotle, the Stoics, the medical tradition (especially Galen), and the Neoplatonists. Publication Date: 2007 Publication Name: Seeing the Face, Seeing the Soul (Simon Swain et al.)

Seeing the Face, Seeing the Soul Polemon’s Physiognomy from Classical Antiquity to Medieval Islam Edited By SIMON SWAIN With contributions by George Boys-Stores, Jas Elsner, Antonella Ghersetti, Robert Hoyland, lan Repath, and Simon Swain OXFORD UNIVERSITY PRESS

Physiognomy and Ancient Psychological Theory George Boys-Stones Of all animals, human beings display the most remarkable tendency towards perversity of nature. But more than this, human beings display the most remarkable variety of perversion. It is as Aristotle said: virtue is simple, but vice takes many forms.! No two people are unnatural in quite the same way (one is more cowardly, another more lecherous, and so on), and while one can generalize about the way humans ought to be, there is only a limited extent to which one can generalize about the way they are. In a sense, then, it is easier to talk about virtue than to talk about vice, and this is reflected in ancient discussions of ethics, which naturally focus, first and foremost, on what it would be like to be a well-functioning individual. But, particularly in a context in which virtue, happiness, and what is ‘natural’ for a human being coincide, as they tend to do in ancient philosophical thought, there is a real and interesting philosophical question raised by the fact that people fail so regularly to achieve the ideal, and specifically a question about what it is that directs them down their particular path of vice. Is this to be explained purely by the education and habituation people receive in the course of their upbringing? Or is the inclination to form a certain kind of moral character to some extent innate—part of what makes them individual humans in the first place? And if innate, is there a link between physical peculiarities and moral character? Again, many ancient philosophers think that there must be, since our souls (as the locus of impulse) are intimately linked with our bodies, and the greater part of our moral character is constituted by our attitude to specifically bodily demands. When one looks at the question from this perspective, it is not so hard to see how the distinctive tradition of Greek ethical philosophy opens the way quite naturally to what we might describe as physiognomical speculation. Depending on the explanation given for the link between the body and the soul, a philosopher might easily find himself committed to the position that there is a regular correspondence between the two such that one could, in principle at least, tell what a person is like from the way he or she looks. It is important to recognize how naturally the course of ancient psychological thought leads into the territory of the physiognomist because it is something that existing studies of physiognomy in antiquity have tended to ignore. | Aristotle, NE 2, 6, 1106°35; and cf. Anon, Lat. 44.

Hd ni e form or other that ‘physiognomy’ in som her rat is n y, tio ump ass The typical ice) influenced philosoph ty or a sociological prejud ivi act and cal ed cti ept pra a acc as be er to (wheth cal truths ted itself as a set of empiri a in the sense that it presen s tends, then, to be either die stu h suc of us foc The . ers in oph on los ati fic phi nti by explained ough the ide luence of physiognomy thr ‘demonstration’ of the inf erial from the surviving mat h wit de nci coi es which theoretical texts of passag o the particular question int n or else an investigatio ;? oks dbo ter han al mic gno physio m appearance to charac ogy—how inferences fro iol t sem tha n al tio mic ump gno ass sio phy the of philosophers.” But by d ure uct str at e and wer are developed iously, or that they to take physiognomy ser philosophers felt compelled one that stands up to any not is my; d’ by physiogno loall in this sense influence of fact that not every phi a plain and simple matter commital tic ore the prolonged scrutiny. It is any yed pla school in antiquity dis t sopher or philosophical t makes the difference, wha all; and when one asks wha at wer ans the , ers oth not ment to physiognomy my and an interest in physiogno gnomy led some philosophers te tical possibility of physio ore the the r the whe to wn do es com e l. cas sou ry in eve nature of the y already held about the t followed from beliefs the at physiognomy in ancien k loo to is dy stu t sen pre a the t in hou n, wit the , My aim l discussion, and it arose in philosophica as ts t tex er ugh oth tho m al fro tic ed ore the ’ import t constitutes ‘physiognomy to prejudicial notion of wha as much a contribution be to ed end int is dy stu the h, All suc ) my(As gno ts. and contex t about physio ught as to ancient though ancient psychological tho ch might seem eccentric whi it of es tur fea to explain existing literature on this, I hope, will help the of by the standards ged jud of is it if ent or defici about the earlier practice t, I have nothing to say e hav ht mig one se: physiognomy. For à star s might seem perver Thi s.” gin ori its and physiognomy for Apuleius); midorus); Opeku (197% 9); Pack (1941; for Arte (196 -400 for Dio 1), 399 (194 393, 5), (193 esp. ns 0, (199 leg. Eva Apuleius again); Gleason for 4; Aurelianus): (198 on ius Mas Cael ; er André {1981) 9-15 9; for the medical writ ected at 6); Horstmanshoff (199 e of passa ges coll Chrysostom); Weiler (199 pare the wide rang m). Com 3; for Dio Chrysosto Bost-Pouderon (200 i (2001), 352, 7); Gleason (1990); Sass Foerster (1893) ii. 237narolo (1985); Dagron (198 228-30; Gra > ag, Burnyeat (1982), 9), 103-13, 145-63. gnomy in Athens (199 t Vog Cf 1. n) practitioners of physio 35-6, 69-7 for early {pre-Aristotelia ence us of DL. 2. evid mag an the rd, Syri reco the s perhap 4 For the yrus (see Section I), and Zop al activity at of t mic oun gno acc sio phy do’s includes Phae al or not, evidence for e’ {both of them, historic e of two or three animals thos to s face an 45 = Aristotle fr. 32 Ros hum all the man who compared d to go. Physiognomy is the time of Socrates); also this it is in any case har at GA 769°20-1. Beyond of Thersites in Iliad 2. ion ript desc s er' mentioned by Aristotle Hom ably ier Greek literature (not identifications are earl in such ted But spot 3}. ly 162uent : freq 1996 his lmann 1988; Weiler take case of Thersites, one might 211-19: cf. Mis tion-begging. (In the specific of his on ques cati and jndi ous an enti n tha tend ly er always high disfavour towards him, rath ’ their of gods h the trut of the of ay pro betr s her se look ugliness as furt ion can compass men who be inat ht mig imag s ch er’ whi Hom to , ff; case character. In any and Odyssey 8. 169 : 358, citing Iliad 2. 673-5 y physiognomicel activity earl for ence character; cf. esp. Sassi 1992 evid icit expl ener 1924: 103-4; Tha 216-24.) The fact that our 9) lends some credibility added Niad 3. 43-5 and ly Persian Zopyrus (cf. n. an magus and the putative Syri the Near East, On this, a in s lves tice invo prac ce ate Gree in k to cogn tice of physiognomy bac 100; Böck (2000); prac the 45), e (199 trac to on Bart mpts 9; to atte 9), (1974); ‘André (1981), (196 éro ms. Bott clai 5); such (193 in us lved see E.R. Kra difficulties invo for the general theor ical but also Lloyd (19918) 21 to take into account prethought that a history of physiognomical theory ought might take primitive ht thoug al medic of y histor a as just existing practice, ognomical theo physi and al medic medical practice as its starting-point. But was built on the mably presu theory al Medic cases. ent turn out to be rather differ an attem t sense some in ents empirical success of practical therapy, and repres argument My failed. others while d worke ents to understand why certain treatm starting-point cal empiri the never was hand, other the on y, is that physiognom appears to have assumed that for philosophical speculation: no philosopher ever , in its first appearance in Indeed ation. explan d neede it worked, such that it s of laughter from a guffaw philosophical literature, physiognomy attracted sed to be told that it surpri rather were mably sophisticated. audience, who presu philosophers who accept could be justified by Socrates’ views of the soul.” Those within a theoretical from so -do the possibility of a physiognomy consistently had physiognomists oped devel have would they model of human nature which physiognomy as a of e istenc never existed. For these philosophers, the pre-ex lary in which to vocabu -made ready a de practice does little more than to provi between body onship relati the on ation specul their of e quenc describeone conse | e people’s deduc ple inci princi in can " e that one quenc ance e conse namea «pear oul— n n i sols philosophers an Secondly, and relatedly, my study of individual it (explic m o that role the ing stand not conducted with a view to under reason, again, is that I do not or implicit) played in shaping their thought. The role in shaping philosophical active assume that physiognomy in fact played an l theories of various thinkers logica psycho the at looks thought. Rather, my study for their view about the s with a view to understanding what this implie and whether it allows ance, appear and ter relationship between individual charac a physiognomy. For as be descri to red prepa be would they that them something t mentions of explici of list a with this Teason too, I preface each section sion; and discus under s school and duals indivi the physiognomy relevant to a category as ative inoper mÿ is although I by no means assume that ‘physiogno discussion my , absent is word the se in the mind of a writer simply becau for any wider tone touchs the form ce perfor which texts, privileges these ssarily austere approach judgements. If all this seems to amount to an unnece of the study as a whole is sions conclu ry prima the of I can only respond that one everyone whose philosoph a direct consequence of it: for it turns out “hat not be predictable from his will ter charac ’s implies the expectation that a person ancient philosophical In al. gnomic physio as it of think does or her appearance that appearance belief the for cally contents, the word seems to be reserved specifi I shall argue) (as Thus ter. charac ed acquir to ed oppos we guide to innate as aedo of Elis, Aristotle, Posidonius, Galen, and (at least some) later Platonists l thought, but Plato accept physiognomy as a consequence of their psychologica the Stoics (unlike and Plato both h thoug ven not—e do Stoics and the earlier p characters develo often people that r example, the Epicureans) accepted ® The narrative is from the Zopyrus of Phaedo of Elis. For more details, see Section 1.

their appearance. This is not which, non-coincidentally, are predictable from wider understanding of our into only an important conclusion to feed sidelight on aspects useful a es provid but world, t ancien physiognomy in the of ancient psychological theory too. tical discussions of Finally, I appeal to my focus on strictly theore of discussion in the h breadt and depth e relativ the excuse physiognomy to Plate and the early about say to Sections that follow. I have, for example, a lot of themselves as think not do they that be will sion Stoics even though my conclu ely because precis say a lot to accepting the possibility of physiognomy; but I have hand, my other the On not. do they it is interesting and illuminating to see why ng than rangi wideless much is V n Sectio in ht discussion of medical thoug ly) entire quite not gh (thou might have been expected—indeed, it is almost before the One day, a foreigner named Zopyrus arrived in Athens. Like visitors to Athens in the Secratic dialogues of Plato (think of the Protagoras, for example, or the Ion) this foreigner had a wonderful skill to exhibit: he claimed that he could divine a man’s character from his physical appearance. Finding himself somehow faced with Socrates amid a group of Socrates’ friends, he demonstrated his art on Socrates and offered a diagnosis. Socrates was, he declared, a man possessed of ‘many vices’; the thickness of his neck indicated that he was ‘stupid and dull’; his eyes showed that he was a womanizer—or else, depending on whom you read, a 22 concerned with Galen. There is a simple reason for this, which is that, ssed in Ch. 5 by Antonella subject was taken up in the Arabic tradition (discu whom we know to have writer l Ghersetti), Galen was pretty well the only medica nd then scarcely qua all—a at level tical theore a engaged with physiognomy at the material is my to ing accord sion discus my cut to desire r medic, A simila generalizes over a which VI n Sectio excuse for a discussion of later Platonism in ht is concerned, thoug l logica psycho their as far so wide range of thinkers who, discussions of that is fact deserve rather more careful discrimination. But the strictly, no are, there indeed ed: physiognomy in Platonism are not so nuanc this case In lity. possibi its of ance accept nt appare an discussions of it at all, only show that ology psych ist what I have tried to do is to highlight themes in Platon and y, ognom physi s discus who s writer an interest in engaging with earlier nt releva the with agrees ist Platon ular partic any as suggest that, insofar cal— gnomi hysio the—p accept principles, we can understand how they might consequences. ognomy in ancient There is scope for the kind of discussion of physi r examination furthe for here: cover philosophy and medicine that I do not nce of noninflue the of s studie r furthe for ics; semiot of physiognomical ways which in theory l theoretical physiognomical thought on philosophica for other are sions discus such But s. are not necessarily explicit or obviou difference a is there that ish establ to with, begin to vital, occasions: it is implicit, perhaps even between explicit reflection on physiognomy and the ts. My aim has been, contex other in y ognom unconscious, influence of physi ancient philosophers what ish establ to : former the isolate to ly, very narrow those who accepted why and ’, ognomy themselves understood by the word ‘physi it did so. I. THE CIRCLE OF SOCRATES (1): PHAEDO OF ELIS Antisthenes: SSR VA 41. 12 (= DL 6. 16); Occurrences of physiognomon- in Socratic writers. (a) Rossetti] 6 (= Cicero, On Fate 10), 63 (== Athenaeus 14. 74. 11, 656F). (b) Phaedo of Elis: frr. [ed. r, On Fate 171. 11-12 Bruns; quoted at Alexande (= 10 24), 4. Satires Persius, to scholion (= 8 Cassian, Conferences 13. 5. 3), 19 {= Origen, Eusebius, Preparation for the Gospel 6. 9. 22), 11 (= Against Celsus 1. 33). 23 pederast. In either case, Socrates’ cormpanions, and Alcibiades in particular, had reason to laugh. The Socratic dialogue as a genre is not without an element of hagiography: the reader knows that Socrates is a paragon of virtue and abstinence, a seeker after the truth, and the wisest man alive. Zopyrus’ diagnosis must be wrong: the whole episode a typical example of a sophist deflated, of false claims to knowledge exploded, But there is a surprise in store. The onlookers laugh at Zopyrus, and the reader laughs with them; but Socrates telis us all to stop: ‘This is how I am,’ he said (or something like it: see further below); ‘but through the practice of philosophy 1 have become better than my nature.’ Such is our evidence for the most famous incident in a dialogue, the Zopyrus by the Socratic writer Phaedo of Elis. Apart from the intriguing, but oee, title of a work by his fellow-Socratic Antisthenes (the ‘founder’ of Cynicism) ; it constitutes our earliest evidence from the Classical world for the encounter between philosophy and physiognomy.® And the grounds for Phaedo’s interest in the subject are very important. In the course of this study, I hope to show that physiognomy attracted the interest of ancient philosophers not as a matter of idle curiosity; and not (for example) because of the value they or their culture might have placed on the ability to ‘read’ the character or intentions of others. Such an ability would clearly be a wonderful thing to have, and the aspiration to it is something on which the surviving handbooks of physiognomy trade. But, as I suggested in the Introduction, philosophers were interested in physiognomy (when they were) insofar as it represented a consequence of their central concern with the nature of the soul and its relationship with the body. It is possible that the character of Zopyrus represents the arrival in Greece of = Testimonia and fragments of Phaedo’s works in SSR HIA, with discussion at iv. 115-27. en is rigorous in excluding texts that do not name Phaedo; a more generous collection of testimonia for his Zopyrus :s collected and discussed at Rossetti (1986) 183-98. With most commentators (including Giannantoni himself: SSR iv. 126), 1 assume as the only serious hypothesis avalible for their provenance that the fragments describing the famous encounter between Zopyrus pe en {= frr. 6-11 Rossetti) ultimately derive from this work—although it is true that Phaedo ala unone of them. On Phaedo generally, cf. von Fritz (1938); Rossetti (1973) (a revised (1996) a = 1977: 121-53); (1974); Toole (1974-5); Montuori (1976); Duëanié (1993); Kahn the sta 3 Nails (2002) 231; Boys-Stones (2004). On the afterlife of this particular anecdote in © Denn tradition, see Ghersetti below, pp. 282-3, 465. : ystognomical Treatise On the Sophists (Peri ton sophistön phystogndmantkos): D.L. 6. 16 10 us appears actually to have been referred to as a ‘physiognomist’ (physiognômôn): frr. 6, ssetti; cf. physiologos (presumably a misreading) in fr. 6,

t” but a practice born in the Eas en it was) because it was readily taken up {wh ously converged with consequences of taking it seri and insofar as the theoretical less than a coincidence, , appropriate, and perhaps native interests. It is, in fact first surface among the uld with physiognomy sho t men age eng ous seri a t -tha tradition looked as the om the later psychological followers of Socrates, to wh ed in physiognomy, it iry. If Phaedo was interest proximate source of this inqu w of the soul which vie a d , and because he hel atic Socr a was he use beca was ld be held to support. physiognomical evidence cou iously’ engage with t of all, that Phaedo aid ‘ser But how do we know, firs ironic; that (did was s yru Zop of e Socrates’ defenc Zopyrus that physiognomy? It could be that of cule ogue) it would be the ridi and it so ext; we but have the whole dial cont d nee wer this question we ans To ds. min our in ed remain have might provide it. fragment of the Zopyrus we happens that the one other following story Ur. 1 in the Zopyrus told the We know that someone = SSR IITA 11}: Rossetti of Persia] made a pet of a hon youngest son of the King [sc. d that ‘They say, Socrates, that the was brought up with the chil that it was because the lion me to ians ms Pers see it the And that .. 80 cub. ng man, went even when he was a you it followed him wherever he the boy. said that it was besotted with grammarian Theon better reason than that the The story is preserved for no to direct speech in rect tion of the change from indi thought it a happy illustra Zopyrus seems the h oug (th it not know whe told narrative. It is lacunese; we do have nothing at all '° and, most importantly, we best guess on the face of it); in relation to the od sto the telling of this story to suggest where and how ore the powerful ign to d har is it , despite all this physiognomical episode. But mic gno al episode. In this narrative to the physio thematic connections linking ity that a creature’s ibil poss moral is, or involves, the le nsib oste the s, case h bot h inciinations can be icular that savage and brutis nature can be changed—in part licentious as Socrates that someone as stupid and tamed. In the one, we learn ber of the wildest mem a opher; in the other, that los phi ste cha a ome bec can otted lover. as tame and broken as a bes species of animal can become readership might his a moral is one that (Epistle 82, Evidence that just such eror Juan edo comes from the Emp have expected from Pha ates] was 9 Rossetti (‘When he [sc. Socr ide Athens is clear from fr. e “from Syria’ if he is to be 9 That Zopyrus was from outs com have d woul He ..”). came to Athens. totle fr. 32 Rose = DL. alive, 2 man called Zopyrus d a violent end for Socrates (Aris predicte identified with the magus who ght him a Persian on the 2. 45), Others have thou basis of his al detail that name, and the circumstanti ec. below in ian prince. (The story is quot a story concerning a Pers someone in the Zopyrus told ian name is not distinctively Pers Rossetti 1977: 145-6.) But the between Zopytus and the story cf. , ment argu the for text; link the pendently tells in favour of a (cf. LGPN s.v}; and nothing inde indication note). e to Zopyrus presence as an ia could as well be in deferenc identification of Zopyrus the 1 But the reference to Pers e (sinc al dent inci be d then again, it coul d out by the that he was the narrator; and . Socrates himself is not rule on the reference to Persia here) ble of relating capa e quit is ay, nationality itself rests in part anyw tes, d to him: the Platonic Socra y addressed fact that the story is addresse s. He is, for example, regularl ussions he has had with other ff. D 201 stories as told to him, or disc osium ussions with Diotima at Symp within his own account of disc of the Prince (see next 25 445À = SSR IIIA 2): but that i a nothin, was beyond the cure of philosophy, that : Phaedo. „supposed can Purified of Fr u of life through it—of their behaviour, tes, everything, in:à | helped only the well-born and well-b ; the sort. If it of he the light, à en tteto a bad state bre brings people in such it did, but ifitit brings a about what ne umassing nothing nn amazing i i There is no way of telling on which of Phaedo’ Rn Test se offered pr the that case assessment; | but it does seem to be the candidate to whom such a moral would be particularly appropriate. For there is one further link between the two episodes of the Zopyrus I have been considering: the person of Alcibiades, According to frr. 6 and 8 Rossetti, it was Alcibiades in particular who laughed at Zopyrus diagnosis of Socrates; and. although we have no information that he was there to hear the story of the li m cub, we might remember for ourselves that Alcibiades compared himself and was compared by others to a lion.'? The possibility of this connection is surel sent by the fact that Alcibiades had a tutor who shared his name with 7 e dialogue s eponym: Zopyrus. Indeed, it might not be too fanciful to suggest at the first readers of the Zopyrus were supposed to assume from the title that they were-purchasing yet another dialogue about Alcibiades and his education. ? Whatever the truth of the matter, the important conclusion for now is that such evidence as we have for the Zopyrus aside from the physiognomical episode suggests a remarkable degree of convergence with a reading that treats Zoe us diagnosis seriously, not ironically, It starts, in other words, to look as if one of me themes of the Zopyrus was the transforming power of philosophy, and that :7 ‘i pyrus as a whole is best served if Zepyrus really did get Socrates right aedo’s Socrates, then, was not being ironic in defending him and confessing t ; a wicked nature: the point is his reform through philosophy. But if this should be accepted, then we can go further and ask by what mechanism Phaedo might have explained all this. What might Phaedo have believed about the soul to lead to the conclusion that ‘natural’ character manifested itself in physical m but was the kind of thing which philosophy could overcome? zo ere — course, no reason at all to suppose that Phaedo ascribed to 2prs a theoretical view of the soul’s relationship with the body, or that this is ocrates was supposed to be in agreement with him about. In fact, the ‘ ynamic of the dialogue would be better explained if Zopyrus had no theory (1938) 1540, and Kahn (1996) 12; but Nails E u " The Zopyrus is a possibility:4 so von Fritz cibiades 2 123 SFth Tine nature eg by Plt(Aebides A: om, Arsophanes (Frogs 1431-2), and Alcibiades himself at Pleterch, pe Denyer 2001: 146) was the dialogues ro supposes that Alcibiades’ tutor (for whom see Plato, Alcibiades 12248) “Zopyrus’) Nails, 2002: 305 s.v. bncspor ym el more cautiously at 1977: 145; cf. also at the time of his encounter Athens in stranger a was Phaedo’s Zopyrus with Socrate (and i hy if Ache nn impossible if one thinks that Zopyrus was a Persian: see again n. 9 above), For gure of the Socratic dialogue, cf. Denyer (2001), 5. Denyer notes that Alcibiades played © él) Amisthene eid alrole in Plato's Protagoras and Symposium; and that Aeschines (DL, 2. (Sue cay Phan 4 also D.L. 6. 18}, and Buclides (D.L. 2. 108) as well as Plato, and Phaedo himself lén= > 154) are all credited with dialogues called Alcibiades.

spent hing like Plato’s, Socrates would have at all. If Paaedo’s dialogues were anyt d ahea ran ties sely to people whose abili a good deal of his time talking preci and s’ ‘skill se —who ning etical underpin of their capacity to give them a theor ”* should have been knowledge-based. they e wher l, rica empi ‘virtues’ were e thos like but, t omis ble enough physiogn Perhaps Zopyrus was like this: a capa 271 cp), e was also about the soul (cf. sophists in Plato whose discours Phaedrus judged. virtues and vices of the souls he lacking ia an understanding of the ates to Socr do’s Phae for es what Zopyrus leav Knowledge about the soul, then, is omy (as iogn phys t rejec to cted expe be not explore. Socrates for his part need show how t thetoric), but he might want to rejec tly exac not does ates Socr ’s Plato nded on grou l, hica osop phil e it can be mad it is only a worthwhile pursuit if ns from anio comp his ped stop he why is this knowledge about the soul. Perhaps laughing. tion is ates think about the soul? The ques But what, then, did Phaedo’s Socr ied repl he ly, sources differ over what, exact a complicated by the fact that our is e ther , cular parti In ras diagnosis. to his companions’ mirth at Zopy acter char ral natu his ged chan had he ght difference over whether Socrates thou er thought that is changeable) or whether he rath (and so, whether one’s nature be something to out s turn re’ one’s ‘natu he acted in spite of it (so that e). immutable but non-determinativ g to former model in his source. Accordin Cicero certainly talks as if he saw the rus, Zopy by ified ident vices the with horn him, Socrates admits to having been tice of himself of them through the prac but states that. he managed to rid philosophy: [Of the vices ascribed to Socrates natural causes; but it by Zopyrus:) It is possible that they were born from al causes that they were is not due to the power of natur and altogether removed so that rooted out vices to which he he himself was called away from those Cicero, On Fate 10). had been prone (fr, 6 Rossetti= vices had been ... who said that, although those o, ti = Cicer [Zopyrus was defended by Socrates] sset Ro 7 (fr, on reas out of himself by implanted in him, he had cast them Tusculen Disputations 4. 80). ates was get at the nature with which Socr The fact that Zopyrus was able to isposes pred that e forc a r othe or w is somcho born would suggest that the body ult’; ‘defa its n dow in a certain way, or lays e one’s itrational nature to develop uenc infl ing heal the to n ope re, which is but it cannot determine one’s natu of reason as well. e majority of our fragments; -and, whil Cicero, however, is at odds with the no way have {we s case e thes in t emen judg a head-count is no way to make a of the fragments are based on a reading of knowing whether or when our of cui e cipl prin the , not) are they source if original, and no evidence for their by ied impl gy thing, the bipartite psycholo bono? speaks against him too. For one 14 The bravery of Laches, despite example. C£ for the theoretical point same title s, would be a good his inability to define bravery in the Lache the two types of physician at Plato, Laws 720 AB, who share the are slaves and acquired their skill by ‘whether they are free men, or whether they obeying h is how study of nature, whic experience—rather than from the and observing their masters, and by own sons.’ their to it teach they how and free men learn it themselves 27 Cicero may come close to his own preference in the area.” For another, such a psychology provides him with the most robust stand-point in the particular polemical context from which his evidence comes.'® It seems more plausibl all round, then, to prefer the testimony of the remaining fragments, which al 2 the evil ates somehow retainedlife: imply that Socr nature identified by Zopyrus, but did not let that interfere with his Bomen fr. 8 (=10Scholia overcome to Persius that very 4. 24; libido, Socrates speaks): speaks): ‘ ‘I am indeed libidinous; ibidi but it is in my fs 9 < Ps.-Plutarch, Peri Askéseds £. 179; Socrates speaking): "7 ‘In actual fact, this man no got it wrong, because by nature I am very much inclined towards Just [sc. for women]; but through careful practice I am as you know me.’ | fr. 10 (= Alexander, n On Fate 6):2He would h: ave been like i this, i as far as his nat concerned, but through the practice of philosophy he became better than this nature. fr. 11 (Cassian, Conferences 13. 5. 3): T am [sc. what Zopyrus says I am], but I abstain.’ ‘I am’, says Socrates (frr. 8, 11); ‘ incline’ B betier than a nature which he ventielen retains Ge. 11 Socrates has become y suggestion, then, is that we can, on the basis of : i i the following claims to Phaedo: (1) that each person has res which gehis or her irrational impulses; (2) that this ‘nature’ is related to the el y in such a way that an expert in the matter could deduce the former from the appearance of the latter; and (3} that one’s ‘nature’, tho ineliminable, does not determine behaviour, for people who en a en philosophy, at least. If this is right, then it suggests that Phaedo is workin with a bipartite model of human behaviour of a kind that, in its general outli : is quite familiar from post-Socratic philosophy: a model according to which a person's actions might be explained either by non-rational desires (nature) orby rational decision, operating independently of desire. What is most strikin about Phaedo s model, though, is that it does not seem to allow that the non, ational desires are susceptible to training or rehabituation. Their persistence is etrayed even in the case of Socrates—whose success as a philosophe someone who has brought his behaviour under the control of econ is presumably not in serious doubt. The non-rational impulses are tied 7 : les15ee an we know that Cicero shared Carneades’ inclination towards a more or + ics (eg. On Duties 1. 2 with Academica 2. 139 and Goul “Calli Pepe etten with towards a more or less mentioned there): this mayy well ha:ve been associated i i an inclination inclinati opde Oyle _ Cicero, Cicero, at On Fate 10, h is guing argui against i the Stoics’ i claim that action is predetermi ee of one's nature. His reply is not to deny that mitre determiines action deny that our x rok at least, is the guarantee that an agent is responsible for what he does), but to oe a are themselves predetermined. We have, he says, the freedom through reason Th what we wish—witness Zopyrus and Socrates. See again Bobzien (1998), 297 | e work is preserved only in Syriac. Rossetti pri enderie Gildemeister andaehe Bücheler (1872), emd Nater nine ih 1872), 527:527: “In Wiiren t hat dieser Mann nicht gelogen, denn u hi: zur Begierde, durd durch angewendete So rgfalt aber bin iniich, wie i ihr hr mi ne i might be relevant that Cassian gives Socrates’ response in Greek {einti gar, re translating it into i think thet he icy Latin, 1 which suggests that he is quoting the original (or thinks he is, or wants us to

the eyes or the shape of e educable than the set of the body, in fact, and no mor Phaedo claim this? the neck. But how could rtite model of human suppose that Phaedo’s bipa ng It might be tempti to the sense that he thought rtite model of the soul, in behaviour must be a bipa from the body; and inct dist entity substantially an was l sou the t tha (a) al and one not. But in this ed of two parts, one ration (b} that the soul was compos that there was something pose that Phaedo thought case, we should have to sup g so closely to the body: clin it e of the soul which mad t par al tion irra the ut abo a perfect inertia against any s with the body, but also it derived from the not just an affinity of sort let alone eradicate, the habits , nge cha to n eve son rea of attempt have believed this. For ont unlikely that Phaedo could hly hig r me to ms see It y. bod but for another, it is unclea h no parallels in antiquity; re whe e cas a im , ned thing, it is a position wit gai e advantage Phaedo might hav what possible explanatory from asserting that any determined by the body, y full so was al tion irra the n them existed at all. held something like substantial difference betwee then, would be that Phaedo ity, ibil poss ve acti attr e A mor be that ‘psychological the soul. The idea would erent an ‘emergentist’ view of are not substantially diff h desire and reason) vity. acti l ica log sio functions (including bot phy on e’ ven , they somehow ‘super her Rat e all. at mpl y exa bod (for the s m fro qualitie e that some supervenient l ica log sio phy the A case could easily be mad of ined by the nature erm det y full ain ree rem deg a the desires) g reason) acquire end, while others (includin activity on which they dep existence and capacities its owe ays alw son would of causal independence. Rea might possess a degree l factors, but in operation to the relevant physiologica that desire altogether my, ono aut y large degree of ver a s hap per my, ono aut of ity; and indeed lacks. greater philosophical credibil has ly ted oub und w vie a Such But it also has historical ly ruled out for Phaedo. J think it cannot be positive ce for emergentist theories for the only (other) eviden problems to contend with: and, more than this, t they post-date Aristotle tha ts ges sug ity iqu ant in of soul nomenalism known to him. The form of epiphe that they were inspired by have relied on a simple the other hand, seems to Plato and Aristotle, on elements; to have been al sic h a ‘harmony’ of phy wit l sou the of on cati identifi independent activity of ch left no room for the a form of reductionism whi .”” reason so essential for Phaedo t to hus, if Sharples (2001) is righ ld be Aristotie's pupil Dicaearc on 2001; and cf. Gottschalk 1971). 19 The earliest example wou nst this Cast tion to him (but see agai below, Section Ill ascribe an emergentist posi soul more generally, see the of re natu over the tes deba by Galen). n elia lied stot supp -Ari t evidence Fot post Section V. ii (for the importan Socrates at (esp. references in n. 69), and posited by Simmias at Phaedo 855-860 (refuted by ry m his lost dialogue, the 20 For Plato, see the theo 0; also fr. 45 Rose? (fro 08°3 o e, see DA 1. 4, 407°27-4 an emergentist, and Plat 91 5-95 A). For Aristotl ; but if Phaedo had been ex silentio are never ideal t which T have more abou sion deci (a Eudemus). Arguments do Phae narrator of the e Phaedo the important for knew it, his decision to mak attack on reductionism so very strange indeed. The been reminder that have tant ld wou cons w) the by belo d ine to say soul would be fatally underm the of h of Socrates’ ity muc rtal to immo le the erab establishing henomenalism less vuln rnative form of epip Phaedo himself held an alte argument. 29 Physiognomy and Ancient Psychological Theory George Boys-Stones If this is right, then we are left with a third possibility, whi between the first two and has, it seems to me, the ay wae eileen and hsore plausibility. Itfous be that Phaedo believed in an independent ‘on the one substantially distinct from the body; but na. Such pe epi explained desires on the other as physiological epiphenome wouldEko physical pra an be used I a guide to the Fer ure’): they are little more than an a one’s ich It would also explain why reason could have no effect otic oat ba ee more ange one’s nature’ through reason than one could improve one’s PE eo by thinking about it. But what reason can do, as an ape au pe to take charge: people can make a rational choice to organize their lives in any way they sec fit, despite the predispositions writt a’ physiology: reason is precisely not determined by the body in which it i pie ial it seems to me that there is a sub ercly) resides. pi pt oe this to which lends its support evidence En is generally ignored in discussions of Phaedo. For the Position ear ed jo Phaedo as the most plausible explanation of the fragments ue g from dialogues written by him converges to an extraordinary degree the psychological model championed by Socrates in another dial h Labs which he is the narrator: I mean, of course, Plato’s Phaedo heSeemsRE the possibility that the Phaedo can be used as evidence for the dialogue is narrated.D Prado though a draw attention to the fact that it is even king Pla en dei es fom À Lis oem eener the discussions narrated*!—but also because of the extent it psychological model assumed by Socrates within the dialogue 5 Beins oo. ks) rather different from anything we get in other Platonic texts. This is tr 4 al particularly in Socrates’ treatment of desire—desire, that is for co stimulation or satisfaction. According to the ‘standard’ Platonic account"his esire forms a i tinet part of the soul, of which another part is reas ae reason (with wie it may conflict), such desire is a Psychological What makes this a plausible account of desire is, first, the vey am mo I tion. it is one source of impulse for a body whose life and activity = epend on the presence of the soul; and, secondly, the fact that the pleast posited by desire as the end of human activity is itself something that sters pehi je In the Phaedo, however, Plato appears to be trying something the = desire is not of the soul at all, nn ze ne wen ingto in of be to ee i Pa idea àseems en Theiden of it. The bu:Has as a a distraction to it,t, notnoi a wayward part onanimated by the directive presence of reason, the needs and the eten ody assert themselves as appropriate objects of reason’s care. In many Phaedovedo isi made to =that Plate was prevented by illness from being present (59 8 10). Note also the emp 21 pei ere ertion at the very opening of the dialogue of Phaedo’s right to narrate as a witness of scussions: “Were you yourself there, Phaedo?” “I was there myself,, Echecrates id i 22 Compare esp, Philebus 21 A-D (for (f the mental dimension i i to pleasure); 35 cp (for the location of

divine rnation to forgetfulness of its proper, cases, reason (brought through inca body. the of e thos with ify its own interests sphere) actually goes so far as to ident to alien erly prop are it, from e es that com Nevertheless, the body, and the desir his to man a or as a guard to his post (cf. 62 p), the soul, which stands to them Socrates to the cell in ned dem con the as aps, perh ed, cloak (cf. 87 B-E)—or, inde which this dialogue is set. and ial resemblance between the Phaedo jt would be wrong to deny the famil , cular parti In soul. the of r usses the characte other dialogues in which Plato disc drus, Phae and ias Gorg the in if, and on? Plato never denies the primacy of reas tial and the soul that it seems to be an essen desire is so far from being alien to might what s take o Plat eus blic and Tima immortal component of it, in the Repu do. Phae the of that and view this een betw be thought of as the middle ground not dent acci as but desire is of the soul, indeed, According to these dialogues, h whic on reas rtal immo to t imen accompan essence, so that it becomes a ‘mortal’ ly tual even dominant theme of the Phaedo) might (in images convergent with the played Plato never denies the crucial role tly, rtan impo be ‘purged’ of it 2* Just as if the Even ally. gener e mor soul irrational by the body in shaping desire, or the h the whic by soul, the body, as the means desire for pleasure springs from the an by n take e shap the significant input into pleasure is attained, naturally has a onic Plat d’ ndar ‘sta the in do can the body individual’s desires. But this is all shape the in more detail in Section II). It can t poin the to rn retu shall (1 unt acco more soul ional irrat the ; but desire (and direction that irrational impulse takes ily ssar nece not and , body the nct from generally) consistentiy remains disti determined by it. e is, ented in the Phaedo is unique. Ther This, then, is where the model pres over d min his ged chan lly actua o Plat that to be sure, no clear case for thinking are we l mode the be, for example, that the nature of the soul. It might ures feat fic speci in certa ore expl nded to presented with in the Phaedo is inte ry, not to challenge it altogether.” theo l gica holo psyc d’ ndar ‘sta ’s of Plato is all remain striking enough, even if that But the differences in presentation d as ente pres are es desir do, Phae ing of the they are. On a straightforward read ng duri not, but else; desires can be resisted, functions of the body, and nothing or on reas with m the ’ for ‘harmonizing life, eliminated. There is no scope 23 Johansen (2000, at 109 with n. 34} notes Republic (611 p) and Timaeus (42D, 90D). 24 Republic esp. 10. 6115-6124; Timaeus nal’ in the references to our rational nature as ‘origi just soul from esp. 41: 69 cD (423 for the escape of the incarnation). mature differently in its to accept that the soul manifests its in 25 Perhaps, for example, we just have ht of as the operation of the soul thoug be can e’ ‘desir that so incarnate and discarnate states but one which is, as it were, soul, the of ity capac tal immor respect of the body—an essential and in perspective, be thought of as n 50 that it can also, from a certa actualized only during incarnatio . In this case, Plato is not body) the with outlast connection mortal (its actualization does not es as it fits the needs of ectiv persp rent g the soul from diffe contradicting himself, metely considerin Gerson (1987); Rowe 2-5; ) ind (1973) 27-31; cf. Bluck (1955 different dialogues. So eg. Archer-H the Phaedo has an intertextual that ent, argum my affect not does (1993) 9; Johansen (2000). Tt sophically substantial or o, whether these differences ate philo relationship with the historical Phaed explained. be to need e ectiv persp of s difference simply a matter of perspective. Even 31 subduing them by it, because there is, more generally, no ibili al the uprooting them from the body. And this suggests à Free standard’ Platonic model. For according to the standard model, in which desires are properly part of the soul, one’s natural character can be worked on and improved: desire can come under the influence of reason and be trained to a better state. But as far as the Phaedo is concerned (at least on a straightforward reading of it), one’s natural character is ineliminably inscribed in one’s bod: Of course reason can resist: desire does not determine behaviour.”* But reason cannot eradicate inclination. In these terms, people have control of their behaviour, but not their character or ‘nature’. But, according to my earlier reconstruction, this is exactly what Phaedo of Elis seems to have thought In other words, it seems to me that what is distinctive about the Phaedo might be ap ica e by the views of the historical Phaedo whom Plato chooses to It will be objected that my use of Plato’s Phaedo as evidence for the views of in that Phaedo of Elis cannot be quite right—or what becomes of Plato himself dialogue? Surely Phaedo is some kind of spokesman for Plato? But my answer to this is that we need take seriously both the fact that Plato in fact wrote the work and also the pointed and emphatic manner in which he places his authorial Vie in the mouth, not just of a friend, of another member of Socrates’ circle, but actually of a fellow-writer of Socratic dialogues. This suggests to me that we need to read the Phaedo as a strongly intertextual dialogue: in other words, that the question of whose view is expressed in Socrates’ theory of the soul in the Phaedo is crudely put. It is not Plato’s or Phaedo’s view: it is Plato's view in engagement with that of Phaedo. A significant number of commentators have explained the idiosyncratic psychological theory of the Phaedo as Plato’s exploration of his psychological model from a particular point of view?” All I am suggesting, in effect, is that we can be more specific about what that point of view is: that the Phaedo is written from the point of view of Plato’s engagement with Phaedo in tackling the issues raised by those who deny the soul’s immortality. As I have suggested elsewhere (Boys-Stones 2004), the reason that Plato might want to invoke Phaedo in this arena is precisely because his main interest is in provin the immortality of the rational soul in particular, and especially in proving ite aga against the ‘harmony-theorists’ who take the soul in its entirety to a epiphenomenon of the body. Phaedo provides an ideal standpoint from wi “ to conduct his argument: for he agrees with Plato that reason is immortal; ut also agrees with the harmony-theorists that some psychological functions are epiphenomena of the body. In Phaedo’s voice, then, Plato criticizes the way in which the harmony-theorists extend a plausible claim that desires are epiphenomena of the body into an untenable claim for the mortality of reason : 26 nature of desires which leave no room for Despite the apparentl H i j N y compelling phil ps . and the unavoidable demands of pleasure and pain which ‘rivet’ the soul u eu , son maintains control in the end. To think elevated thoughts in the grip of acute agony, = un Le might not be within one’s power; but reason can remodel one’s life by sponding =~ ie — DE and pains are likely to feature (B3 B), so that one’s behaviour e co urage, temperance, and justice justi (68 B-69 c). 2 See apna à. 25 above

nomena of the believe that desires are epiphe as well. Plato himself does not r works; but othe in c topi in his investigation of the rather as (Le. body, something that is clear way that m the of ks thin even if one on is an in the Phaedo he can argue that reas that e weight to the suggestion Phaedo does) it gives no mor well. epiphenomenon of the body as ussing Plato this point partly because in disc e mak to s pain I have been at be taken not can d to claim that the Phaedo between in the next section I shall nee ship tion rela the ut for Plato’s views abo course, of , straightforwardly as evidence for now physiology and character. But be can the body and soul—between est, sugg I , t Phaedo believes; and this his of nts the important issue is wha gme fra h wit ng alo of the Phaedo taken summarized from my survey res and emotions are According to Phaedo, desi ows. foll 35 own lost dialogues, can understand the l states. To this extent, Phaedo epiphenomena of physiologica t not be misled mus one that es t claim. But he argu force of the epiphenomenalis reason comes about the thought) into thinking that (perhaps by the economy of feature of human t son is a qualitatively differen in the same way. In fact rea ent on, ox even end dep or by, is in no way defined activity, and something that ously are. When y—as emotions and desires obvi (therefore?) confined to the bod s its capacity to ider cons ectly analysed, when one one read the the activity of reason is corr ght (mi on acti g l desires in directin oppose the body and corporea , mind?) one sees that h the moral of the Zopyrus in argument at Phaedo 93.4 wit , and free, and immortal. it must be something distinct abilities must be is that Zopyrus diagnostic this of all The conclusion of like in the sense just by looking what people are limited: he cannot, after all, tell their physical in g hin not behave because there is But Phaedo of telling how they tend to nt. inme atta l hica osop cation and phil he will be appearance 10 betray their edu that ion underpinnings for the expectat acter of char ’ can give him the theoretical ural ‘nat —the non-rational inclinations r thei tly exac ern disc iology to phys y able the ver this is an epiphenomenon of the irrational soul—given that d appearance as well. which explains their outwar with those who temporary scholarship to side con in There is a tendency ul essf only when it ume that physiognomy is succ laughed at Zopyrus: to ass n (if this should tha er someone actually is, rath accurately captures the way cy is supported den ten s Thi re. or underlying natu be different) his or her original siognomy is useful point of the handbooks that phy by the fact that it is a sellinguse of the word al sion this; also by the occa precisely because it can do innate character, to not r, refe to ) hical contexts physiognomy (outside philosop a h use of the word However, we shall see that suc but to current dispesition.” dbooks improve han al mic secondary. Physicgno ctly stri , hers osop phil for , was 28 And perhaps Phaedo spoke of h the terms used by Socrates its “purification” (katharsis) in muc do believed Julian (as cited above) that Phae case, we have the evidence of nd the cure beyo was ing passira in the Phaedo. In any noth that ed philosophy: ‘Phaedo...suppos in the purifying power of atresthai) of any kind of life that everyone can be purified {aniaton) of philosophy, but (kath m will be on, 98: the crowd outside the courtroo 2 CE Demosthenes, Against Aristogeithow the jurors have voted, through it.’ nesousi) by physiognomy (physiognémo able to tell ME EE ge traits 33 expression and movement, for example— odifications of 1 innate character. . But such modificatit ; : aEr heieof Be. properly (or anyway philosophically) so called. It increasingly apparent that, at least in philı i titom aancan tell i that one i hysio to the claim is restricted precisely contexts, ‘physiognomy’ appearance the innate character of aoP person1 or irrationalsoul. The es anparent fom the fact or of of hihis or her irrational This at claims to physi pe in philosophers who do not believe that he soul has rs eq den psychological theory leads them to expect that evelop in ways predictable from the individual “my, conversely, psi conversely,there there isi a Among those philosophers who do emb race physiognomy, der, even in consistent tendency to suppose that the soul hi thoughJ me be non-determinative for behaviour. innate character, even n what follows,> I shall gather evidence for this position dei by lookin king i ap glancing brielly at Antisthenes and Xenophon on the way) Austotle, and return to the question at the end of Secti IV, before traci ; oics; I shal bout de: of later antiquity. In order to be cl physiognomy through the evidence Phaedo, Hower I shal SUE te. here my conclusion: all the Gentes jos 1 reflect ct on the link between appearance and character; but ae ca ne to Phacdo) aa claim to physiognomy; and the on al , to this extent like Phaedo, thinks that the linkis with i \ d pende f deserves ridicule,, itit is 'only becausehe Zopyrus If character, observed “hehehas & il connections has the: theoretical grounding for the empirical This theoretical grounding, I suggest, was supplied by Phaedo, for om physiognomy—the diagnosis of specifically irrational character from outward appearance— was a natural consequence of his psychology. 7 ® * ob: II. THE CIRCLE OF SOCRATES (2): PLATO to be int i Phaedo of Elis was not the onlly Socratic i philosopher i nature ofthe soul and its relationship with the body far from itDat ae far aswe can t , he was the only Socratic philosopher who made the soul—at Ì ; rational part of the soul, the locus of moral character — directly a Tmaetion of ir oe= | =then, I wish to suggest, no coincidence that for Phaedo alone een my jan philosophers we can reconstruct a positive attitude towards . It is true that Kenophon, while conceivi something essentially distinct from the body | n its irrational nal as wellas it : in rational aspects, has Socrates encourage the painter Parrhasius to think thet he apture the character of the soul (to tés psyches éthos) in his portraits.” 30 pes Thischapter (whose dualism, by the way, is seen as anomalous by Claus 1085, physlognomy ‘a valet e only passage on which a serious argument for a philosophically reflective Pa smophon cone be based. So Foerster (1893) ii. 249, inviting us to compare with de ee: mn B28 and Anon. Lat. 50; also Métraux (1995) 3, suggesting ‘affinities pognon] iat Physiognomy, Evans (1969) 46 and Misener (1924) 106 also influence in Xenophon’s descriptions of Virtue and Vice in his narrative of the

34 a natural psychological suggesting that a person has But he does not do this by ical characteristics that is linked to the natural phys character that in some way argues that a person's an individual. Rather, he distinguish him er her as h physical expression oug thr lf itse tend to manifest character or state of mind will Xenophon does not t proposition, and one that and posture—a very differen e that Antisthenes tru also siognomical.”” It is phy as of nk thi to us age encour cal Treatise On the omy’ in the title (Physiognomi wrote a work with ‘physiogn t of this work that could know nothing of the conten Sophists DL 6. 16). But we ught he could supply tho es dence that Antisthen evi as e titl its ng usi our justify the likelihood is that a physiognomy.” In fact theoretical grounds for altogether, in which tence of the irrational soul Antisthenes denied the exis ance was à guide to ear app ed that physical pos sup e hav dly har ld cou case he ” innate character or inclination. think that the evidence of Phaedo is enough to to , then ng, wro be It would mber of Socrates’ n that Plato, as a fellow-me tio ump ass the to us e pos predis has to say about beauty gnomy. Indeed, what Plato circle, is interested in physio thought. For Plato i-physiognomical strain to his already suggests rather an ant the body and the of uty bea that exists between the makes great play of the gap of beauty are no guide that outward manifestations beauty of the soul—the fact unusually ugly to look : the supreme example of this to the inner mar. Socrates is compares him to a ly ous fam of soul. (Alcibiades at, but unusually beautiful God inside: Symposium ns up 10 reveal an image of statue of Silenus which ope er seem to me allegorical. (In lia 2. 1. 22, though they rath choice of Heracles at Memorabi particular, ical decadence understood as indicating phys arkia, of Vice is probably better e.g. Ps.-Aristotle s suggests on the basis of on OT intelligence, as Evan 2. 19; 4. 18) are cf. 5-7; 5. m osiu than a deficiency in percepti Symp {esp. us descriptions of Socrates is made at that t poin fic speci 807°19-28.) Xenopaon’s famo The (pace again Evans, ibid.). the ‘plumpness’, polys acter cyes linked ta no analysis of his char Socrates finds his snub nose (for smelling) and bulging ul usef how is 5-7. 5. m one should judge a body by that Symposiu sely preci age, mess l osophica serious phil (for seeing all around); the arance (and cf. next note). of the soul, not by its appe its suitability for the purposes that humans cast sympathetic e in turn to the propositions: (1) ow 31 Socrates gets Parrhasius to agre by the painter as well as the eyes cans (2) that joy and sorr depicted and finally, (3) that ed; oduc or baleful looks which can be repr be can too essions, 50 that these violence and cause different facial expr temperance and wisdom, , baseness and illiberality, The “magnificence and liberality her they are at rest or moving.’ whet and the attitudes of men een soul and body, but cidence of character betw ophon is (not the coin Ken for this uses it as a tool. In the all of soul t poin s of the body that the can tell from modification of the soul’ in his work rather) the fact that one he can capture the ‘activities vulgarity shine though the face a sculptor that is not the same passage, Socrates teaches maker that the ideal breastplate 3. 10. 8), and a breastplatelia rabi Memo (3. 10. 9-15). The erge: body cular (ta tes psychés parti the to ed suit er the one best it is for the most obviously beautiful, but rath ld be judged by how useful one, that the body too shou moral is obviously the related attractive it is. we have is soul's purposes, not by how e fragment of the work that oni, SSR iv. 281-3. The singl ‘For they r: 656 11, 74. 14. 32 Por discussion, see Giannant tae VA 62= Athenaeus, Deipnosophis omy ultimately unilluminating (SSR ence of a claim to physiogn evid no is e ther that ng is worth noti to Antisthenes. force-feed their porkers’). It ment, which traced its origins back Cynical move among ail the adherents of the Stoics, An isthenes thought virtue sufficient for happiness (DL nal. (For the link between 33 As later Cynics, and like the saw the sou} as essentially ratio h suggests that he, like them, ) of anything they would — whic ght 6. 11 thou their from soul and the absence the Stoics’ denial of an irrational ion IV.) recognize as ‘physiognomy’, see Sect l and internal 215 AB.) And Socrates is not the only person for whom externa dialogues Plato's e populat who youths the beauty fail to coincide. The beauty of l beauty is is often rematked on;”* but quite apart from the fact that their physica reflection of subject to change and decay, and so could not, as such, be a reliable that their means youth very their D), 439 s a settled good nature (cf. Cratylu physical beauty fact, in them, For looks. their behind lags bly inevita achievement is not a mark of virtue, but at best an emblem or symbol for the potential that > (Socrates finds all youths beautiful they, and all the young, have for virtue al in all: Charmides 1548.) Whether potenti the see can he because presumably it was not in the this potential is fulfilled is another question altogether: glamorous Alcibiades;* it might have been in the ugly Theaetetus (cf. Theaetetus a fine body’ (Gorgias 142 co with 143 £-144 8). ‘An evil soul might well reside in boy receives, not his a ion educat the 523 c}: the determining factor in any case is him that Charmides after named e dialogu the in hint birth. There is, to be sure, a looks: at least good his with along er charact noble a ed inherit might have s as well as virtuou is he that likely it Socrates suggests that his ancestry makes tradition of a has family the that not is point the if But beautiful (1570-158 8). position is official The raising their children well, there is a touch of irony here. ent, virtue in all its forms is that, while one might be born more or less intellig the case, of course, there being This a). 19 5187-5 ic (Republ acquired, not innate exists (Meno 89 as) is ne reason to hope that any expertise for its recognition ble with the daim compati is this all that so, bly It will be objected, and reasona after all, is not so gnomy, Physio gnomy. physio of form some d that Plato accepte vice. Beauty with s uglines and simply a matter of aligning beauty with virtue in the ities: peculiar l physica ual individ of supervenes on the arrangement c 474 DE) Republi esp. (cf. er behold the of eye the in is it least at sensible world gnomical sign. and its very relativity thus rules it out as a potential physio an individual’s of er charact the shing establi of Physiognomy, rather, is a matter r, there Howeve ves. themsel ities peculiar l physica the of basis the on psychology ingly surpris has Plato that is are two reasons whyI make the point. One reason sense strong very The nce. appeara of ities peculiar little to say about individual away with from the of a fully developed visual world that one might come ound of interest backgr general of dialogues is conjured up precisely by the kind of detail. The touches lightest the by off set ed, in beauty that I have discuss s of appearance feature d detaile discuss does Plato when that, is reason second them—and it seems impossible to derive a physiognomical conclusion from 34 1 To‘ cite just those after whom dialo gues are named {references are to the eponymous dialogue): (76Bc), Phaedrus (28 " Charmides (154 ED); Euthydemus (2718), Lysis (2048, 207.4), Meno | a Cf. also — (104.4). Phaedo has beautiful hair (898). i agai should be a law ausanias (Symposium 181 DE) rather piously suggests that there precisely on the ground that it will not zn ayouth before he is old enough to grow a beard, van ee whether the boy is destined for virtue or vice. He misses the point, of course: the right lover the youth has lost his physical beauty, ka nine rnLe to tip the balance, and ensure that when might be significant that Socrates pat head vit (the much more valuable} beauty of soul. (It i most the pattern recommended by Pausanias in pursuing bis ve followed ; failure—aAlcibiades: ef. Alcibiades 103 an with Protagoras 309 An} For a précis of Alcibiades’ career, cf. e.g. Denyer (2001) 1-3. oe GORP

antisay that they carry with them an it might not be going too far to physiognomical message. where Plato indulges in a description To make the point, I here list all cases a stic that could conceivably provoke of the kinds of physical characteri le peop just ion, I mean to exclude not physiognomist’s interest. By this qualificat , but in tiful (for reasons already discussed) who are said merely to be “beau ve markers of age or status or fashion general those whose descriptions invol features. This exclusion would embrace rather than their essential distinguishing ’ is contrasted with the more ‘developed the ‘thinnish’ (sklephros) Critobulus who speed at the be to s seem t poin (the 8 271 us (prospheres) Euthydernus at Euihydem who is d physically); the aged Parmenides which the two boys have develope ful’ grace and ‘tall the with g alon es 127 s),-” ‘very grey’ (sphodra polios. Parmenid Protagoras at d bear first his in iades Alcib Zeno who accompanies him (ibid)? ya iful hair (Phaedo 895: it was obviousl 309 A; Phaedo with his long and beaut mother on that he wore it long).” matter of choice or conventi Socrates’ te kai tough’ (Theaetetus 149 a: gennaias Phaenarete is described as ‘noble and to refer to osed supp is nd of these words blosyras), and it is possible that the seco (and does it if even But n). latio ble trans her appearance (“burly’ would be a possi did she r little: was she naturally burly or rathe us tells it * ely), unlik s seem it ! become so as a result of her profession sts dialogues whose appearance sugge the in le peop are there ly, Final d ribe desc is n) when Parmenides (agai something like ‘respectability, as at us ydem Euth arly simil ; menides 1275 as kalos de kagathos tén opsin (Par grey a). Perhaps in Parmenides’ case, his 207 Lysis at Lysis and », 271 us Euthydem say to fic speci ing otherwise Plato has noth hair contributes to the impression; e nctiv disti the on ents comm in fact, Plato about what does. By my reckoning, gues; dialo characters mentioned in the physical traits of just three of the god stereotypical human being, one ly pure one of in addition to this, (imaginatively rendered), and two horses (likewise imaginary): is accusation brought against Socrates 1. Meletus. Anytus partner in the to me s seem “He 25): ro hyph (Eut s described by Socrates in these term deme is Pitheus, in case you know someone young and unknown. . „His ess in later nides’ greyness shows that he avoided baldn 27 jt could not even be argued that Parme s, e.g. about bis sexual vigour: h in itself might have led to conclusion cf. Aristotle, HA 3. 11, yet have little enough hair to qualify as ‘bald’ man might life (whic 519°24-5; GA A. 5, 774°36-P2). A man witha completely bald man might have a grey beard. enough to show that he is greys and even 1044 5. 38 CE Denyer (2001) 86 on Alcibiades of the dialogue te that Phaedo was still a boy at the time followed by # Convention if it is supposed to indica x, but against this see Robin 1926: p. 1955: 34; (e.g. Burnet 1911 on 895 2, and Bluck partan affiliation (e.g212); choice if it is an indication of pro-S Rossetti 1977: 122-6, and Rowe 1993: Perhaps it is not significant of 231). 2002: Nails 35-6; 1976: Parmentier 1926: 22-3; Montuori ility that it is a hangover at SSR 4. 119). Less likely is the possib anything in particular (Giannantoni 1997: 332 n. 56; for the dson (Davi tute prosti a as to have spent from the time Phaedo is supposed TILA 1-3). story, see the accounts at SSR Phaenarete, and there he at Republic 535 8 as is used here of Plato uses exactly the same phrase d by those due to be neede cter about the psychological chara is talking precisely and explicitly tough (gennaious te kai blosyrous, in and e ‘nobl be must they . things educated as rulers. Among other their character (&thé). 37 ellie ee mon) one Meletusnose of Pitheus: he à long straight hair, i not much of a beard, and 2. Theaetetus. Theodorus enthuses to Socrates about a remarkable youth he has met. He worries that Socrates would suppose that he was in love with the youth—if the youth were beautiful. ‘But,’ he continues, ‘as it is—don’t be annoyed—he isn’t beautiful, but his snub nose (simotéta) and bulgin eyes (fo exd ton ommatön) make him look like you. But these features are less pronounced than with you’ (Theaetetus 1432). Later on, Socrates mm notes that the description ‘snub-nosed and with bulging eyes’ (simon fe 20 — po could apply equally to himself or to 3, Socrates. See above on Theaetetus for the oniy direct mentions in Plato of Socrates’ physical attributes; but the implication of Phaedo 1025 may be that Socrates was, in addition, short in stature. Elsewhere, Meno compares him to a sting-ray ‘in appearance (eidos) and in other respects’ (Meno 80 a); and Alcibiades says he is just like Silenus, or the satyr Marsyas ( Symposium 215 An). For completeness (and because physiognomists are Particularly interested in the eyes), it might be added that Socrates had, at vsane way of looking at you (Phaedo 86D); perhaps, at times, 4, A stereotype. At Republic 495 £, Socrates is explaining the bad r ion i which philosophy is held. It happens, ie De because the best=m corrupted away from the practice of philosophy, which gets taken over by pretentious second-raters. ‘Do you think,’ Socrates continues, ‘that they [ie. pretentious second-raters] look any different from a bald-headed little coppersmith who has come into some money, recently come out of prison has washed at the bath and put on a new cicak like someone intending to 41 Note i is Meletus’ sparse beard re: Meletus’ that feature here: i an indicati À tl the nose is the one relevant Vte age, his hair-length a matter of choice; neither tell us about his physical type. en prejudice (er perhaps an assumption about ancient physicgnomical prejudice) often misleads Se ors into representing Plato's description of Meletus as inherently unattractive (e.g. Nails | : 202: long hair, scragely beard and somewhat hooked nse’). But nothing in the Greek gives scence tp cken point of Meletus’ long hair and aquiline nose is rather that they suggest his cratic : background. d It is true: that long hair could have ti ral (or political) iti connotations for Athenians (cf. Manitheus’ concern that his |: ir vl oren iat hirn: Lysias, ; Pro Manitheo 16. . 18-19). . B But this is is precisely isch See it Me diets beelt ! because is a deliberate signal 3 v . r = . . f TT affation. something which might well provoke suspicion (cf. Hall 1995: 51-3, ep. 52) Hi ug p et ‘ratsis playing to thieprejudice in drawing specific attention to Meletus’ ae ; remembered that he is very much of the class himself, i to (above present ni a sympathetic p = c portrait portrai of at least one other long-haired i young aristocrat, istocrat,namely na an Phaedo | Relatively short, anyway,> since what Socrate s tells us is that he is shorter than Simmias, who i a aemu Zn 102 ec). It should be remembered, however, that Socrates is:70 i , je Simmias and Phaedo are nearer their physical prime, | so it 4 ch oD etabe used as absolute evidence for Socrates’ diminative. stature (paceBars 1969: 21) tdowered a . ng the jailer if he might pour a libation from his cup of poison we are told that he though de t he man, as he used to do, like a bul? (Phaedo 1172). The context makes this difficult. Dien only at this man (the jailer) that he used to look thus? Is the gravity implied by Plato's o be taken seriously? (The question which accompanies it here is not without humour}

because she to marry his master’s daughter get himself a wife, and is about is poor and abandoned?’ describes Eros in speech on the god, Agathon 5. Eros. In the course of his (hapalötatos); in a): ‘He is very young and soft these terms (Symposium 196 if he were stiff for supple (hygros to eidos}\— addition his appeatance is escape notice or y tel ple com to enfold a soul (skléros), he would not be able good looks withdrew. Besides, his graceful when he first entered it or Everyone re. natu his is balanced and fluid in (euschemosyn&) prove that he ness and ugli n wee bet and s,“* inary good look knows that Eros has extraord famously Love there is unceasing wat.’ (at 24648) Socrates in the Phaedrus ged 6. Metaphorical horses. lty consisting of a pair of win orn facu compares the soul to: ‘an inb s are all es and charioteers of the god hors The er. iote char a and horses mixture. a has else ne ryo eve d parentage, but themselves good, and of goo one of his , ndly seco ; pair and iot of a char First, our driver is in charge of noble, k, but the other is the opposite stoc le nob of and e, nobl is in the horses ined outl el mod l gica ms of the psycholo bad the and of opposite stock.’ (In ter , soul esents the spirited part of the e littl A Republic, the good horse repr on.) reas ts the charioteer represen horse the desiring part; and horses. the of ions ript desc get more detailed further on (at 253 DE), we has a high neck with well-articulated limbs, ight ‘upr is se hor The good with eyes that ce, grypos), white in appearan (hypsauchen), is aquiline {epi ; a friend of est mod and te era our, and mod are black; he is a lover of hon command only.’ whip, he is guided by verba! true glory; not needing the ill-made, with a , ing r hand, is ‘crooked, lumber The bad horse, on the othe s), black, with sdpe opro (sim d ose n), is snub-n stout, short neck (kraterauché ; is shaggy ness nd of outrage and boastful grey, bloodshot eyes; is a frie bined.’ com d goa and p cely yields to whi around the ears, deaf, and scar of the ion ript desc these characterizations? The t an esen What should we make of repr t mus but my, g to do with physiogno r, not just coppersmith, first, has nothin clea very is This . ence audi his bbery of appeal on Plato’s part to the sno short stature) al text that links baldness (or mic gno sio phy no e hav we because ersmith: that copp the s rize mobility that characte with the desire for upward vant, however, is vant. What certainly is rele might or might not be rele k (the Republic). wor says later in the very same something that Plato himself indicated by 4 is ce eren diff l that no psychologica For Plato explicitly tells us ublic 454 co) they have the same nature (Rep larly beside man’s being bald or long-haired: simi is d wor crudest sense of the in form’. ‘Physiognomy’ in any but the ple ‘sup as Eros of ion ript thon’s desc the point in understanding Aga In fact not everyone does know s that Eros, as desiring beauty, this, as it happens: Socrates argue ma Te arm ugly (Symposium 202AB). is himself neither beautiful nor characterization of beauty. 43 Cf. Lysis 216CD for a similar way, is to suggest that there is no more basis for discriminating the its 39 Physiognomy and Ancient Psychological Theory George Boys-Stones by Taken 2 46 The point of this passage, een the bald and the hirsute. the division of tasks than betw the ‘nature’ of a person is between men and women in that sts sugge ady alre ment the argu a whole, it might be noted that differentia they might possess. independent of any physical The word hygros (literally ‘moist’) does, to be sure, have a certain currency in medical circles, where it is a quality taken to be more characteristic of the flesh of women, for example, than of men;* and there are physiognomical consequences one can draw from it.“® But much more to the point is the fact that the flesh of children is also taken to be ‘moister’ (or ‘more supple’) than that of adults 4 Rather than constructing a physiognomical picture of Eros (there is too little said about his appearance for this to be a realistic motive in any case), it is clear that Agathon is providing us with an interpretation of Eros’ iconograph Eras is depicted as young and (therefore) supple (or moist): why? Because ifhe were stiff (skleros, or ‘dry’), he would not be able to enfold a soul completely or escape notice ‚when he first entered it or withdrew’. Eros’ youthful supplene é — aliegorical, not physiognomical. °° The descriptions which have received the most attention, as evidence fo Plato’s physiognomical interests, are the horses in the Phaedrus.”* This, as I shall suggest, is in fact entirely appropriate—though not for the Fenn usualt adduced. For it seems to me that to see Plato's account of the horses as physiognomical once again would be to miss his point. In particular, it misses the subtle blend of convention and subversion in the horses’ portraits For these portraits are boringly conventional, insofar they do little more than to establish the horses as, respectively, good and bad horses. In particular, its ‘upright’ bearing, high neck, and aquiline nose would all serve to make the good horse recognizable as such to any rider.** Given this perfectly innocent explanation for Plato s descriptions, it is impossible to say without begging the question that he intended their details to apply to good and bad human beings as well The fact that the terms of his descriptions turn up in later physiognomical handbooks only tells us that the Greeks used the same vocabulary to describe the physical attributes of humans and animals. The fact that some of them answer in those 47 3 809°11); i (and cf, Ps.-Aristotle DP istotle, HA 5389-10 pe AWP 10. 32-3; Aristotl tneHippocrates, Dap eg. i the same combination of ‘moist’ eg. e 807°12-13 (with je His It is mostly a good sign in Ps.-Aristotle: dd oe Se 1397-55 (with Vogt 1999: 459-61 ad loc), where moistness of flesh the fia. is one of the factors that affects the flow of the blood, and so the character of 5 “8 Hippocrates alia 1. 33. 1 (and cf. Galen, On Medical Method, x. 195. 9-11 Kühn). on. oe a hadi} soft and supple is hinted at in Plutarch (Table-Taik n falls in love, he melts like iron in the fire, and becomes ‘soft and N : ad more passt (hapalos kaf hygros kai hedion). (Other examples of hygros erds, oe 65, is à ik be in a literally humid context, are Oppian, Halieutica 4. 2; Nonnus, Dionysiaca 16. Me = ee different interpretation of Eros’ iconographical youth, cf. Comutus, étions of Greek Theology 25, 47, 4-5 Lang, where it symbolizes the folly roden a the Traditio 8 CE esp. Misener{ 1024), 106-7, and Frans (1969), 21. Poerster cites the passage at (1893) ii, 252-3 31 a ippiatri (cf.| auth of the Hippiatrica was in fact read this way by the author en Beaedrus quis 2). Cf. already e.g. Xenophon, ij 1. 8 (and note tha nophen, On On Hilorsemanship oes u vj listed here are marks only of a good physical type: this is, as Xenophon. says at 1. 1 ne a un on with an unbroken horse, for whose ‘soul’ there are ‘no clear signs’). | at the ‘form of the description is that commonly used in later physiognomical handbooks’ is neitherer hhere nor there. It iis more likely, in any case, that both rely on standard 21)rns s 1969: (Evanostic patte diagn

e to the corresponding horse” tells handbooks to features of character appropriat us no more than the fact that some of them do not.” er, the portraits he paints of the In the context of Plato's own writing, howev —subversive, that is, of any rsive Phaedrus horses are also strikingly subve character and appearance ascribed to thought that one can take the link between arison and into the human world, these horses beyond the frame of the comp is one t make with later textbooks) there For (whatever comparisons one migh suggests a link with human characters outstanding feature of each horse which good horse, we are told, is aquiline described by Plato himself: the nose. The pos). Yet, as have seen, it is the one prosö (epigrypos), the bad horse is snub (simo is aquiline (epigrypos), while Socrates evil-doing youth in Plato, Meletus, who acquaintances, Theaetetus, are snub and the most promising of his young human features there is in Plato, of n (simoi).°° Given how litde delineatio it is hard not to see here almost and given the fame of Socrates’ snub nose, ognomical conclusions. The truth, in a deliberate warning against drawing physi tes, Theaetetus, any case, is that Meletus and the good horse look noble; Socra and the bad horse do not. be a stable relationship between the For physiognomy to work, there has to emerge, they have to be resolved tions excep sign and the signified; if apparent do points up the apparent exception (as we have seen that, in the Zopyrus, Phae virtue allows beauty to be emblematic of of Socrates in order to resolve it). Plato iptions arc all exceptional. Personal descr ete or promise; but his (three) concr res its description as ‘upright’ (arthos) with 54 For the good horse, Evans (n. 31} compa res ‘crooked 21 n 31). For the bad horse, she compa Ps.-Aristotle 807°31-2 and g08°19-20 (1969: further suggests Evans 36). (n. ?) b29-31 (sc. ‘807b’ e istotl and lumbering’ (skalios, polys) with Ps.-Ar e ‘8072’ (ie. 807° of insensibility, and compares Ps,-Aristotl that polys ‘is related to” polysarkia, a sign 19-28?) (n. 37). own references 55 The following are exaraples from Evans's where the handbooks suggest an of the good Plato says of the horses. First, the ‘high neck’ association of character at odds with what B39-40 (1969: 21 n. 32). The ‘stout, Ad. at nce arroga of signs res compa horse (hypsauchén): Evans compares conversely, suggests bravery: Evans {n. 38) short neck’ (kraterauchén) of the bad horse, ice in both horses coward of signs see also would ooks handb Ps.-Aristotle 811°10, But then again, the because of its e even of the bad horse): the good horse (though cowardice is not obviously a featur er: ern black eyes (melanommatos. Evans n. h she also notes Ad. B36, 35 compares Ps.-Aristotle 8121; thoug s: Evans bad horse because of its grey eyes (glaukommato where a black eye is a sign of tameness); the 41 cites Ps.-Aristotle ‘812b’, though there n. (Evans y’ ‘shagg is it e becaus B33), n. 40 compares Ad. because it is specification ‘around the ears’), and finally is nothing here corresponding to Plato’s , °1-2 would be more 81212 gh (thou ‘811b’ e istotl Ps,-Ar res compa ‘black’ (melanchros: Evans n. 39 B25). to her point; cf. also 808°17), and Ad. occurs at of Meletus and the bad horse—though grypos 38 The word epigrypos is vsed in Plato only teristics charac ing oppos quite find will lover how a keen Republic 474 D, in a list demonstrating only in used is ise otherw but list, same appears in the attractive in different cases, The word sirtos part of the compound word simopros@pos as and etus, Theaet and es Socrat of Plato’s descriptions ations might che bad horse. Two more speculative observ (literally ‘snub-faced’} used to describe es. First, the iugula concava Socrat like look to sed suppo is horse bad support my suggestion that the neck’) s fr. 6 Rossetti; Kahn 1996: 11 translates ‘bull observed in Socrates by Zopyrus (Phaedo. Zopyru icant that one (kraterauchen). Secondly, it might be signif might recall the bad horse's thick neck horses; his the of ption descri the into way its not find distinctive physical feature of Socrates does be that they are absent because they would it Could ), halmon exopht c: 209 etetus (‘Thea bulging eyes )? phon, Or Horsemanship 1. 9: exophthalmor have to be a feature of the good horse (Xeno 41 appearance is a shifting, unstable, unreliab i | i but a fatally anti-physiognomical, =o fo the trut à very Platonic, I take it, then, that Plato was resistant— and perhaps consciously so—to the idea of physiognomy. But there is more to this resistance than a desire to reinforce the familiar lesson that, as philosophers, we should shed our fixation with the perceptible and search for a grasp of the underlying realities in their own terms. A sceptical attitude towards physiognomy is in fact a natural consequence of Plato’s psychological theory, and in particular his beliefs about the relationship between soul and body. The reason for this is that physiognomy acquires theoretical credibility only if moral character and physical development are supposed to be related (e.g. at the level of physiology). But in the normal run of Plato's theory, there is no causal relation between body and soul which makes this inevitable. The growth and character of the body can be explained b physical causes; but the soul pre-exists (the soul, in fact, is immortal), and comes into the body with its own history, and a (moral) character developed through its previous experiences of incarnation and the world beyond. There is actually no reason why the body should look like the soul, or why it should be any guide to the characterful individual which inhabits it. TE It is always dangerous to generalize across Plato’s works, and it is certainl possible to appeal to texts in which this rather severe generalization is qualified; but none challenges the basic premise. In fact it is only in the Phaedo that we are presented with a psychological mode! according te which moral character (that is, roughly speaking, non-rational inclination) is rooted in and derived from the body—and the psychology of the Phaedo, as 1 argued in the last section, seems Der this Fe at ren to Plato’s engagement with Phaedo of Elis fr that er ialogue, and cannot therefore be read as a straightforward account of his i own Nevertheless, Plato clearly did take seriously the effects that the have on the soul. A small but significant number of texts state or EE east de body and the physical environment more generally can affect the character of the soul, A temperate climate can be used to explain the superior intellectual powers of the inhabitants of Greece, and had formed the backdrop to the excellent natures of the inhabitants of Atlantis.” Despite his commitment io the transmigration of souls,”® a number of passages attest to Plato’s belief that 7 Timaeus 24 C = Foerster (1893) ii. . 251-2 (on the character of the Greeks); Critias iti BE ir also Republic 435 E=Foerster (1893) ii. 249-50 (of northern in such En ; cythians); and ct. Laws 747 DE=Foerster (1893) ii. 251. In this context, Foerster lists Laws 05 u Foerster (1893) ii. 250-1, where the Athenian Stranger warns against the foundation of cities near the sea, But this passage makes no allusion to the effects of the climate on character: rather, it argues that people living on the sea-board are unduly exposed to, and tend to become duly ged > novelty, luxury, and alien patterns of thought. j j nr The Stoics at least appear to have thought that the phenomenon of inherited charact an ee with mire (Plutarch, On Stoic Self-Contradictions 1053 c}; but in fact there we pallies ing the two, as Neoplatonists proved (see further Section VI). A propos: it has vovide an assumed that Passages in Plato which suggest transmigration between species — ; En or his interest in physiognomy (e.g. Phaedo 81 E-82 5 = Foerster 1893: ii. 253; and : 22). There is of course (and was already in antiquity) a question over whether Plato

as physical character is, and he might psychological character is inherited, just cal basis for this. Indeed, in an article have thought that there was some physi argued that Plato in that dialogue has on the Timaeus, Christopher Gill ings—from madness to sexual excess— systematically ascribes psychological fail one precisely can talk of character in to the constitution of the body, so that cter’. (Gill, in fact, appeals to the chara d reductionist terms as ‘body-base Galen.) reductionist reading of this dialogue by ion to be made of Gil’s account There is, however, an important qualificat usion applies to all the texts mentioned, of the Timaeus, and it is one whose concl to physical factors in the development In brief, whatever influence Plato ascribes that they determine its character (or of the non-rational soul, he never says Indeed, they may never, in fact should in any sense its nature) from the start. are counterbalanced by the positive never, have any effect at all, since they person's upbringing. Plato is careful a ng effect on character of reason in shapi (ie. acquires a particular irrational to say that a person becomes bad and an ‘upbringing without education’ character) because of a bad state of body nes with a poor government and combi body of (868); or where a poor state ness, conversely, involves both physical poor parenting (87 as); the route to happi t of all this is that Plato is not quite and intellectual training (88 zc). The resul see in Section V), the n (as we shall so Galenic as Gill implies. For Gale other) a function of the temperament or ehow (som is soul character of the the character of the of the body; for Plato, the body is only one influence on non-rational soul. 2 meant this literally; but even if he did, the point which one comes is that one’s soul migrates into the bady of an (Timaeus 420; and note to resemble in behaviour, not looks tionally just man: of forms which might fall to the conven in Timaeus 91 DE). 59 So Charmides 1570-2588 Cratylus 394 animal the diverse range a wasp or ant in Phaedo 828, but a bird A; hence also the possibility of breeding for good 310 and Laws 773 AB; cf. Critias 221 AB, character: Republic 375-6, 459; Politicus educator: that likely. The parent, qua begetter (not qua seems this us Timae & Ty the context of the ter of the child than the child charac evil the for ty sibili respon is considered separately) takes more itself (Timaeus 87 B). see further Section V-) ionist reading of Plato, & See Gill (2000), 62, (For Galen’s reduct vam ee ee a foolish soul); around the spinal marrow leads to Cf Timaeus 86C-8 (an excess of seed like maleducation, might be the cause of ion, condit al physic (poor 252 ii. 86 E=Foerster (1893) s, stupidity in melancholy, rashness, cowardice, forgetfulnes vice); 87 A (humours cause ill temper, ignorant). Cf. also Laws and stupid and dull soul r weake a g the soul); 88 As (a strong body makin condition which on soul); 790 E (fearfulness is a morbid 728 DE (a moderate body has best effect es good proof provid , course of l, Alcoho body). ent of arises when the soul is dominated by movem y Laws 645 DE (and cf, Ps.-Aristotle, Physiognom body: the by ed imped be can ies activit that the soul’s 805°2-4). (2000: 60), but plays down (61) the significance ® CE Lloyd (1987), 28-30 n. 89. Gill recognizes l development. I also disagree with him about logica psycho on of educational and political influences one’s assumes that the state of reason, and so Gill . reason the effect of the physical environment on of mercy the at itself is ter, charac one’s of t opmen ability to apply a positive control on the devel n pathoiogical states are able to disrupt the case that certai physiology. It seems, however, onl; to be , think that ss and ignorance. (Plato would, I assume the normal course of reason to cause madne placed, like children, under the be d shoul and can kind this of es people suffering mental illness ted from strictly way, even their character can be protec protective rationality of others. In this unnatural influences.) 43 | It might be objected that this misses the point: that ‘reason’ (particularly when it is someone else’s reason, e.p. that of our parents) is, as it were, a corrective. and that the physical forces on our development represent what would be the natural’ state of our character were the artifice of reason not to be applied. But Plate is, on the contrary, clear that philosophy is an entirely natural factor in our development: a natural counterweight to the influence of the body on the irrational soul. This is so not just in the sense that it is a naturally occurring feature of the environment, but in the very strong sense that a tendency to do and to be receptive to philosophy is itself part of the physical construction and development of the individual’s body—every bit as much as (and probably, in fact, more than) an inclination towards the various forms of vice. As a human being grows, the overwhelmingly disturbing influences of the infant body tend to decline, and education (physical as well as philosophical) is supposed to kick in to ensure a healthy state of soul and body (448c; cf 88nc); and man’s philosophical destiny (cf. 42 Am} is an explanatory factor behind, for example, = ne of the sense-organs (473), the mouth (755), and even the gut The point of all this, so far as I am concerned, is that, while the character of the irrational soul can certainly be influenced by the body (even, when unchecked, to the extent of disabling reason}, and while it might be true that every negative characteristic of the soul has a physical cause, it would not be right to say that the natural character of the irrational soul is determined b the body (nor, then, by any external factor which might have an influence on the body). It is not the case, for example, that the body determines the nature of the sou (as it does in Phaedo), and philosophy is awakened later on in reaction to the psychological ‘diseases’ this entails. Rather, an individual’s natural state consists in a proper equilibrium of the irrational soul as it is held in balance between, on the one hand, the requirements of physical existence and, on the other, the controlling force of reason. (It makes no difference that this en in the firs: place, is that of the educators.) Rather than saying that the ETE: of the soul is naturally determined by the body, the most we can say is that if something goes wrong, in particular if the appropriate (natural) educational forces are not in play, then the soul will tend to acquire a character which can be explained by the individual's physiology.“ Still, it might be said, this is not bad going. If we assume that appearance ts in any case determined by physiology, then we should be able to make a reasonable judgement of a person’s character in every case where the proper educational system was lacking. Since the proper educational system is (in our sorry times) almost always lacking, our success rate is likely to be high, A refusal to credit 63 me : “4 or on Wis Johansen (2000) 109-10; Steel (2001). ote® that |this is subtly,2 but importantly,y, different from 1the common Neoplatonist line iplored a Section VL according to which irrational inclinations are written into the body, to mani a emselves as psychological affections in the absence of reason’s control. Plato merely says ost, that the body tends to influence the character of the irrational soul,

forced. But it looks Plato with a physiognomy starts, after all, to look rather the meaning of the about forced, as I am arguing, only if we make assumptions right that ancient am I If . ophers philos t word ‘physiognomy’ for ancien the claim that a natural philosophers reserved the notion of physiognomy for appearance of the body l natura the state of character could be deduced from ance are linked) then appear and state mental that (and not just for any claim consider himself, an not Plato is not properly considered, and certainly would end of Section IV, the at t sugges shall I as ally of physiognomy. In this respect, Stoics—and earlier the of that with ls paralle ting interes his thought has some not think of did they lar, particu when we have seen what they think, and why, in position Plato’s tand unders to le possib be will it themselves as physiognomists, the what about be clear more clearly. However, it is also necessary first to acceptance of physiognomy does involve for a philosopher; and this is something of ancient physiognomical we can see if we turn now to the most influential theorists: Plato’s own pupil Aristotle. Ill. ARISTOTLE AND HIS SCHOOL . Physiognomy (805°1-814°9 Bekker}; Loxus Peripatetic works of physiognomy: Ps.-Aristotle, 117, 133).% References Against Celsus 1. 33; Anon. Lat. 1, 2, 12, 48, 80-1, 89, (fragments: Origen, Lat. 1, Vita Menagiana Aristotelis A, 97 Rose; Anon. to an Aristotelian Physiognomy: DL 5. 25; Anon., Eclogue i. 47. 6. 1-20, , Stobaeus Müller; 7-9 57. QAM Galen, 48, 55, 72, 83-9, 117; Adamantius Al; Onomasticon g Physiognomy 805°1-18); Pollux, i. 301. 23-302. 17 Wachsmuth (quoting the survivin 70 Taillardat) (cites 81117). Occurrences of 2. 135, 7 Bethe (cf. Suetonius, de blasphemits 4. Prior Analytics 2. 27, 70°7, 13-14, 25-6, 32; romy): Physiog the than (other e Aristotl in némonphysiog from works not named: Pliny, Natural gnomy physio lian Aristote of s Citation GA 4. 3, 769°20. On Animals fr. 10 Seel [not naming Trogus, us Pompei cf. Rose”); 286 fr. e, History 11. 273 {= Aristotl in De anima xv, 155. 22 Hayduck [commenting on DA Minio-Paluella [Prior Analytics 70°7, 11]. 408°34-®2]; Latin Scholia to the Prior Analytics 370-1 Alexander, On Fate 171. 11-12 Brans writers: tic Peripate other in nomonphysiog of Occurrences 4 | | > irr : when engaging in his own theoretical discussion of physiognomy, and may even have intended to invoke it in the opening sentence of the Powers of the Soul [QAM] (as I suggest in Section V n. 184 below). Antonella Ghersetti shows, in Ch. 6, how the work, read through Galenic lenses, was received in the Arabic tradition as the foundational text of a ‘scientific’ physiognomy. But it is not only the case that the later tradition looked back to Aristotle for the definitive, or at least the seminal, account of physiognomy as a practical science; it is also the case that most later theoretical discussions of physiognomy in the Classical world in particular depend more or less directly on a reading of Aristotle for their account of how body and soul interact in such a way that physiognomy becomes a possibility in the first place. This might seem slightly more surprising, since Aristotle’s hylomorphic model of the soul did not have a significant take-up among post-Aristotelian thinkers—at least not in the terms we would find it historically appropriate to reconstruct for him. But it happens that Aristotle came to be read in some quarters as an advocate of epiphenomenalism, the theory that psychological functions are explicable in terms of a person’s physiological state.% Since, as I shall argue, such a reading was made possible by those very features of his psychology that open the way to physiognomy, it is after all no surprise, even if it is not exactly inevitable, that 493°32). The immediate object of Pliny’s concern here are signs of iongevity, not character, but he goes on at 11, 275-6 to assimilate these to more conventional physiognomical observations which he quotes from Pompeius Trogus, but which mostly draw on Aristotle, HA 491P11-26. (Pliny’s distaste for physiognomy here is not, by the way, inconsistent with Natural History 11. 138 = Foerster 1893: 769°20-1 in mind}; cf. Commentary on the Hi ' | anonymous Latin writer, draw heavily on it; Galen seems to have had it in mind 484. 8-11, 485. 5 Wallies Philoponus, On the Prior Analytics xii, 2 483. 22-4, xiv. 3, 185. 13 Hayduck [en GA 769°9, with 70°7-38]; Commentary on the Generation of Animals 49223403, 572°8-13]; Galer, QAM 114a-c Giannini [not naming Aristotle; but cf. HA 491°11-26, 492°12, *34-53]; Ammonius, On the 491°11HA 651°17; 650°14, 6482-13 [PA Müller 6 51. 12-57. s 70°7-38 in mind]; Analytic Prior with proem, the [from Wallies 16 7. Prior Analytics iv. 6. [Prior Analytics 4 right to advertise and lend credibility to the study of physiognomy in Classical antiquity and beyond,® the work apparently asserted itself as the principal work of reference for all subsequent writing on physiognomy. The two later handbooks which survive from Classical antiquity, those by Polemon and the nemy. Antigonus, Collection of Amazing Stories known Aristotelian works other than the Physiog 4 Apart from the fact that the name of Aristotle did a certain amount in its own it should be noted that, conversely, Aristotle’s forays into physiognomy can be used to undermine his credibility—as in the case of the (Epicurean?) critique of Aristotle’s natural histo whica deploys physiognomical material from HA 1. 1C (see Section IV n, 132 below). Even among ion to physiognomical passages Aristotle, but cf. HA 491°11-26]. References under this descript k 45 (also quoted at Eusebius, Preparation for the Gospel 6. 9. 22,7 written in antiquity Was The single most influential work of physiognomy the name of Aristotle. under s) survive (and ted circula the Physiognomy which once offers us an example of a & One should, of course, keep in mind the fact that Plato never position would ar psychological trait; the link between a particular trait of appearance and a particul have to be that physiognomy is implicit in Plato. below. 5 por Loxus’ Peripatetic credentials, see subsection (iii) on Aristotle's Physignomonica: & Alexander was once erroneously credited with a commentary gnomical’ passages Foerster himself identified in see Foerster (1893) i, pp. xexiiif. (The ‘physio a spurious work: see Rose Alexander, given ibid, 283, are also drawn from 126-35.) 1863: 216-19; Cranz 1960: those who are sympathetic to Aristotle, not everyone finds this material easy to swallow: Pling pele it with an exasperated. miror (Natural History 11, 2734 = Aristotle, fr. 286 Rose’: of HA il ae which is about facial expression, not physiognomy as such.) | " So already, in the generation after Aristotle, Dicaearchus (frr. 14-28 Mi i); Aristoxenus (fer. 118-20 Wehrli); Strato (fr. 118 Wehrli). ie on, oelpe ne by Galen at QAM 44. 12-20 Müller); Boethus (see esp. Gottschalk 1986); Nicolaus an pu L 353. 12-354, 18 Wachsmuth); Alexander of Aphrodisias (see his surviving work (aaan) Ho generally Caston (1997). For Dicaearchus in particular: Caston (2001) and Sharples “ . For ronicus, Alexander, and their relationship to Galen: Donini (1971); (1974) 127-85; Moraux (1973-84) i. 132-6 (Andronicus); ii. 740-2, 773-85 (Galen); Thillet (1981); Sharples (1987) mn (1987); Blumenthal (1996) 15-16; Sorabji (2000) 254-63, 266-7. Aristotelian 7ee include Theophrastus (cf. fr. 269 FHSG with commentary at Huby 1999: 18; also (Comm 1996: 10; Lennox 1985) and Eristratus (cf. von Staden 1997}; lazer on, cf. Aspasius u mentary on the Nicomachean Ethics 1. 14-19 Heylbut) and Xenarchus of Seleucia (see Stobaeus, one a 5 . 5-8 Wachsmuth). Cratippus does not count here: although he supposed that the Pain was prior to the body and separable from it, the non-rational soul, as being concerned with physical functions, is inseparable, and perhaps not prior. See Cicero, Or Divination 1. 70-1.

enomenalist thought who were it should be philosophers influenced by epiph a serious theoretical possibility.” inspired by Aristotle to consider physiognomy a certain amount to say about has tle In the Prior Analytics (at 2. 27), Aristo established the right sort of has one once how, y: the practicalities of physiognom go about isolating signs for the connection between body and soul, one might r. I shall have a little to say forme the of condition, of the latter in the appearance I consider the affinities when n sectio nt prese the to about this in the Appendix Physiognomy. (For telian Aristo between Aristotle himself and the authors of the ing reasons discussed in subsection (ii) below and in the Appendix, I am assum is pseudonymous, and in fact with the majority of scholars that the work tetic writers.) In what follows a composite of works by two different Peripa sts me. Rather, my immediate intere immediately, hawever, this is not what are prior to this, and actually which ions quest two intention, is to consider ing of theoretical discussions more important for the broader understand tle himself conceived of the Aristo y, exactl of physiognomy in antiquity: how, would lead one to hope that soul such that its relationship with the body and how the underpinnings physiognomy might be a practical possibility; led to an association between have might y ognom Aristotle thereby gave to physi minds of later philosophers.” physiognomy and epiphenomenalism in the Following this, in parts (ii) and (iii), I shall look at the evidence we have for the Aristotle’s school who wrote on psychological models assumed by thinkers in such thinkers: not only the two physiognomy. 1 shall argue that we know of three to produce the surviving ned anonymous authors whose work was combi used as a source by the was who ian’, physic Physiognomy, but also Loxus ‘the The evidence for the psychological author of the anonymous Latin Physiognomy. poor and does not allow for basis of their interest in physiognomy is at least already exemplify the them of two certainty; but it may suggest that an interest in physiognomy and tendency towards the association between in later Sections we shall see in an epiphenomenal view of the soul which later authors. ee Bert amer ee: a 70 ‘There are two exceptions to this generalization: one may be the author of the first part of the t ion (ii) below); the cther certainly ıs the Platenis Aristotelian Physiognomy (see further subsect astus was a teleologist Theophr pupil e’s Aristotl V. Section in discuss I Aristides Quintilianus, whom his Characters is frequently used to support (cf. n. 69 above}, but not a physiognomist—albeit (1924), 107; Currie (1985); Vogt (1999), 196; a claim that he was: e.g. Petersen (1859), 92; Misener di (1999). In fact nothing in the Characters Lombar 28; , (19965) tz Degkwi 170; (1991). Herzog cf. in physiognomy (as opposed to an ethical interest cal theoreti a had astus suggests that Theophr such an interest. interest In manners, for example); nor is there any other evidence that he had Evans’s rather catholic view of physiognomy Indeed, Theophrastus is excluded even from significance occur’). (cf. 1969: 39: 'no passages of strict physiognomical ity of physiognomy, given that I de not 47 (i) Aristotle If it is a matter of historical fact that most philosophical accounts of physiognomy in antiquity explained the relevant features of the soul by the peculiar character of the body, it is not a matter of theoretical necessity that this should have been so. An equally systematic basis might be given to physiognomy on the assumption that the peculiar character of the body was determined by the nature of its soul. Indeed, in Section VI we shall meet one thinker who took just such an approach (the Platonist Aristides Quintilianus); and at first glance, we might be tempted to think that this was precisely how Aristotle himself was able to justify physiognomy. Aristotle, after all, took the soul to be a form-—‘the form’ (as he says) ‘of a natural body potentially having life’ (DA 412°20-1). The very word ‘form’ (eidos) already suggests appearance; and one might readily assume that, just as the form of a statue both describes and explains the way a statue looks so the form of a human individual both describes and explains the way that the individual looks. In this case, Aristotle would be committed to physiognomy for the simple reason that the individual’s soul is precisely what determines his or her appearance.”” Things are not quite so simple, however, because if Aristotle can be thought of as extending the explanatory purpose of the soul to cover appearance as well, there is, paradoxically, a certain restriction in the relationship Aristotle posits between the soul and a person’s psychological character. It is in fact one of the central insights of Aristotelian psychology that the soul is not an independent locus of psychological functioning or psychological character, as one might think of it in Plato, for example. Rather, the ‘soul’ explains what it is to be something that has these functions. Aristotle is clear that particular events that we might have thought of as ‘psychological’ events are not activities of the soul itself, but are the activities of an ensouled body (DA 408°11-15). It is a person, not a soul that!has a ai character. is might seem a nice distinction; but it is one that rules out t ibili that Aristotle took the short route to physiognomy when he radeeal the form’ of a human being. The reason for this is that, even if we were tempted to maintain the identity of ‘soul’ and physical shape (though here too one should be careful to say, not that the soul is a human being’s shape, but an explanans of it) we can no longer be sure that it is the only determinant of his or her character. There is scope to think that there might be aspects of human beings and therefore of their psychological character, that are to be explained by something else—for example, the material out of which they are made. It is true that it must be the soul that explains why there is a human being here at all (and not, for example, a dog), and to do this, the soul must explain everything that is characteristically human about both their psychology and their appearance. But the lower-level characteristics that mark a thing out, not just as a human being, 71 Evidence that Aristotle was committed to the possibil of surviving Physiognomy, include his discussion arsume him to have been the author of the in character: cf. Ross 1949; 498-502; tical hypothe (albeit 27 2. cs Analyti Prior at physiognomical signs assertions in his biological works (e.g. HA 1. 8-10, Burnyeat 1982); and the many physiognomicel 7; GA 2. 7, 747°15-1% 4. 5 491°10-492°12; 15, 493°32-494"1; 494°16-18; Pl 4, 10, 686°2123-4). 1983: 77423622; 5, 7, 786°34-787°2; and cf. Lloyd 72 = sue indivieh more sophisticated terms, the idea would be that the soul, as the form of the rb explains his or her peculiar physical disposition (and so appearance) insofar as that position realizes the natural behavioural capacities specified in the form.

Socrates as opposed to Callias) might but as this human being (for example, s activity required by the particular form’ the of be explicable as modal variations matter out of which Socrates is made. her the soul-form is supposed to In brief, then, the question for us is whet ce that distinguish an individual as explain all aspects of character and appearan in only those that distinguish it as a such, or whether it is rather meant to expla n being).”* This matters a great huma a member of a particular species (e.g. as Aristotle and beyond, because if in omy iogn phys of deal for our understanding on the basis of appearance than that we wish to be able to say more of a person idual characteristics in which we indiv he or she is human, it is precisely these ish, then, is how Aristotle explains shall be interested. What we need to establ to look more closely at his account of these; and in order to do that, we need place. first how an individual comes to be in tae rsial that species traits, both trove uncon is it ated, As I have already indic ’s form. This, at least, does thing a by ‘psychological’ and physical, are explained of a body potentially ‘form the as soul the follow from Aristotle’s definition of responsible for all such the form is as having life’,”* since what this tells us is that to be alive, Since order in res requi it those traits that specify for a thing what shment, growth nouri selfas s thing such of s terrestrial life is defined in term (and decay), locomotion (and rest}, perception and thought (DA 413°22-5), it is a formal cause. But there is more than for these things that Aristotle must posit ss the capacity for self-nutrition posse ot cann this that needs to be said. One nourishing oneself in a particular, unless one possesses the capacity for number of models for a selfany determinate way—given. that there are e soul was merely nutritive, whos exist d coul nourishing organism, and nothing determining characteristics must, then, Le. nutritive without qualification. These ability to photosynthesize, for example; also be explained by a thing’s form: the animals, the ability to feel, to see, etc. of case the in or to feed off plants, or flesh; that a different kind of psychology But they will also include characteristics raised by Aristotle’s discussions of form in 73 The question is related to the question, supposed to have Metaphysics, whether each individual is the its own (‘individual’) form, or whether s, y, it is the case that if all forms are species-form rg Te de mek forms exist only at the level of the species. Clearl species will be open in common by all members of a given only those psycho-physical traits shared any such entity as exists there that s doubt one if even to teleological explanation. Bu: note that, net follow that does it dual, indivi each for one particulars, a ‘species form’, so that forms exist as identify them to us allow would that way each other in a they are formally or essentially distinct from , one might think thet such Indeed s. eristic charact ng guishi distin dual's as the cause of the indivi in essence—OF members of the same species are the same a step would undermine the notion that individual forms are that be might ption assum l natura even the very notion of a species. A more they are the forms of incidentally—that is, by the fact that distinguished one from another only are instantiated in they that fact the by way, put this another numerically different individuals; or, to to say that the characteristics of the able be case this ia might One . matter different parcels of Aristotle is able to of the individual's form (as, analogously, individual are ‘incidental’ characteristics of which it is the soul: of the movement of the composite say that soul moves incidentally in virtue they shared as such provide the explanation only for traits Patzig cf. DA 1. 3, 406°5-10); but the form would , esp. 77; (2000} 21-6; also Frede and with the rest of the species. Cf. on this Frede (1987) (1988), i. 48-57, 74 Though cf. e.g. Caston (1997) 331, arguing that the Aristotelian soul ‘supervenes’ on the body. 49 might distinguish from these as ‘character-traits’. The essence or form of a lion, for example, must explain its characteristic ‘bravery’ 7 since a lion is not merel appetitive, but bravely so: it is prepared to fight for its food, and its preparedness to fight is every bit as important as its having teeth to eat the food with. Brave then, has a formal cause for a lion.”® But since we know that a lion has just that physical structure necessary for the realization of those characteristics for which there isa formal cause, we can expect that it will be possible, at least in theory, to identify physical features of a lion that will bear witness to its bravery. Th y il be just those physical features which enable it to be brave, or are a iR een % Py ok features which allow it to be brave. (In just the came , one can i bear witness i M oe andider antusii features which to the fact that a lion Aristotle’s commitment to physiognomy involves him in the belief that one can perform very much the same sort of exercise at the level of the individual: that is, one can identify physical traits in the individual which are associated with particular traits of character. This above all, perhaps, is what makes it tempting to suppose that Aristotle simply extended down to the level of the individual the scope of what formal causality explains in a creature. The idea would be that the physical and psychological traits of an individual have the same cause as their species traits (namely, a formal cause), so that we should expect to find exactly the same correlations at the level of the individual as we find at the level of the species. In both cases alike, the relevant aspects of the animals physical structure would be due to the teleological activity of their form. It is true that this move has certain metaphysical implications which not all readers of Aristotle would want to accept: it implies (a) that forms exist as individuals, and (b) that individual forms are formally distinct from one another (distinguished, that is, by characteristics that explain the variations between individuals), But against these concerns, it might be argued that Aristotle’s account of natural generation shows directly and explicitly that he did in fact make this move. For Aristotle is clear about one thing in the process of generation: that the father provides the form for the new child, while the mother provides only the matter. As he spells out elsewhere, one of the things that this means is that any resemblance a child might have to its parents must be due to the action of the father.”’ The obvious way of understanding this is to suppose at the child’s appearance quite generally, both as a human being and as this particular human being, is due to the activity of form, so that its individual characteristics are part of its form and have a formal explanation just as its characteristics as a member of the (human) species.” 7 Cf. Prior Analytics 7014-15. w ies is explicit in thinking that animals share not only psychological affections (cf. e.g. NE ii + 1111712), but also ‘moral’ traits with humans. See HA 8. 1, 588°16-26; 9. 1 60811-17 \ ith ‘ortenbaugh (1971); Lloyd (1983), 24-6. En = 7 Fin 1. 20, 729°9-11; 4. 1, 766°12-14; cf. Cooper (1988), esp. 19-21. inber me (19874) 1, 18-19; (1987c); Dean-Jones (1994) 193-5 (suggesting that, in explainin, Da viene Aristotle ‘has to depart from the theory upheld in the main body of hie phy’ to assert the existence of individual forms}; Cooper (1988) (arguing at 35-7 that form is

member of a particular species, and so form part of its essence. Individual characteristics, on the other hand, are incidental at least in the sense that one difficulties, however. One is a Such a position quickly runs up against provides the form for the child, straightforward biological difficulty: if the father ics, then we and the form accounts for individual (as does net have to have them in order to be a particular kind of thing. But if individual characteristics are ‘incidental’ rather than essential, this does not of course mean that an individual can avoid having any such characteristics at all. On the contrary, every individual must have some individual characteristics, for just the same kind of reason that an individual wall, though its essence makes no reference to its being red or brown or grey, must nevertheless be some colour or other if it is to take its place in the world. In the case of individual human beings then, we should perhaps accept that it is no part of their formal account that they have the particular characteristics they do have; but they must nevertheless have some such characteristics. To borrow the example Aristotle gives in his discussion of physiognomy at Prior Analytics 70°14-17, it is not specified in the form of any human individual (as it is for every lion) that he or she should have large extremities, or should be brave. However, each individual must nevertheless have extremities of some size, and lie somewhere between the extremes of cowardice and rashness in order to take his or her place in the world. well as species) characterist . But children often look like their would expect every child to look like its father tle is well aware.”? The second Aristo mother as well or instead—a fact of which (5. 1, 778°30-5) which seems GA the of ge major difficulty is textual: a passa can give a formal explanation for to rule out explicitly the possibility that we ics: terist individual (as opposed to species) charac works of nature nor proper to each species, Of those things which are neither common of anything. The eye is for something, but it none is or becomes the way it is for the sake a condition proper to the species. Neither do is not blue for any purpose— unless this is e but, as some such cases® have anything to do with the account of the animal’s essenc be referred to the coming about by necessity, their causes must ). moving principle (fén kinesasarı archen matter (hylé) and the ee end teristics are not only denied Just to make the point clear, individual charac explanation involving a ative altern formal explanation here, but are given an In other words, the ple’. princi ng movi ‘the kind of necessity: ‘matter’, and a teleological explanation seems to possibility that individual characteristics have might note, the possibility that be explicitly ruled out—and with it, we ogical model suggested by the teleol the on ed physiognomy can be found individual characteristics to be explanation of species-traits. But bow, then, are explained? stand why Aristotle might It is possible on metaphysical grounds to under n to be given to species natio expla want to make the distinction in the kinds of just quoted. Species ge passa GA the by and individual traits suggested l as a particular thing, namely a characteristics, after all, determine an anima There is something, in other words, that determines, that makes concrete, what the form explains at a more general level; something that explains why, when a human being is produced, it is a human being of a particular size, shape, and behavioural disposition. And Aristotle, in the passage just quoted, specifies what: it comes down to the matter from which the new human will be made, and the moving principle which does the making. | It is important to note that there are two additional factors listed here. Focus is frequently directed onto matter alone. In Aristotle’s example of an eye which is blue, but not through formal causality, the point is taken to be that it is blue just because only a blue eye could be made out of the material available for making the eye. (Technically speaking, the matter is said to make the blueness of the eye and so eristics like eye-colour, pitch of voice’, etc., not responsible for ‘biologically superficial charact father’s body the the of ies propert ever ‘which for ible is not exactly individual form; but is respons of a much ation for’ (his italics), and so are ‘universals movements in his blood contain specific inform objection to these positions is not, it should be My ’). thought been has than ity general of lower order to the are, substantially speaking, individual, but rather noted, an objection to the claim that forms al (and hence that we can appeal essenti are that nces differe by duated indivi suggestion that forms are nces between individuals). See again n 73 above. to formal or final causality to explain differe being to one's mother is strictly unnaturat—even 7 eg. GA 4. 3. (Even though resemblance explained.) C£ Dean-Jones be to has still ---it 767°8-9 al: unnatur of female is unnatural in this sense her h it will become apparent that I disagree with (1994), 195 for a statement of the problem (thoug nts of moveme the by ed’ ‘master is semen the which to evaluation of Aristotle’s solution, according i Ee En Se mn the menses, as ‘ad hac’). 50 he word ‘neither’ (oute) which introduces : this clause might be read, as it were, epexegetically to the species; which ‘Blue eyes do not have a final cause unless common 51 is to say that they have no tion to. essence” But why, if this is correct, the restric connection with the account of the animal’s it, then it would apply to ail es preced what es amplifi merely clause ‘some cases’ (ep’ enidn)? If this a further qualification of vrhat has just been said, cases covered there, So the clause must be read as n to the species; and not even then will they commo viz: ‘Blue eyes do not have a final cause unless that it be ary condition for a trait to have a final cause necess a is It cases.” all in cause final a have condition. (As a matter of interest, Aristotle ent suffici a r, howeve not, is it ; species the to common respect nt within species: humans are unusual in this supposes that eye-colour does tend to be consta HA 1, 10.) an issue of ‘hypothetical necessity’.)*! In the case of a child, similarly, we can say that it is the character of the particular blood out of which a child is to be made that helps to explain its individual characteristics. But a little thought in this case will show that matter is not nearly enough on its own. If matter were the sole determining factor for individual characteristics, then one of the original problems we faced would simply be turned on its head. Instead of expecting all children to look like their fathers, we should now expect them to resemble their mothers, since it is from the mother that their matter derives.” This is why Aristotle reminds us that we have to take into account the activity of the father as ‘moving cause” as well. For the active partner in the process of generation is not (as one might be tempted to say) the form, but the father. To be sure, he acts in virtue of having the form (this goes back to the point I made at the beginning BL <p . an un because there is no absolute necessity for there to be an eye; but if there is to be general PA ca = en the nature of the matter makes it necessary that it should be blue. See in Core ed a7 ki Metaphysics A5 Prior Analytics 94°36 with J. Barnes (1994), 232-3; cf. Balme , with the corrective remarks of Gotthelf (1987} 234-5 and Cooper (1987) 267. Galen already noted: On Szed 2. 1. 54-7, and further Section V below.

individual“? And being of the discussion); but he acts as himself a determinate of course has his own a determinate individual, not a mere form, the father way in which he individual characteristics. These, of course, will colour the es. Thus a child activiti his all colour they as tion procrea of s conducts the busines is made; but it it which out of might look like its mother in virtue of the matter that matter. shaping in activity his of virtue might also look like its father in here that the ize emphas to ry necessa is it shortly, emerge will that For reasons Aristotle is seen, have we As matter. father is responsible for the shaping of the child, even the of ristic characte every of g craftin the clear that the father does tion therefore involves those which it shares with its mother, and whose explana then, turns out to be ce, influen s the nature of her matter too. The mother’ slightly indirect. Rather than acting directly to bring about replicas of her own to, as it were, features in the embryo, Aristotle argues that the mother has instead that the mother’s blood persuade the father to do it for her. The idea seems to be be no need of the would there could, it (If . cannot initiate generative activity re or determination father at all.) But it nevertheless has its own structu in virtue precisely the mother’s of which it is naturally inclined to support an organism of semen is too weak to l paterna the that out turns it characteristics. In some cases word is (epi)kratei: see ‘master’ or ‘prevail over’ this inclination {Aristotle’s his design in the modify to has who GA 4, 3, eg. 7687341), Like a carpenter able to him, the intract prove knots and grain whose face of a piece of wood It modifies its blood. semen is instead ‘mastered’ and ‘prevailed over’ by the taken by the form the as just , activity activity—though the activity is still its wood did (the er carpent the by it to given form the ly carpenter's table is precise none of the work).” 83 “For,” as Aristotle cf. 194°29-31 (and says, ‘it is a man that generates a man’: Physics 194°9_13; cf. Metaphysies H4, 1044°34-6, where the ‘moving cause’ is distinguished from formal and final: two only). In more detail, Aristotle thinks listing them, Aristotle posits an identity between the latter exactly as a carpenter uses his tools to that the father uses his semen to craft the blood in the mother that the semen is produced by the father as craft wood: GA 1. 22, 730°11-22. The idea behind this is Since bloed in general has the job of concocted blood with its own generative principle of motion. nd the semen to be doing soraething understa can one l, individua the g sustainin and building capable of it, just the same as the father. See analogous in crafting an individual which is, as far as it is e of movement’; cf. 1. 20, 729*10-12; ‘principl e generativ the for 21 all of GA 1. 19, 726°13 ff. (726”19there is a difference of emphasis that g suggestin (1991e), 4. 3, 767°18-20). CHin ‘general Lloyd the status of the moving cause. between Aristotle's biological and metaphysical works over for Cooper's claim. (1988: 35-6) that # References in n. 77 above. I can, by the way, see no basis mentions eye-colour, pitch of voice, and the (he stics’ characteri al superfici ally ‘biologic are there responsible. natural quality of the hair) for which the father is not the movements required to make a 85 So when Aristotle says that semen contains ‘in actuality’ ts ‘of the fernale and her remoter movernen the ly’ ‘potential contains also copy of the father, but ancestors? (GA 4. 3, 7681114), what he must mean is that the semen is capable of reproducing traits r is capable of producing features that are not from the mother’s side of the family (as the carpente normally, and has to be prevailed upon to do part of his normal style)—only that it would not do so whom the es 2000, esp. 109, according to so. CH Dean-Jones (1994), 195-8 (but pace Dean-Jon semen once it has initiated the process of movements of the mother compete with those of the 154. 9-10, suggesting that the mother’s ad 222 1992: Lacy De also pace embryonic development; prevail). The notion of ‘prevailing to fails semen the movements ‘take over’ in cases where is helpfully assimilated by Cooper (1988: 31-3) to the notion 4 of ‘reciprocal action'—for which 53 All this might seem to amount to an unnecessary complication to the central and important point, which is just that every case of human generation involves a father and a mother, some semen and some blood, and that the particular nature of the child that results depends to some extent on the particular nature of each of these. But part of the reason for the complication is very important for the present study: Aristotle’s need to maintain the priority of form in the process. For it is in virtue of form (namely the form of human being) that there is any case of generation at all. But what the form does is to explain the activity of a determinate individual, namely the father, as Aristotle supposes. So the father must be thought of as doing the generating—all of it. The mother needs to have her say as well; but her contribution cannot be in any sense the source of this activity, This is why Aristotle supposes instead that, just as the father’s activity qua generator is filtered through his own nature as an embodied individual, so it is modified in the face of the particular characteristics of the matter with which he has to work. And this, at last, allows us to understand how it is that physiognomy becomes a live possibility for Aristotle. As we have seen, Aristotle has a ready account for the congruence between the psychological and physical traits of species in the fact that they both have a formal cause: an animal is structured in just the way necessary to realize its nature as a member of a particular species. Individual characteristics, on the other hand, are explained by ‘hypothetical necessity’: the characteristics of the determinate individuals responsible for the generation of the individual. But if species and individual traits have different explanations, they nevertheless come about as the result of the same real-world causes. In particular, both are due to the craftsmanship, that is, the generative activity, of the father. He acts primarily (in virtue of form) to create a certain kind of creature; but in order to do this, and in order to do it with the matter to hand, he sets about to craft an individual with certain determinate characteristics. And the fact that species and individual characteristics are unavoidably consequences of the very same causal activity, because the father’s production of the latter is the means by which he produces the former, is what guarantees that the correspondences between psychological and physical characteristics at the level of the species will obtain at the level of the individual as well. This is no kind of coincidence: as resulting from a single creative activity they are the very same correspondences. This not only explains in general terms how it is that Aristotle can consider physiognomy a theoretical possibility, it also underpins the practical investigation of physiognomical signs that he suggests in Prior Analytics 2. 27. For he says in this passage that physiognomy will be a possibility precisely if one can isolate the psycho-physical traits which differentiate species. The idea is that, when a physical trait distinctive of one species crops up as something that distinguishes cf, Generation and Corruption A 324°5-"13. Note, by the way, that it is only ‘things whose form is in matter’ (324°5-6} which are reciprocally affected in action. In practice, this confirms that we should focus our attention on the father/semen as moving cause, and away from the ‘form’ considered in isolation.

an individual member of another species, we can expect that individual to 26): possess the corresponding psychological characteristic (70°14as bravery to lions, it is For if a characteristic [pathos] belongs to one individual species, along with each change they i hypothes ex well: as sign necessary that there will be some belong to any other other. Let this sign be having large extremities—which cannot in this way when it is complete species, for (as we are used to saying) a sign is proper this sign appears in when And l, proper to a whole species, and not to an individua animal). But he will of type other r whateve (or man brave a have will you another species, right, and we can is this all If ndence, correspo have the sign: for there is a one-to-one and each has its istic, character one only have which animals those [or signs gather such -. nomy. physiog do to able be shall we sign, since it is necessary to have one, mend collecting One might have wondered why Aristotle does not recom r there is a whethe see to lity groups of men known for (e.g.) their irascibi makes the he general in why or n; commo in share distinctive physical trait they gation from investi the starts and uals individ and species n betwee distinction l: for one mer ar ee eee ee the svecies.®* Part of the answer to this will, no doubt, be practica But there may be thing, individuals can change or dissemble their character. (the difference status in nce differe the zes recogni that one too, reason another traits. For ual individ and species n in their respective explanations) betwee are species the of level the at traits l physica psychon the connections betwee is because a lion is (it ed ground ically teleolog are they that fact the by eed guarant man with large to be brave that it has large extremities). The fact that a hand, more or other the on is, men other than braver extremities will tend to be extremities. large having his to led that es process l physica the on ent conting less process that that fact the just is case What guarantees the connection in this it is, Since traits, species of ion realizat the to led that process is the same as the level the at ance appear in this sense, the correlations between character and the level of the of the species that guarantee correlations that can be made at starting suggest to le Aristot for sense of amount certain a individual, it makes with species traits. the second To spell this out in slightly different terms provides an answer for then that noted I well. as section this of ing beginn the at raised I n questio st, omenali epiphen an with ted physiognomy after Aristotle tended to be associa if emerge to starts this for reason e possibl A soul. the of not a teleological model ee for Aristotle guarant e ultimat the s remain gy teleolo while that, rs conside one link between that physiognomy can be given a secure explanatory grounding, the indirect. best at is ristics characte ual individ of tion produc teleology and the are in which es At best we can say that teleology explains the causal process in about come ristics characte ual operation such that particular individ #6 In fact he does appeal to individuals in the practical investigation of signs, but only when if the whole species has two “proper” brought to aporia by the examination of species: ‘But shall we know which of the signs how generous, and brave be might characteristics, as a lion if both occur in part of some follows: as Perhaps stic? characteri following ‘properly’ signifies v-hich in the whole of it, and other species, and in those species in which each is found though not brave but not liberal, and has just whenever a member has one but not the other: for if someone is the lion’ (Prior Analytics 70°26-32). one of these two signs, it is clear that this is the sign of bravery in 55 parier instances of generation. But for someone who focuses on individual 8 § (ie. the kind with which the physiognomist deals), this is not very far om saying that the proper explanation for the psychological and physical traits possessed by individuals is a ‘bottom-up’ explanation, one grounded i the physical processes of generation and growth. I shall defer until Section V a fuller account of the considerations moved by each side of the argument {epiphenomenalist and teleological) in the interpretation of Arist de’ psychology. It becomes most relevant in the case of Galen, who is the clearest example of someone who associates physiognomy with an epiphenomenalist reading of Aristotle, and whose work in any case provides us with the best available evidence for the clebate. But to anticipate the conclusion I shall come t there: teleology is most seductive to those most impressed by the constan and functional adaptation of species as such; but those who consider that an individuals character and capacities are explained by a physical proc conducted between two physical individuals, and that, on the physical lane, these are the means by which species traits are generated, might start to wonder what space is left for the supposedly prior activity of a metaphysical ‘form’. It turns out, in other words, that the attractions of epiphenomenalism are bound up with just that aspect of Aristotle’s psychology which explains individual characteristics and so, in the end, underpins his physiognomy. 7 (ii) Two Anonymous Aristotelians: the Authors of the Physiognomy I have suggested how Aristotle’s psychological (and, for that mat iologi theory provides the theoretical opening for his investigation into en insofar as it explains why we should expect there to be correlations between the psychological and physical traits of individuals. What is more, we have seen that these correlations are exactly the same as correlations between physical and psychological traits characteristic of different animal species. This fact all Aristotle to see the way forward to identifying which physical characteristics signal the presence of which psychological traits. I shall return to the subject in he appendix to this section but, in brief, the idea is that if two or more species ere just one physical characteristic (Aristotle’s example at Prior Analytics N u‘s large extremities) and one psychological trait (bravery), it can be in erred that wherever in the animal kingdom that characteristic is found, it will e indicative of the same psychological trait--whether it is found in another Species or (more to the point) in an individual member of a species not erwise distinguished for large extremities or bravery. Whether Aristotle imself made a systematic attempt to follow through the procedure he recommends to produce a systematic catalogue of physiognomical signs is a ES point there is a scattering of physiognomical observations through his 7© ogic works, and one might be tempted to put some faith in the fact that a rysiognomy was ascribed to him in the ancient catalogues of his works t atever one thinks of the pseudonymous treatise that survives under his name ut see further the Appendix below, nn. 119-20). We do, however, know of CE again n. 71 above.

ced works of physiognomy. In three members of Aristotle’s school who produ is Loxus, one of the sources tion (iii) below I shall argue that one of these subsec ognomy. Before that, though, used by the compiler of the anonymous Latin Physi whose works survive in the I shall consider the two anonymous authors was already circulating in (and us composite Physiognomy which comes to tical underpinning rather theore the is it Since name. antiquity) under Aristotle’s sts me in this study, intere that than the practical conclusions of physiognomy any of these writers of s aspect cal practi I do not intend to examine as such the er aspects of practice in the (though in the appendix to this section I consid at this level do not undermine Physiognomy precisely to show how differences they share with Aristotle that ion direct the essential unity of purpose anc er whether anything of their himself}. But it does seem worthwhile to consid psychological models) can be theoretical background (ie. their respective ce is not all that good; but eviden The discerned in what survives from them. g Aristotle’s immediate successors, what there is might suggest that already amon to an anti-teleological view of his physiognomy was tending to become linked psychology. the Aristotelian Physiognomy is One of the clearest reasons for supposing that I shall refer to as [Treatise] A er, a composite of two texts (which following Foerst at 808011, — spel TE mn TE and [Treatise] B respectively) is the fact that there is a very clear break again with a new preface which, where the treatise begins as it were all over ground as the preface at the though different from it, covers much the same es cover the methodological prefac both ular, beginning of the work. In partic which physical traits are signs for question of how one goes about establishing this extent, with Aristotie’s own which psychological traits—they converge, to have more to say about this in shall I , (Again 27. discussion in Prior Analytics 2. of this is that, very much like the appendix to this section.) The importance composite treatise is interested the of Aristotle in the Prior Analytics, neither part en body and soul such that betwe onship relati a is in establishing why there on has no bearing on the task physiognomy becomes a possibility. The questi signs only requires that we ical ognom at hand: the attempt to identify physi In other words, although two. the en betwe onship accept a close enough relati views about the nature of the the authors of the two treatises must have held them to an interest in led which relationship between body and soul why they should here betray what physiognomy, there is no particular reason on is not immediately relevant, those views were. Indeed, given that the questi have had a positive motive to would B and A one might think that the authors of they might have to explain or avoid putting themselves in a position where of r fact assume. Aristotle himself justify the psychological model they as a matte g at any tendentious views of his hintin in Prior Analytics 2. 27 studiously avoids and soul. All he says is that own about the causal hierarchy between body in respect of their natural least at ‘physiognomy is possible if one allows that, So when Treatise À and -9). (70°7er togeth e characteristics, body and soul chang a relationship whereby they are Treatise B insist that body and soul are in A at 805°6; B at 08°11, 19) heia: mutually affected (the word both use is sympat ents a particular theoretical repres this that se suppo to we should not be tempted Physiognomy and Ancient Psychological Theory 57 position on the narure of the relationship between soul and body. It is rather an assertion that body and soul are related in such a way that physiognomy becomes possible. In fact Aristotle too talks of the sympatheia of body and soul in this context (Prior Analytics 70°16 sympascheir: allélois\—and apparently means nothing more than he meant earlier by saying that the two ‘change at the same time as’ each other (hama metaballein at 70°7-8). | The examples of sympatheia offered by A and B confirm the non-theoretical intent of the word: they largely consist of particular events which one might normally have thought of as primarily physical events (illness and drunkenness at A, 805°2-4; diet and medication at B, 808°21—4) or as primarily psychological events (e.g. experiences of love, fear, pain, and pleasure at A, 805°5-8; grief and joy at B, 808°4-15), but which turn out to be associated with (respectively) psychological and physical changes as well. They talk, in other words, about operations of the soul and body, but say nothing about the causes of the characters or capacities in virtue of which they operate. But this does not quite exhaust the examples we are given. In fact each ‘Treatise offers ene example or claim which is of a rather different order, and may after all hint at their respective authors’ underlying psychological position. Alongside the examples of psycho-physical events, Treatise A offers, in fact begins with, what is apparently a perfectly general statement about the nature of the mind (dianoia): That minds follow their bodies and are not isolated and unaffected by the changes of the body is something that becomes very clear .. aa On its own, this would not be very significant, since the Treatise goes on to prove’ the point by appealing to physical events which are associated with psychological change, and then to show that ‘the opposite’ is equally true (there are psychological events, such as occasions of emotion, with physical implications). But a little further on, we are told something that makes it look as if these examples are misleading if read as a gloss on the original statement. For we are told that the type of dianoia any arimal has as a member of nd of body it has a“nay particular species ‘follows’, is a consequence of, the kind No animal has ever been born that has the appearance of one animal but the mind (dianoia) of another, but soul and body are always of the same animal—so that, necessarily, such-and-such a mind follows such-and-such a body. ; 3 The passage in fact reads, impossibly, hot hai dianoiai hepontai tois sömasi... touto de delon (‘That minds follow their bodies... and this is clear...°}. For discussion, cf. Vogt (1999) 289, Our ane ‘ation fus Foerster, who athetizes the ‘and’ (de); but it is just possible that more has dropped = en this opening statement and that the examples that follow were never intended as a direct onstration cf the opening statement at all. (This would be quite consistent with the understanding of the passage I develop in what follows.) verb that the word ‘follows’ (hepesthai) (also at 805°1 as quoted above) recurs centrally, and u u not coincidentally, in Galen’s discussion of the soul’s zelationship with the body. For its plication and use by Galen, see below, Section V, esp. n. 184. (Treatise B also uses the terminology of following’ b but claims that the body . ‘follows’; the soul as well as vice versa: ‘each follows en at £08°20,

are associated Of course there are physical and psychological events which only the most and that, deny to tried with each other: no theory of the soul ever might have soul the that deny to wish would lism nomena reductionist epiphe is a causal there that trate a casual impact on the body. This does not demons capacities the r whethe is n questio real The soul. equality between body and by the body or ed explain are soul the of ristics characte uishing or disting us, the author of themselves explain the way the body is. From what he tells as such ‘follows’ soul the line: former the take to clearly fairly Treatise A seems the body. My claim, then, is that the author of Treatise A was an epiphenomenalist of tory of his or her some kind: that he took the individual’s body to be explana that he would basis this on is it that psychological capacities and character, and e B on the Treatis . possible be might gnomy physio that have justified his hope Having view. e opposit the other hand, looks as if its author might have taken of the (hexis) state the in change that show asserted and provided examples to s in change that and body, the of } (morphé ance appear the in soul are reflected soul the of state the in changes the form of the body are, equally, reflected in (808°12-14), B goes on to make this rather striking claim (808°27-9): to supervete on (epiginontai) It is quite clear that the appearances (morphai) which come soul. the of powers the bodies are similar to natural reading of the This is, to be sure, very little to go on; but the most the soul, by whose sentence is that the appearance of the body is posterior to powers it is explained. treatise hang on Y emphasize that neither A nor B seems too keen to make his and neither soul; and body n betwee nship relatio the on the position he takes conclusive. is B} from one of the passages I have highlighted (particularly the it looks that note, to e possibl is it that all as Nevertheless, it is worth noting, least in his at ist teleolog a be might B e Treatis of author the if as a little bit t of Aristotle’s own account of species traits—perhaps then, an arthodox adheren e for a trend we evidenc already is A e Treatis of author the hylomorphism—while philosophical among shall start to see of an attachment to epiphenomenalism because the , perhaps , already ‘trend’ A writers who deal with physiognomy. e seems evidenc have we whom for writer al gnomic physio elian one other Aristot ‘the Loxus mean I well: as logy to have been epiphenomenal in his psycho physician’. (iii) Loxus ‘Medicus’ nomy and from Loxus, known to us as a source of the anonymous Latin Physiog in studies of a passing reference in Origen, has always been a difficult figure iate enough appropr is it though y,” curiosit a physiognomy. His name alone is Callimachus, Hymns 4. 292). Rose % Foerster (1886: 6 n. 1} compares the female name Loxd (e.g. 82, and further below), though this is conjectures that it represents a corruption of Eudoxus (1864: Celsus 1. 33: ‘Suppose that the made unlikely by its appearance in Origen as well as Anon. (Against or Polemon. .. that all bodies conform views of the physiognomists are granted, of Zopyrus, Loxus, to the habits of their soul... ”). 59 since it is itself a term of physiognomical analysis,’ and also suggests a connection with Apollo Loxias—-Apollo being patron of medicine (supposed to be Loxus’ own profession: cf. Anon. Lat. 1, where he has the epithet medicus) and of the interpretation of signs.” But if his name has exercised commentators, his philosophical affiliation has been the subject of even greater controversy. Rose (1864: i. 82) suggested that ‘Loxus was a corruption of ‘Eudoxus’, and that our author was the famous and wide-ranging scientific writer Eudoxus of Cnidus, a rough contemporary of Plato’s;~ but this view was attacked by Foerster (18884, esp. 509-10), primarily on the grounds that Loxus’ use of ‘animal’ signs was not cruder (so Rose 1864: i. 81) but actually richer and more subtle than that of Ps.-Aristotle. Since (Foerster continues) Aristotle himself treats the use of such signs as a practice in its infancy, Loxus must be later than he; and Foerster in fact compares Loxus’ psychological theory (I shall return to the details myself below) with that of the Stoics Cleanthes and Diogenes of Babylon, which placed him squarely in the later part of the third century sc (1888a: 510-11).°* Misener (1923) returned to Rose’s view that Loxus’ theory was ‘antiquated’ (showing in the process how subjective such judgements can be) and placed him once again before Aristotle: more precisely, between Empedocles and Aristotle (1923: 9-10). As I have already indicated, however, it seems to me that we do not need to look so far afield as either for the key to Loxus’ position and age. To judge by the evidence supplied by Anon., Loxus stands every bit as close to the Aristotle of the biological works as the Aristotelian authors of the Physiognomy. Loxus, 1 contend, is a writer of clear Peripatetic affiliation. The first twelve chapters of the anonymous Latin Physiognomy constitute an eclectic ‘introduction’ to the work as a whole, and much of the material has close or exact parallels in Polemen and Adamantius (cf. on this Misener 1 Laxos means ‘slanting’ or ‘ambiguous’ in Greek. It occurs in a physiognomical context, head at Ad, B30, 60; but cf. also Gleason (1990) 413, who thinks that it een to mean he es looks sidewise suspiciously” (and so is appropriate rather to the practitioner than the subject of physiognomy). ™ In particular, of course, Apollo is associated with the prediction of the future, and one of the things we learn from Anon. is that Loxus (like Polemon) believed that physiognomical signs could be used to make such predictions (Anon. Lat. 133): ‘Polemon and Loxus advanced this discipline to such an extent that they affirmed it could even predict some things in the future.’ (However, the Ka ends—er, rather, breaks off—shortly afterwards, in the middle of giving examples from polemon s treatise. There is no more information on the nature of Loxus’ predictions, and it is worth En in Fra that for all Anon. tels us, his prognostications might have been of a different order Bodom Ei = datesktare not on certain, bat he app appeam to have hora born between between400 400 and and 390 390 BC.Se Ser See on on I: e generally on Eudoxus, Schneider (2000), 293-302. of Ms als notes that Loxus uses the example of the Celts (cf. Anon. Lat. 9) in place the Ca cythians who appear at Ps.-Aristotle 805°27. He suggests that this might be linked with tic incursions into Greece in 280 BC (1888a 510-11; cf. also Evans 1969: 11}. However, out, the ‘Celts’ were known already in the 5th c. (1981: 25, citing Herodotus 2. 33

61 1923: 16-17). There are, however, two passages in which Loxus is explicitly representing features of physical constitution, and mo rti istics of the blood. So 648*2-11:7” we particularly characke second of those two passages (Anon. Lat. 12, tr. Repath)-—in which, by the way, The thicker and the hotter blood is, the more conducive is it to strength, while in introduced.” For start with the reasons which will become apparent, I shall us (here ) to be synonymous with anim 1 take anima (translated here as ‘soul’ ‘spirit’): But before we come to individual which Loxus placed the seat redness, the signs or signals, it must be known that the blood, in body with bulk and of the soul, if it is copious, provides the ucts the but constricts, dulls, anc. obstr hair with strength and thickness, senses and the sharpness of the human ant, it intellect, whereas if it is thin and not abund and diminishes all bodily parts, ns the body, mars the colour, thins but fosters the its ness of the spirit and does not block off character of the mind and excites the sharp s of sense which are organ the all on speed s confer it y finall mobility and movement, and vessels there are. tongue, nose, and whatever blood situated in our faces, the ears, eyes, spirit, and that spirit is t perfec a tes indica body the of t And so a balanced temperamen soul] so it is that the whole character [sc. of the perfect which is as brave as it is wise. And . rament of the blood takes its rise from the type and tempe weake Loxus, one right well have thought If one had not been told that this was whose author shares the thought that immediately of Ps.-Aristotle 813°11-35, in determining intelligence. Heat, we até the heat of the blood is a crucial factor ‘movements’ of the blood reach the the told, assists blood flow, and so helps out distortion (cf. 81320). It is as such ‘organ of intelligence’ quickly and with cold tends to produce dullness of the conducive to speed of thought— hile term of comparison, this time a factor nd senses. Ps.-Aristotle introduces a seco , but nonetheless essential to its flow: strictly extrinsic to the quality of the blood to go, the size of the body.” The larger the body, the further the blood has while a small one speeds it up. (A small 50 that a large body slows thought down be equally ineffective with a large cold hot man is so flighty in his thought as to ws (813°28-35): man.) Ps.-Aristotle concludes as follo in physical size make for effective or So we have discussed how excess or deficiency these points has e falls in the mean between ineffective bodies. But someone whose natur to bloed does not have far to go in order the For ive. effect the keenest senses and is most is why This again. mind the from away it force reach the mind; but nor does lack of space and the keenest senses. a moderate size makes for efficiency Before passing further comment, we should also look at what Aristotle st two axes also correlates character again himself had to say; for at PA 2. 2, he proportion to its thinness and its coldness is its suitability for sensation and intelligence... Best of all are those whose blood is hot, and at the same time thin and pure. For such are suited alike for the development of courage and of intelligence. Each of these three passages differs from the other two, but it will be seen that Loxus is if anything closer to Aristotle than is the Ps.-Aristotelian writer of the Physiognomy. Aristotle names three qualities of the blood that are relevant (its thickness, heat, and clarity: cf. also PA 651*12-17): Ps.-Aristotle shares with him only one (its heat; his other consideration is the length of its course as determined by the size of the body), but Loxus shares two (in referring to the quantity and thickness of the blood). Indeed, there might be evidence that he shares all three. If we take it, as commentators generally do, that Loxus is behind the discussion of hair types in chs. 14-15 of the anonymous Latin Physiognomy (it follows directly on from the passage of Loxus currently under consideration) then Loxus too associated ‘cold’ (and thin) blood with intelligence, and hot blood with bravery (and lesser intelligence). There is something further to note. if this is right. Aristotle, Ps.-Aristotle and Loxus would agree on this single fact, that ae temperature of the blood is relevant to the intelligence of the person; that colder' thinki n thinking against Ps.-Aristotle in line up agai ns, and Tonsxus stotle, an u Aristotle, but On the evidence of Loxus’ physiognomical analysis of blood, there are, then no grounds for supposing that Loxus is any less Aristotelian than Ps.-Aristotle. The conclusion to which his analysis points is soundly Aristotelian as well Loxus, like Aristotle and Ps.-Aristotle, looks for his ideal in someone who combines a quality somewhat like strength with a quality somewhat like intelligence; like them, he looks for this ideal somewhere in the middle of the physical extremes outlined. It is sometimes suggested that a characteristic feature of Loxus physiognomy is his claim that ‘that soul is perfect... which is as strong as it is wise’ fas above), a claim echoed in an intrusive line in ch. 10: ‘a good character is constituted in no other way than with the strength of the masculine and the wisdom of the feminine” We can now see, however, that this conclusion is as Aristotelian as can be. | 4 Cf. also 2. 4, 650°19-.651°19; cf. HA 3. 19. EE ekthong that Aristotle himself can be selective about the range of relevant physiological not ie Anein =, given context. See e.g. van der Eijk (1997), showing that the blood is itself the eyes), 117-31 is also mentioned explicitly at ch. 81 (on 95 Specificaliy: Anon. Lat. 2, 12. Loxus able conjecture that chs. 13-15 (en reascn a is It tion). predic in belief his (on animal types), 133 (on of blood on Loxan terms of analysis (e.g. the paucity the hair) also derive from Loxus: they draw Occasionally, . 597-8) 18884: er Foerst see and 12, cf. ch. which explains thin, pale hair in ch. 14: the terms retation difficult’ (ck. 1), Anon. Lat. notes ‘where I have found the translation or interp ing terms are quoted from factor relevant to the determination of psychological character. It is worth on oe : the only FA e way, that Aristotelian discussions of blood quality. and its effect on character and appearance makes it all the easier for Aristotelian physiognomy to be received into h al medicine later on: cf. Section V; also Ghersetti below, p. 289. “mee fem: aioe a Loxus en one side with Polemon and Aristotle on the other for according the how peve role in the mix (1990: 391). The thought that intelligence is a feminine quality is, alektrudn (131). in the text of f does something similar: cf. below % ] shall suggest that Loxus himsel Anon. Lat. 2. nur it is worth, the follow which he found in his Greek sources, For what (129), (81); Hchnos, philenklémén (123), glaux traulos s, aigöpo os, Loxus: oindpos (14); charop oe a; : = en (whether or not Mayhew 2004: 94-6 is right to say that Aristotle Fragt pe en ar, and rather low-grade, form of intelligence in mind). Se Ha 9, 1, 608°23-8: Fi emales are softer, more quickly tamed, inclined more to allow themselves to be „and easier to teach (mathetiköteron). Female Laconian dogs, for example, are better-natured

But this, I said, was the second passage of Loxus. The first is Anon.’s introduction to mention. ied): as follows (Anon. Lat. 2; tr. Repath, modif Loxus established that the blood is the ver that dwelling-place of the soul,” and moreo the give signs give different signs according to the whole body and the parts of it which to when ding accor or r thicke or er thinn is it er liveliness or inertia of the blood and wheth w channels, (But others think that, just as the it has free and direct or crooked and narro s its appearance from the quality of the derive soul the soul is the shaper of the body, so derives its appearance from the vessel, and body.) Just like liquid in a small vessel which et: although the breath is uniform, a trump or aulos, pipe, just like air breathed into a s, sound trumpet, pipe, and aulos make different i s on the one hand, and others (ceter Loxu een There is a contrast here betw ‘others’? It looks very much as if quidem) on the other. But who are these he sees as the characteristic what e Anon, is here attempting to summariz the passage just quoted with are Comp y. ognom Physi doctrine of the Aristotelian Ps.-Aristotle 808°11-15: Soul and body seem to me to affect each other sympathetically (sympathein allélois), A rance of the body, and, conversely, when change in the state of the soul alters the appea the state of the soul as well. es chang it es, the appearance of the body chang sted with a position like that of It is, I assume, the fact that Loxus is here contra rs to his theoretical affinity ntato Ps.-Aristotle which has tended to blind comme show that this conclusion ions derat consi two But with Peripatetic physiognomy. ent members of the same school, is hasty. One is the simple fact that differ ent things about the relationship particularly of the Peripatos, might think differ a teleologist, but we know that ally) (basic was between body and soul. Aristotle were interested in physiognomy) some of his followers (including some who sted that the two parts of the sugge were epiphenomenalists. Indeed, I have differ from each other in elves thems might y surviving Peripatetic Phystognom us that Loxus and the ‘others’ are this matter. Anon. himself does not tell d not just assume that they are. members of different schools, and we shoul sted by Anon, between Loxus and sugge ast contr the It might be argued that he does think of them as coming the ‘others’ is strong enough to show that even if is (HA 9 is certainly Aristotelian, by the way, [or ‘cleverer’: euphuesterai] than the males. for a more robust defence of its 21-2; 1983: at ion discuss us cautio not by Aristotle: see Lloyd's e PA 648°2-11 (imelligence linked with thin coal authenticity, Balme 1991: 1-13.) One might compar not cooler than that of men—though admittedly is women of blood) with GA 765°15-18 (the blood s a blend of involve nature best the that le princip thinner according to HA 521°21—4). The general the Politicus 309 E-311 c; Laws 773 au; later on, e.g. oppesites was widely accepted: cf. already Plato, Stoic Cornutus, Introduction 16, 19 (26. mous Platonist 1-2, 34. 12-20 Lang} and cf. the anony 7. Commentary on the Theactetus at cols, 9. 25-10, 100 Note that Lexus does sof say that the blood is the seat of consciousness or of thought—an over Aristotle, Misener, to argue against the connection with translation that led Foerster, and after him 9: “With Loxus we return to the 1923; r Misene (so earlier rather him and ultimately to locate cf. Evans later (Foerster 1886: 6 n. 1; 1888: 510; and primitive conception cf Empedocles’) or rather ng substantially different anythi means Loxus that suggest to here g 1969: 11). There is, in fact, nothin physical es the psychological as well as the from Aristotle’s belief that the quality of the blood underli below that Loxus was a psychological argue characteristics of an individual. (I shall, however, level in the explanation of the claim.) epiphenomenalist, which would introduce differences at some 63 from quite distinct philosophical backgrounds. But this lea consideration, which is that we have a very good reason not to trast pone judgement in this matter. This is not because of any obvious failing on Anon.’s part, but simply because he presumably does not have better evidence for the relative positions taken on physiognomy by Loxus and the ‘others’ than the two textshe has in front of him. On the reading of these that would be most natural to him they would indeed appear to be taking rather different theoretical positions. I suggested in subsection (ii) above that the two parts of the Peripatetic treatise in fact held different views on the question of the relationshi between body and soul, but that they were careful not to make the relevance of their respective discussions rest on the particular view one takes. Both parts then, talk about the relationship in terms of sympatheia—a word which, I have suggested, is precisely supposed to avoid signalling the theoretical mem that underlies the relationship, and to focus instead on the simple fact that the soul and body do (one way or another) interact in the relevant way. But a reader of the treatise who did not recognize its dual authorship, a reader like Anon, who thought that the whole treatise was by Aristotle himself, would naturally get the impression that the term sympatheia was meant to be a term eh arn it named the theory upon which the physiognomy of the treatise This is why Loxus looks so different from Anon; for Loxus avoids the term sympatheia altogether. More than this (and the reason for it), Loxus seems to have been specific in aligning himself with an epiphenomenalist view of the soul This, at any rate, seems to be suggested by his claim that the blood is the dwelling-place’ of the soul. He appears to mean by this that the quality of the blood is responsible for the character of the soul: the blood is, as it were not just where the soul lives, but actually where it comes from. This er to be what he means when he calls the blood the ‘seat’ (sedes) of the soul in Anon. 12 (quoted and discussed in the foregoing), and it seems to be confirmed in ‘the examples that follow.'°? These examples express the differences in character . 1 {assume that Anon. used a text of Ps.-Aristotle that wa:; at least a versijon 6 text that we know. For evidence that makes it reasonable to assume this, hl nn erdeA haere n. 119. | in the ° g these examples to come from Loxus, rather than the ‘others’, I am follow: Abbe) 51 1, pace e.g. André (1981) 51; Raina (1993) 127; and Repath in this volume. Fi in met extthiswor is the fact that the images in question illustrate the effect specifically of body sr, mie e mutual interaction of body and soul which was the point of introducing the Seren ie oen en sees a resemblance between these images for the soul and an image attributed by man = he — eanthes {Seneca, Letter 108, 10 = SVF i. 487), a fact which he uses to support g of Loxus to the 3rd c; but it is important not to exaggerate how close they are, Loxus umKE ishe) says that the effect of the body on the soul is as that of a (musical) pipe loudervps 2 5 ri same breath sound different; Cleanthes says only that our breath sounds thet bees ar i rough a trumpet, There is certainly nothing here to encourage the thought possiblity: 1as u rie and the second image (again, if it his) positively militates against the aire vn Fi Le compared to liquid held in a vessel (vasculum) which changes its shape the soul a : evene. Tris metthat a Stoke ee A that the body is metaphorically the ‘vessel’ of ; : , but only ins location. The manner of the souls relationship within the =ROES cee eee B

that blood of a given kind takes on as it flows through the bady;! and the d here seems to be point of the metaphors of ‘shape’ and ‘tonality’ invoke tand the epiphenomenal precisely that this is one way in which we can unders n individuals are betwee nces nature of the soul. Psychological differe flows through the it as blood the of quality the expressions of differences in of a reciprocal talk to body. Where the Aristotelian Physiognomy appears has causal body the that clear is Loxus relationship between body and soul, priority. In short, Anon. has no reason to think that Loxus is coming to physiognomy considers to be) Aristotle. On from the same theoretical position as {what he lism suggests to Anon. the contrary, his clear commitment to epiphenomena ’ of sympatheia. But ‘theory elian a striking contrast with the apparently Aristot good reason to deny very no is own its on this that (and quite apart from the fact nt psychological differe the Loxus a Peripatetic affiliation) if I am right about sees between Anon. which t contras positions in fact taken by A and B, the very ly with the precise up lines Loxus For tes. evapora Loxus and the Physiognomy faith in his ing ground position I attributed to the author of Treatise A by as we this, in And soul. the of view physiognomy on an epiphenomenalist ian istotel post-Ar of ty majori the unites that trend a of shall see, he is part philosophers who discuss physiognomy. Appendix: The Aristotelian Physiognomy most, though not all In the body of this section, I have assumed (with gnomy that comes Physio the of author the commentators) that Aristotle was not down to us under his name; and I began to argue a | ee Ter oe hd | that the author of Treatise A seems more interested in developing a catalogue of signs based on a study of animal species, while B sets out by making a grand division of animal characteristics into those characteristically ‘male’ and those characteristically “female’.' This fact is, it seems to me, significant not only in confirming the impression that two distinct works went to form the survivin Physiognomy, but also that they were written by two different people,!0® It does not strike me as being of any special theoretical significance, however. 7 On the contrary, it is quite possible to read the different ways in which A and B go about isolating physiognomical signs as variants of each other, and actuall complementary rather than contradictory. (Their being complementary could even explain how it is that Treatises A and B came to be put together, despite the superficial differences between them.) And if A and B go about their work in different ways, what is much more striking is the fact that they seem to share a common starting-point for this stage of the work; and, furthermore, that they seem to share this starting-point with Aristctle’s discussion of how one would go about establishing a physiognomy in Prior Analytics 2. 27.'°® Given that all three do believe that body and soul are related in such a way that physiognomy becomes, at least in principle, a possibility (despite, that is, the differences I have mooted in the details of their underlying psychological models), their views on how one would go about translating this possibility into practice are remarkabl convergent. To make the point, on the following pages, side by side, I sive Aristotle, Prior Analytics 2, 27, and the prefaces to Treatises A and B (For reasons of space, some sections have been summarized rather than translated; in these cases they appear in square brackets.) (in line with most, though any case represented two again not all, commentators) that the Physiognomy in authors, who might be etic Peripat nt differe works on physiognomy by two logical beliefs. In this psycho ying underl their by other each distinguished from assumptions; but these for cation appendix, I want to provide some further justifi there is a very them, n betwee nces differe the I also want <o show that, despite and B in the A es Treatis close affinity between Aristotle and the authors of gnomy. physio al practic a of direction each takes towards the development begin with often gnomy Physio the of s Studie e. (i) From Theory to Practic by which les princip al apparent differences between A and B in the methodologic on to move and signs, al gnomic they seek to construct their catalogues of physio noted often is it lar, particu In ves.’“4 themsel gues differences between the catalo GD devant to the explanation of psycho-physical characteristics (see again IN. ii above). Vogt 0 1 notes in more general terms that Treatise A focuses more on ‘theoretical’ issues, which on the whole avoids, and expresses greater caution. me B at 809°28 ff. (and see further below). both CE. Rose (1854) 221-5, (1863) 697-8; Foerster (18885) 285-90, (1893) i, p. xix (concluding in ii ee at both authors must have read Aristotle), pace Vogt (1999) 191-2. Vogt deduces from : e lack of theoretical contradiction between A and B to the possibility that they are the works of the ame author working on different occasions with different purposes. I agree with her (and argue ner below) that there is no methedological conflict between A and B; but I think this to the PO in the body like liquid in a vessel (IV. 4-5); example, explicitly says that ‘the soul is not contained makes clear, it is the soul that holds that ius Posidon as that, bly presuma is this for and the reason position and language of the Epicureans: the contrast EK; (F149 versa vice body together rather than 936). 555, 440, 3. s e.g. Letter to Herodotus 65-6; Lucretiu here—the nature of its course 103 For the introduction of extrinsic characteristics of the blood above. text the in d discusse as 35 813°11through the body—cf. Ps.-Aristotle in underlying psychological theory I 104 Other noticeable differences, apart ftom the difference terminology (see further below), and suggested in III. ii above, are differences in expression and apparent differences between Aristotle and Treatise 65 B (at 813°7-35) over the qualities of blood each eocannot see how aa difference in purpose would explain the differences in practical ppr f y them. This, in addition to the apparent, and if real then theoreti sigrificant, divergence between A and B over the nature of the soul discussed in IT. ii above, |eed me to oe that the authors of A and B are in fact distinct. | ° (2001), 2 35 35 n. 2:: ‘The differences between the two sulbsectiions Ps Ar! ace eg. Sassi + D herae he ene are also significant from a theoretical point of view, are rightly ween a significant à germes is to Raina 1993: 24 ff). This is not te say that there are no theoretically em ifferences: the difference between their respective psychological models would be an mpns if my reconstruction in JIL. ii above is right ace Vogt (1999) 1.91, who sees a parallel only between Treatise A and the Prior Analytics.

extremities—which cannot Prior Analytics 2. 27, 7067-32 TREATISE A TREATISE B me to That minds follow their bocies and Soul and body seems to Physiognomy is possible if affect each other ted unaffec and d isolate not are one allcws that, at least in sympathetically. A change by the changes of the body is respect of their natural in the state of the soul something that becomes very characteristics, body and alters the appearance of clear in cases of drunkenness soul change together. For the body and, conversely, mind of states for illness: and perhaps semeone learning when the appearance of deal great a change to appear music changes his soul a the body changes, it through the affections of the bit; but this characteristic changes the state of the body. And correlatively, the is not one of the ones soul as well, [Emotional body is clearly affected along natural to us, as are things states involve changes in ons of the soul in affecti the with like (araong natural the body, and there are and fear and grief love of cases movements) cases of anger physical cures of ‘mental’ and pleasure. Again, among and desire, illnesses.] things that come about by nature one might especially see that the body and soul are connate with each other to the extent that in the case of most 4 | So if this is allowed, and there is one sign for one characteristic, and we can grasp the characteristic affections they are causally active on each other. [An animal never came about such as to have the form of one animal and the mind of another; and there are experts in judging the character of animels from their appearance.] If these things are true (and these things are always true), there can be physiognomy. do physiognomy. complete species, for (as we are used to saying) a sign is ‘proper’ in this way when it is proper to a whole species, and not to an individual. And when this sign appears in another species, you will have a brave man (or whatever other type of animal), But he will have the sign: for there is a one to-cne correspondence. If anirnals which have only one characteristic, and each. has its sign, since it is necessary to have one, we shall be able to do physiognomy. [Three methods of physiognomy have been tried: (i) from resemblance to animal species; (ii) from resemblance to human races; (iii) from appearance of emotional states. Difficulties à a ee EE eS Surely not, for he cannot tell what they are signs of. They ought to be signs of peculiar characteristics, but we have no right to assume that there are any mental characteristics peculiar to the different kinds of animals that we examine in physiognomy. Courage is not confined to the lion, but is fourd in many other creatures; nor Emidity to the hare, but it shares this quality with numberless other creatures. Thus it is equally fruitless to select the common with all of these.) But is he any better off if he takes every genus by itself and selects signs that are For if a characteristic belongs to one individual species, as bravery to lions, it is necessary that there will be some sign as well: ex hypovhesi they change along with each other. Let this sign be having large in all animals with bushy gather such signs for those signifies which characteristic? peculiar to each? insolence, which is found all tais is right, and we can signs following ‘properly’ and sign proper to each species, we shall be able to of animal behaviour, some traits are proper to a particular species, others common to several; and the proper activities of the soul are accompanied by proper physical traits, common activities by common traits. Examples of common characters are belong to any other But if the whole species has two ‘proper’ characteristics, as a lion might be brave and generous, how shall we know which of the Perhaps as follows: if both It is quite clear that the appearances which come to supervene on bodies are similar to the powers of the soul, so that all the resemblances in animals are indicative of some identity. Again, in the case Occur in part of some other species, and in those species in which each is found though not in the whole of it, and whenever a member has one but not the other: for if someone is brave but not liberal, and has just one of these two signs, it is clear that this is the sign of bravery in the lion. 67 and the peculiar features, and we must abandon the attempt to proceed by an examination of every kind of animal singly. Rather, we ought te select our signs from all animals that nave some mental characteristic in commen. For instance, when investigating the external marks of courage, we ought to collect all brave animals, and then to inquire what sort of characteristics are natural to all of them but absent in all other animals. For if we were to select this or that as the signs of courage in the animals chosen in such a tails, and violent sexual excitability, which is found alike in asses and in pigs: whilst on the other hand railing is a character peculiar to dogs, and insensibility to pain is peculiar to the ass. It has already been explained how common and proper characters are to be distinguished.

At a linguistic level, there are differences and similarities here which one might way as not to exclude the possibility of the presence in all these animals of some other mental characteristic, we should not be able to tell whether our selected marks were really sigs of courage or of this other note.° These, of course, are not without significance. But what is much more character. The animals from which we choose our signs must be as numerous as possible, and they must not have any mental characteristic in common except that one of which we are investigating the signs. [Additional remarks: {a) Transitory physical signs of no use the physiognomist. {b) Changes in the soul which leave no observable mark in the body: e.g. the acquisition of knowledge or skills...) Here I may mention a possible method of physiognomy which has never yet been tried. Suppose, e.g,, that irascibility and morose sulkiness necessarily involve an envious disposition, and that the physiognomist could, without any bedily signs of the last character, deduce its presence from the presence of the other characters, we should then have a method peculiarly appropriate to masters of philosophy, since it is, we [Additional remarks: (a) The need for experienc? and subtle discrimination; {b) the importance of epiprepeia...] Deduction also should be used in the selection of signs, whenever possible. In the deductive procedure we attach to our data known attributes of them. For instance, if we have it given that a man is sharneless and niggling, we can add that he will be a thief and a miser, the one as a consequence of his shamelessness, the other as a consequence of his suppose, the peculiar mark of niggardliness. In all such philosophy to be able to tell cases we ought te include the deductive method in our procedure that, when certain premisses are given, something necessarily follows. But this method which considers the interrelations of mental characteristics and that which proceeds by observation of animals sometimes arrive at 69 contrary conclusions &c.'°° trait x necessarily is accompanied by 109 The idea (explained at 807°3-10) is that if character presence of y ie. even if there he " the signal also will x of s indicator physical then y, character trait characteristic y}. For a prac ical psycholog the and x of sign physical causal link between the interesting is the degree of convergence in the argument of each of the writers. I de not wish to exaggerate the extent to which they overlap, or the extent to which the overlaps I have suggested constitute exact parallels. But it is significant that they overlap as much as they do; and, more than this, that where they do not overlap, or where the overlaps represent a similarity rather than exact coincidence, they complement each other rather well. There is a close similarity in the overall structure of their respective arguments (all three passages begin from an assertion of the intimate connection between body and soul, and move towards a physiognomy based on signs derived from a study of animal species): and it is, in fact, remarkably easy to imagine a single, coherent Ur-discussion of which they might all be thought of as descendants: 1. There is a sympatheia of natural physical and psychological characteristics (‘natural’ here excluding acquired characteristics such as learnt skills). 2, To appreciate how intimately the two are connected, consider: (a) the physical effects of occasions of anger, desire, etc. which we normally regard as ‘psychological’; (b) conversely: the ‘psychological’ effects that can be brought about by physical changes—as madness can be cured through the use of drugs. 3. If there is a one-to-one correspondence of observable physical sign and psychological correlate, physiognomy will be possible. example, cf. perhaps PA 3. 4, 667°19-21 where, if we follow Peck’s Loeb translation (in Peck and Forster 1955), animals ‘whose cowardice is either outright or else betrayed by their mischievous behaviour (my emphasis) have big hearts. (However, Aristotle might have meant something more like “animals which are clearly cowardly or made mischievous through fear have large hearts’: megalas tas kardias echoust...panth’ hosa phanerds deila € dia phobon kakourga.} N° Differences include the following: Treatise A alone talks of dianoiai (‘minds’: e.g. 805°1) and psychological éthé (‘characters’: 805°27 &c.); Treatise B alone of the soul’s hzxis (‘state’: e.g. 08°22) and dunameis (‘powers’: 808°28). B is also unique in using the notion of epiprepeia (first at 809°13; cf. n. 113 below). Both parts of the Physiagnomy use the language of ‘following’ to describe the relationship of body and soul (albeit the relationship they describe is rather different: see again UL ii above): A at 805°1, B at 808°20-1. As to similarities: all three texts use the word sympaschein! sympatheia to describe the relationship between body and soul (Aristotle at 70°16, Treatise A at 805°6, B at BO8P11, 19}—a word which, in this usage, is first attested in these texts. All use the notion of a sign ‘proper’ to a species (idion; as opposed to a sign ‘common’ to many species: koinen); and both Aristotle and Treatise B excuse their failure to explain the meaning of the term: Aristotle appeals to school usage (70°20), B appeals to previous discussion (‘it has already been explained’ at 809°—though in fact it has not been ‘stated’ anywhere at all in the Physiognomy). And there is an unmistakable resemblance between the conclusion to Aristotle’s opening remarks (‘if this is allowed, and there is one sign for one characteristic, and we can grasp the characteristic and sign proper to each species, we shall be able to do physiognomy’: 70°11-14) and the comment made by Treatise A in the same context (‘If these things are true, and these things are always truc, there can be Physiognomy’: 805°17-18).

been tried, all with their 4. Three previous approaches to physiognomy have difficulties: y, and is prey to these (a) from animals; but this has been done crudel nm objections: "? species, it will (i) if you appeal to signs shared in common with other like one more is sign the g sharin man be impossible to say that a | r; anothe species than ible imposs be will it , species (ii) if you appeal to signs peculiar to one characteristics to know what they are signs for--unless there are are; ever if rarely there which well, as s specie the to peculiar ible imposs be will it , species (iii) if you appeal to signs peculiar to one for. sign à is sign the ics terist charac to know which of its same kinds of theoretical (b) from human racial types, but: < understand the | problem as with (a)?> ns, but: emotio ular partic with ated associ sions expres (©) from on to different characters; (i} the same facial expression may be comm j his own; (ii) a man can assume an expression not normally sion expres facial from drawn be can that nces (iii) the number of infere is small. wey forward: one must 5. A refined form of animal physiognomy is the only have one and only one choose a sample group of animal species that only one psychological and one physical characteristic in common, and characteristic. 6. Further considerations: ae {a} signs liable to alteration are of no help; on in signs; (b) there is a practical need for subtle discriminati note on epiprepeia; . syllogistically. ta * hysiognomical conclusions can be extended tetic Peripa a of sketch (ii) Physiognomical Signs in Ps.-Aristofle. The above periment than as serious Ur-discussion should be regarded more as a thought-ex le makes very clearly the reconstruction. Nevertheless, the fact that it is possib s a physiognomist who compared U Cf perhaps GA 4. 3, 769°20-1, Where Aristotle mention f two or three animals. i ae Non lanl contradiction between Treatises A and B ocr whether there are = but note that Treatise A does not denyi psychological characteristics peculiar to individual species; no right to assume their existence. In! have we that says existence of such characteristics: it merely n lying behind both passages 8: PE one could easily imagine a perfectly Aristotelian reflectio characteristics peculiar to the different have no right to assume that there are any psychological eitat railing is peculiar to dogs, an one unless nomy-— physiog in examine of animals that we i ivi in is peculiar to asses; though such cases wil be rare.” 71 point which seems to me most significant, namely that there is a strong familial resemblance between A, B, and Aristotle himself in their approaches to physiognomy as a practical art. And this resemblance continues to link the specific recommendations each makes for establishing physiognomical signs. In saying this, I do not, again, wish to deny the obvious differences identified here by many commentators. In particular, the author of Treatise A confines himself to the comparative study of animal species suggested in the text already quoted, while B goes on to introduce the idea that the relevant differentials can be classified according to a gendered polarity (809°26—30): Now, first of all, I shall try to distinguish among animals which ones stand out as being brave or cowardly or just or unjust. The whole animal kingdom is to be divided into two types (morphai}: male and female, and what is appropriate (prepon) associated with each type... B, then, does something rather different from anything we can find in A or the Prior Analytics.\'* But while fully recognizing this, what I again wish to focus on is the fact that differences between A and B are not of a kind which suggest that they are the product of distinct theoretical traditions. Indeed, as I suggested above, it is possible to read them as positively complementary. Take, to begin with, the approach apparently preferred by Treatise A (and perhaps Aristotle himself}, which is to classify psycho-physical traits on the basis of a study of animal species. In this case, the idea is one should try to find sets of non-human animal species (i.e. at least two of them) which have one, but only one physical trait in common, and one, but only one psychological trait that they share to exclusion of all other animal species. The resulting psycho-physical correlation can be applied to individual humans: any human with the physical trait identified, will, so far as the natural state of the soul is concerned, possess the corresponding psychological trait as well. This is A’s approach, and it makes sense as far as it goes. But it should be clear that its execution would be by no means a simple task. Not the least of the problems facing someone who wishes to construct a physiognomy this way is the fact that not all physical similarities between any given species will be of any physiognomical significance at all. There is a striking example of this in Aristotle, who gives the hairiness of the ears as an example of a physical trait that signifies nothing about character!" It would be easy to multiply examples by looking at smaller and smaller details. My suggestion is that B approaches the subject of physiognomy with difficulties such as this in his mind. The way he tackles them is as follows. He realizes that the only secure way to identify physiognomically relevant signs will involve an understanding of what links a given sign, a given physical trait, with the psychological trait that it signifies. He also knows that this will involve an analysis of the blood. The reason for this is that, in Aristotelian biology, it is . It is used in slightly different senses by Pe Ariston Per the tevm epiprepria, see Appendix to refer to the fit or ‘congruity zZ and later physiognemical authors. For Ps.-Aristotle it seems Lat. and Adamanüus, on the other u h individual signs and the traits they signify; for Anon. ng that supervenes on the totality of his or it refers to the ‘overall appearance’ of a person, somethi BIJ. Ad. 45; Lat. Anon. (cf. istics her physical character "4 Tt should, however, be noted that B's generalization about male and female types has parallels elsewhere in Aristotle: cf. e.g. PA 3. 1, 661°32~4; also HA 9. 1 (though cf. n. 99 above on the authenticity of this book). F5 Ha 492324; though other physiognomical authors do not agree: see Anon. Lat. 82.

n of the quality of the blood that determines both the psychological operatio and a creature (as we have already seen to some extent in IIL. iii above), also its in fact, the physical appearance"! Whether this is made explicit or not, blood is, nomy. physiog lian Aristote in e common term of all inferenc In principle, then, we know that any physical trait explained by qualities nce for of the blood (as its heat, consistency, etc.) which also have significa do we how But trait, relevant omically physiogn a is ing psychological function the establish correlations between blood-type and psycho-physical trait? Surely access direct have even not do we that reflect we matter becomes hopeless when hysical to the quality of the blood, but can only deduce this from its psycho-p effects? that nature There is, in fact, just one way of making progress: we have to hope sets of psychoing contrast clearly of example one least at with us provides blood quality physical attributes that might reasonably be thought to track between the possible extremes of (say) hot and thick at one end to cold and thin traced by the at the other. Treatise B thinks it does: this is precisely the polarity difference the iation, different of terms other Like sexes. difference between the quality (cf. blood in es differenc to directly relate must sexes the between Aristotle himself, e.g. HA 521°21-4); but unlike other terms of differentiation, it 9). In every constitutes a polarity common to just about every species (809°28iation between the species, he thinks, one encounters the same kinds of different example, are psycho-physical properties of males and females: males, for way, sexual this In tamer. and smaller females fiercer, and generally larger for the basis l empirica e accessibl differentiation provides a secure and easily and physical relevant omically physiogn out mapping construction of a scale ce across psychological attributes, a scale that can be employed with confiden species. ed in sexual Commentators sometimes talk as if the author of B were interest theory nor in neither true is this but differentials instead of species differentials; his anyway) account, my (on because true, ally theoretic in practice. It is not species with concern a as interest in sexual characteristics is the very same interest differentiated differentials. The reason for this is straightforward: the sexes are es are differenc the as insofar least (at species are as ways same the in exactly to referred be can es differenc the physiognomically relevant). In both cases, psychosame the exactly in es themselv manifest and quality, differences in blood in fact B still physical correlations. But it is practically untrue as well, because hysical psycho-p r particula ne determi can cannot, and does not, think that he how he d suggeste have I far So species. animai of study a without correlations ne’ ‘masculi istically character might have constructed a scale stretching from this but ’; ‘feminine istically character are that those to attributes of body and soul indicative of does not suffice to establish which particular physical attributes are to follow has B this, do to order In s. which particular psychological attribute A Treatise in ied exemplif and Aristotle by ended recomm re procedu precisely the shared uniquely find to order in as well: he compares different species s between 116 gee B, 813°7-35 with IIL ifi above. For Aristotle's use of blood to explain difference different species, see e.g. PA 648°2-11, 651°12-17; GA 765°15-18. 73 psycho-physical traits. Something else which commentators often assume is that B makes an arbitrary assignment of different animal species to the masculine or feminine ends of his sexual polarity: the evidence for this is his claim that the lion is the most perfectly masculine of animals, while the leopard is the most perfectly feminine (809°14-810°13). It is true that he calls the lion the most perfectiy masculine type; but in fact he does not say that the leopard represents a‘ony perfect ‘feminine’ type. What he actually says of the leopard iss this thi Of all animals accounted brave, the leopard epproximates most closely to the female type—save in its legs, which it uses to perform any feat of strength. All he means by this, of course, is that the physical characteristics of the leopard generally stand in the same relation to those of other animals ‘accounted brave’ as typically ‘masculine’ characteristics stand to typically ‘feminine’ characteristics. To this extent, then, we can expect the leopard to exhibit psychological traits which relatively speaking are more like those characteristic of females in general. But this is not the point of the example: for one thing, it is something we could learn from observation of the leopard’s behaviour without reference to its appearance; and for another it tells us nothing about which of the leopard’s ‘feminine’ physical characteristics correspond to which of its ‘feminine’ psychological traits. The point of the passage lies rather in the qualifications: ‘of all animals accounted brave’; ‘save in its legs’. What this tells us is that there is just one physical characteristic which the leopard shares with other species united by the psychological trait of courage, namely it has large extremities. It is from this that we can draw a useful physiognomical conclusion: that, of all ‘masculine’ (and to this extent physiognomically relevant) traits, it is powerful legs which betray the psychological quality of courage. This, it might be noted, is a conclusion that is not dissimilar from the connection Aristotle himself pesits between courage and large extremities at Prior Analytics 70°14-17. In the end, then, B’s procedure is not so very different from that of A, or from that recommended by Aristotle in the Prior Analytics. All three seek to isolate pkysiognomical correlations by a comparative study of animal species: all three look for sets of species which share a single physical and a single psychological trait in common. The difference with B is merely that it offers a common scale of empirically accessible and physiognomically relevant terms by which his study can be guided. (iti) Conclusion In subsection (i) above I showed that the prefatory discussions of Treatises A and B seem to converge with Aristotle’s discussion of a physiognomical methodology in Prior Analytics 2. 27; in subsection (ii), that the systems built on the back of these discussions in A and B are perfectly complementary. I repeat that, in making these observations, I do not wish to play down the differences between the three texts under consideration: they are differences which (quite apart from the tentative evidence I adduced in II. ii Contrast Polemon, who does say, simpliciter, that the leopard is the ‘most feminine in appearance’ (Leiden, ch. 2; Ad. A2; cf. Swain below, pp. 180, 189).

logical models) seem to me to above for differences in their underlying psycho of anyone who should wish ders shoul the on ly place the onus of proof square distinct authors. Nevertheless, to deny that the three texts are the works of three more in common than is share rs I think it just as clear that the three autho made that they represent be easily could case a d, sometimes appreciated. Indee al discussion of the topic of three developments of the very same origin above would seem to rule texts three physiognomy.''* My comparison of the e (it emerges from the sourc al origin the itself as out the Prior Analytics er broad discussion). 1 doubt too comparison looking itself like a fragment of some se written by Aristotle himself treati ical that there ever existed a physiognom no external evidence for an which we have now lost. There is, in any case, the surviving composite from Aristotelian Physiognomy significantly different came into circulation mably presu se treati of A and B,!!° and this composite to think that A, B, and reasonably early on.'? But it is perfectly plausible Prior r (references in n. 106 above). 18 Of egain the conclusions of Rose and Foerste Physiognomy which have no parallels in the 119 14 is true that we find references to ‘Aristotle's are to chapter and fine number as printed nces (refere e surviving text. See in Anon. Lat, for exampl 35. 15-86. 25 {cf Ps.-Polemon 428. 12-13 [Foerster]); in this volume): 83. 3-8; 85. 1-8, 9-13 elian texts, are Anon. 86. 7-13; 87, 6-9; 86. 3-5. Further references, with parallels in different Aristot lemon 431. 3-5 (cf. Lat, 86. 4-7 (cf. HA 4922-3); cf. Ps.-Po HA 494716); Pliny, Natural Histery ively, the core of Polemon’s treatise (and so, detivat 11. 2734 := Aristotle fr. 286 Rose”. Nevertheless, the surviving of parts both in l materia with overlap that of Anon, Lat} shows significant n, Leiden chs. 53-66 and Anon. Tat. 90-9 and Physiognomry: the list of character types at Polemo ion -808"9 (cf. Swain below, p. 193), while the discuss 807°31— A e Treatis on relies tially substan 105-15 with inversions of Polemon’s order) though 49-72, Anon. (cf. 3-52 Leiden n, of body parts at Polemo elian below, p. 179). Other references to an Aristot reflects Treatise B, 810°14-813°6 (cf. Swain B indifferently. For and A like) hing (somet to access pose Physiognomy in antiquity similarly presup 808°17-19); 87. 3-6 (cf 808°32-3); 88. 3051-18). Cf, Ps.-?olemon 427. 1-10 F (808°19-24, 32-3); 428. 8-9 F (808°18 1-3 (cf. 808° 31-2); A, see Anon. Lat. 84. 6-14 (cf. perhaps muth Stobaeus, Eelogue i 47. 6, 301. 23-502. 17 Wachs Analytics 2 27 might all depend on unpublished work conducted by Aristotle in the Lyceum: that they are different recensions or versions or records, developed to practical conclusions in different degrees and in different ways, of lectures or notes by Aristotle which were available to his earl followers. 2! j Iv. THE HELLENISTIC SCHOOLS Occurrerces of physiognômon- in evidence for Hellenistic philosophers: Posidonius, F169. 85 EK (= Galen, PHP 5. 5. 22. 2). Occurrences of physiognömon- in fellow travellers of the Stoa: Philo of Alexandria, On Dreams 1. 164; Dio Chrysostom, Orations 4, 87,2% if the frequency of occurrence of the word ‘physiognomy’ in. our surviving evidence is a reasonable measure of interest in the subject, it will be immediately apparent from the list at the head of this section that the Hellenistic schools provide thin pickings for the historian of the subject. Physiognomy was clearly not unknown in the third and second centuries: at least some of the work that came out of Aristotle’s school must have been done in this period, But there is little evidence of interest in the subject on the part of the new movements and schools which dominated Hellenistic philosophy—especially the Stoa, the Epicurean Garden, and the Academy in its Sceptical incarnation. An interest in physiognomy is explicitly ascribed to only one philosopher of the period, namely the Stoic Posidonius. For the sake of completeness, I have included the {single} occurrences of the word in the later writers Philo of Alexandria and Dio Chrysostom, since both authors had an interest in Stoicism and sometimes provide valuable evidence for earlier Stoic thought. But their references to physiognomy, as it happens, suggest distance rather than (quoting, ‘from Aristotle's Physiognomy -19). (The opening n to the opening cf A, and so the line of Galen’s QAM also looks to me like an allusio Ancn. Lat, 83. 1-3 (cf. 8126346); 8-9 a whole: see Section V, n. 184.) Por B, see Physiognamica (811°33-6}; 9-10 rem inder of how difficult 5 to th e Aristotelian to: cI qquite gen erallyy it can be be toto relatee th the ancient catalogues 1-3 (cf. 812°24-5); 3-4 (B13°4-6); 4-6 (ef, 811°23-4); 10-11 (812°6-7); 11-13 (812°1)—12); 84.(cf, 810°28-30); 9-10 (812P13-14); 10-13 7-9 -8); (811°26 1-3 87. -15); (813°14 13-15 (811°17); 85. 6-8 (813°16-20); 11-13 (813°20-3); -16); (813°11 5-6 ; (B13°7) 5 -2); (808°31 {810°1-4): 88. 1-3 ; cf, also Suetonius, de blasphemüs 4. 70 references; cf. Kenner 1939: 455), in which case cf. neces 2. a 10. But the en 5 1-2 (cf, B11°29-30); 429. 34 12); 428. 10-11 (cf. 808P6-7); 428. 16-18 (cf. 811°24-6); 429. without serious support: see Flamand (1989) 612. it might also be worth noting Polemon’s Stoic connections in this context: he was taught by Dio Chrysostom (Philostratus, Lives of the Sophists i, 539, 23-5 Kayser), and could trace as 13-15 (813°23-5), Pollux, Onomasticon 2. 135. 7 (811°17 ate the 1st-c. BC grammarian Didymus as his immedi Taillardat, leading Vogt 1999: 200-1 to suggest °3, 811°3714, 812013 1, 810°30, 808°4-6 order: source). Cf. Ps.-Polemon 427. 11-428. 7 (cf. in 812011 ); 429, 11-17 (ef. 8114-9). (ef. 811728-9); 429. 6-8 (cf. 811°2-3); 429. 8-10 (cf. g11°3-4by the independent circulation of either a ee 75 predated 120 At Jcast there is no solid evidence that it was and ascribed to Aristotle. (See again n. 119 above, part of tie compasite—or of any other treatise of material kind the of and late, ly relative are treatise our note that the citations of material not in nt and they are not, in any case evidence for a differe which might have crept into ‘cur’ text-—that is worth notng that a Physiagnomy is listed It nomy.) Physiog elian Aristot earlier an for certainly not 97 5. 25; Vita Menagiana (ie. Vita Hes;chit} no. in the catalogues of Aristotle’s works: cf. DL. the former in one, and books; two in being as work the gives latter (Düring 1957 at 80-92). The bipartite treatise we have was compiled as late as Moraux (1951) 203 uses this as evidence that the two catalogues simply counted the same work the the 2nd €. AD. One might think, however, that salutary Cf. in any case J. Barnes (1997), 39-44 for a two}. or book one is it think we (do ntly differe = And for the (congenial) relationship between the Physiognomy and undisputed works of pea shoes see Lord (1983), 24; Vogt (1999), 195. emidorus of Daldis has been suspected of Stoicism (e.g. Hopfner 1937: 2240 wi his pedagogical lineage back through Timocrates and Euphrates (535. 32-536. 20; cf. Frede 1997, See Swain below, p. 187) to Musonius Rufus (F:onto, Letters 135. 3-4 van den Hout?, with poe 7 1-2). (In the other direction, his pupil Herodes Atticus—e.g. Philostratus, Lives 536. don D 1 —taught the Stoic Emperor Marcus Aurelius.) There is, however, nothing, either Fön olemon's writing a Physiognomy, or about the Physiognomy he writes, that could support the ver that he is himself influenced by Stoicism. Within what we know of the Physiognomy, = u ly tendentious doctrinal position Polemon adopts is the claim that reason lies in the ast: Leiden ch. 1, A5, cf. ch. 15. This is indeed something maintained by the Stoics (against le); but it more probably comes to Polemon from Aristotle, who believed for=ple); Plato,

in some way different affinity: it is invoked as a study comparable with, but far from providing are they word; the n mentio They do.'?* from, what ‘we’ or generally of Stoic, of y evidence that ‘physiognomy’ was an active categor Hellenistic, research.'*° e of reference to In the case of Epicureanism, the reason for the absenc soul are closely and body that insists physiognomy is fairly clear. Epicureanism tial indepensubstan their on insists also it but intermingled in operation, nature of its the by ed dence from each other.’2” The soul’s character is explain is true that It * body.’” the of those by tly) own atoms, and not (not even indirec of soul powers and growth the n betwee tion correla there is a perfectly regular e becaus not ically emphat and body at the level of the species; but even this is either in any sense causes the other to be a certain way. It is rather the case that a certain kind of soul just ‘naturally’ belongs in a certain kind of body, as trees belong on the earth, not in the sky.!?° (This in turn will be explained by no mechanism beyond the fact that, at an earlier stage of world history, the combination of this kind of soul with this kind of body happened to enen successful inthe fight for survival, and the two have since been reproduced in tandem} ”" Nothing in any of this suggests reasons, even incidental reasons, why below the level of the species a soul of this particular character should reside in a body of this particular appearance.’*' It is striking, anyway. that in the course of his attack on the belief in the transmigration of souls, Lucretius appeals to species-level correlations between body and soul but not correlations at the level of the individual. Strictly speaking, this men that he shows only that souls could not migrate outside their species. If he could have used physiognomical evidence to show that they could not migrate bejween individuals within species, one might have expected him 76 of physiognomists: in both cases, 2% philo compares his work of textual exegesis with the activity t (On Dreams 1. 164). For apparen is what from deduced is invisible and ing something underly nomical handbooks, see Chrysostom convergent with observations in the physiog remarks in Dio (on Gleason (1990}, 393; Bost-Pouderon (2003) Barton (19945), 104; Evans (1941), 101 n. 21; acquired or concern these of most that though, Oration 33 in particular). It is worth noting, the Christian Clement of Alexandria advising acquirable traits icf. Bost-Pouderon 2003: 160, quoting with n. 133, on the sneeze at Oration 33. to acquire them at Pedagogue 3. 11. 61. 8; and further below, hical or theoretical commitment to philosop Dio’s argue to nothing general, in is, there 53); and hysical interaction. In other psycho-p of theory ar particul a of grounds the on physiognomy, e-g. physiognomical practice, with familiar is and nce appeara in interest s words, if Dio has the Sophist’ place in Stoic thought. its for alonc let nding, understa hical he provides no evidence for its philosop Dio refers explicitly to physiognomists, It might also be noted that, in the one passage in which while they deduce a man’s character he is contrasting himself with them. This is partly because, of men accerding to what he knows of their s portrait present to intends rather he form, his from this character symbolically (see Oration 4. character; but it is also because his portraits will capture approaching the visual representation painter a to himself s compare 85-8). To this extent, he of a divinity. In other words, he presents what he is doing here a kind of phystognomy. as a Kind of allegory rather than 125 And cf. perhaps Epictetus, Dissertation 4. 5. 20, making the point that being virtuous (being 126 gee Lucredus, On the Nature of Things 3 passim {e.g in particular 168-76, 323-09, 445-525 kind of eyes or nose one has: you must see a human being’ in a strict sense} is not a matter of the human being. This in fact echoes Cleanthes a to belong that nts judgeme the has person the whether in form, only animals from differ people ated at SVFL 517: ‘uneduc , the question, Lucretius comes quite close at 558-79, 688-97, 741-71, 824-9). In arguing this side of the ‘sympathy’ between soul and body: prove to nomy times to the attempt by the Aristotelian Physiog drunk or sick (3. 463-86; cf. 805°2-4), he draws attention to the disturbance of the soul in someone ns can be cured (3. 510-25; cf. 0821-7). and to the physical medicines by which mental afflictio at the level of species between an animals He also appeals to the perfectly regular correlation the argument, and even the example, namely of form the {That r. characte physical psychological and to and so perhaps awareness of the allusion suggests -dog, hunting the the invariable spirit of Aristotelian Physiognomy is no argument that Lucretius followed that treatise to its conclusion: see 805°11-17.} On the Nature of Things 3. 741-53 with Ps,-Aristotle ts) who view the soul as a ‘harmony’ at 12T See esp. Lucretius’ attack on those {epiphenomenalis now, it turas out that the soul can suffer and 398-135. In contrast to the Aristotelian Physiognonty (or indeed spirit) at all (3. 145-51; contrast act without any corresponding affection in the body 805°1-2). lions, the velocity of deer, and the placidity of 128 So Lucretius 3. 258-322, where the ferocity of as Aristotle would have said, but of the blood, the (e.g.) of cattle are expleined by the quality not disrupt the activity of the soul (cf. eg. atoms that make up the soul. It is true that the body can Timaeus, something which has the ability Lucretius 3. 463-505); but, as we saw in respect of Plato's to disrupt the soul is not ipso facto the explanans of its character. 77 Things are slightly more complicated for the early Stoics, however, because they do have a certain amount to say about the correlation between individual appearance and character--enough to make a plausible case for supposing that the absence of the word physiognomy from our evidence might in this instance be accidenta‘. This, then, is something I wish to consider in subsection (i) below. And although I shall conclude that the silence of our evidence is, in fact, very eloquent (that the early Stoics were not and sould not have thought of themselves zs ‘physiognomists’), the study will lead to two important results. One will be, at last, an understanding of what the word ‘physiognomy’ was in fact taken to mean by ancient philosophers, such that it would be wrong to think that the early Stoics were physiognomists. The second result will be to show the danger to us of using the word more freely than the ancient philosophers of whom we use it. For in subsuming every case in which appearance and character 129 At Lucretius : 3. 94-7, the soul is said to be part of a person rather as a hand, foot, or eye is. These too, I suppose, are not caused by the body (of which they are organs), but have their place in gare as ijs of it For the trees that could not be found in the sky, see Lucretius 3. 784-9 (where ey are in fact used to make a broader point, viz. that souls quite necessarily found in bodies). a CE. Lucretius 5. 837-77. a te A A de It is true that both physical (see Lucretius 4. 1208-31) and psychological (cf. 3. 741-7) racter are inherited from one’s parents; but again there is 10 reason to suppose that there is a correlation be:ween which psychological peculiarities and which physical peculiarities one inherits from a given parent. | The closest Lucretius comes to such an argument is On the Nature of Things 3. 760-4. But the Point here is one concerning character. It is not, in other werds, the point that metempsychosis might force a soul into a body physically suited to a soul of a cifferent character. Rather, Lucretius’ en Be is with the absurdity of thinking that a rational soul would of necessity end up in the body a pre-rational child, There might be evidence of positive Epicurean opposition to Aristotelian gogo in the mocking use of physiognomical material from HA 1. 10, 491°34492"12 at , i Deipnosophistae } 8. 48, 353 BC. 7The provenance of this polemic, , which is part of a larger broadside against Aristotelian natural history, is not given; but it is immediately followed by anti-Aristotelian remarks which are ascribed to Epicurus.

we are at are said to be linked under the single heading of ‘physiognomy’, t and statemen the behind n risk not only of misunderstanding the intentio T3. Chrysippus says that goods and evils are perceptible, writing this in his 78 out to be (from exploration of such links, but also of confusing what turn a theoretical point of view) very different phenomena. (i) The Older Stoics to the The superficially plausible case for ascribing an interest in physiognomy closely very form whose Stoics at large rests in large part on a series of fragments of the art suggests the kinds of judgement made by attested practitioners (numbered for ease of reference in what follows): s, who TL. It is said that some young men set out to play a trick on Cleanthe from grasped be could (&hos) character that claiming in Zeno followed deviant, appearance (ides), They brought to him a man who was sexually d that but toughened up through his work in the fields, and they demande and out, it make not could He . character his Cleanthes should declare sneezed, he told the man to go away. But as the man was going, he = SVF i. 618; and Cleanthes said ‘I have him! He's dissolute.” (DL 7. 173 part =i. 204). image T2. Zeno of Citium seems to sketch a beautiful and properly loveable be pure; of a young man. He sculpts him like this: Let his countenance neck not his brow not relaxed; his eye not wide open nor half-closed; his thrown back; nor the limbs of his body relaxed, but keyed up like strings movement under tension; his ear cocked for the logos and his bearing and upon giving no hope to the licentious. Let modesty and a manly look flower hs goldsmit and shops s’ perfumer the of nt exciteme the with away but him, the spend women where shops other the and wool shops—and indeed all a brothel whole day adorned like courtesans, as though they were sitting in (SVF i, 246). * 133 ‘Sexually deviant’ = kinnidos; ‘dissolute’ = malakos. (I owe to Thomas Hubbard the observapresumed homosexuality were the tion that it is tendentious to translate this passage as if the man’s Anon. Lat. 11, ad fin.) (which Foerster and 53, 33. Oration Dio, at parallels see story, the For point.) 121, tracing the sneeze back to 1893: i, p. bezi, thinks might come from Loxus; cf. also Pack 1978: the fact that he attributes the and Tarsus, in set oration an from Aristotle), Dio’s version comes diagnosis to a ‘local’ led von Arnim to identify Chrysippus (who came from in this case (see SVF ii. 10a with comment on i 618). nearby Soli) as its author But the explicit naming of Cieanthes in the present text has led to much greater speculation about his interest Barton (1888a} 511; Misener (1923) 3 (cf. 17); Evans (1969) 10-11. in physiognomy: see eg. Foerster (1994F 101) uses the present text tury B.C. Stoic physiognomist as the point of departure for her imaginative claim that the ‘third-cen s, and adulterers’. Cleanthes’ Cleanthes makes it his business to unmask charlatans, effeminate a superficial similarity between an association with physiognomy has been further supported by (probably from Loxus) at Anon. Lat. image he uses for tne soul at SVF i. 487 and an image reported 2. The similarity is not, however, as close as it at first seems: see again Section II n. 102. 79 On the End, book 2: ‘The following is a sufficient basis for us to be able to say that goods and evils are perceptible: for not only are passions perceptible, along with their species, such as grief, and fear, and the like but so also is it possible to perceive theft and adultery and similar; and, a general, folly and cowardice and many other vices; not only joy and good works and many other right actions, but also wisdom and courage and the rest of the virtues’ (SVF iii. 85; similarly SVF iii. 84). Ta. The Stoics think that the sage can be grasped as such through perception, using his appearance (eidos) as evidence (SVF i. 204). On the face of it, the language in these texts obviously does coincide with language found in unambiguously physiognamical texts: note especially, for example, the claim in TI that character (&thos) can be discerned in physical = (eidos). But it ly quickly becomes evident that none of these texts refers to, or seems to rely on, am hing about the natural ge physical body from which their sedvetions are drawn—the shape en pe the breadth of the forehead: the sort of evidence on which Phaedo and Aristotle relied for their physiognomy, so described. T1 actually seems to rule such evidence out (Cleanthes was baffled when presented with the mere appearance of the man}; and both T1 and T2 appeal to movement or posture rather than the shape of bodily parts. As to T4: we know that there are ne such physical attributes that distinguish a sage. Apart from anything else, one’s physical features do not change when one becomes a sage (despite the fact that one is supposed at that moment to become ‘beautiful’ where one had been ugly before: see Plutarch, That the Stoics Talk More Absurdly than the Poets 1058 A). | | The key to all of these texts, then, seems to be T3: the quite general claim that virtue(s} and vice{s) can be discerned. It is tempting to explain this in turn by the fact that, according to the Stoics, qualities are corporeal: we might suppose that they can therefore be seen in, as it were, their own right. This cannot exactly be wrong; but at the same time it is important to remember that qualities are not entities separable from the bodies in which they occur. It is probably best, then to think of them in terms of secondary physical characteristics, encompassing such things as expression and posture (T2). This would fit in with the implication in T3 that they are perceptible in just the way and presumably for Just the sort of reasons that emotions are manifest in a person’s appearance. 5 C£ also Marcus Aurelius, Meditations 11. 15 (the eyes and voice of i goodness; see Rutherford 1989: 95-6 with n. 13). It ige: be proper to en testers as the manifestations of an individual’s tonos: compare the youth with limbs ‘keyed up = under tension (T2 above). Cf. also Hierocles Y. 11-15 (which, as the example shows, is ir Kin out caring not physical features as such): ‘It is not impossible that someone sufficiently ii = a judgement of chatacters {pr(os) e(pijkrisin étho(n}) could, standing near people when ‘one of dewe aiom themiatiner in which a person sleeps what sort of disposition (dia)thelsis]) the ly questionable by the fact our best pieces of evidence for Stoic physiognomics’ is made immediate on the ‘tension’ of the limbs) that it describes a pose that anyone (more or less; but cf. n. 135 below Nearer the point might be could adopt without sharing the character it is supposed to imply. how a “well-schooled young man Gleason (1990) 403, who sees it as advisory: the passage shows availability’. would know how to avoid body language that might signal sexual iden iwas strong and full of tension (tonow plfejres) or softer than it ought to be. thew x expectingJig el.. J even have concern for their corpses to fall decorously, as the girl in on i ab so much more will the proofs of their disposition (tekméria és) d(ia)th[fesjeds) pass wit ne bodies of those who sleep.’ (The “girl in the tragedy’ is Polyxena: cf. Ruripides, Hecuba 569 astianini and Long 1992: 427-8.) that the passage is 134 The translation is taken from Schofield (1991), 116. His claim at 115

My conclusion about these texts is that, if they are to be considered evidence for ‘physiognomy’, it is physiognomy in rather a different sense from what we have seen in Phaedo or Aristotle, in particular because they none of them concern the shape of the body or its parts. But if these texts can fairly quickly be dismissed, they are not the only fragments which constitute our evidence for the Stoics’ thinking about the relationship between appearance and character. It happens that the Stoics did also have things to say about the relationship between character and the shape of the body and its parts. In particular, we have clear evidence that Cleanthes, Chrysippus, and Panaetius, at least, recognized and addressed the fact (much discussed, as we have seen, by Aristotle) that children tend to resemble their parents in appearance and character. Here is the evidence for Cleanthes and Panaetius— shall quote and discuss our evidence for Chrysippus at greater length below: theory of generation has ever allowed. The soul of a child is made out of a piece 80 Cleanthes has it that the parents’ likeness is reflected in their children—not only through tics of the soul the cast of their body (corporis lineamentis), but also the characteris 5). Soul the On Tertullian, from 518, i (SVF notis) (animae Not only, says Cleanthes, do we become like our parents in body, but also psychologically, ) (SVF i. 518, from in our passions (pathé), characters (éthé), and dispositions (diatheseis Nemesius, On the Nature of Man 2, 20. 14-17 Morani),’* clear—a Panaetius has it... that souls are born, as our similarity to our parents makes (fr. 83 van bodies our only not ingeniis), (in characters our in appears which similarity Straate=n Cicero, Tuseulan Disputations 1. 79). It is nowhere explicit in our texts how the Stoics accounted for the psychological resemblance of children and their parents, but on the basis of what of we have seen in Plato and Aristotle there would seem to have been two kinds its approach available to them, One would be to say that the soul acquires in individual character from the body. (There is no problem, at least in principle, soul tue if and parents; its to ce resemblan physical has child a why explaining d takes its character one way or another from the body, then we can understan I fact In well.) as them to ce resemblan ical psycholog acquire to how it comes of face the on But below. think this must be right, and I shall return to the detail it, it might seem that the other kind of approach has more going for it. This of approach is to say (rather more in the terms of Aristotle) that the lines parallel are parents one’s with causality which lead to psychological similarity with the lines of causality which lead to physical similarity with them. This looks like a plausible way of understanding the Stoic claim that children resemble their parents in character as well as appearance because the Stoics did, after all, think that a child’s soul is a chip off the old block in as near a literal sense as any 136 Yon Arnim includes a third passage at SVFi. 518 (like the first passage, from ‘Tertullian —this between parent and child, not for the view that the similarity is traced to the seed. Apart from l character that, time On the Soul 25): ‘How can we explain in addition the similarity of psychologica we do not appeal ta the seed according to Cleanthes’ testimony, a person shares with their parents if ta psychological affinity of the soul? But note that Cleanthes here is only being invoked: as a witness of the soul at conception, anything else, this view is associated by Tertullian with a belief in the origin a belief not held by the Stoics (as we shall see). 81 {or ‘fragment’: apospasma) of the very ‘breath’ (pneuma) from which its father’s soul was made (SVF i. 128). It is true that the Stoics talk as if there were an element of discontinuity here as well: the pneuma inherited from the parents’ soul does not, properly speaking, constitute a soul in the embryo, which is conceived of as a kind of ‘plant’, sustained by ‘nature’ (physis SVF ii. 806). (We cannot call the pneuma which directs the development of the embryo and foetus a ‘soul’, presumably because the foetus does not have the capacities of perception and impulse characteristic of an animal.) But one might casily think of this as a technicality. When a child is born, its ‘soul’ is the result of the ‘tempering’ of its nature, of this very parce) of pneuma as it comes into contact with the cold air outside the womb. It is ‘tempered’ (the comparison is with steel} in a way that allows the creature to take control of itself (SVF ii, 806 again), and it changes from plant to autonomous animal. The idea would be that in some way the unique character of the individual soul is captured and fixed in this process as well. A more elaborate account might describe this character in terms of the ‘tension’ (fonos) transmitted as part of the pneuma from parental soul through the embrye’s ‘nature’ to the ‘soul’ of the child. In any case, there is very clearly a line of physical continuity which the Stoics might have exploited to explain the inheritance of psychological as well as corporeal character. Whatever the attractions of such a story, however, it is clear that it cannot be right. There are, in fact, two reasons to think that the Stoics did not believe that a person’s character was somehow inscribed in the physical nature of the pneuma that makes a child’s soul. One is a text (not quoted above) which shows that Chrysippus, at least, did not as a matter of fact think it; the other is a consideration from Stoic psychology which makes it impossible that any of the older Stoics might have done. Consider, then, first of all, what Plutarch says of Chrysippus: Chrysippus adduces as proof that the soul has come into existence, and comes into existence later [sc. than the body], the fact that children become like their parents both in their manner (tropes) and their character (éthos) (SVF ii. 806= Plutarch, Stoic Self- Contradictions 1053 cp). | Attempts to deal with this rather obscure passage have tended to operate as if it presents us with two psychological facts which we need to reconcile: (1) that the soul comes into being later than the body;'?” and (2) that children resemble their parents. But these are not two facts in search of reconciliation: they are parts of an argument. The fact that children resemble their parents is evidence that the soul comes about later than the body, and this alone seems to forbid us to explain the resemblance by the material substrate of the child’s soul. If the character of the soul were explicable by the material from which it was made, then similarity between parent and child could be made consistent with the later genesis of the soul, but there is no reason at all why it should be taken to prove it. On the contrary: the fact of such similarity would encourage one to suppose that 137 fll It is not part of the present argument, but we know from the passage of Plutarch immediately ollowing that it is as late as the moment of birth: Stoic Self-Contradictions 1053 cp.

the soul was engendered at conception, or a little later, or at any point in the process whatsoever. I conclude from this that the character of the soul is not straightforwardly to be explained by (or as) some quality of the material rom which it is made. There are, in other words, no psychological continuities between the character of the child and that of its parents at all despite the fact eir respective souls are made out of the same material. how does a child acquire the character of its parents if not from the stuff of its soul? I take it that the answer comes in Chrysippus’ emphasis on the fact that the soul comes to be after the body. if this is supposed to be implied by the fact of psychological resemblance, then it looks very much as if Chrysippus must have thought that the soul derives its character (at least insofar as the chi . character is like that of its parents) from the body. The idea would have to be that there is some link between the appearance of the body and the character of fixed in a way which makes it clear which are in the ascendency. If the character of the soul were to be determined at any point before this (e.g. at conception) then it would be hard to see how a child might be equally disposed to become like its mother or its father. In order for this to be possible, it must take the body for whose character they are jointly responsible as its reference-point; and so it comes about later. It will be objected to my claim (ie, that the soul must derive its character from the body, at least insofar as its character is like that of its parents} that it leaves open the possibility that Chrysippus might have thought that the soul comes about with the body; that its character is determined in parallel with the bedy through the very same process of gestation. Why does Chrysippus insist that psychological resemblance proves that it comes to be afterwards? The answer to this involves the consideration that I alluded to earlier as ruling out the possibility that any Stoic might have thought that the material substrate of the soul carries encoding for its character. This is the fact that the Stoics gave a purely cognitive account of character. It was simply not open to the Stoics to say (as 82 the soul so that, insofar as a child’s body is like that of its parents, its character be like theirs as well. One can see the attractions of this solution for him. For one thing—and I take it that this is a large part of the point behind his argument—it would explain why a child comes to resemble its parents; that is ta say, its parents rather Fir just its father. According to the Stoics, it is from its father s semen that ac grows.'® But, like Aristotle, the Stoics know that a child can resemble its mother instead of (or as well as) its father, and the way they address this, at least as far a physical resemblance is concerned, is to say that the mother also produces seee which plays a role in the process of gestation. When the male seed is receivec into the womb, it is brought under the influence of this female seed: Zeno says that it is ‘taken hold of by another preums, part of the female soul (SVF i. 128). The female seed is not capable of initiating embryonic construction (or o becoming 2 child); however, it does have its own ideas of how one might about building a human being and, much as in Aristotle, it occasionally ‘prevails over the father’s semen (SVF ii, 749): The woman too emits seed: and if the woman’s seed prevails (epikratesei), the offspring resembles the mother; but if the man’s prevails, it resembles the father. A child, then, might look like its father and/or mother. But it is presumably a little time after conception that the influences of their various principles start to operate; and a while longer before the direction of embryonic development is BE ‘Seed’ is carefully defined by the Stoics not in terms of its material substance, we is terms of the capacities or principles (e.g. SVF i. 128) for which the moisture acts as a vehicle; = = is = 128; i. the seed might properly be thought of as a fragment of the father’s sou) or preuma ( 7 1" bs presumably, is the answer to the conflict Diels saw between DL. 7. 159 (where aoa un is said to be ‘unproductive’, agonon) and the fragment that follows (= Aetius 5. iL. 3; see Diels a u 423, 5-8 with n.; cf. also von Arnim’s n. on SVF ii. 749, ga). Note that al of this is only the start “ Stoics adopted a system very closely analogous to that of Aristotle which involves not only the idea of ‘prevailing’, but also of ‘regression’, according to which the characteristics of remoter ancestorsca skip generations to recur further down the line. This works for the Stoics because every u ue h and so, of course, the seed of every individual, contains the principles of his or her ore cars and siblings: to be this person is precisely to be the result of a specific line of couplings, so at the principles of all one’s ancestors are implied in one's own unique blend. So anyway SVF il. 747. 83 Aristotle, or again as Phaedo of Elis said) that character is imprinted in the irrational soul at the same time as the body is formed, because they believed that to have any sort of psychological character at all—in the sense of moral character, in any case'*°—is to have concepts and beliefs. And concepts and beliefs are not inherited or inheritable or part of your physical development as such, When you are born, your soul is a fabula rasa, identical, as far as character is concerned, with everyone else’s (SVF ii. 83). And there is more. Not only is the soul empty of cognitive content at birth, but we also know that there is nothing about its physical structure (e.g. its ‘tension’) that determines its cognitive development—certainly not, for example, anything that would determine whether or not a child became vicious in the first place. The Stoics are adamant that the causes of vice are external (SVF iii, 228-36; esp. e.g. 235: all evi! comes into the soul from outside).'*' The provenance of a child’s soul turns out to be of no assistance in explaining how the child comes to resemble its parents in character. And yet we are told that children resemble their parents in psychological character as well as physical appearance, It seems to me that there is only one way left to explain this. If it is due, not to the soul’s provenance, but to the beliefs it goes on to form, and given that we can explain by straightforward inheritance how a child comes to look like its parents, then it must be the case that the individuals body has an influence on its cognitive development, How this happens is what I now want to investigate. The body does not generally play a conspicuous role in accounts of the Stoic theory of rationality. When we talk about the development of rationality, we focus on the terms of their empiricism most familiar, perhaps, to us: the tabula rasa, the soul’s acquisition of primary concepts, and finally the articulatio n of #8 This is not to say that the Stoics did not recognize ‘natural’ differences between people in respect of their mental abilities. So SVF iii. 136-7: among ‘preferable’ features of the soul are such things as memory and mental acuity, To talk about the distinctive character of an individual is, of course, to talk of his or her vices: there is only one way of being virtuous for the Stoics as for Aristotle (cf. again NE 110635).

these concepts into a cognitive network capable of active rational analysis (SVF ii. 83). Everyone recognizes that the body has some role to play in all of this: it provides the organs of sensation and the means of locomotion: it gives the soul access to the external world on which its cognitive development is based. But I wish to suggest that it does rather more than this too. For a start, it does not merely provide organs of sensation, as if it stands as the soul’s window on external events. The body is itself the first object of the soul’s perception. Furthermore, it is constantly present as an object of perception in all (further) acts of perception. The Epicureans, with whom the Stoics explicitly contrast themselves on this score, do indeed believe that perception implies the impact of for this, but our feet and the whole of our legs. In the same way, we don’t use our feet but our hands when we want to take or give something. Hence the first proof that an animal perceives its whole self is co-perception (syn(ai)sthésis) of the limbs and the functions for which they are provided. 84 the external world on an animal’s constitution so that it is, from the very first, perception of the external world (cf. Cicero, On Ends 1. 30). But the Stoics argue that self-perception is a necessary precondition for perception of the world (cf. Cicero, On Ends 3. 16). In order to register visual information, for example, I have first to perceive my eyes (and perceive them as mine), and I have to keep perceiving them as I perceive the world through them (Seneca, Epistle 121, esp. 5-13). The Stoics have a technical word to refer to perception which involves selfperception in this way (i.e. all perception of the external world): syraisthesis. The importance of this word, and the theory it identifies, is sometimes obscured by commentators who take it mean ‘consciousness’ ot ‘self-awareness’ (concepts for which there is in fact little evidence in ancient thought). '* More recently, A. A. Long has created something of a consensus with his argument that synaisthésis is a straightforward synonym of self-perception, explained by him as the Stoic answer to the modern technical notion of ‘proprioception’ .’” But the Stoics have other, clearer ways of saying “self-perception”. The very formation of synaisthésis already suggests something more like ‘co-perception’, Le. the perception of one thing along with another, and such a meaning is in fact demanded by the principal evidence we have for the word’s use. Consider, for example, the following passage from the Ethical Principles by the Stoic Hierocles (who was probably writing in the second century ap). It comes in the context of a discussion by Hierocles of the importance of self-perception. Every animal, he says, perceives that it has certain limbs; but it also perceives what these limbs are for. He invokes features of ordinary experience to support his point (I. 55—II. 3): We ourselves [perceive] our eyes and ears and the rest. And so, when we want to see something, we strain our eyes towards the visible thing, not our cars. And when we want to hear, we direct our ears, not our eyes. When we wish te walk, we don’t use our hands 142 <Consciousness': e.g. Pembroke (1571) 119; cf. Sorabji (1990) 312; ‘self-awareness’ e.g. Watson (1971) 220. 383 See esp. Bastianini and Long (1992) 396 on i. 38-9 (‘sincnimo’); and wider discussion of the concept of self-perception at 380-90. Cf. Long (1993). 44 “Perception of self being the most obvious: e.g. (ai)stha[njet(ai) [hea]utou at Hierocles I. 36, etcy sensus sui at Cicero, On Ends 3. 16. The Stoics also talked about an animals perception heaut(ön) of its constitution, or of its limbs (or ‘parts’), So in Hierocles: to t{6n)] melr(ôn)j est. (ai)sthanesth(ai) (L 48; cf. 51); cf. Seneca, Epistle 121. 9 omnibus constitutionis suae sensus Finally, Hierocles talks about the ‘apprehension’ an animal has of itself or its limbs/parts: [tés heaut}(on) ... afntiljepseds at V. 30 (cf. L 446). 85 Synaisthesis is introduced here as a proof of the primacy of self-perception— not as a synonym for it. And the proof is precisely that our perception of external objects involves or could not happen witheut perception of our organs as well!" Co-perception, then, is an important concept in the Stoic understanding of perception quite generally. But, as will quickly become apparent, there are implications here that go well beyond the science of perception. In particular, co-perception might also help us to explain the characters that animals display by helping to explain the evaluations they make about the world. Cicero notes at the beginning of his account of Stoic ethics (On Ends 3. 16) that an animal is no sooner born than it is concerned for its own preservation; by extension it is attracted to things necessary for its survival (the Stoic term for this attraction, applied primarily to the animal’s constitution, and derivatively to the things that benefit it, is ofkeidsis, usually translated as ‘appropriation’). Correlatively, an animal hates and avoids things which are harmful to it, But how does it know which things are harmful and which benefñcial®! It must be through co-perception, since the notions of ‘benefit’ and “narm’ are, of course, relative to the individual constitution. The impression a panda receives of bamboo, for example, will be a favourable one, because the external data reporting the bamboo registers along with the panda’s perception of itself as a constitution in need of the sustenance bamboo can provide. There is nothing mysterious about this: a panda is constitutionally adapted to benefit from bamboo, and its being so adapted (its constitution) is the very thing that makes the external data (the appearance, texture, taste, etc. of the bamboo) register as a favourable impression. Or, to put it another way (and from the subjective viewpoint), the panda ¢o-perceives the bamboo as something beneficial. A fox has a different constitution, and bamboo registers less favourably with it than chickens... and so on. It turns out, in fact, that one has a remarkably rich sense of one’s “5 A further demonstration that synaisthésis means ‘co-perception’ rather than ‘self-perception’ can be found in the fact that the word can have as its stated object either the animal’s constitution, or the feature of the external world perceived in relation to it. So at Ethical Principles 11. 37-8, the toad is said to have synaisthésis of a distance: tou poson ek{tetnct} (at) d(ia)stéma syn(at)sthanet(at) (the idea being that it can perceive without calculation whether it is capable of leaping that far). In this case, one would produce nonsense if one translated the word as ‘self-perception’. It is true that, when Hierocles introduces a subsection purporting to show that co-perception continues uninterrupted (UL 55-6), his conclusion is that self-perception continues withcut a break (IV. 44; cf. V. 30). But this only shows that self-perception is of real interest to Hierocles insofar as it is entailed by co-perception: hisdemonstration that co-perception continues shows a fortiori that self-perception does. Here and in what follows I am speaking a bit loosely: animals do not strictly ‘know’ anything at all; nor do they conceptualize the objects of the perception in the rational terms implied by my deseription (eg. SVF ii. 83). On the other hand (as the same fragment makes clear), ‘animal’ Perception is the first stage on the route to the acquisition of human rationality, and we might think of the mental disposition of the animal relative to the object of its (co-)perception as the primitive precursor of rational concepts.

relationship with the external environment famous i This qualification is very ry important; but it by no means undermines th =Perception ofIhe body plays in:shaping the character of someone who las by , me vicious. There is an interesting parallel here with which may help to understand the point. As we saw in Section II, Platotoo denied ridvied infant was in any absolute sense determined by the body in virtue of co-perception. In a need to learn that weasels and example from Hierocles, a chick does not such from the start—though it hawks ate dangerous, but perceives them as ? remains unflustered in the presence of a bull in the elemental In general, it seems that the ‘character’ of anirnals, expressed | + certain things, can be understood terms of their inclinations to pursue and avoid world—which is to say, in terms in terms of the co-perception each has of the viewed through the filter of when of the irapressions that external data make is true, do not have moral it ls, Anima . tution their particular physical consti to make judgements about what characters: they lack the rationality necessary a small step to suppose that the is it is, in a strict sense, ‘good’ or ‘had’, But the good and the bad are about make s being n judgements individual huma perception of what is harmful and based in precisely the same way on their (co-) at last how the moral character see can beneficial to them. In this case, we precisely on their experience of their developed by an individual might be based physical constitution.'** des an opening for a Stoic It might seem that this immediately provi nt features of the individual’s physiogaomy (assuming, anyway, that the releva rd appearance). But before outwa constitution have manifestations in his or her n to be made. J have icatio qualif vital a is hastening to this conclusion, there s allows them to being in human suggested that the development of rationality is, and that vice alone is bad. To this make no difference to the moral extent, differences in physical constitution can that they cannot determine sense the in character individuals develop—at least all. at s viciou or ous virtu es whether a person becom 149 Seneca, Epistle 121. 18 (and cf. perhaps Aelian, 148 Fence there are texts which say that one’s physic] temperament influences one’s characte 1993: 331 for the alternative suggestion that the constitutive of Fi educatie Re that a better answer will be found in the very features of the person’ vicious but pty be o! ‘course in the first place: as, for example, a person might become not ju : er cens Ù Ee were brought up in a society that placed a high value on wealth. This cont hee he a i whole story, or even the greatest part of it. For one thing it fails to r; er is a matter of the ais m acuparibus = their parents; for another, it leaves us to trie thee es 10 es greedier than another. In general, i their insistence thet charact and we can now see why this would not contradict ppus at SVF fi. 950-1 (= Cicero, rational beliefs one holds. So e.g. Chrysi heppens and one grows up with right reason, the Stoi i room at all for the development of vice. We | should ul expect in thi everyone would grow up with an identical moral dere namely a virtum er. But suppose now that something goes wrong: th: i 1 i j to, for for ıexample, ‚aa bad education which ae ely leads h ade him him or oFher awe fen the insight that virtue alone is good (seeain again SVF ci SV iii. 228-36 as as cited oben), What À fills the void?? What does such a person start to think is ‘good’? Presumabl an et oe things that were the object of appropriation in his or her pre-rational (eh general, what such people in their pre-rational lives co-perceived ik as en and beneficial with respect to their physical constitution will =me iat —— = ofLes ‘evil’ and ‘good’ respectively. This will se range of moral characters that manifest themsel among groups aa people whose educational influences have been "broadly . of them have been diverted from achievi standing3 of of themselves them as essentially rati tional creaturesnde (that i See So direct their activities accordin;g to a Fat rational understanding ‘ofthe of the wider wi courmos), This means that 1their self-understanding elf anding is retarded in a certain Bain way: way alth ahai has become rational, it remains fixated on the local needs ofthe tical cons i ution, not the capacities engendered in the organism by the very presence son. But, aspawn physical constitutions differ from each other, so then h individuals start to form about what is good and b: perceive ; 1themselves as s rather weak might become more rat © timid, timid, beee they al i © Lapa more things in the world as potentially harmful; edlen (one Em of the example in Plato, Timaeus 86¢ of a person suffering from a cial aad harmful for them as what point). On the Characteristics of Animals 9. 40 for a similar character were (supposed to b case ght Sois: In the ‘normal’ case, ie. the case in which eh that rationality, when it develops, respectively, ‘good’ and ‘bad’. The trouble is something that enters into their is itself part of the nature of human beings and part of their nature, but (once a co-perception of the world. And it is not just e human beings are supposed matur such, As e. they have it) their defining featur elves as rational beings—even if to care above all for the preservation of thems s. What is co-perceived as tution this is at the expense of their physical consti life: and what is conducive al ration a to cive condu ‘good’ ought to be just what is then, are supposed, by the very to a rational life is virtuous action. Humans, been outlining, to develop the have I ion process of co-perception and appropriat 147 Hierocles LIL. 40-3; the example recurs at ences on the activity ; of reason. . But he also thought thatdthe bodyoe b ini factor in the development of moral ch aracter if and whenease reas wae undermined by,offor example, le, a bad educati ation, Something i similar, ‘inn Lemen I isthe redescribe what is constitutionally benefi belief that virtue, and virtue alone, is good, 87 On Fate 7-9; cf. Sedley (de only) à ecKeine a person from reaching the ‘natural’ erenis atmosphere may have a direct effect on the preuma do soe ne oaly be sue ithas allthe years until the development of his or her rationality to ii 76), Ite wane aoeKast years (SVF ii. 83), and perhaps 14 (when rationality is ‘perfected’: SVE rather lees Bee e between a person and his or her self-perception as a physical tur i owever, since this is something that begins from the moment of birth, “ Panaetins fr. 76 van Straaten = Proclus, one’s soul and thereby on one’s character); for a literal interpretation of Plato’s claim that On the Timaeus i, 162. 11--15 Diehl (arguing of the environment); Seneca, On Anger 2. 19 natare the to related y causall be intelligence might ky, Panofsky, and Saxl 1964: 56). Klibans cf. faci, torum elemen (et morum variztates mixtura

lustful if they see the world super-abundance of seed) might become rather to sexual activity... and ation inclin al tution consti through the filter of their As we saw above, the earlier Stoics thought that appropriation was directed towards a single goal: the preservation of one’s nature. This is the object of appropriation for every animal, and it is the object of appropriation from the moment of birth. But in the case of human beings, most strikingly, this has the effect (as we have seen above) that the particular objects of our striving change as our nature, or more precisely our perception of our nature, changes This is most striking in the case of human beings because the development of rational between the ages of 7 and 14 constitutes the most radical change undergone by any animal in the course of its natural development. The result of it ought to be that we come to perceive ourselves as rational beings and to value only what reserves our rational natures—whatever the consequences for our bodies. It is true that as a matter of fact most adults care as much about their physical well being as any animal or child; but this does not undermine the position that appropriation is directed towards the preservation of one’s nature: it only sho that they have somehow become confused about their nature.!”? Aa Posidonius, it seems, did not agree with this analysis of appropriation. From the first, he argued, appropriation is directed not towards nature, under that description, but towards pleasure and victory (158, 160, 161 EK). Assumin: that Posidonius based his understanding of appropriation on co-perception, 7 the | earlier Stoics did, we can rephrase this by saying that, accordin | to Posidonius, an animal does not co-perceive things as beneficial to its nature as a certain kind of animal; rather, it co-perceives them as satisfying specific constitutional needs—the need to make good some lack, for exam le: or t secure itself against a potentially hostile environment. | _ ° The difference in Posidonius’ analysis of appropriation is no doubt to be explained in part by reflection on the grammar of impulse and appropriation the proper terms in which to describe their objects;'?? but it has advantages for his ethics more widely as well. For one important consequence of replacin appropriation towards ‘nature’ with appropriative inclinations towards victo ; and pleasure in explaining the behaviour of young children is that one does not have to think of the changes that come about with the development of reason in 88 50 on. 150 to stray from the path of virtue; Ones constitution, then, cannot cause one from it, it will, through the but if external influences nudge one away g role in shaping the vic.ous mechanism of (co-)perception, play the leadin y featured in the thought ognom physi if And moral character one forms instead. g; this is the only way that of earlier Stoicism, this must have been its openin between the characters of a Stoic could have supposed that differences spondence with differences corre r regula of sort individuals might exhibit any themselves recognize the in their physical attributes, But did the early Stoics answer will be that they My y? ognom opening and commit themselves to physi nating reasons. But in order to see did not—and {as it will turn out) for illumi er the different approach to what these reasons were, it is first necessary to consid told, did make use of are we who, psychology adopted by the one Stoic nc). -51 (c.135 s Rhode of onius physiognomical data: Posid (ii) Posidonius The extent of Posidonius’ disagreement with earlier Stoics has become a matter y dependent for our evidence, has of hot debate. Galen, on whom we are heavil rate any difference there right an axe of his own to grind and a motive to exagge to which his testimony can degree the over on have been, leading to a real questi Galen has me {and unkind) view that be trusted.'?! But unless one takes the extre existence of any theoretical not just etaggerated, but actually fabricated the at all, there is one respect essors predec his and difference between Posidonius only, or the simplest way to in which it seems that he must have differed—the ed to him. On my view, ascrib ly explain all other differences rightly or wrong on’. priati ‘appro of is analys his he must have differed in 150 Another example: anger is defined as the desire to punish perceited offenders (SVF iti. 395, 886 blood about the heart (SVF ii, 878: 3. 416; cf. 2. 398), It is also understood to be a boiling of the ns of anger). So what makes one occasio during chest the in activity of ésis and 900 for our synaisth ? (A sage, as it happens, never thinks anger non-sage more inclined to anger than another his or her blood is, for example, hotter--in any that fact appropriate.) [ am suggesting that it is the ly more inclined to initiate themselves the utional case, readier to boi. This makes such folk constit wish for their enemies. For this reason, they value might aded cool-he the even which ment punish phrase punishing. It might be thought tempting te enacting the punishment: they desire to do the utions by different ed in different constit these examples more simply in terms of the pleasure produc activities. (Cf. perhaps Cicero, On Fate 8; SE, PH 1. 80; also Aristides Quintilianus, On Music 2. 8, 67. e in could say that the coward takes greater pleasur 15-68. 13 Winnington-Ingram.) In this case, we forth. so and , activity l physica e extrem and of danger comfort, the pugilist in the adrenalin rush meaning of ‘pleasure’ in the context of ancient However, this would, according to the normal ed that even fools act to preserve their nature suppos Stoics the fact In ing. mislead philosophy, be wrong). Only incidentally is it ever pleasure is g tandin unders their (albeit as they understand it they seek. (2000) 93-32, 255-60: Gill (1998); Sorabji 151 See especially Kidd (1971); Cooper (1998); most extreme in denying significant doctrinal Tieleman (2003), esp. chs. 4-5. Of these, Tieleman is divergence between Posidonius and Chrysippus. 89 terms of their modification.’ Rather, Posidonius is able to suggest that what happens at this point is that the creature (now an adult human) acquires an additional object of appropriation, namely the good, In virtue of their 152 se N oh an that Tam not ascribing to Chrysippus a view that one ever loses concern for one’s own prea, constiation or self preservation, in a way sometimes suggested by his critics, as in Cicero, == , a (Galen 5 daim that Chrysippean oikeidsis is only directed towards the good one cheeae vt eie distortion: cf. Tieleman 2003: 221-2.) But what does happen, ideally, is that Pope een constitution is for, what | constitutes its ges and enlarges one’s view of what thee physical physi itution itutes ofthe =a considered, for example, that although the term ‘natural’ is an important part tote sion Sescription of the objects of our appropriation, it is wrong to think of it as relevant Soie ea description. One desires things under the description of them as ‘pleasurable’, or wh te as 'natural’—even though it might turn out that it is their being natural that explains 7 = seem pleasurable or noble to us. = En ey can be modified, , though, viz byy habituati habituation (e.g. F165. £72— indirectly by reason (among other things). 289 HO Tand ence

not only in terms of rationality, adult human beings co-perceive external objects l constitution, but physica their to bring the pleasure and security that they might good (what is wider the satisfy will them of pursuit their r also in terms of whethe y concerned, properl is ity good for the cosmos as a whole) with which rational come to see should being l rationa a that Like the earlier Stoics, Posidonius thinks despite the ‘best’ g, speakin vely objecti is, what do to him that his nature enables of our perception of the world, a real bias in the information i of rational creatures, as the basis for decision-making. This ie why hg no need to appeal to external pressures, by which the earlier Stoics (as we 90 subjective inclinations of his constitution, and that to live consistently in this unlike them, he thinks way is the only path to happiness (ef. F185-7 EK); but a sage, will retain his even this, doing in s succeed who e that even someon and victory, his or her e pleasur or her original senses of appropriation towards terms. these co-perception of the world in torily (as he thinks) All this enables Posidonius to explain much more satisfac lar the vexed particu in lives, some key phenomena of our mature ethical is supposed to vice when vicious become ently consist question of why people so wete right, Stoics earlier be the one evil recognized by a rational nature. If the than to vice into people corrupt to said Posidonius, it ought to be no easier convince them that pain is desirable (cf. £35, saw) had to explain the birth of vice in an individual: the ‘affective movements’ the aa se to action born of our co-perception of the world exert ie an internal pe pressure u on rea son to accede in valuing the objects j of their i But there is more to be said than this. If it turns out that there is an internal cause of our deviation from the path to virtue, that very cause also explains the character of our deviation. As in general co-perception means that we see certain things in the world as pleasurable and glorious, so the particular things in which individuals glory or take pleasure will depend on their particular way of co-perceiving the world—the variable factor being their self-perception, the character of their physical constitution as they perceive it. In this case, it u no surprise if Posidonius thought that the environment could affect character, Justae it affects our constitution (F169. 88-96; tr. Kidd 1999: 234, 169, 1-77 EK). Posidonius’ uous appropriation explanation for this simply involves appeal to our contin in these terms as world the of eption towards pleasure and victory, our co-perc continuously, always, we that fact the For ‘good’. is well as in terms of what the world rceive perceive the world in these terms just because we (co-)pe tion inclina an have thereby we that through a particular constitution, the fact g tandin unders our er whatev nable inelimi is which victory towards pleasure and ve ‘affecti calls ius Posidon of what is good and right, leads to our having what ng like inclinations movements’ (pathetikai kinéseis: F169, 101-3 EK)— somethi or child would animal an which in way the just namely way, in a certain to act h, on my act in pursuit of pleasure or victory.'** It is important that althoug account, these movements must be strictly nort-rational ( animals and children ally es, specific have them toc), they are nevertheless based in cognitive process the 91 to think of them as process of (co-)perception. It would be wrong, then, constitute distinct distinct, ir-rational powers, as Galen implies. Nor do they agreed with earlier ius Posidon that know we : humans adult in impulse sources of remained squarely extent this to Stoics that impulse required rational assent and ent part in the hands of reason.!® But they do form an important and persist al pull’: so Galen at F 169. 80 EK 155 posidonjus seems himself to have talked of an ‘emotion animos sed adtrahentem). CH. tantum n irritanter (non 16-17 F170. at Seneca also sut (pathetiké holke), is ‘led (agomenor) by the one that virtue of n definitio the of part it makes F186, where Posidonius this enables Posidonius all that way, the by believe, can One irrational part of the soul in nothing’. ‘emotional pulis’ into account when to refine the Stoic account of emotions (ie. by taking these seduced by Galen into thinking that explaining their origin, object, and intensity} without being given by Chrysippus. To this extent, emotions the of account the with broke actually us Posidoni that there is, in effect, no difference however, argues, o 4-5—wh chs. (2003), I agree with Tieleman between them at all. 37, 186 OF F154 EK with Tieleman (2003) 278-83; also Cooper (1998), esp. 87-8 with note V 5, 21, 320. 23-8 De Lacy against emendations at F169. 77-82 defending the received text of PHP movements could, after all, initiate BK which force it to provide the only evidence that affective judgement in some cases ofemotion, ‘denied us Posidoni that suggests 104-5 (2000) Sorabji impulse. And environment contributes to considerable differences in human character with regard to cowardice, daring, love of pleasure or toil; the grounds for this are that the affective movements of the soul follow always the physical state, which is altered in no small degree from the temperature in the environment. For he [sc. Posidonius] makes the point too that even the blood in animals differs in warmth and coldness, thickness and thinness, mn a considerable number ofother different ways, a topic which Aristotle developed at ength. If all this is right, then the psychology of Posidonius seems to be a little closer to that of Phaedo of Elis (at least as described in Section I) than to that of Plato It is not a thoroughgoing reductionism, since reason remains free; but the appropriative inclinations towards pleasure and victory are determined by the body. And in the passage immediately preceding that just quoted, we learn that Posidonius produced evidence to support his revised psychology— evidence that is, from physiognomy (F169. 82-8 EK, tr. Kidd): | Posidonius said that, while a creature’s impul: i pulse was sometimes born in the jjudgmental = of the rational faculty, most often it occurs in the movement of the emotional culty (kinései tou pathétikou). Posidonius reasonably attaches to this discussion physiognomical evidence (ta kata ten physiognomonian phainomena): all broad-chested but saw judgement as being present in the standard cases’; but the text on which he i kie pear (Faa, 9-12 EK) only says that emotions are not, and do not supervene on, ennn flee jd wu Fon thought This is quite right, since emotions as such (and as opposed to merely Fe u m nt} are ra re(and always) determined by the presence of an affective movement. en animals and children are subject to emotions without judgement—cf. Sorabji (2000) — = prove that the same will be true of adult humans, Aristotle at least supposed that non rational parts of the soul which govern animal behaviour cannot dictate impulse in th Presence of deliberate choice—hence the problem, for him, of akrasia. J ° 157 plien ard 153 EK. One might—though Tam suggesting there is no need to—be suspicious of the Ka ot on that this represents Posidonius fairly, not least because of Galen’s own sympathy for the of reductionist psychology implied here (cf. Section V).

-hipped and more spirited by nature, the broad and warmer creatures and humans are t... colder, more cowardly, And environmen (iii) Physiognomical Conclusion us’ psychology. But note that, on Physiognomy, then, supports Posidoni my een different nding of the differences betw account, Posidonius’ understa tend to align with e differences will characters, and the reasons why thes the same as that of the earlier Stoics. not differences in physical appearance, is f that an individual has an itted to the belie Posidonius, it turns out, is comm it would ed in the body. (This for him is what root r acte char onal innate, non-rati done, that our senses of on he might have mean to say, as on My reconstructi the are based on our co-perception of ory vict and appropriation to pleasure cier, chara te’ ‘inna an of on noti gether the body.) The earlier Stoics rejected alto individual’s physiology. All they were the in d ribe insc was that one e let alon the n of the body will often help explain prepared to say was that co-perceptio ire acqu we ng, ingi upbr d upte g to a corr vicious character we acquire—if, owin the question on them een betw rence diffe the any vicious character at all, In fact seen akin to the difference we have already of character-formation is very closely er; and both otle (and Phaedo, for that matt between Plato and Aristotle. Arist this r was innate, and Aristotle’s thinking like Posidonius) thought that characte (and hand r othe the on , iognomy. Plato is at the root of his exploration of phys in body could influence the direction the tnat eved beli s) Stoic er like the earli in vice of ce not account for the appearan which the vice developed, but could citly expli otle Arist and us that Posidoni the first place. Is it coincidence, then, e of their respective psychologies, whil nce eque cons a as ’ nomy siog accept ‘phy ? cism Stoi er ence for Plato and for earli the word is absent from all of our evid d the ability to say what a person woul , Plato ng ussi disc n As I observed whe ingstart bad a not is uous virt be to not tend to be like if he or she happened from e a living by predicting character point for someone wishing to mak have fact in do we And all, ous, after appearance: most people are not virtu ictions pred e mak to sage the of ty abili the evidence that the Stoics recognized on the basis of physical appearance about how a person might turn out (SVF iii. 716):'° nature weldisplaying through their appearance à The sage will love young men, those s ten pros eidou tou dia ön inont empha e (tén disposed towards the acquisition of virtu in On Lives 1, and ippus Chrys lic, Repub his in says aretén euphuian)—as Zeno Apollodorus in his Ethics. which the sage is said kind from those (discussed earlier} in 158 This fragment, then, is different in ht that the term euphuia itself thoug be might It vices. and s to be able to perceive actual virtue a rational adult, But the pre-rational analogue to virtue in represents 4 quality—perhaps the iii. 650 suggests that the SVF at love of tion defini Stoic similarity of language with that used in the 14) to be intellectually least at (ie. h are themselves old enoug children the Stoics have in mind here ): ‘Anyone who is intelligent will adults to ing referr inly (certa 650 iii. mature, Cf. perhaps also SVF and a love of learning who displays both a good nature rule, at times, and live with a king 120-3. (1980) lo Ioppo term the on Cf ian). (kai eiphuian emphainonti kai philomathe 93 The idea, , I take it, is | that the sage would tell from the physical appe en = particular vices towards which the youth will tend (given that he is ede of such a character or of such intensity as to prevent a realistic hope of abilitation. ” {we shall see something very similar among Neoplatonists u er oo ) and again, this fragment could be adduced in support of the position : s were physiognomists. Yet tae more such evid i (and we have seen quite t a lot by now, albeiti drawn in fact ‘hom from rather rather digerent diff contexts) the more striking is the absence of the word ‘physiognomy’ and the comparison with Plato, and the contrast with Phaedo, Aristotle and Posidonius y now suggest a reason ! why we should never have expected ed itt in the first place.x Seems mt Physio suomy (etymoiogically speaking, the Knowledge of e) is restricted to cases which involve the discussi f t moral ' character—something whose existence is denied or innate i iedby Plat Pl o and De Seien el is about the way someone is ont mal at tl and not, as we might have been incli 8 pan Fr in een inclined to suppose, whatever character is Phaedo well knew,vs a person’s character, or at any rate behaviour, mi change “someting which the handbooks of practical physiognomy have a unt, which is why they do indeed encourage us to look i as gait and gesture which might be thought to develop : inin the the v wake vee ofof, and aDaears for, the development of character. The fact that Pinedo, , osidonius accept ‘physiognomy’ does not h ictions of character made on the basis of thei r psychological i model ito would b dictions made on th i of successful (that is, , successful more often) th icti the basis Plato’s psychology,A or that of the Stoic s. Nevertheless, werthe it mak, vomplete sense from a theoretical point of view that the word iss restricted restricted tto “hex eyes thei isnot not so much the What is philosophicallyy interesting about physiognomy i H the appearance of the body, bat question of whether the state of soul can affect if fe body hequestion of howsand how closely, the individual soul is bound up with the indi ody at a’ ental level. In this case, of co hat is intrigui is precisely the | possibility that there mi ght be some© correlation | correlationbetween betw: varybates characteristics {the size of limbs and the colour of the eyes) 1 the ‘or aracter of the soul, It is the elucidati i which distinguishes physiognomy properly speaking ® of such correlations 150 . are hon Soles debated among themselves how quickiy to give up: SVF iii. 366 asks whether sai we should understand the question whether a abilizy to or made as well (by which, presumably, present in everyone), and suggests a division equally Is society of prion th who between Stoics and ‘good breeding’ (eugeneia) are disposition’ eg who eas virtue could be taught as well and towards steps innate, first these t h a t À ought . pq. those 160This hni then, be compared with cases of_Physiognomy might, in the eS a ased on si omic inference that do not belong to the realm of temperament’: see Ghersetti below, p. 304. beb on signs

V. MEDICAL WRITERS the relationship between the physical and the psychological is explored: certain Against Celsus 1. 33; Anon. Medical writers of works on physiognomy: Lows (fragments: Origen, es of physiognöman- in Occurrenc s. Adamantiu 133); 117, 89, Lat. Physisgnomy 1, 2, 12, 48, 80-1, 5 and 6; fragment ap. Ps.-Galen, medical writings: Hippocrates, chapter titles to Epidemics 2. logical activity, conversely, can have an effect at the physiological level!“ The treatise On the Sacred Disease famously provides a physiological basis for psychological activity quite generally; and it is not the only work to suggest that aspects of psychological character are linked to particular physiological 94 Prognostica de decubitu xix. 530. 5-6, 9, 10 Kühn; Galen, physical pathologies have psychological symptoms, for example;’® and psycho- On Temperaments i. 624, 3 Kühn; QAM 54. VI, 7-8 (53) 20, 55. 8, 63. 1-2 Müller; PHP 5. 5. 22. 2; cf. Commentary on Hippocrates’ Epidemies on Chersetti 1999a (=426. 37-8 Wenkebach and Pfaff) (firâsa). De Lacy. Palladius, Commentary Galen’s Book on Therapy Hippocrates’ Epidemics ii, 124. 32, 195, 11 Dietz. Stephanus, Commentary on derived from medical ‘For Glaucon’ i, 252. 2 Dietz. Occurrences of physiognömon- in contexts possibly s of phystogndmon- in contexts sources: SE, PH 1. 85; Suda s.v.diathesis (6 556). Other occurrence 3; Fulgentius, Contents of Virgil 2. 135. 21. 1. Stromata Clement, medicine; and omy linking physiogn 84. 10-11 Helm. (i) The Hippocratic Corpus the notion of The meagre evidence we have might be taken to suggest that as carly as the Athens in familiar bly reasona was art l practica a as physiognomy fifth century ac;'® but in the course of this study, I have argued that it became ments an object of theoretical interest only in the fourth century, when develop the between ship relation the in interest new a to led ogy psychol hical in philosop itself present not did gnomy physio soul and the body. (Even then, I have argued, Rather, as a given, a system of data to which philosophers felt they had to argue. certain a of ence consequ a as d emerge gnomy physio of ity possibil cal the theoreti its which to ng accordi soul the kind of view of the soul: specifically, a view of of r characte natural the with way ntal fundame a in ‘natural’ character was linked the body.) of the matter. There is, however, a case to be made for a slightly different view gnomy in a It is, in fact, commonly assumed that the emergence of physio context into the that and argued, have I than older rather is context cal theoreti actually which it emerged was not (as such) philosophical, but medical.’ Galen ‘physiog word the rates; Hippoc by d invente was gnomy tells us that physio nomy’ appears to have its first attestations in Hippocratic literature; and there out the is a wealth of apparently ‘physiognomical’ material scattered through angles t differen us numero from ns occasio us numero On Corpus. Hippocratic 161 See again Introduction n. 4. . . Evans (1969) 17-18; 12 eg, Foerster (1893} i, pp. xiv £; Misener (1924) 104; Megow (1963) 220; 169-70; Degkwitz 97-9, (1994) Barton 13-19; (1993) Raina 506; (1992} André (1981) 9, 16; Stok realistic (because the most non(19966) 25-7; Böck (2000) 62, Vogt (1999) 108-13 gives the most medicine. committal) overview of the relationship between physiognomy and Waters, 188 Ror the claim see QAM 57. 11-13. 6 Müller. For the highly selective reading of Airs, 27; Singer Places [AWP] that follaws, see esp. Lloyd (1988) 23, 28-31 {and 1992: 136). cf. Flashar 1966: titles to Epidemics 164 “repeats because none of the (three) occurrences of the word (in chapter 9, 10 Kühn) are 2.5 and 6, and in a Hippocratic fragment ap. Ps.-Galen, de decubitu xix. 530. 5-6, beyond the suspicion of inauthenticity. See further below. 355 og Internal Affections 48; Aphorisms 2. 33, 6. 23. Cf. esp. Di Benedetto (1986) 35-69; (2000), esp. 30-1 with notes. 95 Gundert dispositions.” On occasion, Hippocratic texts even tell us how aspects ofa person's character cen be discerned from rather specific physical traits.’ There is, however, a significant gap between the material identified as phys:cgnomical’ in the Hippocratic Corpus and the philosophical discussions af physiognomy which are supposed to follow in its wake. For one thing, this material is largely premised on an interest in correlations between rental activity and physiology, and it is not at all clear that this is the same thing as, or even very similer to, the physiognomist’s investigation of links between psychological charaster and outward appearance. It is casy to underestimate the importance of this point, because one might think that it is precisely the suggestion that there is a link between mental activity and physiology that represents the theoretical breakthrough for physiognomy. The trouble is that this would be so—that it would De physiognomy for which it represented a theoretical breakthrough—only if the Hippocratic texts also offered a general explanation of physical appearance in terms of a person’s underlying physiology. In this case, of course, we should be able to see that something linked appearance and mental state (namely the underlying physiology of the individual). This is just the sort of explanation we find later on in Galen and after him (as Antonella Ghersetti shows in Ch. 7 of this volume) in the Arabic tradition. But the truth is that there is nothing of the !* sort in the Hippocratic corpus. There are, to be sure, occasions when (quite apart from claims about the person’s underlying physiology) outward appearance is linked to psychological characteristics; but these cases, now, are too isolated to amount to anything 166 og. Epidemics 2, 4, 4; 6. 5. 5, Fuller lists at Gundert (2000) 29-30 with notes. * According to The Sacred Disease, thought (phromésis) is a product of air, whose primary seat i the body is the brain (esp. 19); £ the brain is diseased, then Psychological functions ts rm on this eg. Hitffmeier 1961, esp. 51-61). In Regimen 1, the soul is a blend of fire and water (1. 7, 25) which determines both how ‘feminine’ or ‘masculine’ the soul is {the moister it is, the more feminine: 1. 27-8, 34}, and how intelligent (1. 35). A looser correlation between racial characteristics of body and soul is suggested in the second part of AWP—at least, the environment is a factor in shaping each; though the author of AWP does not invoke a mechanism connecting psychological and Physiological changes (cf. Lloyd 1988: 28-9), Further on the relationship between the ‘psychological’ and the physical in the Hippocratic corpus (though emphasizing that the categories are not clearly distinct for its authors): Miller (1948); Pigeaud (1981), (1987); Gundert (2000). ken. eg. Epidemics 2. > 16; 2. 6. 1 (unblinking eyes: the same word, askardamyktos, occurs in Ad. 7 ne 6.14 (but only ifone takes it, as some in antiquity did, that the word katakorea, applied here to 5 ste est, means ‘hairy’ rather than ‘congested’ or suchlike: see Galen, Commentary on Epiderics 1 and 4, 11-14. Wenkebach end Pfaff); 6. 4. 19 (cf. prominent veins at Ps.-Aristotle 812729); 6. 7. 6. humo For Galen (as for the Arabic tradition which relies on him) it is the (temperament of) the Fear pa is the denominator common to physi(ologi)cal and psychological types; but no such on is made even in those Hippocratic works which employ humoral theory. See on this eg. Ibansky, Panofsky, and Sa:d (1964) 8-15; Elashar (1966) 21-46 (esp. 35).

interest. They are neither one might think of as indicative of ‘physiognomical’ Of course, it would be 7° cant. signifi ically empirically systematic, nor theoret other contexts (for from ’ ognomy ‘physi with nted acquai ne possible for someo s) to use the word Zopyru as example the systematic claims of practitioners such n to resemble happe they as r insofa ne medici in describing aspects of Hippocratic the word when ning happe is what ely precis be might it. Indeed, this (2. 5 and ics Epidem the of rs ‘physiognomy’ is used in the titles of two chapte patient’s the which in sis diagno of es exampl n 2, 6) which happen to contai the best be also .’ It might mental state can be deduced from their appearance 5-10 530. xix. tu decubi de len, way to explain a notorious passage in Ps.-Ga of those ment judge ‘the that said having as d quote is Kühn, where Hippocrates > ië) ognomy (physiogno<m on who practise medicine but have no share in physi e becaus lt, difficu is e passag The h’.'”? rambles in the dark getting old and sluggis But of course none of this makes Hippocrates (or any Hippocratic writer) prsysiognomist ‚ any more than members of the crowd outside the at Sy are made physiognomists by Demosthenes’ remark that ey physiognomize’ the expressions of the jurors as they 1 ee Against Aristogeiton 98. 4). yee There is an obvious objection to all this: that we philosophical standards to judge the physiognomy of veces medal texts Te original claim was only that physiognomy took its rise in the medical tradition not that it became systematized there. Instead of expecting a fully-fled d Hippocratic physiognomy (the objection would continue), one should Father be looking to see how Aristotelian physiognomy was only made possible by th plonsering pork a Hippocratic writers on human. nature. ° e trouble with even such a moderate view, however, i it implies ei that medical writers established an interest in the dende [emere example) the body and the character of a person, and that this interest was taken up and pursued by philosophers; or at the very least that medical writers somehow established a way of looking at human rature such that philosophers following in their path naturally turned to consider the relationship been body and character in a certain light. But neither of these claims can be justified on the basis of our evidence. In the first place, it seems fairly clear that philosophy and medicine, particularly in their approach to the analysis of human nature, developed as theoretical disciplines in parallel with each othe from an intellectual background which, in crucial respects, they shared in common. Alcmeon, Diogenes of Apollonia, Empedocles “Anaxagoras: the same figures stand behind medical and philosophical thought about the material world in general and the nature of man in particular.'”° The importan | of this is that it is hasty to suppose that all similerities between medicine nd philosophy in their theories of human nature must be the result of direct influence between the two later traditions. In particular, there is nothing very 96 ce for Hippocrates’ Ps.-Galen takes this reference to ‘physiognomy’ as eviden is going on if we what astrology, but we could perhaps understand interest in did as a matter of fact link imagine that early physiognomical practitioners chart. a person’s character and appearance with his or her astrological word ‘physiognomy’ In this case, ‘Hippocrates’ could have used the in a loose in the effects of the environment, sense as a kind of shorthand for his interest induding the heavens, on the psycho-physical nature of an individual.’”* gnomical’ inference in Hippocrates are either 170 Particular examples of supposedly ‘physio 5. 1 (examples as in n. 168 above; also Epidemics 2. impressionistic (as Regimen. 1) or else sporadic a physiognomy might be based, see which on ism mechan cal theoreti a of lack the For 6). and 2. 5. the question of how the physical characteristics of again n. 169 above. Hippocratic writers who tackle resemblance to its parents through the theory of child’s a of terms in them an individual arise explain is drawn from all parts of their respective parents the from ‘pangenesis’ (according to which seed (On Seed; cf. AWP 14, Sacred Disease 55 theory this bodies}. However, none of the texts which discuss l 162 n. 46) links the child’s particular physica possibly Regimen 1. 7; cf. Dean-Jones 1994: state. ogical psychol ar particul its with ristics characte (Littré’s apparatus says ‘puotoyvupia: aut 171 The title to Epidemics 2. 5 is Physiognömonië (and in the apparatus: ‘uLOYv@ovıRög on nömonik Physiog is 6 2. to that quo voywopodivy L'} form, but it nömonie has authentic dialectal C+turoyvupovirdg Beutepog D'). The word Physiog original. In particular, Galen shows no indication is nevertheless possible to doubt that the titles are have appealed to them in support of his claim in of knowing them. One might think that he could but he mentions them neither in his commentaries gnomy; physio d invente ates Hippocr that QAM gnomy. on the Epidemics, nor in his discussions of physio suspicion {Diels 1910; Heeg 1911); though 172 Again, the authenticity of the quotation is open ta not an invention of Ps.-Galen, at least, ably presum is it and it, there is no urgent reason to doubt rate the importance of astrology in demonst to ing attempt is since is not obviously to his point. He that enthusiasm for ‘physiognomy’ further argue to medicine, and to make use of this quotation has 1916; Wellmann 1913}. Schéne (cf. too gy astrolo for asm enthusi implies be DL 2, 45 = Aristotle fr. 32 Rose”, 13 An early example of predictive physiognomy might end for Socrates. Later on, Loxus violent a ed predict things’ other ‘among of a Syrian magus who s fate as well as character person’ a reveal could gnomy physio that and Polemon both thought (cf, Anon. Lat. 133). climate, and astrology, cf, Ptolemy; 17% Bor a later way of making the link between appearance, the word ‘physiognomy in the loose a parallel use of ius, Contents of Virgil 84. 10-11 Helm: his sense I am suggesting for Hippocrates here, cf Fulgent and medicine’ (omnis physiognomonicus gnomy physio about ‘all is 2 s Georgie daim that Virgil, type of men bred by Italian soil (cf. esp. the of tion descrip s Virgil’ on et medicinalis) must be based Tetrabiblos 1. 2; cf. also Section VI n. 227, For iad—— 97 rag 7 medicine), Finally, for an interest in the (psycho-)physical effects moment in the Hippocratic Corpus, see eg. Nature of Man and A not, coincidental’ being an important text in Galen’s representation of re p ysiognomist). It should, however, be noted that in all this I am taking the reference to p oman ;n the Ps.-Galenic fragment much more seriously than most scholars think it can be en ae . 7 höne already suggested that it is meant as a simple equivalent for physiologie, the ra ne (1910: 415; see now esp. van der Eijk 2000-1: ii. 136 on Diocles fr. 64. 9-16, with his u . Suza: a position s would, of course, 2;be entirely consiistent with i my overall u eenent; but it meerie 4ma that it goes too far in suggesting a meaning for the word ‘physiognomy’ here that is y different from, but actually cannot be related to any other attested occurrence of it 175 . . : . ‘potes etn DE hi — Medicine deplores the influence of philosophical thereby gives divert online cine, wl ot hed. a long and independent traditi i ind ep radition of practice behind it—but YE e importance of Presocratic thought in medical i recognition by medics ich proem Longs (1909) ce themselves. . Cf.CE variously usly Celsus , On Medicine, ij7-8; Edelstein Sa (1952); sen, gee cos p. 13, cae 16, and 19 GE for the philosop. philosophical deb ebt owed by the author of Ancient’ ret à H| H 7 er Eijk Eijl (1999)(esp. esp 9-11for thethe relationship i i between mediici lo mnt Fc medical histeriography, including the passage of Celsus cited) Ee anil “opment ant of philosophy Ain) na«ine and phy is argued diin Orellii (1998;H cf. 138 “Affinitä ‘Affinität statt

phers develop surprising at all in the fact that medical writers and phileso similarities. formal have which bedy and mind accounts of the interaction of of the medical aware were phers philoso that deny to course, of not, This is influence as medical tradition and influenced by it; it is only to say that such must have been, at nature human about t though phical theory had on philoso philosophers which under best, incidental to the development of the conditions became interested in physiognomy. directly into the As to the possibility that medicine fed physiognomy n about the questio a as form nic philosophical tradition (perhaps in embryo number of a are there states): mental and l physica n relationship betwee e that evidenc no is there indications that this cannot be right. Most importantly, l physica and s pattern ural behavio n betwee ce it was a perceived corresponden and soul. On the body of s account their develop to phers philoso led traits which emerged on the back contrary, this seems to have been a secondary question that ion and activity motivat human explain to ped develo s of psychological theorie physiognomy in interest more generally. In any case, we have seen that an s it can be emerge it where that and was not systematic among philosophers, particular their for) tion motiva the (not of uence conseq understood as the do have we where that, psychological theory. The next point to consider is not for is it tion, inspira for texts l medica to evidence for philosophers looking are looking. they tha: soul and body n betwee nship relatio the about inspiration sensitive to medical Indeed, we have seen that Plato, who was both unusually s a distinctively toward work s’ Socrate ing develop in e thought, and decisiv avoids, or even y carefull , philosophical tradition of psychological analysis Finally, all our ® work.'” own his in gnomy subverts, any suggestion of physio on) comes traditi phical philoso the in (i.e. gnomy physio ical theoret for evidence different ively distinct already is that in the context of a model of human nature to rely on seems it ar, particul In n. traditio atic Hippocr the in from any found soul as the locus of a systematic distinction between body on the one hand and to be related to turn in seems which tion distinc moral agency on the other—a was in any case and er, charact moral human on ophy philos c Socrati in the focus by Plato;!7* and it is actually the basis invoked by a medical writer of later antiquity precisely to explain why it is that Hippocrates, despite the occasional coincidence of material, cannot be considered a real physiognomist Commenting on a passage from the Epidemics (6. 7. 6) in which Hippocrates says that a person who naturally has the kind of voice or eyes that one has when one is angry is naturally irascible, the sixth-century physician Palladius says this (Commentary on Hippocrates’ Epidemics, ii, 195. 10-17 Dietz): " 98 not clearly drawn before Socrates.'”” It might, of course, be a freak of evidence associated with that theoretical discussions of physiognomy are always see a dichotomous model of body and soul along these lines; but one can Socratic philosophy, how the two go very well together. Physiognomy, like ans might talk Physici er. charact moral human of ent judgem the has as its object because they not but ena, phenom about (what we should class as) psychological surviving the from clear is This such. as ‘soul’ are concerned with the medicine Greek of ristic evidence; it is explicitly remarked on as a characte tradition on Plato’s psychology, ¢.g176 See again Section JI. For the influence of the medical (which give an interesting argument Poschenrieder (1882); Jouanna (1966); Claus (1981), esp. 177-8 development of a dichotomous Plato’s towards impulse indirect the provided thought that medical model of the soul). not made until Plato's Middle Period. 77 Claus (1981, esp. ch. 5) argues that the distinction was the absence of a systematically For ‘soul’ generally in the Hippocratic corpus—or, rather, for (1973-84) ii. 781; Pigeaud (1981) 32-47; dichotomous notion of soul—see e.g. Miiri (1947); Moraux Gundert (2000), esp. 33-5; Singer (1992). 99 There are real physiognomists (eisi gar tines ontös 7 physiognömones) íognd who make predicti about the character of the soul from the face, If they see someone with eyebrows drawn up, they call him shameless’; if with eyebrows drawn down, they call him ‘liberal’.!7* They ook for these things so that they can grasp the character of the soul (to tes psychés ethos); but a) Hippocrates looks for them in order to grasp the disposition ispositi of the body (tér 7 tou Iam assuming that the practice of physiognomy in Greece be; i theoretical context either of medicine or of philosophy, and 1 Me ae became an object of theoretical study in the hands of philosophers, who were motivated in particular by their interest in ethical character. That there is a certain overlap between philosophy and medicine in the treatment of the mental and physical aspects of human nature tells us something about the direction from which philosophy was coming when it started to explore physiagnomy: it does not tell us where it was when it did, The specific interest of philosophy in physiognomy was, in the end, all its own; and whatever one thinks about the earlier practitioners of physiognomy (such as Phaedo’s Zopyrus), it is, I suggest no coincidence that we first hear about them in philosophical texts, Nor5 it coincidence that, when a medical writer finally turned to discuss physiognom: it was on Aristotle that he relied for the basis of his theory, * (ii) Galen Galen’s reflection on the nature of the soul is not, or is not presented as, the starting-point for his theory of human nature. Rather, he starts from a physiological theory: a general theory of the composition of bodies. Famousl Galen developed from his reading of Hippocrates an explanatory model based on the belief that all organic bodies were composed of a certain ‘mixture’ or temperament’ (Greek krasis, Latin tempcramentum) of four fundamental 178 ; body res De{Socrates citing a Thracian physician): ‘One should not try to cure... the out ae Tk siis also why most diseases esczpe Greek physicians: they don’t look to the Whee ae ch one oe e subject of their care... (This is not contradicted by Phaedrus 270c, Mona ole’ in wl ich Hippocrates is said to be interested is the whole body—used, indeed, as frequents r investigation of the whole soul in this case. In general, it might be noted that Plato's oat invocation of medical science as an analogy for what a science of the soul might be is its eeent for the difference in emphasis between the two.) den „Perhaps Aristotle ap. Anon. Lat. 85; though Palladius’ point is presumably that this kind of Aristotleana P Ysiognomical ‘standard (and it might not be coincidence that we find Millen ana ab on eyebrows at HA 1. 9, 4911418 quoted both by Galen, at QAM 55. 17-21 by the paradoxographer Antigonus, at Collection of Amazing Stories 14a).

| Roughly pas , and phlegm. 186 ‘humours’, viz. blood, yellow bile, black bile ours was as it should be if the balance of these hum would ipso facto ‘ in an in iw wal, a er hand was to be exp be healthy; disease on the oth onstrated to the point ne a‘Galen thought, be dem so im ical er7 vien,= nions 15. 2); and, from a clin certainty (On My Own Opi en a zn En art orw the doctrine are straightf immediate consequences of = or hum any tify and rectify wished ie the doctor would be to iden done be to ded nee k e wor eo en atients. However, rather mor per tem l ora sal processes linking hum cau e cis pre the a an nd Pat sta n der lon, e of the individual's ont kind e i exp development and sustenanc be ld shou an organic constitution r the whe ure re uns wi was Galen peran en having a certain humoral tem al e ap 7 natural product of matter Y ura of such a nature would nat n. f thing into which matter appealu eee ° n necessary was develop), or whether it matter © i the existence inci true explanatory cause for the be ald e rat ope nd ce | prin u icular kind. im e some i r e a oon is that attraction of the teleological 2 sec 7a a M t wha ing iness of nature, includ 8 i of a u s anic and purposiveness of the org Be is fo he the a PE ty blis esta ing it seemed to Galen. Hav things. This, anyway, is how e could, wi r mixtur tattheie con the Tumdeournts 7Sexpelainn thegintr satisfactionevthat struction ofa living icat et be se as the eeri can view a hou ionism would imply. One on ism, as a strict reduct this organiza ion se materials; but to explain how ee nization of some basic body, ue arly, nor the For to its builder. abe ut requires reference in a ve to the tike dynamis) seems needed in} “constructive power’ (diaplas form a body to e rat ope n how they mix and humours themselves to explai d to a teleology, = s takes Galen along the roa ins en these line Kaeo our the , er pow e tiv ruc t the ‘const wake it makes sense to suppose tha u that as its instruments, so ; P mens bodies, uses the humours (On m the to or nt of, and pri chological force is independe or he nod , well as immediately a problem here be pl ek te 1635-6 Kühn). Yet there is int ar ne lligence E the soo attributes an extraordinary inte icular e iving body according to som n in ie see be to re the not is e enc nine ehtellig D y exhibited by ly understanding of the bod at all. Indeed, the craftsman m o3). on (199 e and Nuttc a in n n. 169 aabov ral theory see references in 180 For the development of humo its rela and soul fthe ici ’ entary 1°u ons commme ), withyoNoti ton 1999 es al3-6). re 26 700. iv. In generaforilatC tion of the Embr C£ also Construcctio 15 (on 60 nism; in others (eg. Usefulness ofikae ngly for redu de 6-10), a aAM} Galen argues stro (on 78. p n t works Saale strong case for teleology. A work like On Temperaments lies somere es on ( Cast ); ents eram Temp On er (19978) (53640 for ners Loyd (1988); Sing 101 constructive power remains inaccessible even to the most thoughtful and reflective of adults. If the teleologists were right, basic self-awareness should do perfectly for all of us what the science of anatomy imperfectly does for very few (Construction of the Embryo iv. 692. 18-693. 9, 700. 15-17 Ktihn). Teleological explanation has another great weakness in Galen’s mind as well— one more directly pertinent to the present study—and that is that the explanatory power that makes it so attractive reaches down only as far as the level of the species. When we marvel at the ‘craftsmanship’ apparent in nature, at how well constructed and well adapted to living a creature is, we are not focusing on characteristics peculiar to the individual. It is the species that is well constructed and well adapted to life: the individual is so only incidentally. (It is not, for example, the irascibility of individuals which inspires a teleologically inclined wonder, but, more generally, their capacity for thinking and moving in ways which will procure nourishment and self-preservation.) Yet the world is populated by individuals; and when one considers what makes them individuals (ie. their individual peculiarities) the attractions of teleology start to pall. In my account of Aristotle, I argued that he, at least, felt able to provide an explanation for individual peculiarities within a teleological model, and that he did it through a framework of hypothetical necessity. But Galen (and one might smagine that his arguments are based in the strong tradition of epiphenomenalism within the Peripatetic school itself) baulks at this account. He interprets Aristotle’s invocation of (a) teleology qualified by (b) hypothetical necessity as the claim that (a*) form is responsible for species characteristics (e.g, a nose), while (b*) matter determines what kind of a nose it is going to be. But in this case there are two serious problems. First, form could not even start building a nose without any idea of what sort of a nose to build. The sculptor Polyclitus might change his designs for the nose on a sculpture in the light of the materials he finds himself working with; but he needs to have designs before he can begin to wield his chisel. The second problem is that, if matter is supposed to be responsible for the characteristics of the child, we should expect (since form comes from the father and matter from the mother) that a creature would always look like its mother, and never like its father.’® This, of course, is an issue of which Aristotle was well aware; and on my reconstruction of his account, Galen’s dilemma seems to rest on a more or less crude misunderstanding of his hylomorphism. Galen talks as if the form were supposed to be the thing that does all the work (like Polyclitus}, whereas actually Aristotle knows very well that it is not the form of human being that does the work, but a particular human father. And a particular human father does aim to create a determinate kind of nose—namely, just the kind of nose that he has himself, But we can still explain how reserablance to the mother comes about: the matter supplied by the mother has some influence on the new child, insofar 12 See On Seed 2. 1. 51-3, iv. 605. 16-606. 10 Kühn for the former point; 2. 1. 54-7, iv. 606, 10-607. 6 Kühn for the latter (apparently using Anistctle’s example): if the artisan is responsible for kind, the matter for qualities (‘in the same way that the eye is made by the moulding power but is made either blue or black by the matter’), then all children would resemble their mothers; and if the artisan is responsible for the qualities too, then all would resemble (only) their fathers.

to it, or fails to ‘master’ it. as the father (or, more precisely, his seen) ‘gives in’ like both its father and less or more look It is thus that a child can come to mother. response precisely conPerhaps, however, the move Aristotle makes in this learn from Aristotle’s we What gy. teleolo from stitutes for Galen the move away ular cases of generation, the account (as I have presented it) is that in partic say that in order for there to protagonists are already determined. Indeed, we can already be determined. must onists protag the be a particular case of generation, and pure matter; it is form pure of dence coinci the from A child is not made another’s blood. Its and semen made from the encounter between one person's c than in terms more Galeni constitution, in other words—or, to speak temperaments of mother Aristotelian, its temperament—is determined by the position. Aristotle could ionist and father, And this begins to sound like a reduct action of form under gical teleolo the of terms maintain his analysis of events in does act outside never form that given but ty; necessi etical conditions of hypoth it is supposed which y activit the the conditions of hypothetical necessity, i.e. that individual, inate determ a of y activit the rily to explain is always and necessa as a genuine nce existe its posit to licence any have we that r it becomes unciea at all} It looks more entity or as something prior to the activity of individuals it; it certainly does ning explai of than y activit like a way of describing their e power’ whose ructiv not have the substantial status accorded to the ‘const for Galen. lity possibi ive attract hypothesis is needed to make teleology an seriously the take to ed prepar was Galen then, lties, In the light of these difficu Aristotle’s of those from possibility of a non-teleological biology, taking his jead to quote was, He onist. reducti a as pupils who saw him not as a teleologist but subject, the on works ant import most s Galen’ of the opening line of one soul the of eis) (dynam s prepared to entertain the possibility ‘that the power follaw the temperaments of bodies’ (QAM 32, 1-2 Müller}. The idea, in a 183 One af Aristotle’s own pupils, Dicaearchus of Messene, seems to have made a cognate point in Mirhady): ‘The Eretrians abolish qualities, as arguing his own reductionism (fr. 8g Wehrli=26 but subsist by “hemselves and in combination , common in e substanc of nature the in nothing having that there is an animal, bat abolishes the agrees reason, same the for hus, Dicaearc in particulars. And of universal: it is susceptible to just the kind a into es evaporat ul form-so soul as its cause.’ Aristotle’s be worth noting that there are might It forms. Platonic to type of argument he himself applies have substantial or prior existence. Ci. e.g. modem commentators who deny that Aristotelian forms ulos argues that ‘forms’ do not account for Caston (1997), 33%; or again Paviopoulos (2003). Pavlopo principle which te a metaphysical why creatures of a certain kind exist in the first place, but constitu ce lasting a very explains why ‘once life has occurred—be it a unique occurren short while—this same the disagreement between Galen life should go on indefinitely’ (2003: 180). If Pavlopoulos is right, could form the basis for a meanirgful and Aristotle here would be over whether such a model e to embrace reductionism. teleology, or whether it should rather have led Aristotl ntly, used in this sense (of psychological 14 The terminology of ‘following’ is, perhaps significa deed, it would net be absurd to nomy—in Physiog lian Aristote the by dependence on the body) Enovran TOG aupaan KTA.) lies r Sudvora ai (brt work that of imagine that the opening line oúparog Kpdceaty ÉTEUBE Tag vo (relic quoted somewhere behind the first line of the QAM as word ‘follow’ is, as Geoffrey Lloyd has Suvdperc TAG puxfic). Exactly what Galen intends by the shown, subject to ‘a certain vagueness’ (see Lloyd 1988: 33-7; and cf. Singer 1997. 535; also Caston onism as an approach n about reducti 1997: 352). It turns out that Galen is not only uncertai nutshell, is that individual creatures can be 103 lained from the botto appeal to the matter out of which they are made. Matter of the ar complexity is supplied by the parents and, given the kind of matter it is, it undergoes certain physical changes: it develops into a ‘construction’ (diaplasis) of a certain kind.'*° But it happens that one set of properties possessed by this sort of material is what we think of as its ‘psychological’ properties, including (most importantly) psychological powers or capacities. These psychological properties vary according to the temperament of the material, and so (a fortiori) vary in step with the different biological constructions produced by materials of different temperament. Indeed, one might think of them as marks of the kind of material it is, such that it develops in a certain way. The construction of the animal thus appears to work in the light of its psychological capacities. In fact. the psychological capacities are expressions of the material which forms the basis for just that construction. Although Galen’s explorations of psychological teleology and reductionism find their‚Erpression in both cases as interpretation of Aristotle,!®* Galen does not go so far as to attribute to Aristotle his humoral analysis of biologi Nevertheless, he thinks that Aristotle held a view that, intTheory at leat worked to much the same effect. When Aristotle said that the soul was the form’ of the body, we are now supposed to understand him to have meant that the soul is the ‘temperament’ or ‘mixture’ of hot, cold, wet, and dry in the body (QAM 37. 5-24 Müller). In this way, ‘Aristotle too believes that the soul's faculties depend upon the mixture of the mother’s blood from which, in his opinion aur blood derives’ (QAM 51. 12-15 Müller). To support this claim, Galen appeals to HA 1 (54. 21-55. 4 Müller). As becomes apparent there (says Galen): Aristotle would have it that the construction (diaplasis) of the whole body comes about as appropriate to the characteristics and faculties of the soul in each genus of animal-—e. the birth of blooded animals is from the blood of the mother, and the Psychologie l characteristics follows its mixture (as has been shown in the foregoing). But the m. of the organiz limbs comes about as appropriate for the characteristics e soul. to psychology, but also (and less overtly) inconsistent in how strong he believes a reductionism has to be: at times he is eliminitivist (the soul is nothing more than temperament}, at times emergentist (ee soul somehow emerges from the constitution of the body and is perhaps influenced, though not termined in its character by it). For details of the stronger line as Galen derives it from Plato and Aristotle in the QAM see 37. 5-24, Müller (Aristotle; and cf. PHP 5. 5. 27. 1-28. 1 De Lacy); 42, 3-45 3 Muller (Plato and Aristotle). (Gill, 2000: 66-7 with n. 30, reads the latter passage as a rejection of ai reductionism, but whatever the details of interpretation, this conclusion cannot be right, since en is quite explicit about the matter at the end of the passage: ‘it is not open to us to sa that it [the soul] is anything other than the temperament’. C£ also Sorabji 2000: 255.) mu 185 Galen explains a child’s resemblance to either or b oth parents by supposing i that each Br to the matter (as, in the slightly different explanatory model of the On Seed, each tibu r : D D jd hae a portion of both matter and moving principle: e.g. On Seed 2. 2. 19-24, iv. 614, 5-615. 186 A Aristotle is obviously Galen's starting-poi g-point for thinking inki about physiognomy, but he does not acknowledge the real gaps that also exist between their respective models (a fact which is, of course significant for the way in which i the two were read together in the Arabic i tradition). (ome) zes. See Lloyd

is directed If it seems to be the case that the organic construction of an animal psycho(its way certain a in living for s towards the realization of its capacitie it is But ®” correct.' sense certain a in is this d), construe broad! logical powers matter, the upon acting force not because the capacities constitute a teleological s spell out, or add up to a ‘constructive power’; it is rather because those capacitie it does. as we might say, the nature of the mixture such that it develops the way ar particul a support to y necessar material the A given kind of blood constitutes the ng) explaini extent this to (and at looking of way one is this and way of living; ar kind fact that its inherent movements lead to the formation of a particul living. of way this s embodie that one ly of organic structure—name is the The process so far gives an explanation of two crucial phenomena. One its to exactly onds corresp animai an of r fact that the psychological characte such being ’s material the of on expressi an are powers (Those , physical structure as to develop into this physical structure.} The other (and here the theory starts for all to go beyond what Galen believes a teleology can explain) is the fact that, their generic resemblance to other members of the same species, it is its parents its that a creature most closely resembles,'** The reason for this, of course, is that their d explaine which ment tempera the precisely by d explaine construction is construction, But Galen does not stop here; he also considers what will happen if the temperament of the animal is somehow altered by external factors—something be to which it would seem, as a physical rather than a metaphysical principle, to make could diet person’s a of or example, for climate, of effects susceptible. The a difference to their temperament, and so (independently of their genetic inheritance) to their psycho-physical character. This point might seem innocuous; but, apart from the fact that it has certain physiognomical nt implications, which I shall consider below, it might turn out to be importa this make Galen that large at model nist reductio his of ity to the plausibil distinction between, on the one band, the temperarnent a creature inherits from of some constraining force (such as the ‘constructive power’ posited by Galen in teleological mood), a strict identification of soul and temperament might leave ne room for talking about an individuals nature as anything other than individual—the result of a particular set of circumstances which includes but in ne way privileges tae temperament of the parental material in which it began life. Just as teleolagy might be criticized for failing adequately to explain individual characteristics, so reductionism might be accused of failing to delimit the nature of an individual, to secure its nature a3 a particular kind of thing at the level of the species. To put this bluntly: if soul just is temperament, and the nature of a thing is the kind of soul it has, then any change in temperament would mean a change in soul—and this in turn would mean a change in its nature, the kind of thing it is. The argument takes satirical form in Nemesius 104 its parents in the material from which it starts to grow, and, on the other hand, further modifications to which its temperament might be subject in the course that of its life, The reason that this might be important is because one criticism absence the in that, precisely was y antiquit in nism was levelled against reductio body, devoid of organic 17 Cf QAM 37 Müller: it is to the ‘homogeneous and simple kind of the that activities (energeiai) primarily belong’—i.e., again, the diaplasis is ‘for’ composition... different view of the matter, activities already expressed in the humoral temperament. (For a slightly of the process of see On Temperaments i 566.17-567.14, 635.2-636.2 Kühn.) More details nt of the four ‘principal construction are given in the Art of Medicine, which describes the developme determines the final parts’ of the bady (brain, heart, liver, testicles), whose character in turn of this work is purpose (The tuated. si are they which in body the of character of the various regions parts in an individual from to allow one to apply this understanding to the diagnosis of the principal mical’ external physical signs. Not coincidentally, Galen comes closest here to practical ‘physiogno at i. 336.15-337.2 observations—something of which he seems conscious in issuing a disclaimer ensures that our ‘innate’ philosophy through characters our amend can we that fact the Kühn: character is never fixed.) its parents (the one fact 18 Note that Galen took the fact that a child tended to resemble to be an argument for the explained by the reductionist qualification of Aristotle’ psychology} reductionist position: Construction of the Embryo iv. 700. 12-14 Kuhn. 105 (On the Nature of Man 2. 24. 21-4 Morani):!” Again,: if the soul is a temperament, | and temperaments change according to age, scason, and diet, the soul changes. And if it changes, we do not have the same soul, but, according to the temperament,A sometimes that of a lion, À someti etimes that of a sheep, some i of something else, Which is absurd, ij Fes that Galen already was aware of just such a line of reasoning;!* and it seems to be in an attempt to head it off that he is careful to make a distinction between the temperament of an animal’s original constitution, on which is based its construction (diaplasis), and further modifications to the temperament which might occur during its life. The idea must be that an animal’s nature is determined and delimited by its original temperament, so that we can say that it is just the kind of animal that its parents were, and that further modifications of this temperament (due in primis to climate and diet) can never be so severe as to change its nature. Such a claim would not be ad hoc, but underwritten by sound biological considerations. Once the process of organic construction is under way (ie, as soon as parental material begins the process of development into a new animal}, any influence on the animal which might in principle be strong enough to change its temperament into a mixture suitable for a different kind of construction would bring its temperament into conflict with its actual psychophysical construction. The result of this would not be a change in its nature 1 . en is worth observing here that Nemesius gives no indication that he is aware of, or interested in, theories of physiognomy—despite a common assumption to the contrary. Evans, for example we to On the Nature of Man as ‘one of the most important documents of the fourth century’ u this context (1969: 80). But as her own discussion goes on to show, Nemesius is important rather in mere general terms for his discussions of the relationship between body and soul, and for the evidence he supplies for earlier views on the subject. It is true that, when Nemesius discusses Galen, the views he telks about happen to have been relevant to their author's physiognomical interest as well, But even in this case Nemesius makes no menton of the fact. Against people who think that in the case of crass-breeds it is the mother that determines the en on the grounds that things change according to their nutrition, he says: ‘The Persian plant, Wer ransported to Egypt, did not become different in kind, but on receiving good nutriment bore e fruit, its fruit not being edible before; and sheep when moved at zome time to a different Pasture did not become similar to the goats formerly pastured there, just as the goats did not become De Lacyto190 similar sheep or asses or horses’ { (On Seed 2. 1. 42, iv. 603. 10-15 Kühn as translated by

rament and construction (it is too late for that), but a conflict betweer. tempe death: the destruction of tely, ultima and, e which would manifest itself in diseas psychologically foregoing). But the construction of the organic limbs comes about as appropriate for the characteristics of the soul [: very much so, according to Aristotle]. And, in line with this there are quite a few indications of psychological character and physical temperament. Some of the physiognomical signs indicate temperament directly and immediately—such as whose dealing with colour'”” and hair, but also voice’ and the movements of the 106 l (both the animal.!°! So we can say that the ‘nature’ of the anima physical construction and and physically) is exhibited in and determined by its temperament. its in y iatel immed not that, ie the causes which underl of physiognomy, for we has Galen view the for ations implic has this of All on the basis of the sign ical ognom physi can now distinguish two kinds of reductionist model ated with its he offers. There are, first of all, those signs associ ns of the temperament construction (diaplasis): the psycho-physical manifestatio animal’s nature as a the to spond corre These al. materi of its original (parental) i) its nature as the fortior a member of a particular species, and also (in fact be an example of a might nose its of shape The offspring of particular parents. will have a nose that is like physiognomical sign under this heading: an animal of its parents; and there that like is it as that of others of its species just insofar l character to which logica psycho its of ption descri a in es will no doubt be featur kind of nose will, ular partic a the nose corresponds. (At the level of the species, of the animal’s degree the ps perha and trivially, indicate the power of smell, ’s nose is a bit more snub than sensitivity to smell; if, like its parents, an animal with them.) But then we can usual, it might indicate a lascivious tendency shared those modifications of its of ns statio manife cal also lcok for psycho-physi whatever sort (modifications temperament caused by environmental factors of not come into conflict with do they as r which the animal can support just insofa is a tricky area, but an this that below see shall We is). the needs of its diaplas an individual to caused which diet example for now might be the effects of a inclined to more and face the in redder so and become hotter in constitution point —the very this is it anger than he or she would otherwise have been. In fact d earlier quote e passag QAM the h —whic signs double nature of physiognomical its wider in again, e passag same the is Here ning. explai as on is introduced quotation from Aristotle, context (54. 6-55. 11 Müller, beginning with a PA 651°12-17):° to do with an animal’s character and “The nature of the blood is the cause of many things the whole body (for nourishment is of matter the is it for ble, perception. This is reasona nce whether . So it makes a great differe matter, and blood is the ultimate nourishment) dirty’. Aristotle says other things in the blood is hot or cold, or thin or thick, or pure or to me superfluous to write out seems it but ms, Proble the works on animals and in the le Aristot thinks about the temperaments everything: it is sufficient for me to show what Nevertheless, I shall add things that are soul. the of es faculti the of the body and about to 1, some of which lead immediately spoken about in Historia Animalium book [according to physiognomical signs temperament (krasis}, some through intermediate the construction (diaplasis) of the that it have would le Aristotle himself]. For Aristot eristics and faculties of the soul in charact the whole body comes about as appropriate to s is from the blood of the mother, animal d bloode of birth the —.g, each genus of animal e (as has been shown in the mixtur its follow s and the psychological characteristic 11 CE Art of Medicine i 310.8-311.3 Kühn; On My Own Opinions 15, 2-4. 182 In the translation that follows, the phrases in square brackets are athetized by Müller (but note eldt). pp. 58/26 ed, Biesterf that they are present in the Arabic translation, 107 imbs. Construction (diaplasis) provides indirect signs of temperament (and hence psychological character); what I have characterized as modifications of construction give more direct signs of the same thing.” | Galen, then, gives a solid theoretical basis for physiognomy. But it is Important even in his case not to exaggerate the importance to him of physiognomy as such. For one thing, even Galen is never a physiognomical determinist. He usually allows that rational thought, at least, is an emergent activity, which can overcome the natural disposition determined by temperament with no corresponding trace in the physical character of the individual. This, to be sure, does little to undermine the theoretical status of physiognomy—or even its practical uses, given how little inclined most people are to philosophy (and the same point could in any case be made for Phaedo and Aristotle). There are more significant considerations, however, One is the fact that physiognomy is clearly of interest to Galen only as a consequence ofa reductionist psychology, not as an end in itseli—something immediately apparent in the fact that Galen’s ‘physiognomical’ speculation is all concentrated in reductionist’ contexts.!”” This is a significant restriction on its significance for him, because although Galen has reasons to be attracted to psychological reductionism (for one thing, it promotes his attempt to present medicine as a kind of master science if ethical character turns out to be a function of a person’s 173 C£ On Temperaments i, 627-8 Kühn, on the usefulness of skin in inference i temperament: since skin ‘in our part of the world’ is well-balanced, it ‘gives mare parts beneath—though even here this relationship is not straightforward, the above applying only to ‚nme parts which have the same mixture as that of the skin.’ 7 In Art of Medicine i. 350-1 Kühn, loudness of voice is an innate, not an acquired characteristic: but note that it is there explained indirectly by the influence of heat on the (diaplasis of the) windpipe. So perhaps we should understand that in some aspects the voice functions as a direct si; of temperament, while in others (including loudness) it is an indirect sign. u | It might be worth noting, as perhaps another indication that Galen is aware of the Aristotelian id N n. i above) that 4 in that work considerations of colour, hair, movement, and voice chy 31-32 are cf. Polemon, Leiden order) are ttlisted= to; | er after the survey of bady parts (812"11—end; = eg, at Art of Medicine i. 336.16-337.2 Kühn; cf. On the Passions and Errors of the Soul v. 37.1 1-14 Kühn (nature is dominant in youth, beliefs and habit—ta te dogmata kai ten askêsin —later on). But for the vacillation in the rigour of Galen’s reductionism, see again n. 184. There is no explicit mention of physiognomy anywhere else in his a either in neutral contexts, or where teleology is to the fore. Of the six occurrences of the word ‘physiognomy (and its cognates), four come from the QAM and one from On Temperaments. The remaining example is his reference at PHP 5. 5. 22, 1 to Posidonius, whose appeal to ‘the physiognomists’ evidence likewise cam D Er : abon) e in an explicitly reductionist context (cf, PHP 5. 5. 23 with discussion in Section IV.

of his wider output, physiological state), his commitment to it is, in terms rather limited. theoretical enthusEven within his reductionist explorations, however, Galen’s an argument for into te transla fact of matter a iasm for physiognomy does not as his voluminous in ces inferen al gnomic physio of es exampl and its use in practice, physiognomy a of lity possibi works are vanishingly few.” He supports the ical constraints; based on zoological typology, assuming the appropriate theoret move beyond theory, but he starts to lose interest where he might be expected to (On Temperaments i. le Aristot to d referre and readers who want to know more are 624, 17-625, 1 Kühn; similarly On Epidemics Vi, 7-8 (53) Ghersetti 1999a}. al inference collapses Perhaps it is precisely because in his view physiognomic er the state of the —wheth states ogical physiol ental with diagnosis of more fundam ying temperament underl the of or e, Medicin of Art the in as organs, al princip itself—that it could have very little independent clinical value. 198 “The more one’s character is brought under the influence of physical, and especially dietary and ce of medicine in human life—-to the environmental, factors, the greater the scope and importan net been lost on commentators: see extent that it can subsume even philosophy. This point has This consideration will also help to (1988). Lloyd esp. and 129; (1988), Ballester 144; (1974), Donini Galen is keen in the QAM to though:, ian Aristotel on reliance obvious Galen’s despite explain why, (and through Aristotle himself provide his psychology and physiognomy with a lineage that goes back Plato too}, to an origin in Hippocrates. . First, a quotation from 199 1 know of two candidates from passages in Galen not already discussed derives from Galen’s lost al-Mas‘udi (which, as Vivian Nutton bas suggested to me, Commeniary on the AWP), In the translation of Meynard perhaps and Courteille (1861) 1634 this reads: crépus, Les sourcils rares, “Galien signale dix propriétés particulières aux noirs, à savoir: les cheveux de la peau, la noirceur du teint, la les narines dilatées, les lèvres épaisses, les dents aiguës, la puanteur génitales et une pétulanc: excessive. longueur des pieds et des mains, le développement des parties ation imparfaite de son cerveau, Cet auteur explique cette derniére qualité chez le noir par l'organis substantial to what we have already nothing adds This ce.’ intelligen son de faiblesse la résulte d'où stupidity and an ‘ill-formed neck’ is seen; though it might be interesting to note that the link between fr. 6 Rossetti}. A second passage (Phaeda, Socrates to applied have to one that Zopyrus is supposed is To Glaucon on Medical Method, which might provide evidence for Galen's interest in physiognomy says, ‘to make distinctions in the case xi. 11 Kühn (= On Crises ix. 627 Kühn). ‘Tt is necessary’, Galen of the character (Ethos) of the soul in of the eyes especially. For it is possible to get evidence from them able to see accurately.’ Barton uses those who are healthy; and in the ill, the signs are clearer for one to draw the line between the difficult is ‘it that claim her justify to passage) this only (and this passage at times’ (19942: 98, with omy! physiogn and diagnosis [viz. body the of ion two types of examinat tell us what to look for not does Gaien that fact the by vitiated is n. 18). But the value of the claim would be physiognomical signs in in the eyes. It might be humoral temperament, so that the eyes (contrast Art of Medicine i. 329-31 a sense recognized by Galen’s theoretical model for the science the humorai state of the brain); reflect might eyes the of moisture and colour, size, the where Kiibn, them. in activity mental of signs for look should we that be but it might just that one has a causal explanation to 00 On Temperaments i. 624 Kühn. The constraints are: (a) ical traits (Galen thinks that this is hand for the link mooted between physical and psycholog and (b) that one does not generslize something neglected by many physiognomical practitioners); a whole. This latter point, says too quickly from individual limbs to the character of the body as (ef. n. 113 to Section II, appendix, ners practitio sophical non-philo the on Jost even not was Galen, ion te ; also e.g. Polemon, Leiden ch. 1, A16)—and elsewhere he has his own contribut on epiprepeia he notes that the state of ane of the make towards explaining the problem. In the Art of Medicine, under the more direct influence body the of parts on principal organs can have knock-on effects in the heart can lead to a smaller of another-of the principal organs: e.g. at i, 333 Kühn, heat of the head is the brain. head—though the organ primarily responsible for the character 109 It is not, though, only pragmatic considerations that limit Galen’s interest in making a practical study of physiognomy. For when Galen looks more closely at the detailed workings of humoral theory in a reductionist context, serious obstacles to physiognomical diagnosis emerge. For it turns out that the very same outward, physical effects can be the result of quite diverse, even opposing influences on the temperament. In particular, the effects of climate on appearance and character might not be consistent with the effects of the inherited temperament. So, for example, there are outward signs of a naturally ‘hot’ or ‘cold’ temperament (perhaps darker or paler skin respectively). But the effect of a particularly cold climate is not (as one might have hoped) to make its inhabitants colder than they would otherwise have been in temperament and paler in appearance“! Rather, the cold is supposed to drive their innate heat inwards, making them hotter in psychological character, but paler (as if colder) in appearance. In a hot climate, conversely, people’s heat is “drawn out’, making them lock ‘hotter’ still, but leaving them with little internal heat, and psychologically cooler. (See on this Temperaments i. 627-8 Kihn.) So in general, Galen says, one needs to ask whether the visible signs of heat in a body are signs of external heat or of heat which belongs to the body (i. 628.13-15 Kühn; cf. also Art of Medicine i. 341. 14-18 Kühn). This is a fatal blow. However much theoretical interest physiognomy might have in certain contexts, it could not have significant diagnostic value even for Galen. (iii) Other Medical Writers As Antonella Ghersetti shows later on in this volume, the materials Galen provided in the QAM for reading Aristotle’s Physiognomy through his own humoral theory were eventually adopted and developed systematically in the Arabic medical tradition. But their immediate impact was altogether less impressive, In fact, Greek medicine after Galen proved no more interested in physiognomy as such than it had been before him. Before him, the only medical writer on physiognomy we hear about is Loxus, used as a source by the author of the anonymous Latin treatise from whom he earns the epithet ‘medicus’. But, as we have already seen in Section III. iii, there is nothing characteristically ‘medical’ about the fragments of Loxus’ work: nothing, in fact, that distinguishes it from a work like Ps.-Aristotle’s Physiognomy at al. 7 The only other physician 201 “Cold” and ‘hot’ climates here are relative to the ‘well-tempered’ climate of the middle regions of the world (such as Greece}, which have a neutral effect as such on physiological temperament: see On Temperaments i, 627. 5 Kühn (and cf. Plato, Timaeus 24 cb; Aristotle, Politics 7. 7, 1327°23-33; but contrast on this AWP 12. 6). | , In general, Anon.’s choice of Loxus, Aristotle, and Polemon, and his characterization of them as physician, philosopher, and orator respectively (Anon. Lat. 1}, is clearly meant to impress us with the seriousness of physiognomy and the breadth of its appeal. In fact it tells us nothing about the theoretical scope of physiognomical writing. On the contrary, it only becomes clearer how those who seriously engage with physiognomy as a theoretically grounded discipline rely heavily on Aristotle {or Aristotelians) in doing so. A physician, like an orator, might conceive an interest in physiognomy or find a use for it; but there is no more (discernible) input into the physiognomical tradition from the medical community as such than there is from the rhetorical community as such.

we know to have written on physiognomy is Adamantius; but his surviving work is merely an epitome of Polemon’s physiognomical treatise25. Again, it contains to nothing distinctively medical, nothing that cannot be explained by reference the Peripatetic tradition. Medics writing as such naturally continued to be interested in the psychological effects of physiological dispositions as symptoms of physical d in illness which could be used in diagnosis. They were also, of course, intereste VI. POST-HELLENISTIC PLATONISM 110 mental illness as something which could be explained by physical causes.?°* But and despite the occasional coincidence of material between medical observations omical physiogn surviving the in posited ons correlati the sort of psycho-physical handbooks, despite the opening given by Galen for physiognomy in medical n discussion, there is in all of this scarcely any explicit or systematic discussio 6.7.6, s Epidernic on comment his in tone the sets of physiognomy. Palladius at as quoted above. There, he carefully distanced medicine from physiognomy VI, s Epidemic (On passage same the a point where Galen, in his commentary on 7-8 (53) Ghersetti 19994), had sought to bring them together. Physicians, says not Palladius, sometimes use the same materials as physiognomists; that does this in often so (as is thing striking make them physiognomists. Indeed, the most c study) the absence of explicit reference to physiognomy, the absence of systemati discussion of evidence we might class as physiognomical. Even when lly physiognomy and medicine are spoken of in the same breath, as occasiona be can both where contexts, ical non-techn t somewha in be to tends it they are, authors medical the as far 2% As linked with other forms of predictive art as well of antiquity themselves are concerned, physiognomy as a category of thought seems generally to hold very little interest.” 203 Adamantius is normally identified with the Adamantius called ‘Sophist’ (at Actus, Iatrica 163. 1, 8. 29. 2) who wrote On the Genesis of the Winds, On the Origin of given by Stars. Cf. Follet {1989}, 3. Dreams, and On the Signs 208 For the debare over whether physical indisposition always underlay mental illness, see esp. the or lesser degree on surviving works of Caclius Aurelianus in the 5th c. (though he relies to a greater the history of melancholia the 2nd-c. physician Soranus), with T. M. Brown (1993), esp. 429-40, For and Saxl (1964); (an archetypal mental illness, along with mania), see esp. Klibansky, Panofsky, ry interest in Flashar (1966). (Cf. also Swain above p. 11 for the possible influence of contempora menta! health on Folemon.) 111 Oceurrences of physiognémon- in the writings of Platonists: L. Calvenus Taurus ap. Autus Gelli Attic Nights |, 5. 2; Porphyry; Life of Pythagoras 13. 2; 54. 14 Nauck; Aristides Quintilianns, On Miah 3. 8, 106, 6 Winnington-Ingram; Iamblichus, On the Pythagorean Life 17, 71. 13-14; 74.3 Klein; Ammonius, On the Prior Analytics 7. 16 Wallies; Proclus, On the Cratylus 88. 6 Pasquali; Olympiodorus, On the Alcibiades 13. 19; 97. 6, 12 Westerink; On the Gorgias 5, 3. 14 Westerink; Philoponus, On the Prior Analytics xiii. 2, 483. 224, 484, 8-11, 485. 5 Wallies; On the Generation of Animals iv. 3, 185. 13 Hayduck; On the de anima xv, 155. 22 Hayduck. Occurrences of physiognömon- (and metäposkop-) in the writings of fellow travellers:””® Philo of Alexandria, On Dreams 1. 1642 Galen, On Temperaments i. 624. 3 Kühn; QAM 54. 20, 55. 8, 63, 1-2 Müller; PHP 5. 5. 22. 2 De Lacy; Clement, Stromata 1. 21. 135. 2. 3; Pedagogue 3. 3. 15, 2 (metöposkapos glossed as physiognömen in the scholia ad loc. 336. 10 Stählin-Treu); Hippolytus, Refutation 1. 2. 5, 4. 15. 3 (metóskopië manteia); Origen, Against Celsus 1. 33. 14.219 Physiognomy as a theoretical concern in the philosophy of antiquity sta with Phaedo, flourishes in Aristotle’s school, and ends, en ‘night fairly omy. with Galen, By this I mean that we have evidence for the theoretical discussion of physiognomy as such in no other philosopher. The story has a continuation. however; for while there is no reflection on the nature of physiognomy in the surviving works of later Platonist thinkers, it is clear that many of them accepted not respect, if it shows anything at all} There, is by the way, no evidence, nor is it likely, that the word typos (‘stamp’) used here to describe the relationship between body (as sign) and soul (as signified) was ons used in other contexts by Empiricists to denote the relationship between two things experienced in conjunction (though it dees recur in physiognomical contexts to describe the relationship between soul and body: e.g. Ps.-Aristotle 806°32; cf. Phaedo fr. 18 Rossetti = Maximus of Tyre, Dissertations 25, 3). 208 ‘There are those who might «vant to add Posidonius to this list as well; though his use of Plato seems to me thoroughly Stoic (see Section IV. ii above; cf. Boys-Stones 2001: 99-100 with n. 3}, Philo, however, gives no clear indication that he believes in the art (see again Section IV n. 124}; and the extent to which it is appropriate to see him as part of the mainstream Platonist revival is a moot point (cf. Boys-Stones 2001: 100-1). It is worth noting how little direct, theoretical discussion of physiognomy there is among Christians, Origen is the onl; Christian writer to show awareness of the theoretical history of the subject (see n. 233 below}, Ambrose in the 4th c. believes that gait and gesture reveal the soul, so that correction of one’s psychological nature needs to underlie correction of manners (On the Duties of medicine the Clergy L 18. 71-5; L. 19. 84). This is at a little distance from the tradition of philosophical (1999), on Aurelianus’ and physiognemy in the evolving concept of melancholia; also Horstmanshoff reflection on physiognomy (for one thing, it is manners he is speaking of, not physical features); but even 50 unusual in how near it gets, Clement of Alexandria, who had earlier on spoken about ‘how Christians should order their public appearance (cf, Pedagogue 3, 11-12), was clear that an ability to tell people's character from the way they dress and so on is like, but is not, ‘metoposcopy’ (Le. Physiognomy: Pedagogue 3. 3, 15. 2). That the theoretical issues here are skirted around by Christians is probably to be explained on pragmatic grounds. Not only might Christians find physiognomy Pedagogically unhelpful (at least when the accent is on the possibility of change: cf. Origen, Against 25 Cf again Klibansky, Panofsky, and Sax! (1964), arguing for convergence between treatment of emotion {but note the important qualification in his claim at 267: ‘Ses observations, la première source de sa description, s'accordent parfois mot pour mot avec la tradition direef [my emphasis])physiognomonique, même si Pon ne peut pas fournir les preuves d'un emprunt kai manteis); Fuigentius, 06 Clement, Stromata 1. 21, 135. 2. 3 (hoi physiognömonourtes iatroi te Contents of Virgil 84. 10-11 Helm as cited in n, 174 above. of the soul, as physiognomy 0? Sextus Empiricus at PH 1. 85 says that ‘the body is the stamp sts, that they lack an shows’ and Galen's rather dismissive comments on the mass of physiognomi might be taken as evidence explanatory basis for their predictions {On Temperaments i. 624 Kühn), in their philosophy. But this that most were not only empirical in their approach but Empiricist 99) Sextus himself was would be taking evidence to the limits of optimism; and (pace Barton 19945: positions. (The fact Dogmatic refute to positions Dogmatic using hominem, ad certainly speaking niké sophia, shows irony: that he refers to physiognomy as a kind of ‘wisdom’, physiogndmo a 66, 69), they might shrink from unnecessary association with an art that sometimes ri scorn for is pretensions (cf. Artemidorus, Oneirocritica 2. 69, 10), and may have been ought to rely for its successes on daemonic influence (as in the case of other forms of prognostication: cf, e.g. Clement, Stromata 1. 21. 135. 2; Hippolytus 4. 15. 3). But for a fuller and po positive assssment of early Christian attitudes to physiognomy, see Parsons (2006), esp. ch. 2 ysiognomy in Jewish and Early Christian Literature’).

it as a theoretical possibility.’ There might be no evidence that Plato !? they explained interested in the art, but Platonists knew that Aristotle was; on the body.”"* Galen, indeed, had suggested that one could not be a true Platonist if one thought otherwise (QAM 64. 19-65. 9 Müller): 112 Socrates’ attraction to Alcibiades as an example of physiognomy in practice; and they attributed its invention and use to Pythagoras.” the As with every other thinker I have considered, Platonists could accept beliefs their of nce conseque a as arose it as possibility of physiognomy insofer the about the relationship between the body and the soul. That they did not feel to due bly presuma is beliefs these need to argue or discuss it as a consequence of ly adequate been had subject the d, concerne were the fact that, as far as they as covered by earlier thinkers in whose footsteps they thought of themselves with m Platonis enters nomy following: that a sympathetic attitude to physiog their reception of, in particular, Aristotle. But here we encounter a problem: for, as we have seen, especially in Galen, it turns out that Aristotelian physiognomy quickly came to be thought of as consequent on a specifically epiphenomenalist interpretation of his psychology. According to this interpretation, the body was a guide to the character of the soul insofar as the activity of the soul could be explained as a product of the body’s functioning. The problem comes because one might assume that Platonists would want to be very careful to avoid associating themselves with such a view. It is central, not only to the psychology, but actually to the entire metaphysical system of Platonism, that soul is immortal, independent of the corporeal world, and (at least logically) prier to it. It would be anathema to suggest that the soul relied for its nature on the physical circumstances of its instantiation. Already Apuleius, for example, warning that we should not prejudge a person's character by place of birth, explains that local environmental conditions can make a difference to the taste of vegetables and quality of vines but, since the soul comes into the body from outside, such things could make no difference to its virtue.” This point has a particular sensitivity, since Plato himself at one point appears to explain the superior intelligence of the Greeks by the temperate climate of their country (Timaeus 24c), and this passage was seized on by some commentators as support for claims that the soul was dependent to some extent 211 rare and, in any case peripheral, exception is the Platonizing Sophist Maximus of Tyre who, and asks {rhetorically} to at Oration 25. 3, calls Zopyrus’ art ‘obscure divination’ {manteian asaphé) what kind of co-mingling (epimixia) the body and soul owe their similarity. The mainstream to this question will emerge in what folows. answer 212 In fact they rarely mention Aristotle’s physiognomical interests; but it becomes unavoidable in 16 Wallies; commentaries on the Prior Analytics (cf. eg. Ammonius, On the Prior Analytics 7. Philoponus, On the Prior Analytics xiii. 2, 483, 22 Wallies). 213 Broclus, On the First Alcibiades 94, 4-15 Westerink; cf. Olympiodorus, On the Alcibiades 13. Westerink; also Plutarch, Alcibiades 4. 1. Cf. Apuleius, On Plato 1. 1 for Socrates’ Proclus has a sharp rejoinder to this. In his own discussion of Timaeus 24c, he says that the literal reading of it was upheld by ‘Panaetius and other Platonists’ (On the Timaeus, i. 162. 11-15 Diehl). As we shall shortly see, he was certainly not ignorant of Galen’s position, and it seems odd that Galen’s name is absent here—until one thinks what it might mean for Proclus to appeal to the Stoic Panaetius as a “Platonist’. His point is precisely directed at Galen’s controversial remark: if it is ‘Platonists’ who believe in a literal reading of Timaeus 24 c, they are Platonists like Panaetius: that is, no Platonists at all.?!? | | Not all of Proclus’ attacks on Galen are so indirect. Indeed, Galen becomes for him the archetypal adherent of the reducticnism that Platonists in fact wish to avoid, and Proclus’ attacks on reductionist theories of soul often name him, or allude to the QAM. So, for example, when he is explaining the fact that people become wiser and more considerate as they mature, Proclus contrasts his view of the relationship between body and soul with that of Galen, encapsulated by the iii 349. (On the the TiTimaeus iii, opening (and programmatic) statement of the QAM (On 21-30 Diehl): Galen would say that thc powers of the soul follow the temperaments of the body... But it is not right to make the immortal soul, which exists before the body, think through the body: we should rather say that the body sometimes becomes more of a distraction to the soul in its attempt to live a well-ordered life, sometimes less disturbing. | There is an allusion to the same line in his defence of the soul’s immortality in his Commentary on Plato's Parmenides (1014. 28-36 Cousin): If the soul is not immortal... it will in addition be mixed with bodies and enmattered, and it will not be able to rule them ar lead them where it wishes, but will follow the temperaments of the body. 19 and pupils 21% pythagoras is supposed by Platonists to have used physiognomy in choosing friends ap. Aulus (in line with Ademantius’ recommendations, by the way, A2}. See Calvenus Taurus 13. 2-14. 1, 54 Gellius, Attic Nights 1. 9. 2 (68 for Taurus as the source); Porphyry, Life of Pythagoras lamblichus, Pythagorean Life 17. 71-4 Klein; Olympiodorus, On the Gorgias 5. 3, 14 Westerink (with Refutation 1. 2. 5. 2. ex istis 215 Apology 24 enimvero animo hominis extrinsecus in hospitium corporis immigranti quid addi vel minui ad virtutern uel malitiam potest? Some of those who call themselves Platonists think that the soul can be impeded by the body during illnesses, but performs its proper activities when the body is well, being neither helped ner hindered by it. So I shall transcribe some remarks of Plato’s where it is clear that men are helped or harmed where intelligence is concerned by the temperament of the locality, even when the body is not sick. In the Timaeus, early in the dialogue [ie. at 24e], he writes: “The goddess first organized you according to this complete arrangement and design when she established your city; and she chose the place in which you would be born with an eye to the fact that its seasons were temperate (ten eukrasian tin hôrôn en autdi katidousa) and would bear the wisest men.’ identification of the infant Plato. which cf. Iamblichus, The Pythagorean Life 25, 112 Klein). Cf also Hippolytus, 113 a . 6 Apuleius may be issuing a tacit warning about the passage when he says that the tape determination of moral character is disproved by the fact that not all Athenians are = (as, correlatively, not all Scythians are stupid: Apology 24). an passage then, which appears to give textual comfort tc those who would like to see penn us as a Platonist ‚in fact proves the very opposite (and cf. Boys-Stones 2001: 99 with n. 3). In = ion to giving his own. view, Proclus lists alternative readings of Timaeus 24c from Longinus, ngen (ie. the pagan), Porphyry, and Jamblichus (On the Timaeus i. 162-5 Diehl}.

And again, On the Republic, i. 249. 21-3 Kroll: So let us not accept that atgument—when someone says that the powers of the soul follow the quality of the temperaments of bodies... Compare On the First Alcibiades 226. 16-18 Westerink (against those who have the soul ‘completed’ by its association with the body): This kind of ‘completion’ is corporeal and enmattered, and posits a completion of the soul which follows the temperaments of the body... Proclus expresses in these passages the hostility of Platonism at large to the kind of reductionism in psychology evident to him in Galen’s QAM. But since, as we saw in Section V, the QAM is the very text in which Galen argues for the close association of reductionism and physiognomy, does this mean that when Platonists accept the possibility of physiognomy it must be under the assumption that it is the soul that is somehow responsible for the physical character of the body? There is, in fact, one Platonist thinker who goes some way towards such a position: the music theorist Aristides Quintilianus.?"® According to Aristides, the soul ‘wher: it is far from the things of earth and associating with the almighty’ is purely rational, and free of all desire (On Music 2. 8. 1-3). At a certain point, however, the time comes round for its descent into the corporeal world. A ‘need? arises in the soul for a body—and a ‘suitable’ body at that (2. 8. 3-5). As we shall see, this is familiar enough territory for a Platonist (though there is a slight eccentricity in the way Aristides characterizes the soul's desire as the desire for a specifically masculine or feminine body).””* Aristides is not even out on a limb when he goes on to suggest that things can go wrong for the soul, which raight fail to acquire a body suitable for it. Other Platonists worried about just such a possibility.?° But not only does Aristides seem to think that the 218 Artstides is known only through his work On Music (translated in Mathiesen 1983), from which he appears to be a Pythagoreanizing Platonist, probably writing in the late 3rd or early 4th c., though an earlier cate is possible. See Jahn (1882); Jan (1896); Zeiler end Mondolfo (1932-8) ifi/6. 1 n. 1 (associating him with the ‘scuola di Porfirio o fors'anche di Giamblico'}; Festugiére (1954} (suggesting, at 71, a pre-Plotinian milieu for his work); Zanoncelli (1977); Barker (1982); Mathiesen (1983) (10-14 for his life and date); Centrone (1983). Although, as we shall see, Aristides appears to support the possibility of physiognomy, he only uses the word once, at On Music 3. 8, 106. 5-6 Winnington-Ingram, where he calls the art of painting a ‘conversion (antistrophé) of the physiognomists’ art’. (The idea is that physiognomists use appearance to get at the underlying character, while the painter uses his knowledge of a person’s character in constructing his portrait. Mathiesen (1983} 172 n. 121 compares Ps.-Aristotle, Physiognomy 806°19°3; but Aristides might also, or rather, bave had in mind Socrates’ discussion with the painter Parrhasius described at Xenophon, 4. Memorabilia 3. 10. 3; and there is an even closer parallel with what Dic Chrysostom says at Oration 85-8, For these passages see respectively Section II and Section IV n. 124 above.) 219 On Music 2. 8, 16-18. The idea that a soul might be more or less ‘masculine’ or ‘feminine’ (or, strictly, in Aristides’ case, might be characterized by masculine or feminine desires) is one that appears already in Hippocrates (cf. Section V n. 167 above); within the Platonist tradition, cf. e.g. Proclus, On the Timaeus iti. 283. 12-284, 12 Diehl. 220 Bor the importance of the soul’s choosing a body suited to it (and the possibility of its failure to do so), Plotinus 3. 4. 5. 10-19; Proclus, On the Timaeus i. 163. 31-164. 3 Diehl; iii. 279. 14-15 athetized by Diehl}; Olympiodorus, Ox the Phaedo 7, 4. 8-9 Westerink. (but 115 resulting mismatch is not altogether uncommon, he also thinks (and here he does start to diverge from other Platonists) that when things do go wrong in the allocation of body, the soul can do something to put it right (2. 8, 66. 25-67, 3 Winnington-Ingram):*”" If souls do not naturally gain this kind of body [i.e. the kind they desire], they remodel it themselves according to their own characteristics (tais spheterais tropais) and make it into something like themselves {to hautais homoion). This is how men come to have a feminine appearance, in which one can see the kind of life they lead; and it is how, conversely, women come to have a masculine appearance, by which we conjecture that they hzve that sort of character (to éthos homoion). Thus there are smooth men and hairy women; and a man might have a melting look, a woman a steely gaze—so that you will find that the particulars of people's characters (&thé} are in harmony with their appearance -(morphé). It is, of course, a commonplace of physiognomy that a man might have ‘feminine’ characteristics, and a women ‘masculine’ characteristics”? but Aristides’ development of the theme is distinctive in the way that he reconciles the insight with a Platonic tripartition of the soul. In particular, he associates femininity of nature with the appetitive urges of the soul, a ‘masculine’ nature with the ‘spirited’ part of the soul, and encourages his readers to think about the myriad ways in which appetite and spirit might combine to constitute an individual’s character (2. 8, 67. 3-14 Winnington-Ingram):?*? From its passionate attachment to the masculine or feminine, or both, passions come to be established in the soul. The feminine is highly unrestrained, and the appetitive part of the soul (fo epithymétikon) answers to it; the masculine, on the other hand, is vigorous and active, and the spirited part of the soul (to thymikon) is like this. In the feminine part of the soul, and in feminine people, pains and pleasures proliferate; in the masculine, anger and boldness. Again, there are combinations of these things: of pains and pleasures; of anger and boldness; of boldness with pleasure and pain; and then of anger with each of them. And again, there are mixtures of each with one or more of the others, One could find thousands of images of these passions as they are considered in their variety. . Plato’s apparatus for psychological analysis is, in other words, translated here into the terms by which a soul imposes the nuances of its peculiar character onto 721 Plotinus talks at 3. 4, 5. 10-19 of a good soul having the capacity to improve a poor body {and a bad soul to make a good body worse); and, like subsequent Platonists, he thinks that the soul has come ability to calm the passionate inclinations of the body (e.g. 1. 2. 5, 21-31). This, though, seems rather more limited than the sort of change envisaged by Aristides, who is unique in implying that the appearance in the sense of the structure of the body will be affected by the character of the soul (though Plutarch talks of the soul imparting the colour of its passions to the body: On those to Whom Divine Retribution Comes Slowly 565 c; Table Talk 5. 7, 681 D). For ‘sexual’ characteristics in the Aristotelian tradition, see Ps.-Aristotle, Physiognomy 5, 809°26-810°13; Polemon, Leiden ch. 2, pp. 393-5; Anon. Lat, 3. Aristides does not develop a systematic table of correspondences whose dependence on this tradition we could check; but it might be wort noting that, whereas he opposes the kygros and the gorgos (in my translation ‘melting’ and steely’) as characteristically ‘feminine’ and ‘masculine’ ways of looking, Adamantius thinks that the two adjectives go together as a positive feature of the eyes (A16, B32, 44). syn abes eal Clement, Stromata 3. 13. 93. 1-3, where references to ‘male’ and ‘female’ in a of JeJesque persapel of the Egyptians 1 De Er are understood to mean respectively spirit

the essentially masculine or feminine body to which it was (as such) attracted. Here, for the first time in ancient philosophy, is a system, or the beginnings of a system, clearly designed to support a physiognomy led by the character of the individual soul. It also happens to be the last time: Aristides is exceptional among Platonists, as he is exceptional in the surviving evidence for physiognomical theory at large? This is not an accident of our evidence: most Platonists have very good reasons to doubt whether the soul can affect the character of the body. Indeed, the very independence of the soul makes such a position untenable. For the soul is generally taken to exist not just prior to the body whose soul it becomes, but independently of any particular body. It does not, for example, exist. actually or potentially in the individuals parents (as in some sense the Aristotelian soul might be taken to do). But in this case the soul lacks any real opportunity to direct the development of the new body--or so, at least, Porphyry argues in the ad Gaurum (esp. 2. 2-3). On the one hand, he says, it is absurd to suppose that the soul slips into the parental seed during sex: it would have to come in time to direct the conception and development of the embryo, but without being so respective natures, independently considered, make them ‘suitable’ for each other.?” This is how Plotinus describes it: 116 117 The souls of men see their images as if in the mirror of Dionysus and come to be on that level with a leap from above... (4. 3. 12. 1-2). Each soul comes down to a body made ready for it according to its resemblance to the soul’s disposition (kath’ homoidsin tes diatheseös). It is carried there to that to which it is made like (héi an homotötheisa ei)... (4, 3. 12. 37-8). There is a different time for each soul, and when its time comes, as if at the call of a herald, it descends and enters into the appropriate body (to prosphoron soma) (4. 3. 13. 8-10). Despite appearances, it should be noted that, strictly speaking, the ‘resemblance’ here does not refer in the first place to resemblance between the body and the soul—-that is, if we think of the scul in its truest sense as an incorporeal ‘mind’ or intellect. The soul in this sense does not necessarily have a character in terms of which it would make any sense to talk about its resemblance to this or that body in the first place. (Many Platonists believe that, at least the first time a soul ‘descends’ to a body, i: does so in a state of moral early that it could be thought to have come from one or other of the parents.” purity.)?”° Rather, the resemblance is primarily thet between the body and the already under way, and clearly does not require a soul to continue.” In short, if soul are thoroughly bound together). For the relationship of the individual soul with the preformed If, on the other hand, the soul were to wait until conception before it entered, it would be too late to be of any explanatory use: the physical process of growth is there is commensurability between body and soul, it cannot be (pace Aristides) because the soul in any way influences the natural shape of the body. All this adds up to what, prima facie, appears a dilemma for Platonists, who want to ensure the soul’s separability and autonomy on the one hand, while explaining its compatibility with and responsiveness to its embodied environment on the other; a tension, crudely put, between the best in the opposing models of psychological epiphenomenalism and teleology. Their response to it was to find a middle way. Instead of arguing that the fit between body and soul was explained, in one direction or the other, by their union, they argued that their union is explained by their fit. When a soul ‘descends’ to a body, it is because their 224 Plotinus notoriously appears to equivocate over the possibility that there are forms of individuals (cf. e.g. Blumenthal 1966}; and if (or when) he believes this (cf. esp. Ennead 5, 7), one might be tempted to think that an individual’s form lies somewhere behind an explanation of his or her psycho-physical peculiarities, But in fact Plotinus is clear that individual characteristics are due, not to individual form, but to the particular material circumstances of the instantiation of the species form (1. 8. 9. 11-14; 5. 9. 12, the clearest proof-text for a denial of individual forms; cf. also 5. 7. 2}. 227 Cf. also e.g. Porphyry, ad Gaurum 11. 2 (where sympatheia is the means br which body and body in Plotinus, cf. Blumenthal (1971) with Corrigan’s rejoinder (1985) (esp. 46 for what he argues is the Peripatetic vocabulary of ‘suitability’). Since the cheracter of the body is determined by its temperament, which in turn is susceptible to environmental influences, it is subject to change over time as weil: cf. e.g, Plotinus for the effect of diet (3. 1. 1. 33-6) and climate (3, 1. 5. 24-33; 4. 3. 7. 20-6). It is worth noting, then, that the ‘suitability’ of a particular body for a particular soul relies on a diachronic understanding of what it is to be a body: it is a life, not a moment, to which the soul descends. Cf. Plotinus 2. 3. 15; also Proclus, On the Timaeus i. 163 Diehl. Accounts of the path taken by the soul in its descent towards the body through the zones of the zodiac and the spheres of the planets is a way of explaining (or expressing) how its inclinations are commensurate not just with the synchronic state of the body at which it arrives, but with the diachronic fate of that body (eg. Porphyry 271F Smith; Macrobius, On the Dream of Scipio 1. 12; and for the influence that the stars have at least on one’s physical temperament, cf. Celsus ap. Origen, Against Celsus 6. 21; Plotinus 3. 1, 6. 3-11; 4. 3. 7, 20-3; Porphyry, Cave of the Nymphs 22-3 = Numenius fr. 31. 10-26 des Places; Produs, On the Republic ii, 54-65 Kroll). This explains how Piotinus can know something of a person’s (character and) fate from his or her body (2. 3, 7. 8-11), while at the same time denying (since the practice of philosophy gives one freedom from physical inclination) that astrology is determinative (3, 1. 5. 55-7 and 3, 1. 6. 3-11 with 2. 3. 12; also Olympiodorus, On the Gorgias 48. 5 Westerink; cf. Frocluz, On Providence 3. 14-18, 20). It also explaias why Proclus thinks physiognomy a ‘murky syllogism’ when applied to the newborn, whose temperament has yet to stabilize (amydros spllogismos. On Plato’s Cratylus 88. 6 Pasquali). Even if individual forms exist, A. H. Armstrong (1977) rightly argues that they would individuate at 7% On the image of the ‘mirror of Dionysus’, see Pépin (1970). 7 Cf, Alcincus, Didaskalikos 25. 6, 178. 36-7 Hermann (souls descending according to natural clearly in the case of the true, higher selves of individual men’: 58). They would not, in other words, be responsible for a person’s physical appearance—let alone moral character. 225 CÉ Longinus ap. Proclus, On the Timacus i. 51. 9-13 Diehl (= fr. 27 Brisson—Patillon); and sequence or divine will}; Celsus ap. Origen, Against Celsus 8. 53 (souls descending ‘for the a much higher ontological level (viz. ‘wherever a real formal difference can be detected, as it can be Proclus himself, ibid. iii. 322. 24-9. 26 Like the Stoics, to this extent, Porphyry believes that the embryo is in strict truth a plant, and that the soul does not enter the body until the moment of birth (the proof of this being, in fact, the main purpose of the work ad Gaurum). For the entry of the soul at birth, cf. also Porphyry 266F Smith (from lamblichus ap. Stobaeus, Eciogue i. 381. 2-6 Wachsmuth); Proclus, On the Timaeus ill. 322, 18-19 Diehl. administration af the world’); Jamblichus ap. Stobaeus, Eclogue ;. 380. 6-9 Wachsmuth (some souls “come down in order to bring salvation and purification and perfection to things in this world, and make an unsullied descent’); also Procius, On the Timaeus iii. 295, 15-20 Diehl (the moral purity of the soul on its first descent ensures that it will not enter the body of an animal), lamblichus, however, contrasts Cronius, Numenius, and Harpocration, who think that ‘every embodiment is evil’ (ap. Stobaeus, Eclogue i. 380, 14-19 Wachsmuth = Numenius fr. 48 des Places; cf. also frr. 32 and 38); and Plutarch too perhaps thought that every descent begins with something like carnal desire (Divine Retribution 565 D--566 A; cf. again also Aristides Quintilianus, On Music 2. 8).

soul’s ‘vehicle —the astral or pneumatic body with which. it is provided as a There are, of course, implications here for physiognomy—though somewhat negative, in the first place since, given that the soul might not be morally tainted when it descends to the body, there should be no absolute expectation that one could discern the moral character of the soul from the appearance of its body (though one might expect, less interestingly, to learn something about its 118 means of establishing its presence in the physical cosmos.’* For the body and the soul’s vehicle do share something in virtue of which they can be said to resemble one another: a material and irrational nature.” It is true that the body’s nature is the more intrusive, as its greater density and complexity makes it more sensitive to external forces and more determinedly inclined to act in certain ways in respect of them. The body’s behavioural inclinations are, indeed, strong enough to be considered quasi-moral, and even carne to be classified as ‘virtues’ and ‘vices’ of a kind, albeit a low-level kind.”*? Nevertheless, the vehicle has a certain (irrational) sensitivity too; and it is in terms of the congruence between this and the irrational nature of the body that we can understand the attraction between a particular soul and a particular body. The soul, in short, is attracted to a particular body by the ‘fit’ it recognizes between the body which it is due to contro! and the vehicle through which it will exercise that control??? 230 jamblichus claims that ‘most Platonists’ think that the soul joins directly with the body (ap. Stobaeus, Eclogue i. 385. 1-10 Wachsmuth), though in fact, our evidence includes only two notable exponents of such a view: Plutarch (for whom cf. n. 231 below); and, apparently, Alcinous, 25. 7, who seems to think that the reots of irrational desire are part of the soul itself (cf. Didaskalikos 178. 39-46 Hermann), actualized when it swaps its star for an earthly body as its ‘vehicle’ (16. 2, 172. 2-19 Hermann). Note, however, that where there is no danger of amb:guity it is quite usual for Platonists to apply the term ‘soul’ to the combination of soul (properly speaking) and vehicle; or even to the vehicle on its own (ef. Plotinus 1. 1. 12. 21: alle psychés eidos). For Numenius, at the other extreme, the vehicle actually is another soul, of the same substance as the evil world soul posited by his dualism: see esp. fr. 44 des Places. 351 Cf esp. Iamblichus ap. Stobaeus, Eclogue i, 385. 1-10 Wachsmuth ‘but for the text see Festugière 1953: 237 with nn; cf. Finamore and Dillon 2002 ad loc, with discussion at 184-5); Procdus, On the Timaeus iii, 234. 8-237. 6 Diehl, variously encompasting the views of Albinus, Atticus, Porphyry, and Syrianus. Proclus himself posits two vehicles: one properly irrational and perishable; the other containing ‘roots’ of irrationality (akrotétes tes alogou 2685}, a view which Dodds (1963) 307 thinks might have been suggested by Plotinus 6. 7. 3 ff; but cf. already Alcinous, Didaskalikos 25. 7, 178. 39-46 Hermann. In Plutarch’s distinctive psychology, the individual soul is already involved with the material and the irrational since it represents the organization of true soul, which for him is irrational (On the Gerieration of the Soul 1014 pe). Although its activity is made essentially intellectual through its organization by the Demiurge, it goes on to acquire an irrational nature when and insofar as it descends into the body and becomes like it (exomoiousthat: On Moral Virtue 4508; On the Sign of Socrates 591 DE). 232 ‘The notion of ‘physical virtue’ (and vice) becomes part of a standard Neoplatonic scheme with lamblichus, assuming him to be behind the whole of Damascius, Commentary on the Phaedo (versio 1), 8813843 Westerink. Cf. later on, and just for example, Olympiodorus, On the Phaedo 8.2. 1-7; Nemesius, On the Nature of Man 2, 25. 20-4 Morani; and already Alcinous, Didaskalikos 1.4, 152. 23-9 Hermann; Apuleius, On Plato 2. 6. All of this formalizes a distinction that can be found readily enough in Plato: e.g. Republic 5180-519 A. 233 ‘This is sometimes expressed as the allocation by Providence of a soul to a particular body, which in turn allows the process to be described as a matter of the soul’s ‘punishment’ of ‘cure’. C£ eg. Jamblichus ap. Stobaeus, Eclogue i. 380. 9-14 Wachsmuth; also Proclus, Commentary on Plato’s First Alcibiades 94, 4-15 Westerink quoted below in the text. There is of course no contradiction with of the idea that the process is also natural, since nature is understood to be an instrument providence, Compare also Origen, On First Principles 1. 8. 1; Against Celsus 1. 32-3 (better souls are allotted better bodies according to certain ‘esoteric principles’). Note that this latter passage is supported by appeal to Zopyrus, Loxus, and Polemon, who, in Origen’s report, thought that ‘all bodies are appropriate for their souls’ (1. 33. 14). (This is scarcel evidence for close engagement 119 vehicle). However, it turns out that the soul is rather rarely innocent of moral character when it descends. Most incarnations are reincarnations, and motivated precisely by the fact that a soul in its previous life acquired an attachment to the material—acquired a moral character.””* And in this case, the way to physiognomy seems to be opened up at last; because the moral disposition acquired by the soul will precisely be a reflection of the non-rational character of its vehicle, whose similarity with the new body again explains their union. This is inevitably true, because it is in the soul’s identification of itself with the body it previously inhabited, mediated through the congruent passions of the vehicle, that the soul’s character is established. It needs to be emphasized that there is no external necessity that can lead the soul to acquire a tainted moral character. Indeed, the soul, as an essentially intellectual and incorporeal entity, is normally (certainly after Plotinus) taken to be strictly impassible throughout: immune to the shocks which affect the body and the passions they arouse in it. Nevertheless, as the soul is aware of what is happening with the body, what tends to happen is that it becomes so engrossed in it that it starts to mistake the passions of the body for its own: It is as if someone standing on a promontory should cause his image and shape to appear ina flowing river, keeping his face still while the stream as it moved changed his image every which way so that it presented it in different ways at different moments—skewed and straight, fragmented and whole, as chance would have it. Seeing this, the man might think that he was really undergoing these things, though he was only looking at his shadow in the water; and the thought would make him anguished, disturbed, perplexed, frustrated. The soul is in the same case when it sees its image in the body being carried along in the river of becoming and changed all the time by passions arising from within and by external forces, The soul is actually unchanged, but it thinks itself changed because it does not know itself and thinks that it is its own image. It becomes disturbed, perplexed, and bewildered. (Proclus, On the Timaeus iii, 330, 10-24 Diehl.) When the soul does this, when it comes to identify itself with the physical passions of the body, we can talk of it as having, to this extent, a moral character And the moral character it acquires, its desires and concerns, etc., will of course be precisely those of its body, and (what amounts to the same thing) of its vehicle, through which it perceives the body. Furthermore, since the soul retains its vehicle after death, it will retain the character it acquired when during life it came to identify itself with it. When it next descends to a body we will no longer vei edtheory of physiognomy en Origen’s part, by the way, but it does show an awareness of the = which is unusual among Christians—cf. above n. 210. It also helps in general io confirm the evance of physiognomy within theoretical discussions of body and soul.) 5 _ And so the soul might be said to be inclined to (re)incarnation (sc. in . i a certain sort of bod: u a passions: cf. Plutarch at Divine Retribution 566À and 567 A; Celsus ap. Origen, Against pha 33; cf. Proclus, On the Timaeus iii. 293. 26-9 Diehl,

the soul and the be able to make a distinction between the moral character of The similarity thing. same the become have inclinations of the vehicle: they will not only be a will d attracte next is it which to body the and soul between the but also a body, the of that and congruence between the character of the vehicle moral er—the charact the and body the of ions congruence between the inclinat character—of the soul. l nature Since Platonists came to rely for their understanding of the physica all of Galen, on arly particul and texts, medical on body and development of the on n) directio t this represents a convergence of sorts (albeit from a very differen The QAM. the in Galen by ed describ the reductionist position we have seen n of its physical Platonist position is that the body’s character is a functio is not perfectly it ng temperament, and that the soul’s moral character, assumi then, that follows, It body. the with virtuous, is acquired through identification d by the acquire er charact moral the that say all after can in a certain sense we sees an odorus soul is consequent on the body's physical temperament. Olympi t differen on takes water same the analogy for the process in the way in which flows it which through earth the of nature the to characters according dependence dependen Mule of temperament, and so character, on the environment (cf. QAM 57. 120 (Commentary on Aristotle's Meteora 148. 214 Stüve): ng temperament of the Consider the case of souls, which change according to the underlyi is present, on the other body if they don’t huve reason in charge of the reins. When reason hand, the soul does not serve the body's temperament. 121 Given the explicit aversion we have seen to Galen’s reductionism in the writings of Proclus, this might seem surprising. But the invocation of Galen does not, of course, now suggest approval of the claim that the soul, or any part of the soul, is to be identified with physical temperament; nor, more generally, that the character of the soul is somehow determined by the body. Such a ‘strong reading of Galen’s thesis has been roundly rejected. Nevertheless, it is open to Platonists to appropriate Galen’s language for another, weaker reading of the relationship between body and soul, one which suits their purposes rather well. When Galen says that ‘the powers af the soul follow the temperaments of the body’, it is possible to read ‘follow’ in a sense less than deterministic: to use Galen’s considerations in order to support a model according to which although the soul remains independent of the body, it can nevertheless Choose, by the acquiescence of reason, to ‘follow’ the body—i.e. to identify its interests with it. Proclus himself, therefore, despite his rejection of a strict reading of the opening line of Galen's QAM, can now, with suitable qualification, quote it more favourably (On the Republic L 222. 14-15 Kroll, referring back to Il. 4-6, where the Galenic quotation is unabbreviated): | In the uneducated, the powers [sc. of the soul] foilow the temperaments [sc. of the body]. The same allusion (with the same qualification) is to be found in , more or less, Plotinus too talks about how the soul might lose its freedom n to the body’s relatio in r stronge are s when it enters the body. Then its passion ant soul to compli the compel can which 28-38), 8. 8. (1. physical temperament and power wealth at feel lust, anger, wretchedness in the face of poverty, or pride Olympiodorus (On the Gorgias 49, 6. 7-9 Westerink): in general have their irascibility (4. 3. 23. 42-7; 4. 4. 28. 40-5, 49-52); desires specifically the area 42-5), origin in the body (4. 4. 20. 28-36; 4. 4. 21: 4. 8. 2. Commentary on the De Anima (xv. 50. 16 ff. Hayduck). These doctors are in fact invoked twice: first after a survey of ways in which passions seem to be the result of physical states (including disease or diet at 50. 16-51. 3); and then in a discussion of various ways in which mental operations too seem to be determined or affected by the state of the body and the nature of diet (51. 3-51. 15). On both occasions (with an insignificant change of word-order) Philoponus says: (3. 1. 8. 14-20) A particular quality of the bile and blood might around the liver (4. 4. 28. 17).?° All this (to pick up another explain theme we have seen helps explain why to be important in the discussions of Aristotle and Galen) as physically (2. 3. well as y children tend to take after their parents psychologicall 12, 14. 5-9); it also explains how environmental factors, which, as we saw a person’s moral above, affect physical temperament, might thereby influence in many ways to open is soul ‘the development as well. When Plotinus says that the locations of and air; and waters and places of being moulded by the nature cities differ as do the temperaments of bodies’ (4. 3, 7. 22-5), there might be an demonstrating the allusion to Galen’s reading of the Hippocratic AWP as a text kt is said that the powers of the soul follow the temperaments of the body; but we must add unless philosophy counteracts their effect’. Galen, similarly, must lurk behind the ‘doctors’ attacked by Philoponus in his on why the doctors have said that the powers of the soul follow the temperaments of the Similarly at 155. 314: - souls are suited for different things, and have tendencies to particular passions, is because of their relationship to the body and their sympathy with it—which is why the body | people have thought that the powers of the soul follow the temperaments of d to some degree by the 25 CE Plotinus I. 8. 14; 4. 8. 2. 42-5. Even intelligence can be hampere obstructs body; but it is very important that the embodiment precisely the soul to differing degrees in this case: we should not wish to say that the power of thought was given by the body in various on memory in particular; also Longinus degrees—pace Galen, of course. Cf. Plotinus 4, 3. 26. 12-14 atilion (part); Proclus himself at On the at Proclus, On the Timaeus i, 162. 15-27 =fr. 36 Brisson—P Timaeus iii, 349. 28-350. 8 Diehl. 236 Cf generally 2. 3, 8. 13-16; 4. 4. 28, 40-5; 4. 7. 237 And remember that Galen himself equivocates over the sense he gives to ‘follow’ in claim that ‘ ‘the powers of the soul follow the temperaments of the body’: cf again Section V

And, finally, cf. 183. 30-4, which shows that they are, to this extent and with Socrates {though one might be tempted to find more, unsignalled examples in the biographical works of Plutarch).2® Significant too is the real emphasis placed on the ability of physiognomy to discover the character of someone’s soul in the sense of his or her potential: potential as a Pythagorean recruit, for example; or potential for learning from Socrates. The fact that Socrates picks out Alcibiades as well as Plato as promising youths reminds us (as I said) that potential can go unfulfilled. This is important because there are aspects of Plato’s inheritance which stop the Platonists from spending too much time on the subject. For a start, if I am right that ‘physiognomy’ properly describes the discovery of innate psychological character (rather than, for example, character acquired under the influence of the body), then the term should not be used of 122 the appropriate qualification, right to think it: When intellect, the natural leader, is not in charge, order is overthrown and the powers of the soul follow the temperaments of the body.’”" People who have a peculiar moral character have it because they have acquiesced in the inclinations of their bodies, and to this extent their character is explained by their temperament—by the very feature of their physiology which, as Galen described in the QAM and as the Platonists must therefore have been fully aware, was responsible for the appearance of their body as well. To the extent, then, that most people fail to preserve the original purity of their souls, and identify themselves with their bodies, we can understand how it is that a person's moral character will be predictable from his or her appearance: both are unde:written by bodily temperament through a process similar in crucial respects to the mechanism spelt out by Galen. Of course, a person can improve, more or less, through philosophy, or become worse by embodiment in a life that allows wider scope to develop his or her passions. The change of moral state (of the degree of identification with bodily passions) will, in any case, be carried forward after death, to form the basis for the next choice of life. Perhaps the most striking and specific example of physiognomy in a Platonist text comes (zgain) in Proclus—his description of Socrates’ selection of Alcibiades as a pupil. Alcibiades, it seems, had through previous lives acquired a moral character close to perfection, and in receiving now a body appropriate to its nature enabled Socrates to see that he could be helped towards the ultimate achievement (Commentary on Plato's First Alcibiades 94, 4-15 Westerink): Socrates saw many wonderful indications in Alcibiades that he was capable of virtue. His very size ancl beauty were signs of the outstanding power. strength, and preeminence of his soul; but nature, who crafts the body as a tool {sc for the soul], attached further symbols of these traits to his appearance as well. When Socrates saw these, he understood that the boy was worthy of his care and attention. He derived this custom from the Pythagoreans, who would judge whether those who came to them were capable of the greater life on the basis of physical signs. The idea is that nature constructs bodies which are suitable for the particular souls which will use them as their tools, and sets out images clever at (eikonas} of these souls in the bodies, Through these images, someone recognizing such things can uncover the individual characteristics of particular souls. Socrates was disappointed, of course; but he was na more wrong to see a noble natute in Alcibiades than Phaedo’s Zopyrus had been to see a vicious character in Socrates. Alcibiades really did have a promising soul, which earned hire. a wonderful body. He squandered his opportunity. I: is necessary to end this section with a word of qualification. As I suggested at the beginning, there is not a great deal of physiognoray in Neoplatonist works, and the range of such physiognomical passages as we have is surprisingly narrow, the bulk of it being constituted by reports of its use by Pythagoras and 238 Cf On the De anima xv. 52. 7, 388. 25 Hayduck; On the Physics xvi. 191, 18 Vitelli. For the same claim, but in language not so obviously alluding to Galen, Simplicius, On the De anima 4-4 Hayduck. xi. 19. 123 pure’ souls in their first descent to the body, even when (as often) they quickly succumb to the body’s influence. It should instead be restricted to subsequent cases of descent where the soul has a moral character independently of the body to which it is joined. But it is, of course, hard to know which is which, And in any case, the practical focus of Platonist ethics demands that we concentrate on the soul’s potential for purity, its capacity for free action, undetermined and uncoloured by the body. Physiognomy as a practice thrives on the suggestion that the soul is constrained by the body, but we should actually be concentrating our thoughts on the freedom we have to release ourselves from it: the freedom not to identify ourselves with passions that have their reots in the body (cf Plotinus 1. 2. 5, 12-15; 1. 2. 5. 21-31; 3, 1. 7. 13-20; 3. 6. 3}, a freedom which we acquire when we turn away from the body and back towards the intellectual realm (Proclus, On Providence and Fate 20). The Platonist soul might 2% As do e.g, Misener (1924) 110-11; Evans (1935) 57 with n. 3; {1941} 104-5; Métraux (1995) 3-4, On the other hand, Sassi (1992) concludes that Plutarch is antifistognomico—that he is aware of and uses ‘physiognomical’ associations on some occasions, but deliberately avoids or subverts them on others. Sassi’s view is undoubtedly based on a more sophisticated view of Plutarch; nevertheless it might be too pessimistic. We know that Plutarch had a mechanism by which he could explain the union of soul and body by their similarity (see again n. 231 above); we know that other Platonists accepted the possibility of physiognomy under these conditions; and we know, finally, that it is quite standard for a philosopher to accept physiognomy as a means of ascertaining a person’s innate disposition without believing that behaviour is in any way determined by physical constitution. This could account for the practical ‘ambivalence’ in Plutarch’s use of physiognomical evidence as outlined by Sassi; it would also explain how he could, without inconsistency or opportunism (pace Wardman 1967), argue in one place that sculpture cannot capture a man’s ‘character and way of life’ (to ethas kai jon we Cimon 1. 2-3), yet elsewhere find statues that do reflect his subject's character (as Marius 7 ‚and can even use sculpture as evidence for character (Themistocles 22, 3; Aratus 3. 2; cf. perhaps exander 1. 3). It might simply depend on the person and whether his or her success involved epercoming or exploiting innate disposition. For the record, then, Plutarch finds reflections of Solla (2 Fi the looks ofthe following (with references to the eponymous Life unless otherwise stated): oho „1, 6. 6-7; cf. his meeting with a Chaldean, characterized, though not described, as a u oe at 5. 5-6), Pompey (2. 1), Alexander (4. 1-2; cf. Pompey 2. 1), Timocleia (Alexander (For em meus (2, 2-3), Antony (4. 1}, Pyrhus (3. 4), Marius (2. 1), and Aratus (3. 2). pbsBr nass other characters whose physical features are described are: Cocles, Publicola 16. De mistocles at 22. 3; Pericles at 3. 2; Alcibiades at 1. 3-4; Lysander at 1. 1-2; Agesilaus at 2. 2; es, Aratus 20. 1; Galba at 27, 2; also Cyrus, Precepts of Statecraft 821 8. Less pertinently, or i facial expression Clearchus’ wart at Table-Talk 620D; Sertorius’ single eye at 1. 4; Fabius Maximus’

become brutish, literally or metaphorically;“° but Platonist ethics is peared towards making us see that this is an eradicable fault. of the body which In short, then, Platonists appear to accept that those aspects also give rise to and er charact lar particu a with soul a attract time at the same discernible have und compo living ng particular desires and passions in the resulti to be. tend will people how see to t gnomis physio the tokens, and thus allow body their them: to up ely However, their actually being this way is ultimat r howeve and, tion incarna of time the at soul at best mirrors the state of their and form self-re ogical difficult it makes things, cannot prevent psychol physiognomy in a way purification. Platonists clearly accept the possibility of onal for reasons of provisi be to sions conclu its take they but that Plato does not; on the freedom focus they as which he would thoroughly approve. In proportion gnomy. The physio of lities practica the about little of the soul, so they say very s, not down to upward mind the trate concen to is ophy philos this of whole point the compound about which physiognomy speaks, species boundaries, or whether Plate #9 On the question of whether transmigration crossed ation of the wicked in animals, reincarn the of talked he was speaking metaphorically when 13-20 Norvin; Nemesius, On the Nature of see e.g. Olympiodorus, Commentary on the Phaedo 124. of (rejincarnation outlined in this section, Man 2. 34. 24-35. 11 Morani. Given the general account if it occurs, fulfils the conditions that ation, reincarn animal that case the be to it expect one might arimal psycho-physically suitable for an enter will person a i.e, : possible nomy make human physiog might his or her moral state. If this is right, then Platonists who believed in animal transmigration a study of animal species in collecting have used the fact to guarantee the relevance of anyone did this. physiognomical signs. But we have no evidence that d at seminars in Durham, Reykjavik, Earlier drafts of material from this study have been presente Austin, Texas: my thanks to the and Studies), Classical Warwick, London (at the Institute of as well to Simon Swain for asking me to organizers and audiences in each case. I am very grateful his Rowe, both for his comments, and for think about this interesting topic; and to Christopher Nutton and Philip ven der Eijk Vivien done. was work the of most where support at Durham, ly improved in the light of their generous suffered a too early draft of Section V, which was great comments. 3 Polemon’s Physiognomy Simon Swain This chapter offers an interpretation of Polemon's Physiognomy in i century context. Why was it written and how did contemporaries read À om the perspective of their own cultural horizons? The Physiognomy survives in two complementary versions, Adamantius’ Greek abridgement and the Leiden Arabic translation. The closeness of these makes it reasonable to follow Foerster in holding that Adamantius’ Greek text is very similar to Polemon’s own. In the previous chapter George Boys-Stones has explored the earlier literature em which the Physiognomy builds. Polemon was influenced by Ps.-Aristotle and (perhaps) Loxus and no doubt by other lost writings and oral traditions, But his work is very far from being a summary of his predecessors, Rather, Polemon develops s thoughts within the cultural system of his day. He offers his readers a physiognomical handbook which is illustrated by a range of examples drawn from his peers. He runs through the parts of the human body, paying especial attention to the eye, and maps onto these parts or sub-parts particular moral and political qualities. The examples of real enemies and acquaintances which survive in the Leiden (especially in the section on the eye and in the final chepters only preserved in the Arabic) put flesh on the bones. This observation : morally bad and (to a far lesser extent) good types places the Physiognomy rmly within the culture of inspection and moral evaluation that is visible in so much of the literature of Polemon’s time and which is arguably a product of the competition and rivalry of elite life. By grouping appearances and qualities the 1 siognomy may even be read as implicitly telling readers what they should be = just as contemporary moralists make their descriptions of anger, greed. Im so on the basis of self-reflection and moral improvement.! In addition, the Mae ae provides important information on Polemon’s attitude to Rome vet e emperor Hadrian, and in general terms offers valuable commentary on BRi aracter of political and civic life in the Greek world of the second century. Pr:€ culture of observation and evaluation among Polemon’s peers is the focus a 5 Fes Inge of this chapter. Section 1 introduces the links between the wits i" Politics, the phenomenon of classicism, and the competitiveness of h e. Section II looks at particular issues of moral conduct, specifically the egulation and control of male behaviour including correct sexual relations and , Absence of Anger; below, 138-40, 191-2 for anger, etc. below, 195-7. Moralists: in the Cf.Physiognomy. oralists: Plutarch’s

Bibliography ABBREVIATIONS Ad. = Adamantius, Physiognomy Anon. Lat. = Anonymus Latinus, Physiognomy ANRW= Aufstieg und Niedergang der römischen Welt Aristotle: DA= De anima GA = De generatione animalium HA= Historia animalium NE= Nicomachean Ethics PA= De partibus animalium CMG = Corpus Medicorum Graecorum CPh= Classical Phiiology CQ = Classical Quarterly DL=Diogenes Laertius, Lives of the Philosophers BI=H. A. R. Gibb et al, The Encyclopaedia of Islam’, 12 vols. (Leiden, 1960-2004) EK=L, Edelstein and I. G. Kidd (eds.), Posidonius, i: The Fragments (Cambridge, 1989) FHSG=W. W. Fortenbaugh, P. H. Huby, R. W. Sharples, and D. Gutas (eds), Theophrastus of Eresus. Sources for his Life, Writings, Thought and Influence, 2 vols. (Leiden, New York, and Cologne, 1993) Freytag =G. W. Freytag, Lexicon Arabico-Latinum, 4 vols. (Halle, 1830-7) Galen: QAM = Powers of the Soul (Quod animi mores corporis temperamenta sequantur) PHP= de Placitis Hippocratis et Platonis GALex=G. Endress and D. Gutas, A Greek and Arabic Lexicon (Leiden, 1992-) GAS=F. Sezgin, Geschichte des arabischen Schrifttums (Leiden, 1967-) Hippocrates: AWP:= Airs, Waters, Places IGR=R. Cagnat er af, Inscriptiones Graecae ad res Romanas pertinentes (Paris, 1906-27) IK= Inschriften griechischer Städte aus Kleinasien (Bonn, 1972--) JAOS= Journal of the American Oriental Society Kazimirski = A, de Biberstein Kazimirski, Dictionnaire arabe-frangais, 2 vols. (Paris, 1860) Lane=E. W. Lane, An Arabic-English Lexicon (London, 1863--93) LGPN=P. M. Fraser et al, A Lexicon of Greek Personal Names (Oxford, 1987-) LIMC= Lexicon iconographicum mythologiae classicae, 9 vols. (Zurich, 1981-97) OLD=P. G. W. Glare, Oxford Latin Dictionary (Oxford, 1968-82) Payne Smith =R. Payne Smith, Thesaurus Syriacus (Oxford, 1879-1901) PCPhS = Proceedings of the Cambridge Philosophical Society Philostratus: VS= Vitae sophistarum

Bibliography Bibliography PIR = Prosopographia Imperii Romani (2nd edn.) SA = Quaderni di studi arabi | hare Pauly, G, Wissowa, and W. Kroll, Altertumswissenschaft (Berlin, 1893-1980) RhM = Rheinisches Museum SE = Sextus Empiricus PH = Pyrrhoniac hypotyposes SSR= 6 Real-Encyclopadie _ der classischen | SVF= Stoicorum Veterum Fragmenta, ed. H. von Arnim, vols. i-iii ( Stuttgart, 1903-24); vol. iv (indexes) by M. Adler u TAPhA = Transactions of the American Philological Association (Irvine, Calif, 1995) | Wehr=H. Wehr, A Dictionary of Modern Written Arabic‘, ed. J.M. Cowan (Wiesbaden, 1979) Aba l-Fath al-Karajaki (1985), Kanz al-fawäïd (Beirut). | a Abi Nu‘aym al-Isfahäni, Ahmad ibn ‘Abdallah (1985), Hilyat al-awliya, 12 vols, (Beirut). Abt Nuwas (2003), Diwan, vol. 3, ed. E. Wagner (Beirut). oo Amshatt, al-, al-Qawl al-sadid fi khtiyar al-ima’ wa-\‘abid, Cambridge University Library, MS Or. 1023 (7). . Anonymus, Fi I-firäsa, Damascus, Zahiriyya Library, MS 172 (slim 7840). Aristatalis, see Ghersetti (19994). Avicenna, see Ibn Sind. 2 Pack (1974) i , De physiognomonta libellus, see Pac . \ | tn ‘bn al Husayn al- (1981), al-Frigäd wa-l-hidäya ila sabil al-rashad d. A. Sagr, 2 vols. (Cairo) — (1971), Manägib al-Shafti, ed. A. Sagr, 2 vols, (Cairo). _ Baker, Muhammad ibn Ismä’d al- (1360-84/1941-64), al-Tarikh al-kabir, 8 vols. (Hyderabad). | __ 118621908), Sahih, ed. L. Krehl and T. W. Juynboll, 4 vols. (Leiden). Dhahabi, Muhammad ibn Ahmad al- (1380/1961), al-Tibb al-nabawi (Cairo). —— (1981-8), Siyar alam al-nubal@, ed. Shu‘ayb al-Arna’ut, 25 vols. (Beirut). Dimashqi, Shams al-din, al-Ansari al-, Kitab al-firasa, Bursa Inebey i! Halk Kütüphanesi, MS Hüseyin Celebi 882. Kitab en À tlm al-firäsa, Bibliothèque Nationale de France, ms. arabe 5928, fos, I-45, Huntington 534, fos. 347°-500?, Ibn Abi Shayba (1989), Musannaf, ed. . (1299/1882), Kitab al-siyasa fi ‘ln al-firasa (Cairo, repr. 1332/1914). | (1983), Kitab al-siyäsa fi tlm al-firasa, ed. Muhammad ibn al-Sufi, 5 vols. (Cairo). Fakhr al-Din al-Razi, Kitab al-firäsa, ed. and trans. in Mourad (1939). Ghazali, Muhammad al-, Hidäyat al-murid fi taglib al-abid, see Harun (1411/1991) i 421-42, 7 \ _ Ghazali, Aba Hamid Muhammad b. Muhammad (1992), Ihya’ wlüm al-din, ed. M. al-Dali Baltah (Beirut). | Gilles de Corbeil (1864), De physionomiis, in V. Rose, Anecdote Graeca et Graecolatina Berlin; repr. Amsterdam, 1963) i. 177-201. _ nm an re known as Kätip Celebi} (1835-58), Kashf al-zunün ‘an asämi 4 | wa-l-funûn, ed. G. Flügel (Leipzig and London; repr. New York and London, 1964). (1941-7), Kashf al-zunün, ed. M. S.-D. Yaltkaya (Istanbul). Library, MS Arab. 456, 6 — §. M. al-Lahham, 9 vols, (Beirut). Ibn Abr Usaybi‘a (1884), Kitab ‘uytin al-anh a’ fi tabagat al-atibbä, ed. A. Müller, 2 vols. in 1 (Cairo; rev. edn, Königsberg, 1884; repr. Westmead, 1972), —— (1965), Kitab wyün al-anba’ fi tabaqät al-atibb@, ed. N. Rida (Beirut). Ibn al-Akfäni, Tonal al-siyasa fi ihn al-firasa, Bibliothèque Nationale de France, ms. arabe ——- al-Nazar wa-F-tahaig fi taglib al-ragiq, Bibliothèque Nationale de France, ms, arabe 2762, 2234 (3), fos. 148°_150°, —— Irshad al-qäsid ila asnä al-magäsid, see PRIMARY SOURCES ape Haythami, ‘Ali ibn Abi Bakr al- ( 1987), Majma ' al-zawaid, 10 vols. (Cairo and Beirut). Hunayn b. Ishaq, Risala Fi dhikr ma turjim a min kutub Jalinas bi-Tmihi wa bad ma lam Yutarjam, see Bergsträsser (1925). —— (1978), al-Masaäl À Ltibb Li-L-muta‘ai limin, ed. J. M. Musa (Alexandria). Ibn ‘Abd al-Barr (1991), Al-istt ab fi matifat al-ashab, 4 vols. (Beirut), ibn ‘Abd Rabbihi ( 1983), al-Igd al-farid, ed. ‘A. M. al-Tarhini, 9 vols, in & (Beirut). Ibn Abr l-Ash'ath, Kitab al-hayawan, Oxford, Bodleian eee Socratis et Secraticorum Reliquiae, 4 vols, (Naples, 1990) TLG= Thesaurus Linguae Graecae CD Rom # E 665 Witkam (1989). Ibn ‘Arabi, Muhyi I-Din (1293/1876), al-Fu tühat al-makkiyya, 2 vols, (Cairo). ——- (1919), al-Tadbirat al-ilahiyya ft islah al-ma mlakat al-insäniyya, ed. H. 5. Nyberg, in Kleinere Schriften des Ibn al-Arabi (Leide n), 161-76; trans, M. J. Viguera, in Dos cartilias de fisiogndmica (Madrid, 1977), 31-70 . —— (1957-8), Ahkam al-Qurän, ed. “A. M. Bajawi (Cairo). Ibn (Bar) Bahläl, Kitab al-dala’il, Istanbul, Millet Kütüphanesi, MS Hekimoglr 572/1. ~— (1987), Kitab al-dalail, ed. J. Habbi (Safat) . Ibn Butlan, Risäla jamia H-funan nafta ft shirä al-ragiq wa-taglib al-‘abid, see Harun (1411 /1991) i, 381-420; German translation in H. Müller (1980) ; Italian translation in A. Ghersetti, Trattato generale sulfacguisto e Pesame degli schiavi (Catanzaro, 2001), Ibn Hamdün (1996), Tadhkira, ed, L “Abba s and B. ‘Abbas, 10 vols, (Beirut). Ibn Hazm al-And alusi (1914), Tawq al-hamama, ed. D. K. Petrov (St Petersburg and Leiden); trans. A, J. Arberry, The Ring of the Dove (London, 1953). Ibn Hisham, ‘Abd al-Malik (1858-60), Sirat rasal Allah, ed. F. Wiistenfeld (Göttingen). Ibn al-Jaw zi (1970), Akhbar al-adhkiya, ed. M. M. al-Khuli (Cairo). — (1988), Akhbar al-hamga wa-l-mughaffali Ibn al-Jazzar, Zad al-musäftr, Sächsische n (Beirut), Landesbibliothek — Staats - und Universitätsbibliothek Dresden, MS arab, E 209, Ibn Juljul (1955), Tabagat al-atibba’ wa-Ihukama’ (Les Générations des médecins et des sages), ed. F. Sayyid (Cairo), Ibn Kathir, Ismail ibn ‘Umar (1937), Tafsir al-Qurän, 4 vols. (Cairo). Ibn Khallikän, Ahmad ibn Muhaminad (1842) , Wafäyat al-a'yan, ed, I. ‘Abbas, 4 vols, (Beirut, 1972); trans. Baron de Slane, Ibn Khalli kan’s Biographical Dictionary, 4 vols. (Paris, 1842-75), Ibn al-Mugaffa‘ (1993), al-Adab al-kabir (with al-Adab al-saghir; publ. Beirut, 1960), trans. J. Tardy in Annales islamologiques, 27: 181-223. Ibn al-Nadim, Muhammad ibn Ishaq (18712), Kitab al-Fihrist, ed. G. Flügel, 2 vols. (Leipzig). —— (1971), Kitab al-Fihrist hil-Nadim, ed, R. Tajaddud (Tehran). Ibn Qayyim al-Jawziyya (1961), al-Turug al-hukmiyy “Abd al-Hamid and A. ‘AH. al-‘Askari (Cairo a fi Lsiyäsat al-shar'iyya, ed. M. M. ). Ibn al-Qift i, ‘Al ibn Yasuf (1903), Tatikh al-hu kama Jon Qutayba, ‘Abdallah ibn Muslim (1900), Adab al-kated, J. Lippert (Leipzig). ib, ed. M. Grünert (Leiden).

Bibliography Bibliography —— (1925-30), Uyün al-akhbar (Cairo); book 1 trans. J. Horowitz in Islamic Culture, 4 (1930) 172-98, 331-62, 487-530; 5 (1931) 1-27. Ibn Ridwän, “Ali, (1986), al-Kitäb al-naff fi kayfiyyat talim sinänt al-tibb, ed. K, al-Samarra’i (Baghdad). Ibn Sina (1298/1880-1), F5 agsäm al-ulüm al-agliyya, in Tisa rasaïl ff I-hikma wa-l-tabriyyat (Istanbul). —— (1954), Qissat Hayy ibn Yagzan, ed. and trans, in H. Corbin, Avicenne et le récit visionnaire (Tehran and Paris). (1406/1985-6), al-Shifé, al-Tabtiyyat, viii: al-Hayawan, ed, 1. Madkur et al. (Qum; repr. Cairo n.d.). Ton al-Washshä’ (1886), Kitab al-muwashshä, ed, R. Brünnow (Leiden). Ibshihi, Muhammad ibn Ahmad al- (1999), al-Mustatraffi kull fann mustazraf, ed. 5. M. al-Lahham (Beirut). Isfahänt, Aba 1-Faraj al- (1927-82), Aghani, 16 vols, (Cairo). Jahiz, ‘Amr ibn Bahr al- (1932), al-Tabassur bi-I-tjara, ed. H. H. “Abd al-Wahhab in Revue de l'Académie arabe à Damas, 12: 321-51. —— (1938-45), al-Hayawan, ed. ‘A.-S. M. Harun, & vols. (Cairo). —- (1955), al-Tarbt‘ wa-l-tadwir, ed. C. Pellat (Damascus). (1956), Dhamm al-kuttab, trans. C. Pellat, ‘Un charge contre les secrétaires d’État attribué à Gahiz’, Hesperis, 43 29-47. (1964), Ras#il, ed. ‘A.-S. M. Harun (Cairo), including: Manägib al-Turk, i. 5-86; Kitman al-sirr wa-hifz al-lisan, i. 139-72; al-Jidd wa-lhazl, i 231~76; Fr nafy altashbih, i. 283-307; al-Qiyän, ii, 143-82; H ujaj al-nubuwwa, iti, 223-81; al-Wukal@ iv. 95-105; al-Awtan wa-l-buldan, iv. 109-47. —— (1968), al-Bayan wa-l-tabyin, ed. ‘A.-S, M. Harun, 4 vols. in 2 (Cairo). —— (1992), al-Bursän wa-l-‘urjan wa-l-umyän wa-l-hawlän, ed. M. M. al-Khuli (Beirut). Jälinüs, F anna quwa I-nafs tabia li-mizaj al-badan, see Biesterfeldt (1973); also Badawi (1981) 183-4, Jawhari, “All ibn al-Ja‘d al. (1996), Musnad (riwäyat al-Baghawi\, ed. ‘AA, Haydar (Beirut). Kahhal, ‘Ala’ al-Din al- (1955), al-Ahkarn al-nabawiyya fi U-sind’a I-tibbiyya, ii (Cairo). Khatib al-Baghdadi, Ahmad ibn ‘Ali al- (1931), Tatikh Baghdad (Cairo). Khwärizmi, Muhammad ibn Misa al- (1831), Algebra, ed. and trans. F. Rosen (London). Kushajim, Mahmüd ibn al-Husayn (1881), Adab al-nadım (Bulag). —— (1999), Adab al-nadım (Cairo). Mas’üdı, ‘Ali ibn al-Husayn al- (1866-74), Murij al-dhahab, ed. and trans. C. Barbier de Meynard and Pavet de Courteille, 9 vols. (Paris). —— (1894), al-Tanbth wa-l-ishraf, ed. M. J. de Goeje, 8 vols. (Leiden). Michael Scot (1477), Liber phisionomie (Venice). Miskawayh, Ahmad ibn Muhammad (1968), Tahdhib al-akhläg, trans. Constantin Zurayk, The Refinement of Character (Beirut; text 1966). Mizzi, Yüsuf ibn al-Zaki al- (1980), Tahdhib al-kamal, 35 vols. (Beirut). Mubarrad, Muhammad ibn Yazid al- (1874), al-Kamil, ed. W. Wright, 2 vols. (Leipzig). Mubashshir ibn Fatik, Mukhtar al-hikam, ed. ‘A-R. Badawi (Madrid, 1958); the section which includes sayings of Polemon is translated in Rosenthal (1975) 12444. Munäwi, ‘Abd al-Ra’üf al- (1937), Fayd al-gadir, 6 vols. (Cairo). Nisäbür, Muhammad ibn ‘Abdallah (1990), al-Mustadrak ‘aid L-sahthayn, 5 vols. (Beirut). Nuwayri, Ahmad ibn “Abd al-Wahhab (1923-92), Nihayat al-arab fi funän al-adad, 31 vols. (Cairo). 667 Pietro d’Abano (1474),), Liber compilationis Li physi Physionomie 4 (Padua). oa Ahmad ibn “Al al- (1987), Subh al-ashä (Beirut). = 2 (Baghda 7 7 Jafar (1981), al-Kharäj jn u wa-sina‘at al-kitaba, ed. M. H. al-Zubaydi oen “Abd al-Karim (1948), al-Risäla (Cairo). usta ibn | Lüga (1941), Hr ikhtilaf al-nas ft siyarihim wa-Lakhla À him Le L i Caractères), ed. and trans. Paul Sbath d in Bulletin de l'Institut d'Égypte, a mo “ Razi, Aba Bakr al- {Rhazes) (1374/1955), al-Hawt fi Ltibb, i (Hyderabad). —- (1408; 1987), al-Mansürt Ji 1-tibb, ed. H. al-B. al-Sadigi (Kuwait), — al-Mansart ft Leibb, Bibliothèque Nationale de France, ms. arabe 2866. ae al-Din al-, see Fakhr al-Din al-Räzi, Safiiri, heats “Abd rin al-Rahman ibn “Abd al-Salam al{ (1927 ), Nuzhat al-majal majälis wa-muntakhab Sa‘id al-Andalüst, Aba L-Qasim (1912), Tabagat al-umam, ed a al-asrar, in Badawi (1954), 67-177. 4 7 labari, Muhammad ibn Jarir al(1999), Tafsir, 13 vols, (Beirut) i i ho (Beirut) Tabbakh, M. R. al- (1347/1929), Kitab al-firäsa K-Filtmün al-hakim, wa-yalthi Jumal ahkäm al-firdsa li-Abi Bakr Muhammad b. Zakariyya I-Räzt (Aleppo). Tahanawi, al- (1862), Kashshaf istilähät al-funün, 2 vols. (Calcutta; repr. Tehran, 1947). Tantkhi, al-Muhassin ibn ‘Al al- (1971-3), Nishwar al-muhädara, ed. ‘A, al-Shalji, 8 vols. (Beirut), u | Täshkuprüzäda ” = (Tasköprüzade), > Ahmad b, . Mustafa (n.d.), ah “Hu Miftah EL “i dh al-siyäda fi mawdiat al-ulüm, 7 a 3 vols. (Beirut). Tah al sada wa vak Tawhidi, Aba Hayyan al- (2001), al-Hawamil wa-Lshawämil, ed. 5. Kasrawi (Beirut). = ( 1939), al-Imtä' wa-l-muänasa, ed. A. Amin and A. al-Zayn, 3 vols. (Cairo). en Abd al-Malik ibn Muhammad (1985), Thimar al-qulüb, ed. M. A.-F. Ibrahim ee ibn ‘Isa aleirut). (1978-86), Sunan, ed. A. M. Shakir me,“en „Din al- (1964), The Nasirean et al, 5 vols Ethics (Akhläg-e Nasir), ’ | trans. G. M. Wickens ndon). .. Waki’, Vs |Muhammad ibn Khalaf (1992), ; Akhbar al-auda qudäh, 3 vols. (Beirut). i ,Yahya SH.ibn Griffith ‘Adi (2002), (Prone, Tahdhib nn al-akhiaa ( The Reformation of Morals), ed. and trans. ade a‘qûbr,= Ahmad nme ibn Abi A Ya'qab| al- (1883), » Tarikh, Tatikh, ed. ed. T. T. Houtsma, la ma, 77 vols. | (Leiden). i “uid (1907-31), Irshad al-arib (Mufam al-udaba ), ed. D, 5. Margeliouth —— ~ ¢(1866-73) ; Mujam j al-buldan, , ed. ed, F F. Wüstenfeld (Leipzi g). Zabidi, Muhammad ibn Muhammad (1889-90), Taj al“arüs (Cairo). Zamakhshart, Mahmüd ibn Umar al- (1976), Rabi al-abrar (Baghdad). SECONDARY SOURCES Abbas, I. (1993), Malämih yananiya LES A rare @) f l-adab al-arabi (Beirut). 098), An Epigram of Ausonius (87, p. 344 Peiper)’, Latomus, 42: ine Latin of the Vindolanda Writing Table ts: An 95-109, Interim Report’, JRS 85: "Getto Pelagonius and Lati n Veterinary Terminology in the Rom an Empire —— and Janse, M., and Swain, 5. (2002) (eds,), Bilingualism in Ancient Society (Oxford).

Bibliography Bibliography “Adi, ‘AK. Z. (1982/3), ‘al-Firdsa ‘inda }-"Arab’, Revue de l'Académie Arabe à Damas, 57: 709-28; 58: 161-93, 343-63, 570-631. Agrimi, J. (2002), Ingeniosa scientia nature: Studi sulla fisiognomica medievale (Florence). Ahsan, M. M. (1979), Social Life under the Abbasids (London). Alexander, P. 5. (1996), ‘Physiognomy, Initiation, and Rank in the Qumran Community’, in H. Cancik et al. (eds), Geschichte — Tradition — Reflexion: Festschrift fiir Martin Hengel zum 70. Geburtstag. Band I Judentum (Tübingen), 385-94, Amberger-Lahrmann, M. (1996), Anatomie und Physiognomie in der hellenistischen Plastik dargestellt am Pergamonaltar (Abh. der geistes- und sozialwissensch. Klasse, Stuttgart). 10; Anawati, G. (1951), ‘Essai de bibliographie avicennienne’, Revue thomiste, 51: 407~40, —— (1977), ‘Les divisions des scier.ces intellectuelles d'Avicenne’, Mélanges de l'Institut Dominicain d'Études Orientales du Caire, 13: 323-35. Anderson, G. (1976), Lucian: Theme and Variation in the Second Sophistic (Mnemosyne, Suppl. 41, Leiden), —— (1993), The Second Sophistic: A Cultural Phenomenon in the Roman Empire (London). , André, J. (1970), ‘Philologie latine’, Annuaire de l'École pratique des hautes études 1969-1970, 4° section: Sciences historiques et philologiques, 102: 281-4. — (1981) (ed), Anonyme latin: Traité de physiognomonie (Paris). —— (1991), Le Vocabulaire latin de l'anatomie (Paris). Archer-Hind, R. D. (1973) (ed.}, The Phaedo of Plato (New York). Arias, P. (1986), ‘Analisi critica delle statue’, in L. Satriani and M. Paoletti (eds.), GH eroi venuti dal mare (Rome) 39-64. Armstrong, A. H. (1977), ‘Form, Individual and Person in Plotinus’, Dionysius, 1: 49-68, Armstrong, A. MacC. (1958), “The Methods of the Greek Physiognomists’, Greece and Rome, 5: 52-6, Arnim, H. von (1906), Hierokles: Ethische Elementurlehre (Berlin). Asin Palacios, M. (1911), ‘Noticias de los manuscritos arabes de! Sacro-Monte de Granada’, Revista del Centro de Estudios Histöricos de Granada y su Reine, 1 (1911), 249-77, Asınus, R. (1906), ‘Vergessene Physiognomonika’, Philologus, 65: 410--24. Autuori, A, (1983), ‘Antecedenti greci e arabi del Kitab al-firäsa di Pakhr al-din al-Razt’, Ann. delPInst. Or. di Napoli (AION), 43: 43-60. —— (1984), ‘Testo latino e testo arabo della fisiognomica di Muhammad ibn Zakariya’ al-Razy’, AION 44: 29-40, Avotins, L (1975), ‘The Holders of the Chairs of Rhetoric at Athens’, HSCP 79: 313-24. ——— (1978), ‘The Date and Recipient of the Vitae Sophistarum of Philostratus’, Hermes, 106: 242-7, Badawi, ‘A-R. (1954), al-Usal al-yünäniyya E-Inazariyyat al-sipasiyya fi Lislam, i (Cairo). — (1981) Dirasat wa-nusüs fi I-falsafa wa-l-wlär inda ‘Arab (Beirut). —- (1985), Hunayn ibn Ishag, Adah al-faläsifa (Sentences des philosophes) (Kuwait). Baldassarre, G. (1978), ‘Musonio Rufo e l'Economico di Brisone’, in E, Livrea and G. A. Privitera (eds.), Studi in onore di Anthos Ardizzoni (Rome) i. 37-48, Ballester, L, G, (1988), ‘Soul and Body, Disease of the Soul and Disease of the Body in Galen’s Medical Thought’, in Manuli and Vegetti (1988) 117-52. Balmaseda, L, (1990), ‘Hercules (in periphera occidentali)’, LIMC v/1, cols. 253-62. Balme, D, M. (1972), Aristotle’s De partibus animalium I and De generatione animalium I (with Passages from H.1-3} (Clarendon Aristotle Series; Oxford). 669 —— (19874), ‘The Place of Biology in Aristotle’s Philosophy’, in Gotthelf and Lennox {1987} 9-20. —— (19876), ‘Teleology and Necessity’, in Gotthelf and Lennox (1987) 275-853. —— (1987c), ‘Aristotle’s Biology Was Not Essentialist’, in Gotthelf and Lennox (1987) 291-312, -— {1991} (ed.), Aristotle: History of Animals, Books VII-X (Loeb Classical Library; Cambridge, Mass., and Londen). Baloglou, Ch, and Constantinidis, A. (1996), ‘The Treatise “Oeconomicus” of the Neopythagorean Bryson and its Influence on Arab-Islamic Economi c Thought’, Journal of Oriental and African Studies (Athens), 8: 46-55. Baltes, M. (1975), ‘Numenios von Apamea und der platonische Timaios’, Vigiliae Christianae, 29: 241-70, (2000), ‘La dottrina delPanima in Plutarco’, Elenchos, 21: 245-70, Barker, A. (1982), ‘Aristides Quintilianus and Constructions in Early Music Theory’, CQ, NS 32: 184-97, Barnes, J. (1994), Aristotle: Posterior Analytics (Clarendon Aristotle Series: Oxford}. —— (1997), Roman Aristotle’, in J. Barnes and M. Griffin (eds.}, Philosophia Togata IT: Plato and Aristotle at Rome (Oxford), 1-69. Barnes, T. D. (1981), Constantine and Eusebius (Cambridge, Mass.) Barton, T. 5. (19944), Ancient Astrology (London). (19946), Power and Knowledge: Astrology, Physiognomics, and Medicine under the Roman Empire (Michigan), (1994c) “The Inventio of Nero: Suetonius’, in J. Elsner and J. Masters. (eds), Reflections of Nero (London), 48-63, Bastianini, G, and Long, A. A. (1992), ‘Hierocles’, Corpus dei Papiri Greci e Latini 268-451, Behr, C. A. (1981), P. Aelius Aristides: The Complete Works, ii (Leiden). 1**: Bergstrasser, G. (1913), Hunain ibn Ishak und seine Schule: Sprachund literargeschichtliche Untersuchungen zu den arabischen Hippokrates- und Galen-Üb ersetzungen (Leiden), —— (1925), Hunain ibn Ishag: Übersetzungen (Leipzig). Über die sprischen und arabischen Galen- —— (1932), Neue Materialien zu Hunain ibn Ishäg’s Galen-Bibliographie Berkowitz, L., and Squitier, K. A. (1986), Thesaurus Linguae Bertier, J. (1992), ‘Médecine et philosophie dans l'Ad Gaurum, sur l'embryon s'anime, in López Pérez (1992) 635-45. Bickerman, E, (1952), ‘Origines Gentium’, CPh 47: 65-81 = (Leipzig). Graecae: Canon of Greek Authors and Works (New York). la manière dont id, Religions and Politics in the Hellenistic and Roman Periods (Como, 1985) 399417, Biesterfeldt, H. H. (1973), Galens Traktat ‘Daß die Kräfte der Seele den Mischungen des Körpers falgen’ in arabischer Übersetzung (Abhandlungen für die Kunde des Morgenlandes, 40/4; Wiesbaden). —— (1986), ‘Galinas Quwä n-nafs’, Der Islam, 63: 119-36. Bindman, D. (2002), Ape to Apollo: Aesthetics and the Idea of Race in the 18th Century {London}. Birley, A. R. (1997), Hadrian: The Restless Emperor (London) . Bluck, R. 8. (1955), Plato's Phaedo: Translated with Introduct ion, Notes and Appendices (London). Blumenthal, H. J. (1966), ‘Did Plotinus Believe in Ideas of Individua ls®”, Phronesis, — (19714), Plotinus’ Psychology. His Doctrine of the Embodie d Soul (The Hague).

Bibliography Bibliography —— (19716), ‘Soul, World-Soul and Individual Soul in Plotinus’, in Le Néoplatonisme (Colloque international sur le néoplatonisme, 1969; Paris), 55-63. —— (1975), ‘Plutarch’s Exposition of the De Anima and the Psychology of Proclus’, in H. Dörrie (ed), De Jamblique à Proclus (Entretiens sur PAntiquité Classique, 21; Geneva) 122-47. —— (1996), Aristotle and Neoplatonism in Late Antiquity: Interpretations of the De Anima (London), —— (1997), ‘Were Aristotle’s Intentions in Writing the De anima Forgotten in Late Antiquity?’, Documenti e studi sulla tradizione filosofica medievale, 8: 143-37. Boatwright, M. T. (2000), Hadrian and the Cities of the Roman Empire (Princeton). Bobzien, 5, (1998}, Determinism and Freedom in Stoic Philosophy (Oxford). Buck, B. (2000), Die babylonisch-assyrische Morphoskopie (Vienna). Bonebakker, 5. A. (1984), ‘Early Arabic Literature and the Term adab’, Jerusalem Studies in Arabic and Islam (JSAI), 5: 389-421. Bost-Pouderon, C, (2003), ‘Dion de Pruse et la physiognomonie dans le Discours XXXII’, Revue des études anciennes, 105: 157-74, Bottéro, J. (1969), ‘Antiquités assyro-babyloniennes’, Annuaire de l'École pratique des hautes études 1968-1969, 4° section: Sciences historiques et philologiques, 101: 79-84, —— (1974), ‘Symptômes, signes, écritures en Mésopotamie ancienne’, in J. P. Vernant al, (eds), Divination et rationalité (Paris), 70-196. et Bouché-Leclercq, A, (1899), L'Astrologie grecque (Paris). Bouhdiba, A, (1998), Sexuality in Islam (English translation; London). Bouyges, M. (1924), ‘Notes sur les traductions arabes d’auteurs grecs’, Archives de philosophie, 2: 1-23, Bowersock, G. W. (1965), Augustus and the Greek World (Oxford). (1969), Greek Sophists in the Roman Empire (Oxford). Bowie, E, L. (1970), ‘The Greeks and their Past in the Second Sophistic’, Past and Present, 46: 3-41; repr. with corrections in M. I. Finley, Studies in Ancient Society (London, 1974), 166-209. —— (1978), ‘Apollonius of Tyana: Tradition and Reality’, ANRW II 16. 2: 1652-99, —— (1982), ‘The Importance of Sophists’, YCS 27: 29-60. —— (1997), ‘Hadrian, Favorinus, and Plutarch’, in J. M. Mossman (ed.), Plutarch and his Intellectual World (London), 1-15. Boys-Stones, G. R. (2001), Post-Hellenistic Philosophy: A Study of its Development from the Stoics to Origen (Oxford). (2004), ‘Phaedo of Elis and Plato On the Soul’, Phronesis, 49: 1-23. Braund, 5, and Most, G. W. (2003), Ancient Anger: Perspectives from Homer to Galen (Yale Classical Studies, 32; Cambridge). Bremmer, J. (1991), “Walking, Standing, and Sitting in Ancient Greek Culture’, in id. and H. Roodenburg (eds.), A Cultural History of Gesture: From Antiquity to the Present Day (Cambridge), 15-35, Brisson, L., and Patillon, M. (2001) (eds, and trans.), Longin: Fragments, Art rhétorique. Rufus: Art rhétorique (Paris). Brock, 5. P. (1979), ‘Aspects of Translation Technique in Antiquity’, Greek, Roman and Byzantine Studies, 20: 69-87 repr. in Syriac Perspectives on Late Antiquity (London, 1984) III. —— (1983), ‘Towards a History of Syriac Translation Technique’, Orientalia Christiana Analecta, 221: 1-14 repr. in Studies in Syriac Christianity (Aldershot, 1992) X. —— (1991), “The Syriac Background to Hunayn’s Translation Techniques’, Aram, 3: 139-62 repr. in From Ephrem to Romanos (Aldershot, 1999) XIV. en (1996), “Greek Words in Syriac: Some General Features’, Scripta Classica Israelica, 15; 251-62 repr. in From Ephrem to Romanos (1999) XV. {1999}, “Two Letters of the Patriarch Timothy from the Late Eighth Century on Translations from Greek’, Arabic Sciences and Philosophy, 9: 233-46. Brown, P., and Levinson, $. C. (1987), Politeness: Some Universals in Language Usage (Cambridge). . Brown, T. M. (1993), ‘Mental Diseases’, in Bynum and Porter (1993) 438-63. Brown, W. N. (1922), ‘A Comparative Translation of the Arabic Kalila wa-Dimna’, JAOS 42: 215-50, Burnet, J. (1911) (ed), Plato's Phaedo: With Introduction and Notes (Oxford). Burnyeat, M. F. (1982), ‘The Origins of Non-Deductive Inference’, in J. Barnes, J. Brunschwig, M. Burnyeat, and M. Schofield (eds), Science and Speculation. Studies in Hellenistic Theory and Practice (Cambridge and Paris}, 193-238. Byl, S. (1991), ‘La Physiognomie du Nepi dépuv, bödtwv, témwv dans le Parisinus Lat. 7027’, in G. Sabbah (ed.), Le Latin médical: la constitution d’un langage scientifique (Saint-Etienne) 53-62, Bynum, W. F, and Porter, R. (1993) (eds.), Companion Encyclopedia of the History of Medicine, i (London and New York). Cairns, D. (2005), ‘Bullish Looks and Sidelong Glanzes: Social Interaction and the Eyes in Ancient Greek Culture’, in id, Body Language in the Greek and Roman Worlds (Swansea), 123-35. Campanile, M. D, (1999), ‘La costruzione del sofista: Note sul Bioc di Polemone’, Studi ellenistici, 12: 269-315, (2003), Vivere e morire da sofista: Adriano di Tiro’, Studi ellenistici, 15: 245-73. Campe, R., and Schneider, M. (1996) (eds), Geschichten der Physiognomik: Text, Bild, Wissen (Rombach Wissenschaft; Reihe Litterae, 36; Freiburg). Cappont, F. (1995), L'anatomia e la fisiologia di Plinio (Genoa). Casartelli, A. (1998), ‘La funzione distintiva del colore nell’abbigliamento romano della prima eta imperiale’, Aevum, 72: 109-25, Caston, V. (1997), ‘Epiphenomenalisms, Ancient and Modem’, The Philosophical Review, 106: 309-54. —— (2001), ‘Dicaearchus’ Philosophy of Mind’, in Fortenbaugh and Schiitrumpf (2001} 175-93, Centrone, B. (1989), ‘Aristide Quintilien’ (A354), in Goulet (1989-) i, 369. —— (2000), “Cosa significa essere pitagorico in etd imperiale: Per una riconsiderazione della categoria storiografica del neopitagorismo’, in A. Brancacci (ed.), La filosofia in etd imperiale (Naples), 137-68. Chamberlain, M. (1994), Knowledge and Social Practice in Medieval Damascus, 1190-1350 (Cambridge). Champlin, E, (1980), Fronto and Antonine Rome {Frincetoni). Cheikho, L. (1921), ‘Un traité inédit d'économie domestique’, Al-Machrig, 19: 161-81. ~— (1920-3), Anciens traités arabes ... tirés de la Revue al-Machrig (Beirut). Chejne, A. (1965), ‘The Boon Companion in Early Abbasid Times’, JAOS 85: 327-35, Civiletti, M. (2002), Filostrato: Vite dei sofisti (Milan). Clarke, A. G. (1987), ‘Metoposcopy: An Art to Find the Mind’s Construction in the Forehead’, in P. Curry (ed.), Astrology, Science and Society, Historical Essays (London) 171-95, Claus, D. B. (1981), Toward the Soul: An Inquiry into the Meaning of Psyche before Plato (Yale Classical Monographs, 2; New Haven). Collas-Heddeland, E. (1995), ‘Le Culte impérial dans la compétition des titres sous le Haut-Empire: Une lettre d’Antonin aux Ephésiens’, REG 108: 410-29.

Bibliography Bibliography Cooper, J. M. (1987), ‘Hypothetical Necessity and Natural Teleology’ Lennox (1987) 243-74, , in Gotthelf and (1988), “Metaphysics in Aristotle’s Embryology’, PCPAS 214 (Ns 34): 14-41, —— (1998), ‘Posidenius on the Emotions’, in Sihvola and Engberg-Pedersen 71-112. (1998) Cormack, R. (2000), Byzantine Art (Oxford). Corrigan, E. (1985), ‘Body’s Appreach to Soul: An Examination of a Recurrent Theme in the Enneads’, Dionysius, 9: 37-52. Couissin, J. (1953), ‘Suétone physiognomiste dans “Les Vies des XII Cäsars””, Revue des études latines, 31: 23456, Courtine, J.-J., and Haroche, Cl. (1988), Histoire du visage. Exprimer et taire ses émotions (XVF-debut XIX° siècle) (Paris). Cramer, J. A. (1837), Anecdota Graeca ex codd. manuscripts Bibliothecarum Oxoniensium, iv (Oxford). Cranz, F. E. (1960), ‘Alexander of Aphrodisias’, in P. ©. Kristeller (ed.), Catalogus Translationum ct Commentariorum: Mediaeval and Renaissance Latin Translations and Commentaries, Annotated Lists and Guides, i (Washington) 77-135. Cribiore, R (2001), Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt (Princeton). Currie, H, Mel. (1985), ‘Aristotle and Quintilien: Physiognomical Reffections’, in Gotthelf (1985) 359-66. Dagron, G. (1987), ‘Image de bête ou image de Dieu: La physiognomonie animale dans la tradition grecque et ses avatars byzantins’, in Poikilia: Études offertes à Jean-Pierr e Vernant (Paris) 69-80. D'Ambra, E. (1996), ‘The Calculus of Venus: Nude Portraits of Roman Matrons’, in N. Kampen (ed.), Sexuality in Ancient Art (Cambridge) 219-32. Davidson, J. (1997), Courtesans and Fishcakes: The Consuming Passions of Classical Athens (London). Dawkins, R. M. (1916), Modern Greek in Asia Minor (Cambridge). Dean-Jones, L. (1994), Women’s Bodies in Classical Greek Science (Oxford), (2000), “Aristotle’s Understanding of Plato’s Receptacle and its Significance for Aristotle’s Theory of Familial Resemblance’, in Wright (2000) 101-12. Degkwitz, A. (19964), ‘Der Text der pseudo-aristotelischen “Physiognomonica” Traktat A’, in Campe and Schneider (1996) 13-21. —— (19965), ‘Die pseudo-aristotelischen “Physiognomonica®”, in Campe and Schneider (1996) 23-44, Deissmann, A, (1933}, Forschungen und Funde im Serai: Mit einem Verzeichni s der nichtislamischen Handschriften im Topkapu Serai zu Istanbul (Berlin). De Lacy, P. (1988), ‘The Third Part of the Soul’, in Manuli and Vegetti (1988) 43-63. (1992) (ed.) Galen, On Semen: Edition, Translation and Commentary (CMG V. 3. 1; Berlin), Denyer, N. (2001} (ed.), Plato: Alcibiades (Cambridge), Desideri, P. (1978), Dione di Prusa: Un intellettuale greco nelPimpero romano (Messina and Florence), des Places, E. (1973) (ed.), Numenius: Fragments (Paris). Deuse, W. (1983), Untersuchungen zur mittelplatonischen und neuplatonischen Seelenlehre (Akademie der Wissenschaften und der Literatur Mainz, Abhandhingen der geistesund sozialwissenschaftlichen Klasse, Einzelveröffentlichung, 3; Wiesbaden). Di Benedetto, V. (1986), If medico e la malattia: La scienza di Ippocrate (Turin). Diels, H. (1879) {ed.}, Doxographi Graeci (Berlin), 673 —— (1910), ‘Uber einen neuen Versuch, die Echtheit einiger Hippokrat ischen Schriften nachzuweisen’, Sitzunigsberichte der königlich Preußischen Akademie der Wissenscha ften (Berlin) 1140-55. | Dietrich, A. (1966), Medicinalia Arabica: Studien über arabische Handschriften in türkischen und syrischen Bibliotheken (Göttingen). medizinische Dillon, J. (1971), ‘Harpocration’s Commentary on Plato: Fragments of a Middle Platonic Commentary’, California Studies in Classical Antiquity, A; 125-46, -~— (1996), The Middle Platonists, 80 B.C. to A.D. 220 (London) [repr. of 1977 edn. with Afterword]. Dittmer, H. L. (1940), Konstitutionstypen im Corpus Hippocraticum cingeleitet durch einen kurzen Überblick über ihr Verhältnis zu Hippokrates (diss. Würzburg). Dodds, E‚ R. (1963) (ed), Proclus, The Elements of Theology. A Revised Text with Translation, Introduction, and Commentary” (Oxford). Dodge, B, (1970) (trans.), The Fihrist of al-Nadim, 2 vols. (New York). Dols, M. W. (1987), ‘The Origins of the Islamic Hospital: Myth and Reality’, Bulletin af the History of Medicire, 61: 367-90. (1988), “Galen and Islamic Psychiatry’, in Manuli and Vegetti (1988), 243-80. —— (1992), Majnun: The Madman in Medieval Islamic Society, ed. D. E. Immisch (Oxford), Donini, P. L. (1971), ‘L’anima e gli elementi nel De anima di Alessandr o di Afrodisia’, Atti dell’ Accademia delle Scienze di Torino, Classe di scienze morali, storiche e filologiche, 105: 61-107. —— (1974), Tre studi sull'aristotelismo nel If secolo d. C. (Historia Politica Philosophia, 7; Turin). Doxiadis, E. (1995), The Mysterious Fayum Portraits (London). Drew-Bear, T., and Labarre, G. (2002), ‘Les trois statues de la concorde à Antioche de Druart, T-A. (1996), ‘La philosophie morale arabe et PAntiquité tardive’, Bulletin des Pisidie’, Epigraphica Anatolica, 34: 71-92. études orientales, 48: 183-7. Duff, T. E. (1999), Plutarch’s Lives: Exploring Virtue and Vice (Oxford). Duindam, J. (1995), Myths of Power: Norbert Elias and the Early Modern European (Amsterdam). Du Moulin, A. (1549), De diuersa hominum natura, prout a veteribus comporum speciebus reperta est, cognoscenda liber (Lyon). Court philosophis ex Duncan-Jones, R. (1990), Structure and Scale in the Roman Economy (Cambridg e). Düring, L (1957), Aristotle in the Ancient Biographical Tradition (Studia Graeca et Latina Gothoburgensia, 5; Göteborg). Dusanié, S. (1993), ‘Phaedo’s Enslavement and Liberation’, Hlineis Classical Studies, 19: 83-97, Ebert, J. (1968), ‘Die Gestalt des Thersites in der Ilias’, Philologus, 112: 159-75. Eck, W. (1998), ‘M. Cornelius Fronto, Lehrer Marc Aurels, consul RhM, nr 141: 193-6, suffectus im J. 142’, Edelstein, L. (1952), ‘The Relation of Ancient Philosophy to Medicine’, Bulletin of the History of Medicine, 26: 299-316, Eijk, P. J. van der (1990), ‘Aristoteles über die Melancholie’, Mnemosyne, Ath ser. 43: 33-72. —— (1997), ‘The Matter of Mind: Aristotle on the Biology of “Psychic” Processes the Bodily Aspects of Thinking’, in Kullmann and Föllinger (1997) 231-58, and —— (1999), ‘Historical Awareness, Historiography and Doxography in Greek and Roman Medicine’, in id. (ed.), Ancient Histories of Medicine: Essays in Medical

Bibliography Bibliography Doxography and Historiography in Classical Antiquity (Studies in Medicine, 20; Leiden) Flashar, H. (1966), Melancholie und Melancholiker in den medizinischen Theorien der 674 Antike (Berlin). 1-31. — (2000), ‘Aristotle's Psycho-Physiological Account of the Soul-Body Relationship’, in Wright and Potter (2000) 37-77. (2000-1), Diocles of Carystus: A Collection of the Fragments with Translation and Commentary, 2 vols. (Studies in Ancient Medicine 22-3; Leiden), Elias, N. (1983), The Court Society, trans. Edmund Jephcott (New York). —— (2000), The Civilizing Process: Sociogenetic and Psychogenetic Investigations, trans, Edmund Jephcott, rev. edn. (Oxford and Malden, Mass.). Eisner, J. (1995), Art and the Roman Viewer: The Transformation of Art from the Pagan World to Christianity (Cambridge). —— (1998), ‘Ancient Viewing and Modern Art History’, Metis, 13: 417-37. — (2003), Visualising Women in Late Antique Rome: The Projecta Casket’, in C. Entwhistle (ed.), Through a Glass Brightly: Festschrift for David Buckton (Oxford), 22-36, Emilsson, E. K. (1994), ‘Platonic Soul-Body Dualism in the Early Centuries of the Empire to Plotinus’, ANRW II 36. 7: 5331-62. Enderwitz, 5. (1989), ‘Du fai au zarif, ou comment on se distingue?’, Arabica, 36: 125-42. Endress, G, (1982-92), ‘Die Wissenschaftliche Literatur’ in W. Fischer and H. Gätje, Grundriß der arabischen Philologie (Wiesbaden), ii, 400-506, iil, 3-152. —— (1997) “The Circle of al-Kindi: Early Arabic Translations from the Greek and the Rise of Islamic Philosophy’ in id. and R. Kruk (eds), The Ancient Tradition in Christian and Islamic Hellenism (Leiden), 43-76. Evans, E. C. (1930), ‘Quo modo corpora voltusque hominum auctozes Latini descripserunt’, Harvard Studies in Classical Philology, 41: 192-5, — (1935), ‘Roman Descriptions of Personal Appearance in History and Biography’, Harvard Studies in Classical Philology, 46: 43-84. —— (1541), ‘The Study of Physiognomy in the Second Century BC’, TAPhA 72: 96-108. —— (1945), ‘Galen the Physicist as Physiognomist’, TAPhA 76: 287-98. —— (1945-50), ‘Physiognomics in the Roman Empire’, 45: 277-82. 675 The Classical Journal, | —— (1969), Physiognomics in the Ancient World (Transactions of the American Philosophical Society, Ns 59/5; Philadelphia). Flügel, G. (1835-58), see Hajji Khalifa (1835-58). (1871-2), see Ibn al-Nadim (1871-2). Foerster, R. (1884), Die Physiognomik der Griechen (Kiel). (1886), De Polemonis physiognomonicis dissertatio (Kiel). (1887), ‘De Apulei quae fertur Physiognomia recensenda et emendanda’, Jahrbücher für classische Philologie, Suppl. vol. 15: 557-90. (18884), ‘De Loxi Physiognomonia’, RAM, wr 43: 505-11. —— (18886), ‘De Aristotelis quae feruntur physiognomicorum indole ac condicione’, in Philologische Abhandlungen Martin Hertz zum siebzigsten Geburtstag von ehemaligen Schülern dargebracht (Berlin) 283-303. —— (1890), Quaestiones physiognomonicac (Kiel). —— (1893), Scriptores physiognomonici Graeci et Latini, 2 vols. (Leipzig). _ -—— (1897), ‘Zur Ueberlieferung der Physiognomik des Adamantios’, RhM, nr 52: 298-9, —— (1900), ‘Zur Epitome des Adamantios’, RhM, nr 55: 139-48, -—— (1925-6), ‘Zu den griechischen Physiognomikern’, Philologus, 81: 236-8. Follet, 5. (1976), Athènes au IF et au III” siècle: Études chronologiques et prosopographiques (Paris). (1989), ‘Adamantios’ (A17} in Goulet (1989) i. 51-3. Fortenbaugh, W. W. (1971), “Aristotle: Animals, Emotion, and Moral Virtue’, Arethusa, 4: 137-65. -—— and Schütrumpf, E‚ (2001) (eds.), Dicaearchus of Messana: Text, Translation, and Discussion (Rutgers University Studies in Classical Humanities, 10, New Brunswick ar.d London). Fotinis, A. P. (1979), The De Anima of Alexander of Aphrodisias: A Translation and Commentary (Washington D.C). Foucault, M. (1988), The History of Sexuality, iii: The Care of the Self, trans. R. Hurley (London, 1988). Frank, G. (2000}, The Memory of the Eyes: Pilgrims to Living Saints in Christian Late Antiquity (Berkeley). Franses, R. (2003), ‘When All That is Gold Does not Glitter: The Strange History of Looking in Byzantine Art’, in A. Eastmond and L. James (eds.), Icon and Word: The Power of Images in Byzantium (Ashgate) 13-24. Freds, M. (1987), ‘Substance in Aristotle’s Metaphysics’, in id. Essays in Ancient Philosophy (Oxford) 72-80 [originally Gotthelf 1985: 17-26]. Fahd, T. (1966), La Divination arabe (Leiden). -—— (1977), Firäsa, in EI ii. 916-17. (1991), Malhama, in EI vi. 247. Favreau Linder, A.-M, (2004), ‘Polémon de Laodicée: Pénigme d'un style’, in Giancarle Institute of Classical Studies, Suppl. vol. 68; London) 1-11. —— (2000), ‘Introduction’, in M. Frede and D. Charles (eds.), Aristotle's Metaphysics Abbamonte, Ferruccio Conti Bizzarro, and Luigi Spina (eds), L'ultima parola. L'analisi —— and Patzig, G. (1988) (eds), Aristotcles ‘Metaphysik Z’. Text, Übersetzung und dei testi: Teorie pratiche nell’antichitd greca e latina (Naples), 105-21. Festugière, A. J. (1953), La Révélation d’Hermes Trismégiste, iti: Les Doctrines de Fäme (Paris). —— (1954), ‘L’Ame et la musique, d'après Aristide Quintilien’, TAPRA 85: 55-78. Finamore, J. F., and Dillon, J. M. (2002) (eds), Famblichus De Anima: Text, Translation, and Commentary (Philosophia Antiqua, 92; Leiden, Boston, and Cologne). . (1997), ‘Euphrates of Tyre’, in R. Sorabji (ed.}, Aristotle and After (Bulletin of the Lambda (Oxford) 1-52. Kommentar, 2 vols. (Munich). Fritz, K. von (1938), ‘Phaidon (3), RE xix/2: 1538-42, Frolova, O. B, and Deryagina, T. P. (1996), Arabskie rukopisi vostochnogo otdela nauchnoy biblioteki Sankt-Peterburgskogo Gosudarstvennogo universiteta: kratkiy katalog (St Petersburg). Furlani, G. (1929), ‘Die Physiognomik der Barhebräus in syrischer Sprache, T, Zeitschrift Fittschen, K., and Zanker, P. (1983-5), Katalog der römischen Portrdis in der Capitolinischen Museen und anderen kommunalen Sammiungen der Stadt Rom, i anc. für Semitistik, 7: 1-16. Gage, J. (1993), Colour and Culture (London). Flamand, J.-M. (1989), ‘Artémidore de Daldis’ (A429) in Goulet (1989-) i. 605-14. —— (1999), Colour and Meaning (London). Garnsey, P. (1996), Attitudes to Slavery (Cambridge).

Bibliography Bibliography Gerson, L, P. (1987), ‘A Note on Tripartition and Immortality in Plato’, Apeiron, —— and Lennox, J. G. (1987) (eds), Philosophical Issues in Aristotle's Biology 676 20; 81-96. Getrevi, P. (1991), Le scritture del volto: fisiognomica e modelli culrurali dal medioevo ad oggi (Milan), Ghazi, M. P. (1959), ‘Un groupe social: «les raffinés» (zurafä’)’, Studia Islamica, 11: 39-71. Ghersetti, A. (1994), ‘Una tabella di fisiognomica nel Qabs al-anwar wa-bahgat al-asrär attribuito a Ibn ‘Arabi’, QSA 12: 1547. — (1996), "Firäsa and Intelligence: the Silly and the Intelligent in Arab Physiognomy’, The Arabist: Budapest Studies in Arabic, 17: 121-31. —— (19994), Il Kitab Aristatalis al-faylasüf fi ]-firäsa nella traduzicne di Hunayn b. Ishäg (QSA, Studi e testi 4; Rome). —— (19995), ‘Legittimazione del sapere e cultura arabo-islamica: il trattato sulla fisiognomica delle donne attribuito a Polemone di Laodicea’, in G. Canova (ed), Scienza e Islam: Atti della giornata di studio (QSA, Studi e testi 3; Rome), 59-68. ~—— {2001), ‘De Pachat des esclaves: entre examen medical et physiognomonie. Le chapitre 46 du Kitab al-dalail d’Ibn Bahlül (X* s.)’, in K. Dévényi and T. Ivanyi (eds.), Essays in Honour of Alexander Fodor on His Sixtieth Birthday = The Arabist: Budapest Studies in Arabic, 13: 83-94, Gildemeister, J., and Bücheler, F. (1872), ‘Pseudo-Plutarchos nepi doxrjoewg’, RAM, NE 27: 520-38, Gill, C. (1998), ‘Did Galen Understand Platonic and Stoic Thinking on Emotions?’, in Sihvola and Engberg-Pedersen (1998) 113-48. ——- (2000), “The Body’s Fault? Plato's Timaeus on Psychic Illness’, in Wright (2000} 59-84. Ginzburg, C. (1986), ‘Spie: Radici di un paradigma indiziaric’, in id, Miti emblemi spie: Morfologia e storia (Turin), 158-209 = ‘Clues: Roots of an Evidential Paradigm’, in id., Clues, Myths, and the Historical Method, trans. J. and A, Tedeschi (Baltimore, 1989), 677 (Cambridge). Gottschalk, H. B. (1971), ‘Soul as Harmonia’, Phronesis, 16: 179-98. (1986), ‘Boethus’ Psychology and the Neoplatonists’, Phronesis, 31: 243-57. —— (1987), ‘Aristotelian Philosophy in the Roman World from the Time of Cicero to the End of the Second Century an’, ANRW II 36. 2: 1079-174. Goulet, R. (1989-) (ed.), Dictionnaire des philosophes antiques (Paris). (1994), ‘Calliphon’ (C28) in Goulet (1989-) ii. 176-7. Gow, A. $. F., and Page, D. L. (1968), The Greek Anthology. The Garland of Philip, 2 vols. (Cambridge). Graf, F. (1991), ‘Gestures and Conventions: The Gestures of Roman Actors and Orators’, in J. Bremmer and H. Roodenburg (eds.), A Cultural History of Gesture: From Antiquity to the Present Day (Cambridge), 36-58. Granarolo, J. (1985), ‘Sémiologies non verbales de l'amour dans l'Antiquité’, Sémiologie de l'amour dans les civilisations méditerranéennes (Paris), 85-97. Grignaschi, M. (1974), ‘La « Physiognomonie » traduite par Hunayn ibn Ishaq’, Arabica, 21: 285-91, —— (1976), ‘L’Origine et les métamorphoses du “Sirr-al-asrar’ * (Secretum secretorum)’, Archives d'histoire doctrinale et littéraire du Moyen Age, 43: 7-112. Gunderson, R. (1998), ‘Discovering the Body in Roman Oratory’, in M. Wyke (ed), Parchments of Gender: Deciphering the Body in Antiquity (Oxford), 169-89. Gundert, B. (2000), ‘Soma and Psyche in Hippocratic Medicine’, in Wright and Potter (2000) 14-35. Gutas, D. (1975), Greek Wisdom Literature in Arabic Translation: A Study of the Graeco-Arabic Gnomologia (New Haven). (1998), Greek Thought, Arabic Culture: The Graeco-Arabic Translation Movement in Baghdad and Barly ‘Abbasid Society (2nd—4th/8th—10th centuries) (London). Guthrie, K. S. (1987), The Pythagorean Sourcebook and Library, rev. D. R. Fideler (Grand Rapids, Mich.}. Gleason, M. W. (1990), ‘The Semiotics of Gender. Physiognomy and Self-Fashioning in the Second Century CE, in D. M. Halperin, J. J. Winkler, and F. I. Zeitlin (eds.), Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World (Princeton), 389-415. Habbi, J. (1984), ‘Le Livre des Signes d'al-Hasan b. Bahlül’, Oriens Christianus, 68: 210-12. Habicht, C. (1969), Die Inschriften des Asklepieions: Altertümer von Pergamon VIIL3 (Berlin). —— (1995), Making Men: Sophists and Self-Presentation in Ancient Rome (Princeton). Hadot, P. (1995), Philosophy as a Way of Life: Spiritual Exercises from Socrates to Foucault, Goffman, E. (1982), interaction Ritual: Essays on Face-to-Face Behavior (New York; first published 1967). Goldhill, 8. (2001) ‘The Erotic Eye: Visual Stimulation and Cultural Conflict’, in id. (ed.), Being Greek Under Rome (Cambridge), 154-94. ——— (1998), The Inner Citadel: The Meditations of Marcus Aurelius, trans. M. Chase (Camb. Mass). Gombrich, E. (1960), Art and Illusion (Oxford). Goodman, L. E. (1990), ‘The Translation of Greek Materials into Arabic’, in M. J. L. Young, J. D. Latham, and R. B. Serjeant {eds.), The Cambridge History of Arabic Literature, iii: Religion, Learning and Science in the ‘Abbasid Period (Cambridge), 477-97. Gürler, W. (1997), ‘Storing Up Past Pleasures: The Soul-Vessel-Metaphor in Lucretius and his Greek Models’, in K. A. Algra, M. H. Koenen, and P. H. Schrijvers (eds.}, Lucretius and his Intellectual Background (Amsterdam), 193-207, Gotthelf, A. (1985) (ed), Aristotle on Nature and Living Things: Philosophical and Historical Studies (Pittsburgh and Bristol). cum (1987), ‘Aristotle's Conception of Final Causality, in Gotthelf and Lennox (1987) 20442, trans. M. Chase (Oxford). Halfmann, H. (1979), Die Senatoren aus dem östlichen Teil des Imperium Romanum bis zum Ende des 2. Jh. n. Chr. (Göttingen). — — (1986), Itinera principum (Stuttgart). Hall, E. M. (1995), ‘Lawcourt Dramas: The Power of Performance in Greek Forensic Oratory’, Bulletin of the Institute of Classical Studies, 40: 39-58. Hamarneh, S. K, (1975), A Catalogue of Arabic Manuscripts in Medicine and Pharmacy at the British Library (Cairo). (as $. Kh. al-Hamarna} (1398/1979), Makhrürät Dar al-kutub al-Zahiriyya: Al-tibb wa-l-saydala (Damascus), Hamilton, R. (1988), Walid and his Friends: An Umayyad Tragedy (Oxford). Harlfinger, D., and Reinsch, D. (1970), ‘Die Aristotelica des Parisinus Gr. 1741: Zur Überlieferung von Poetik, Rhetorik, Physiognomik, De signis, De ventorum situ’, Philologus, 114: 28-50.

Bibliography Bibliography Harris, W. V. (2001), Restraining Rage: The Ideology of Anger Control in Classical Antiquity (Cambridge, Mass.). Harrison, §. J., Hilton, j. L., Hunink, V. J.C. (2001), Apuleius . Rhetorical Works (Oxford ). Harun, ‘A.-8, (1411/1991), Nawadir al-makhtütat, 2 vols. (Beirut) . Haussig, H. W. (1965), ‘Der Einfluss der hellenistischen Physiognomik auf die frühchristliche Eildgestaltung’, in Ati del VI congresso internaz ionale di archealogia cristiana (Studi di antichita cristiana 26; Vatican), 199-205. Heath, M. (1995), Hermogenes On Issues: Strategies of Argume nt in Later Greek Rhetoric (Oxford), —— (2004), Menander: A Rhetor in Context (Oxford ). Heeg, J. (1911), ‘Uber ein angebliches Diokleszitat’, Sitzungsberichte der königlich preußischen Akademie der Wissenschaften (Berlin), 991-1007, Helleman-Elgersma, W. (1980), Soul-Sisters: A Commen tary on Enneads IV 3 (27), 1-8 of Plotinus (Elementa: Schriften zur Philosophie und ihrer Problemgeschichte 15; Amsterdam), Herzog, R. (1991), ‘Mnemotechnik des Individuellen: Uberleg ungen zur Serniotik und Asthetik der Physiognomie’, in A, von Haverkamp and R. Lachmann (eds.), Gedächtniskunst: Raum — Bild — Schrift. Studien zur Mnemotechnik (Frankfurt), 165-88. Hesk, J. (2000), Deception and Democracy in Classical Athens (Cambridge). Highet, G. (1983), ‘Mutilations in the Text of Dio Chrysos tom’, in R. J. Ball (ed), The Classical Papers of Gilbert Highet (New York), 74-99. Hinck, H. (1873), Polemonis declamationes quae extant duae (Leipzig ). Hoenerbach, W. (1950), ‘Zur Heeresverwaltung der ‘Abbasiden’, Der Islam, 29: 257-90, Holford-Strevens, L. A. (1997), ‘Aulus Gellius: The Non-Visual Portraitist’, in M, J. Edwards and S. Swain (eds), Portraits: Biographical Represen tation in the Greek and Latin Literature of the Roman Empire (Oxford), 93-116, —— (2003), Aulus Gellius: An Antonine Scholar and his Achieve ment (Oxford). Hopfner, T. (1937), ‘Traumdeutung’, RE? vid: 2233-345, Horstmanshoff, M. (1999), ‘Les émotions chez Caelius Aurelian us’, in P. Mudry (ed.), Le Traité des ‘Maladies aiguës’ et des ‘Maladies chroniques’ de Caelius Aurelianus: Nouvelles approches (Actes du colloque de Lausanne, 1996; Nantes), 259-90, Hout, M. P. J. van den (1988) (ed.), M. Cornelius Fronto, Epistulae Hoyland, R. G. (1997}, Seeing Islam as Others Saw it (Princet (Leipzig), on). — (2004) “Theomnestus of Nicopolis, Hunayn ibn Ishaq, and the Beginnings of Islamic Veterinary Science’ in id. and P. FE. Kennedy (eds), Islamic Reflectio ns, Arabic Musings: Studies in Honour of Professor Alan Jones (Oxford), 150-69, Huby, P. (1999), Theophrastus of Bresus. Sources for his Life, Influence. Commentary Volume 4: Psychology (Texts 265-327 Hüffmeier, F. (1961), ‘Phronesis in den Schriften des Corpus 89; 51-84. Writings, Thought and ) {Leiden}. Hippocraticum’, Hermes, Hurvitz, N. (2003), ‘From Scholarly Circles ta Mass Movements: The , Communities in Islamic Societies’, American Historical Formation ofLegal Review, 108: 985-1008, Thsanoflu, E., et al, (1984), Catalogue of Medical Manuscripts (in Arabic, Turkish and Persian) in the Libraries of Turkey (Istankul). Inan, J., and Rosenbaum, E. (1966), Roman and Early Byzantine Portrait Sculpture in Asia Minor (London). Inwood, B. (1985), Ethics and Human Action in Early Stoicism (Oxford). loppolo, A. M. (1980), Aristone di Chio e lo stoicismo antico (Naples). Irwin, E. (1974), Color Terms in Greek Poetry (Toronto), Isager, J. (1991), Pliny on Art and Society (Odense). 679 Iskandar, 1, Z. (1967), A Catalogue of Arabic Manuscripts on Medicine and Science in the Wellcome Historical Medical Library (London). ~— (1988), Galen: On Examinations By Which the Best Physicians are Recognize d (CMG Suppl. Orient, 4; Berlin). Jackson, D. E. P. (1972), “The Arabic Translation of a Greek Manual of Mechanics’, The Islamic Quarterly, 16: 96-103. Jacquact, D, (1994), ‘La physiognomonie à l'époque de Frédéric II: Le traité Scot’, Micrologus, 2: 19-38. de Michel and Micheau, F. (1990), La Médecine arabe et l'occident médiéval (Paris). Jahn, A. (1882) (ed.}, Aristidis Quintiliani de musica libri IT (Berlin). James, J. (1996), Light and Colour in Byzantine Art (Oxford). Jan, C. von (1896), ‘Aristeides (25) Quintilianus’, RE ii. 894-6. Janes, D. (1998), God and Gold in Late Antiquity (Cambridge). Janssens, J. L. (1991), An Annotated Bibliography on Ibn Sind (1970-198 9) including Arabic and Persian Publications and ‘Turkish and Russian References (Leuven). Johansen, T, (2000), ‘Body, Soul, and Tripartition in Plato’s Timaeus’, Oxford Studies in Ancient Philosophy, 19: 87-111. Joly, R. (1962), ‘La Caractérologie antique jusqu'à Aristote”, Revue belge de philologie, 5-28. 40: Jones, C. P. (1971), Plutarch and Rome (Oxford). —— (1980), Culture and Society in Lucian (Cambridge, Mass.). —— (1999), Kinship Diplomacy in the Ancient World (Cambridge, Mass.). —— (2002), ‘Philostratus and the Gordiani’, Med. Ant. 5: 759-67. —— (2003), ‘Epigraphica IV-V: IV. Philosophers and Sophists at Phocaea; V. M. Aurelius Torquatus, a Ehetor of Trebenna’, ZPE 142: 127-33. Jong, P. de, and Goeje, M. J. de (1865), Catalogus codicum orientaliu m Bibliothecae Academiae Lugduno-Batavae, iii (Leiden). Jouanna, J. (1966), ‘La théorie de l'intelligence et de ’äme dans le traité hippocratique “Du régime”: ses rapports avec Empédocle ct le “Timée” de Platon’, Revue des études grecques, 79: xv—xviii. —— (1992), ‘La naissance de la science de l’homme chez les médecins e: les savants à l'époque d’Hippocrate: Problèmes de méthode’, in Löpez Férez (1992) 91-111. Kahn, C. H. (1996), Plato and the Socratic Dialogue (Cambridge). Karatay, F. E. (1966), Topkap Saray: Müzesi Kütübhanesi Arapça yazmalar katalogu (Istanbul). Kennedy, G. A. (1983), Greek Rhetoric under Christian Emperars (Princeton). Kenner, H. (1939), ‘Oneiros’, RE xviti/l: 448-59. Kessler, H. (2000), Spiritucl Seeing (Philadelphia). Kidd, 1. G. (1971), ‘Posidonius on Emotions’, in Long (1971) 200-15. —— (1999), Posidonius, Volume II: The Translation of the Fragments (Cambridge). Kister, M. J. (1983), ‘The Sirah Literature’, in À F. L. Beeston, T. M. jobnstone, R. B. Serjeant, and G. R. Smith (eds), The Cambridge History of Arabic Literature, i: Arabic Literature to the End of the Umayyad Period (Cambridge), 352-67. . Klein-Franke, F. (1977), Abu Said bin Behtisü, Risäla fi I-tibb wa-l-ahda Über die Heilung der Krankheiten der Seele und des Körpers (Beirut), t al-nafsäniyya: —— (1982), Vorlesungen über die Medizin im Islam (Wiesbaden). Kleiner, D. (1992), Roman Sculpture {New Haven). Klibansky, R., Panofsky, E., and Sazl, F. (1964), Sarurn and Melancholy: Studies in the History af Natural Philosophy, Religion and Art (London). Klose, D. O. A, (1987), Die Münzprägung von Smyrna in der römischen Kaiserzeit (Berlin). Kluge, K., and Lehmann-Hartleben, K. (1927), Die antiken Großbronzen (Berlin).

Bibliography Bibliography König, J. (2001), ‘Favorinus’ Corinthian Oration in its Corinthian Context’, PCPAS —— (1989), ‘Presocratic Philosophy and Hippocratic Medicine’, History of Science, 47: 141-71, Konstan, D. (1994), Sexual Symmetry, Love in the Ancient Novel and Related Genres (Princeton). Korus, K (1981), ‘The Motif of Panthea in Lucian’s Encomium', Eos, 69: 47--56. Kraemer, J. L. (1986), Hurnanism in the Renaissance of Islam (Leiden), Kraus, F. R. (1935), Die physiognomischen Omina der Babylonier (Mitteilungen der vorasiatisch-ägyptischen Gesellschaft (e. V.), 40, 2; Leipzig). Kraus, P,, and Walzer, R. (1951), Galeni compendium Timaei Platonis (Plato Arabus, i; London). Kruk, R. (1986), ‘Some Late Medieval Zoological Texts and their Sources’, in Actas del XII congreso de Ia UEAI (Malaga, 1984) (Madrid) 423-9. : Kullmann, W. and Föllinger, S. (1997} (eds), Aristotelische Biologie: Intentionen, Methoden, Ergebnisse (Stuttgart). Lahusen, G., ard Formigli, E. (2001), Römische Bildnisse aus Bronze: Kunst und Technik 681 27: 1-39. —— (1993), Greek Rational Medicine: Philosophy and Medicine from Alcmaeon to the Alexandrians (London and New York). López Férez, J. A. (1992) (ed.), Tratados hipocrdticos; Estudios acerca de su contenido, forma e injluencia (Actas del VIF Colloque international hippocratique) (Madrid). L’Orange, H. P. (1947), Apotheosis in Ancient Art (Oslo). Maderna, €. (1988), Iuppiter, Diomedes und Merkur als Vorbilder für römische Bildnisstatuen (Heidelberg). Madkur (1406/1985-6), see Ibn Sina (1406/1985~6), Maffei, 5. (1986), ‘Le “Imagines” di Luciano: Un “patchwork” di capolavori antichi, Un problema di metodo combinatorio’, Studi Classici e Orientali, 36: 14764, —— (1991), ‘La copia del pittore e del poeta nel proemio delle Imagines di Filostrato Maggiore’ Annali della Scuola Normale Superiore di Pisa, Classe di Lettere e Filosofia, 21: 291-321, (Munich). Lakoff, R. (1973), "The Logic of Politeness; or Minding your P's and Q's’, Papers from the Ninth Regional Meeting of the Chicago Linguistic Society (Chicago), 292-305, — (1994) (ed.), Luciano di Samosata: Descrizioni di coere d’arte (Turin). Makdisi, G. (1981}, The Rise of Colleges: Institutions of Learning in Islam and the West Lamoreaux, J, ©. (1999), ‘New Light on the Textual Tradition of Bar Bahlül’s Book of Signs’, Le Muséon, 112: 187-90. Manuli, P., and Vegetti, M. (1988) (eds), Le opere psicologiche di Galeno (Atti del terzo Langslow, D. R. (2000), Medical Latin in the Roman Empire (Oxford). Lennox, J. G. (1985), “Theophrastus on the Limits of Teleology’, in Fortenbaugh and Schütrumpf {2001} 143-63. —— (2002) (ed), Aristotle, On the Parts of Animals I-IV (Clarendon Aristotle Series; Oxford}. Levi della Vida, G. (1935), Elenco dei manoscritti arabi islamict della Biblioteca Vaticana (Vatican City). | Lichtenstädter, 1, (1943), ‘The Conception of Adab’, Muslim World, 33: 33-8. (Edinburgh). Colloquio galenico internazionale; Naples). (1988), ‘Scholarship, Authority and Argument in Galen's Quod animi mores’, in —— (1989), ‘The Early Translations from Greek into Arabic: An Experiment in — (19914), Methods and Problems in Greek Science: Selected Papers (Cambridge) —— (19915), “The Deb: of Greek Philosophy and Science to the Ancient Near East’, in Lloyd (19914) 278-98. ~— (19912), ‘Aristotle’s Zoology and his Metaphysics, The Status Quaestionis’, in Lloyd (1991a) 372-97, Loewenthal, A. (1896), Honein ibn Ishak, Sinnsprüche der Philosophen: Nach der hebräischen Uebersetzung Charisi’s (Berlin). Lombardi, M. (1999), ‘Paradigmi caratteriologici ed osservazione empirica nei Caratteri di Teofrasto e nei Physiognomorica pseudoaristotelici’, Rivista di cultura classica e medioevale, 41: 111-21. Long, À A, (1971) (ed.), Problems in Stoicism (London). —— (1993), ‘Hierocles on Oikeidsis and Self-Perception’, in K. J. Boudouris (ed.), Hellenistic Philosophy, i (Athens, 1993), 93-104; repr. in Stoic Studies (Cambridge, 1996), 250-63. Longrigg, J. (1985), ‘A Seminal “Debate” in the Fifth Century B.C.?, in Gotthelf (1985) 277-87. q Manzalaoui, M. A. (1974), ‘Kitab sirr al-asrär Facts and Problems’, Oriens, 23-4; 148-257, —— (1977), Secretum secretorum: Nine English Versions (Oxford), Markus, R. A. (1990), The End of Ancient Christianity (Cambridge). Mason, H. J. (1984), ‘Physiognomy in Apuleius Metamorphoses 2.2’, CPh 79: 307-9. Mathews, T. (1995), The Clash of Gods, rev. edn, (Princeton). Mathiesen, T’. J. (1983), Aristides Quintilianus: On Music in Three Books. Translation, with Introduction, Commentary and Annotations (Music Theory Translations; New Haven and London}. Mattock, J. N. (1972), ‘A Translation of the Arabic Epitome of Galen’s Book Tiepi 106v’, in $. M. Stern, A. Hourani, and V, Brown (eds.), Islamic Philosophy and the Classical Tradition, Essays Presented ... to Richard Walzer ... (Oxford), 235-60. Manuli and Vegetti (1988) 11-42. & | Lippert (1903), see Ibn al-Qifti (1903). Lloyd, G. E, R. (1983), Science, Folklore and Ideology: Studies in the Life Sciences in Ancient Greece (Cambridge). (1987), The Revolutions of Wisdom: Studies in the Claims and Practice of Ancient Greek Science (Sather Classical Lectures, 52: Berkeley, Los Angeles, and London). cl Comparative Assessment’, in G. Endress (ed), Symposium Graeco-Arabicum I (Amsterdam), 73-102. Mattusch, C. (1988), Greek Bronze Statuary (Ithaca), Maxwell-Stuact, P. G. (1981), Studies in Greek Colour Terminology, i: TAAYKOL; ii: XAPOTIOZ (Mnemosyne, Suppl. 65, 67, Leiden). Mayhew, R. (2004), The Female in Aristotle’s Biology: Reason or Rationalization (Chicago and London), McQueen, E, 1, and Rowe, C. J. (1989), ‘Phaedo, Socrates, and the Chronology of the Spartan War with Elis’, Méthexis, 2: 1-18. Megow, R. (1963), ‘Antike Physiognomielehre’, Das Altertum, 9: 213-23. Mesk, J. (1932), ‘Die Beispiele in Polemons Physiognomenik’, Wiener Studien, 50: 51-67. Métraux, G. P. R. (1995), Sculptors and Physicians in Fifth-Century Greece: A Preliminary Study (Montreal). Meyer, H. (1991), Antinoos: Die archäologischen Denkmäler unter Einbeziehung des numismatischen und epigraphischen Materials sowie der literarischen Nachrichten. Ein Beitrag zur Kunst- und Kulturgeschichte der hadrianisch-frühantoninischen Zeit (Munich). winne Ro Saat

Bibliography Bibliography Meynard, C. B. de, and Courteille, P. de (1861), Magoudi: Les Prairies d'or. Texte et Nollé, M. K. (1996), ‘Die Eintracht zweier Metropolen: Überlegungen zur Homonoia von Ephesos und Alexandreia zu Beginn der Regierung Gordians JIL’, Jahrbuch für Numismatik und Geldgeschichte, 46: 49-72. traduction, i (Paris). Mez, A. (1937), The Renaissance of Islam (Engl. trans.; London). Miller, H. W. (1948), ‘A Medical Theory of Cognition’, TAPhA 79: 168-83, Minio-Paluello, L. (1962) (ed.), Analytica Priora (Aristoteles Latinus III. 1-4; Bruges and Paris), Misener, G. (1923), ‘Loxus, Physician and Physiognomist’, CPh 18: 1-22. —— (1924), ‘Iconistic Portraits’, CPh 19: 97-123. Mitchell, 5. (1990), ‘Festivals, Games, and Civic Life in Roman Asia Minor’, JRS 80: 183-93. —— (1993), Anatolia. Land, Men, and Gods in Asia Minor, 2 vols. (Oxford). Möllendorf, P. von (2004), ‘Puzzling Beauty: Zur ästhetischen Konstruktion von Paideia in Lukians “Bilder-Dialogen””, Millennium, 1: 1-24. Montero, S. (1993), ‘Plauto, Mil. 694 y los primeros metoposcopi latinas’, Dioniso: Rivista di studi sul teatro antico, 63(1): 77-82. Montuori, M. (1976), ‘Su Fedone di Elide’, Aiti della Accademia Pontaniana, 25: 27-40. Moraux, P. (1951), Les Listes anciennes des œuvres d'Aristote (Louvain). —— (1973-84), Der Aristotelismus bei den Griechen von Andronikos bis Alexander von Aphrodisias, i: Die Renaissance des Aristotelismus im I. Jh. v. Chr; ti: Der Aristotelismus im IL und IL Jh. n. Chr. (Berlin and New York). Moreno, P. (1998), f bronzi di Riace (Milan). Morphy, H. (1992), ‘From Dull to Brilliant: The Aesthetics of Spiritual Power among the Yolngu’, in J. Coote and A. Shelton {eds.}, Anthropology, Art and Aesthetics (Oxford), 181-208. Mourad, Y. (1939), La Phystognomonie arabe et le Kitab al-firasa de Fakhr al-Din al-Räzi (Paris). Müller, A. (1884), see Ibn Abi Usaybi'a (1884). Müller, F. G. J. M. (1994), The So-Called Pelcus and Thetis Sarcophagus in the Villa Albani (Amsterdam). Müller, H (1980), Die Kunst des Sklavenkaufs nach arabischen, persischen und türkischen Ratgebern vom 10. bis zum 18. Jahrhundert (Freiburg). Miri, W. (1947), ‘Bemerkungen zur hippokratischen Psychologie’, in Festschrift für Edouard Tiéche (Schriften der Literarischen Gesellschaft Bern, 6; Bern), 71-85, (1953), Melancholie und schwarze Galle’, Museum Helveticum, 10: 21-38. 683 Nussbaum, M. C. (1978) (ed.}, Aristotle's De Motu Animalium (Princeton). Nutton, V. (1979), Galen: On Prognosis (CMG V. 8. 1; Berlin). (1993), Humoralism’, in Bynum and Porter (1993) 281-91. —— (1999) (ed), Galen: On My Own Opinions (CMG V. 3, 2; Berlin). —— (2004), Ancient Medicine (London). O'Meara, D. (1985), ‘Plotinus on How Soul Acts on Body, in id, (ed.), Platonic Investigations (Studies in Philosophy and the History of Philosophy, 13; Washington), 247-62, Opeku, F. (1979), ‘Physiognomy in Apuleius’, in C. Deroux (ed.), Studies in Latin Literature and Roman History, i (Collections Latomus, 164; Brussels), 467-74. Orelli, L. (1998), “Vorsokratiker und hippokratische Medizin’, in W. Burkert et al, (eds.), Fragmentsammiungen philosophischer Texte der Antike/Le raccolte dei frammenti di filosofi antichi (Göttingen), 128-45, Pack, R. A. (1941), ‘Artemidorus and the Physiognomists’, TAPhA 72: 321-34. (1974), ‘Auctoris incerti de physiognomonia libellus’, Archives d’histoire doctrinale et littéraire du Moyen Age, 41: 113-38. —— (1978), ‘Aristotle’s Chiromantic Principle and its Influence’, TAPhA 108: 121-30, Parmentier, L. (1926), ‘age de Phédon d’Elis’, Bulletin de l'Association Guillaume Budé, 10: 224, Parsons, M. C. (2606), Body and Character in Luke and Acts: the Critique of Physiognomy in Early Christianity (Grand Rapids, Mich.). Pavlopoulos, M. (2003), “Aristotle’s Natural Teleology and Metaphysics of Life’, Oxford Studies in Ancient Philosophy, 26: 133-81. Peck, A. L., and Forster, E‚ S. (1955) (eds), Aristotle, De partibus animalium; De motu animalium; De incessu animalium (Loeb Classical Library; Cambridge, Mass, and London: rev. edn.). Peers, G. (2001), Subtle Bodies (Berkeley). Pellat, Ch. (1954), ‘Gahiziana, L: Le Kitab al-tabassur bi-l-tigara’, Arabica, 1: 153-65. (1964), Variations sur le thème de l’adab’, Correspondance d’Orient, études, 5-6: 19-37, Pembroke, E, G. (1971), ‘Oikeidsis, in Long (1971) 114-49. Myers-Scotton, C. (1993a), Social Motivations for Codeswitching: Evidence from Africa (Oxford), (1993b), Duelling Languages: Grammatical Structure in Codeswitching (Oxford). Pépin, J. (1970), ‘Plotin et le miroir de Dionysos (ENN. IV, 3 [27], 12, 1-2}, Revue Nails, D. (2002), The People of Plato: A Prosopography of Plato and other Socratics Pernot, L. (1993), La Rhétorique de l'éloge dans le monde gréco-romain, 2 vols. (Paris). (Indianapolis). Najim, W. T. al- (1979), ‘Aristotle’s Book Corpus de animalibus and al-Gahiz's Book al-Hayawan’, Arabica 26: 307-309. Natali, C, (19954), ‘Oikonomia in Hellenistic Political Thought’, in A. Laks and M. Schofield (eds.}, Justice and Generosity: Studies in Hellenistic Social and Political Philosophy (Proceedings of the Sixth Symposium Hellenisticum; Cambridge), 95-128. (19955), Aristotele: L'amministrazione della casa (Rome and Bari). Nelson, R. (2000) (ed.}, Visuality Before and Beyond the Renaissance (Cambridge). Nesselrath, H. G. (1990), ‘Lucian’s Introductions’, in D. A. Russell (ed.), Antonine Literature (Oxford) 111-40. Newton, B. (1972), Cypriot Greek: Its Phonology and Inflections (The Hague). Nisbet, G. (2003), Greek Epigram in the Roman Empire: Martial’s Forgotten Rivals (Oxford). Penella, R. J. (1979), The Letters of Apollonius of Tyana(Leiden). internationale dz philosophie, 24: 304-20. Pertsch, L. C. W. (1878-92), Die arabischen Handschriften der herzoglichen Bibliothek zu Gotha, 5 vols. (Gotha), Petersen, E. (1859) (ed. ), Theophrasti Characteres (Leipzig). Petit, P. (1957), Les Etudiants de Libanius (Paris). Petrochilos, N. (1974), Roman Attitudes to the Greeks (Athens), Philippson, A. (1948), Das Klima Griechenlands (Bonn). Pigeaud, J. (1981), La maladie de l'âme: Étude sur la relation de l'âme et du corps dans la tradition médicc-philosophique antique (Paris). —— (1987), Folie et cures de la folie chez les médecins de l'antiquité greco-romaine: La manie (Paris), Pingree, D. (1963), ‘Astronomy and Astrology in India and Iran’, Isis, 54; 229-46. Plessner, M. (1928), Der OIKONOMIKOE des Neupythagorcers ‘Bryson’ und sein Einfluß auf die islamische Wissenschaft (Heidelberg).

Bibliography Bibliography 24749, Pohlenz, M. (1933), “16 nipénov: Ein Beitrag zur Geschichte des griechisch en Geistes’, Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen, Philologisc hhistorische Klasse (Berlin) 53-92. Pomeroy, 5. B. (1984), Women in Hellenistic Egypt: From Alexander to Cleopatra (New York). —— (1994), Xenophor; Oeconomicus. A Social and Historical Commentary (Oxford). Poschenrieder, F. (1882), Die platonischen Dialoge in ihrem Verkältniss e zu den hippokratischen Schriften (Beilage zum Jahres-Bericht der Studien-Austalt Metten 1881/2; Landshut). für Pradeau, J.-F. (1998), ‘L'âme et la moelle: Les conditions psychiques et physiologi ques de l'anthropologie dans le Timée de Platon’, Archives de philosophie, 61: 489-518. Praechter, K, (1912), ‘Der Topos nepì omou8fc Kal Todas", Hermes, 47: 471-6. Premerstein, A. von (1908), Das Attentat der Konsulare auf Hadrian im jahre 118 n, Chr. (Leipzig). Preus, A. (1977), ‘Galen’s Criticism of Aristotle’s Conception Theory’, Journal of the History of Biology, 10: 65-85. Price, $. R. F. (1984a), Rituals and Power: The Roman Imperial Cult in Asia Minor (Cambridge). ~—— (19846), ‘Gods and Emperors: The Greek Language of the Roman Imperial Cult’, FHS 104: 79-95, Puech, B. (2002), Orateurs et sophistes grecs dans les inscriptions d'époque impériale Qannawji, Siddig ibn Hasan al- (1978), Abjad al-ulim, 3 vols. (Damascus ). Quet, M.-H. (2003), ‘Le sophiste M. Antonius Polémon de Laodicée, (Paris). éminente personñalité politique de PAsie romaine du n° siècle’, in M. Cébeillac-Gervasoni and L. Lamoine (eds.), Les Elites et leurs facettes: Les élites locales dans le monde hellenistique et romain (Rome) 40143, Raina, G. (1993), Pseudo-Aristotele, Fisiognomica; Anonimo Latino, I trattato di fisiognomica: Introduzione, traduzione e note (Milan). Rapisarda, S, (2005) (ed.), Manuali medievali di chiromanzia (Roma). Rashed, R. (1989), ‘Problems of the Transmission of Greek Scientific Thought into Arabic: Examples from Mathematics and Optics’, History of Science, 27: 199-209, Reader, W. (in collaboration with A, J, Chvala-Smith) (1998), The Severed Hand and the Upright Corpse: The Declamations of Marcus Antonius Polemo (Atlanta, Ga). Reardon, B. P. (1989) (ed), Collected Ancient Greek Novels (Berkeley), Repath, I. D. (forthcoming), Playing with Plato: Platonic Allusion in Achilles Tatius’ Leucippe and Cleitophon (Orford). Reydams-Schils, G. (1997), ‘Posidonius and the Timaeus Off to Rhodes and Back to Plato?”, CQ, ns 47: 455-76, Reynolds, G. S. (2000), ‘The Sufi Approach to Food: A Case Study of ada’, Muslim World, 90: 198-217, Richtsteig, E. (1923), ‘Richard Foerster’, Biographisches Jahrbuch (Bursians Jahrbuch), 43: 34-57, Rida (1965), see Ibn Abt Usaybi'a (1995), Ridgway, B. S. (1990), Hellenistic Sculpture, i (Bristol), Rist, J. M. (1963), ‘Forms of Individuals in Plotinus’, CQ, ns 13: 223-3], Ritter, H. (1917), ‘Ein arabisches Handbuch des Handelswissenschaft’, Der Islam, 7: 1-91. Ritti, T. (1988), ‘Il sofista Antipatros di Hierapolis’, Miscellanea greca e romana, 13: 71-128. 685 —— (1997), ‘Antonino proconsole d’Asia: una nuova iscrizione di Hierapolis di Frigia’, in XI Congresso Internazionale di Epigrafia greca e latina: Preatti (Rome), 507-12, Rizakis, A, D, (1997), ‘Roman Colonies in the Province of Achaia: Territories, Land and Population’, in 5, E, Alcock (ed.), The Early Roman Empire in the East Robert, L. (1946), ‘Sur un décret d’Ephése’, Hellenica, 3 (Paris) 79-85, (Oxford), 15-36, —— (19779), ‘La titulature de Nicée et de Nicomédie: La gloire et la haine’, HSCP 81: 1-39; repr. in Opera Minora Selecta, vi (1989), 21149). —— (1977b), ‘Documents d’Asie Mineure’, BCH 101: 43-132; repr. in Documents d'Asie Mineure (Bulletin de l’École française à Rome, 239 bis; Paris, 1987) 1-90. Robin, L. (1926) (ed.), Platon: Phédon (Paris). Robinson, C, (2002), In Praise of Song: The Making of Courtly Culture in al-Andalus and Provence 1005-1134 ap (Leiden). Rochette, B. (1997), Le Latin dans le monde grec (Brussels). Romeo, I. (2002), “The Panhellenion and Ethnic Identity in Hadrianic Greece’, CPh 97: 21-40. Romm, J. (1990), “Wax, Stone and Promethian Clay: Lucian as a Plastic Artist’, Classical Antiquity, 9: 74-98. Rose, V. (1854), De Aristotelis librorum ordine et auctoritate commentatio (Berlin), (1863), Aristoteles Pseudepigraphus (Leipzig). (1864), Anecdota Graeca et Graecolatina: Mitteilungen aus Handschrif ten zur Geschichte der griechischen Wissenschaft, i (Berlin; repr. Amsterdam 1963). Rosenthal, F, (1961), ‘Al-Mubashshir ibn Fätik: Prolegomena to an Abortive Edition’, Oriens, 13-14: 132-58. —— (1975), The Classical Heritage in Islam (London), —— (1997), ‘Male and Female: Described and Compared’, in J. W. Wright and E. K. Rowson (eds.), Homoeroticism in Classical Arabic Literature (New York), 24-54. Ross, W. D. (1949) (ed), Aristotle’s Prior and Posterior Analytics: Revised Text with Introduction and Commentary (Oxford). Rossetti, L, (1973), “Socratica” in Fedone di Elide’, Studi urbinati (Serie B}, 47: 364-81. (1974), "Therapeia in the Minor Socratics’, Theta-Pi, 3: 145-57, —— (1977), Aspetti della letteratura socratica antica (Chieti). —— (1980), ‘Ricerche sui “Dialoghi Socratici” di Fedone e di Euclide’, Hermes, 108: 182-209, Rothe, S. (1989), Kommentar zu ausgewählten Sophistenviten des Philostrato s (Heidelberg). Rowe, C. T. (1993} (ed), Plato, Phaedo (Cambridge). Rowson, E, K. (1991), ‘The Effeminates of Early Medina’, JAOS 111: 671-93. Russell, D. A. (1993), Plutarch: Selected Essays and Dialogues (Oxford). — (1996), Libanius: imaginary Speeches (London). Rutherford, R. B, (1989), The Meditations of Marcus Aurelius: A Study (Oxford). Ryan, W. F,, and Schmitt, C. B. (1982), Ps.-Aristotle, the Secret of Secrets: Sources and Influences (London). Sadigi (1408/1987), see Razi, Aba Bakr al- (1408/1987). Sahlins, M. (1976), ‘Colors and Cultures’, Semiotica, 16: 1-22, Saller, R. (1991), “Corporal Punishment, Authority, and Obedience in the Roman Household’, in B. Rawson (ed.}, Marriage, Divorce, and Children in Ancient Rome (Oxford), 144-65. Salmeri, G. (2002), ‘Dio, Rome, and the Civic Life of Asia Minor’, in $, Swain (ed), Dio Chrysostom: Politics, Letters, and Philosophy (Oxford), 53-92. Sassi, M: M. (1992), “Plutarco antifisiognomico, ovvero: Del dominio della passione’, in L Gallo {ed.}, Plutarco e le scienze: Atti del IV convegno internazionale plutarcheo (Genoa), 353-73, en —— (1967), ‘The Physiognomics of Rhazes and its Influence on Eastern and Western Authors’, in Actes du XI Congrès International d'Histoire des Sciences (Warsaw) Sunes en TN

Bibliography Bibliography 687 (2001), The Science of Man in Ancient Greece = rev, edn. of La scienza dell’ uomo —— (forthcoming) (ed.), The Court and Court Society in Ancient Monarchies nella Grecia antica (Turin, 1988), trans. P. Tucker with a foreword by G. E. R. Lloyd (Cambridge). Sprengling, M. (1939-40), ‘From Persian to Arabic’, American Journal of Semitic (Chicago and London). Sayyid, F, (1955), see Ibn Juljul (1955). Sayyid-Marsot, A. L. (1979), Society and the Sexes in Medieval Islam (Levi della Vida Conference VI; Malibu, Calif.). Sbath, P. (1941), ‘Le Livre des Caractères de Qostâ ibn Lotigd’, Bulletin de l'Institut d'Égypte, 23: 103-69. Schmeling, G. L, (1996) (ed), The Novel in the Ancient World (Leiden). Schmidt, J. (1941), ‘Physiognomik’, RE xx/1: 1064-74. Schmitz, T. (1997), Bildung und Macht: Zur sozialen und politischen Funktion der zweiten Sophistik in der griechischen Welt der Kaiserzeit (Munich), Schmölders, C. (ed.) (1996), Der exzentrische Blick: Gespräch über Physiognomik (Berlin). Schneider, J. P. (2000), ‘Eudoxe de Cnide’ (E98), in Goulet (1989-) iii. 293-302, Schofield, M. (1991), The Stoic Idea of the City (Cambridge). Schöne, H. (1910), ‘Echte Hippokratesschriften’, Deutsche medizinische Wochenschrift, 36: 418-19, 466-7 Schouler, B. (1984), La Tradition hellénique chez Libanios, 2 vols. (Paris). Scourfield, J. H. D. (2003), ‘Anger and Gender in Chariton’s Chaereas and Callirhoe’, in Braund and Most (2003) 163-84. Sedley, D. N. (1993), “Chrysippus on Psychophysical Causality’, in J. Brunschwig and M. C. Nussbaum (eds.), Passions and Perceptions: Studies in Hellenistic Philosophy of Mind (Proceedings of the 5th Symposium Hellenisticum, Cambridge), 313-31. Shahbazi, A. 5. (1990), ‘On the Xwaday-Namag’, Acta Iranica, 16: 208-29. Sharples, R. W. (1987), ‘Alexander of Aphrodisias: Scholasticism and Innovation’, ANRW I 36. 2: 1176-1243, —— {2001}, ‘Dicaearchus on the Soul and on Divination’, in Fortenbaugh and Schütrumpf (2001) 143-73. Shaw, B. (1987), ‘The Family in Late Antiquity’, Past & Present, 115: 3-51. Sheppard, A. R. R. (1984-6), ‘Homonoia in the Greek Cities of the Roman Empire’, Ancient Society, 15-17: 229-52, Sherwin-White, A. N. (1966), The Letters of Pliny. A Historical and Social Commentary (Oxford), Sihvola, J., and Engberg-Pedersen, T. (1998) (eds.}, The Emotions in Hellenistic Philosophy (Dordrecht and London). Singer, P. N. (1992), ‘Some Hippocratic Mind-Body Problems’, in López Férez (1992), 131-43. —— (19974), Galen, Selected Works: Translated with an Introduction and Notes (Oxford). — (19978) ‘Levels of Explanation in Galen’, CQ, ns 47: 525-42. Smith, A. (1993) (ed.), Porphyrius: Fragmenta (Leipzig). Smith, R. R. R. (1998), ‘Cultural Choice and Political Identity in Honorific Portrait Statues in the Greek East in the Second Century A.D’, JRS 88: 56-93, Smith, W. D. (1979), The Hippocratic Tradition (Ithaca and London). Sorabji, R. (1990), ‘Perceptual Content in the Stoics’, Phronesis, 35: 307-14. —— (2000), Emotion and Peace of Mind: From Stoic Agitation to Christian Temptation (Oxford). Sourdel, D. (1952-4), ‘Le «livre des secrétaires» de ‘Abdallah al-Baghdadi’, Bulletin Languages and Literatures, 56: 175-224, 325-36, 57: 302-5, Städele, A. (1980), Die Briefe des Pythagoras und der Pythagoreer (Meisenheim am Glan). Staden, H. von (1997), “Teleology and Mechanism: Aristotelian Biology and Early Hellenistic Medicine’, in Kullmann and Föllinger (1997) 183-208, Steel, C. (2001), “The Moral Purpose of the Human Body: A Reading of Timaeus 69-72’, Phronesis, 46: 105-28. j Stegemann, W. (1952), ‘Polemon’, in RE xxif2: 1320-57. Steinschneider M. (1865), ‘Die “mittleren” Bücher der Araber und ihre Bearbeiter‘, Zeitschrift für Mathematik und Physik, 10: 456-98. —— (1960), Die arabischen Übersetzungen aus dem Griechischen (repr. Graz). Stewart, A. (1993), Faces of Power: Alexander's Image and Hellenistic Politics (Berkeley), Stewart, P. (2003), Statues in Roman Society: Representation and Response (Oxford). Stok, F, (1992), ‘I prologo del De physiognomonia’, in C. Santini and N. Scivoletto (eds.), Prefazioni, prologhi, proemi di opere tecnico-scientifiche latine, ii (Rome), 499-517. (1993a), ‘Physiognomonica Caesariana’, in D. Poli (ed), La cultura in Cesare (Atti del convegno internazionale di studi Macerata-Matelica, 30 aprile-4 maggio 1990; Rorne) i, 59-84 (19936), ‘Physiognomonica maniliana’, in D. Liuzzi (ed.}, Manilio: Fra poesia e scienza (Atti del convegno Lecce, 14-16 maggio 1992; Galatina) 169-84. (1995) ‘Ritratti fisionomici in Svetonio’, in I. Gallo and L. Nicastri (eds), Biografia e autobiografia degli antichi e det moderni (Naples), 109-35. Stroud, B. (2000), The Quest of Reality: Subjectivism and the Metaphysics of Colour (Oxford). Strubbe, J. H. M. (1984-6), ‘Gründer kleinasiatischer Städte: Fiktion und Realität’, Ancient Society, 15-17; 253-304. Swain, 5. (1989), ‘Favorinus and Hadrian’, Zeitschrift für Papyrologie und Epigraphik, 79: 150-8. —— (1990), ‘The Promotion of Hadrian of Tyre and the Death of Herodes Atticus’, CPh 85: 214-16. — (1994), ‘Man and Medicine in Thucydides’, Arethusa, 27: 303-27. —— (1996), Hellenism and Empire: Language, Classicism, and Power in the Greek World, AD 50-250 (Oxford). —— (19994) (ed), Oxford Readings in the Greek Novel (Oxford). —— (1999b), ‘Defending Hellenism: Philostratus, In Honour of Apollonius, in M. J. Edwards et al. (eds.), Apologetics in the Roman Empire (Oxford), 157-96. —— (2000) (ed.}, Dio Chrysostom: Politics, Letters, and Philosophy (Oxford). —— (2002), ‘Bilingualism in Cicero? The Evidence of Code-Switching’, in Adams, Janse, and Swain (2002) 128-67. (2004), ‘Bilingualism and Biculturalism in Antonine Rome: Apuleius, Fronto, and Gellius’, in L. A. Holford-Strevens and A. Vardi (eds.), The Worlds of Aulus Gellius (Oxford), 340. —— (2006), ‘Beyond the Limits of Greek Biography: Galen from Alexandria to the Arabs’, in B. McGing and J. Mossman (eds.), The Limits of Ancient Biography (Dublin), d'études orientales, 14: 115-53. Spawforth, A. J. 5. (1989), ‘The Image of Tradition’, in P. Cartledge and A. Spawforth, 395-433. (forthcoming), ‘Culture and Nature in Philostratus’, in E. L. Bowie and J. Elsner (eds.}, Philostrasus (Cambridge). Hellenistic and Roman Sparta (London), 190-211. —— (1999), ‘The Panhellenion Again’, Chiron, 29: 339-52. and Y. Rapoport (forthcoming), Brysor’s Administration of the House: Family, Society, and Economy, from Rome to Islam.

Bibliography Bibliography 689 Syme, R. (1958), Tacitus, 2 vols. (Oxford). Tabbakh, M. R. al- (1347/1929), see under Primary Sources, Tantum, G, (1979), ‘Muslim Warfare: A Study of a Medieval Muslim Treatise on the Art Vesel, Z. (1986), Les Encyclopédies persanes: Essai de typologie et de classification des Taylor, D. G. K. (1999), The Syriac Versions of the De Spiritu Sancto by Basil of Caesarea Verne, P. (1978), ‘La famille et Pamour sous le Haut-Empire romain’, Annales B‚5.C. 33: of War’, in R. Elgood (ed), Islamic Arms and Armour (Lendon), 187-201, (CSCO 576; Leuven). Thalmann, W. G. (1988), ‘Thersites’, TAPRA 118: 1-28. Thein, K. (2002), ‘Gods and Pzinters: Philostratus the Elder, Stoic avragie and the Strategy of Describing’, Ramus 31: 136-45. Thesleff, H. (1961), An Introduction to the Pythagorean Writings of the Hellenistic Period (Abo [Turku]). —— (1965), The Py:hagorean Texts of the Hellenistic Period (Abe [Turkuj). Thillet, P. (1981), ‘Matérialisme et theorie de l’âme et de l'intellect chez Alexandre d’Aphrodise’, Revue philosophique de la France et de étranger, 106: 5-24. Thomann, J. (1996), ‘Avicenna über die physiognomische Methode’, in Campe and Schneider (1996) 47-63. —— (1997), ‘Studien zum “Speculum physionomie” des Michele Savonarola’ (Diss. Zurich). Thonemann, P. J. (2004), ‘Polemo, Son of Polemo (Dio, 59.12.2)’, Epigraphica Anatolica, 37: 144-50. Tieleman, T. (2003), Chrysippus’ On Affections. Reconstruction and Interpretation (Philosophia Antiqua, 94; Leiden and Boston), Todd, R. B. (1977), ‘Galen's Medical Ideas in the Greek Aristotelian Commentators’, Symbolae Osloenses, 52: 117-34. (1984), ‘Philosophy and Medicine in John Philoponus’ Commentary on Aristotle's De Anima’, in J. Scarborough (ed.), Symposium on Byzantine Medicine (Dumbarton Oaks Papers, 38; Washington), 10310, Toole, H. (1974-5), “Eig notov Zukparikòv mpénet và dnodc6f fj Sugnopeia Toû Zupdroug; Hepì Tfig xpavoAoyrigewg TBV Epyuv 1@v Zukparıküv”, Athena, 75: 303-17. Toomer, G. J, (1984), Ptolemy's Almagest (London). Treggiari, S. (1991), Roman Marriage lusti Coniuges from the Time of Cicero to the Time of Ulpian (Oxford). Trimble, 3. (2000), ‘Replicating the Body Politic: The Herculaneum Woman Statues’, Journal of Roman Archaeology, 13: 41-68, Tsouna, V, (1998), ‘Doubts about Other Minds and the Science of Physiognomics’, CQ, Ns 48: 175-86. Turner, V. (1966), ‘Colour Classification in Ndembu Ritual’ in M. Banton (ed), Anthropological Approaches to the Study of Religion (London), 47-84, Turton, D. (1980), “There's No Such Beast: Cattle and Colour Naming Among the Murs”, Man, 15: 320-38, Tytler, G, (1982), Physiognomy in the European Novel: Faces and Fortunes (Princeton). Ullmann, M. (1961), Die arabische Uberlieferung der sogenannten Menandersentenzen (AKM 34. 1; Wiesbaden). —~ (1970), Die Medizin im Islam (Leiden and Cologne). —— (1972), Die Natur- und Geheimwissenschaften im Islam (Leiden). —— (1978), Islamic Medicine (Islamic Surveys 11; Edinburgh). —— (2002), Wörterbuch zu den griechisch-arabischen Übersetzungen des 9, Jahrhunderts (Wiesbaden). Uri, J. (1787), Bibliothecae Bodleianae codicum manuscriptorum Orientalium... catalogus, i (Oxford). Vadet, J. C. (1968), L'Esprit courtois en Orient (Paris). (1995) Les Idées morales dans l'Islam (Paris). sciences (Institut Francais de Recherche en Iran no. 31; Paris). 35-63; repr. in La Société romaine (Paris, 1991), 88-130. —— (1990), Bread and Circuses, trans. B. Pearce (London). Vogt, S. (1999), Aristoteles, Physiognomonica: Übersetzt und kommentiert (Aristoteles’ Werke in deutscher Ubersetzung, 18/6; Berlin). Walker, A. (1991), ‘Erés and the Eye in the Love-Letters of Philostratus’, PCPhS 38; 132-48. Walker, S., and Bierbrier, M. (1997), Ancient Faces: Mummy Portraits from Roman Egypt (London), Walzer, R. (1949), Galen on Jews and Christians (Oxford). —— (1962), Greek into Arabic: Essays on Islamic Philosophy (Oxford). Wardman, A. E. (1967), ‘Description of Personal Appearance in Plutarch and Suetonius: The Use of Statues as Evidence’, CQ, ns 17: 41422, Watson, G. (1971), "The Natural Law and Stoicism’, in Long (1971) 216-38. Watts, R. J. (2003), Politeness (Cambridge). Wegner, M, (1956), Hadrian (Das rémische Herrscherbild, ii/3; Berlin). Weiler, 1. (1996), ‘Physiognomische Überlegungen zu mens sana in corpore sano’, in €. Klodt (ed), Sature Lanx: Festschrift Werner A. Krenkel zum 70. Geburtstag (Spudasmata, 62; Hildesheim), 153-68. Weiss, P. (1995), “Hadrian in Lydien’, Chiron, 25: 213-23. Weitzmann, K. (1976), The Monastery of St Catherinc at Mount Sinai: The Icons, i (Princeton). Wellmann, M. (1913), ‘Zu Diokles’, Hermes, 48: 464-8. White, L. Michael (2005), ‘Favorinus’s “Corinthian Oration”: A Piqued Panorama of the Hadrianic Forum’, in D. N. Schowalter and $. J. Friesen (eds), Urban Religion in Roman Corinth: Interdisciplinary Approaches (Cambridge, Mass.), 61-110. Wilamowitz-Moellendorff, U. von (1900), ‘Asianismus und Atticismus’, Hermes, 35: 1-52; repr. in Kleine Schriften, iii (Berlin 1969) 223-73. Wilhelm, F. (1915), ‘Die Oeconomica der Neupythagoreer Bryson, Kallikratidas, Periktione, Phintys’, RhM, nr 70: 161-223. Williams, W. (1976), ‘Individuality in the Imperial Constitutions: Hadrian and the Antonines’, JRS 66: 67-83. Winterling, A. (1997), ‘Hof ohne “Staat”: Die aula Caesaris im 1. und 2. Jahrhundert n. Chr’, in id. (ed), Zwischen ‘Haus’ und ‘Staat’: Antike Höfe im Vergleich (Historische Zeitschrift, Beiheft 23) (Munich), 91-112. —— (1929), Aula Caesaris: Studien zur Institutionalisierung des römischen Kaiserhofes in der Zeit von Augustus bis Commodus (31 v. Chr.-192 n. Chr.) (Munich). Witkam, J. J. (1989), De egyptische arts Ibn al-Akfän? (gest. 749/1348) en zijn indeling van de wetenschappen: Editie van het Kitab irSad al-gäsid ila asnä al-magasid met een inleiding over het leven en werk van de auteur (Leiden). (2004), ‘Aflimiin’, EI 12: 44-5. Wörrle, M, (1988), Stadt und Fest im kaiserzeitlichen Kleinasien (Munich). Wooten, C. W. (1987), Hermogenes’ On Types of Style (Chapel Hill). Wrede, H. (1981), Consecratio in Formam Deorum (Mainz). Wright, J. P, and Potter, P. (2000) (eds), Psyche and Soma: Physicians and Metaphysicians on the Mind-Body Problem from Antiquity to Enlightenment (Oxford). Wright, M. R. (2000) (ed.), Reason and Necessity: Essays on Plato's Timaeus (London).

Bibliography Zanker, P. (1995), The Mask of Socrates (Berkeley). Zanoncelli, L. (1977), ‘La filosofia musicale di Aristide Quintiliano’, Quaderni urbinati di cultura classica, 24: 51-93. Zayyat, H. (1939), ‘Adab al-ma’ida fi |-Islam’, al-Machrig, 37: 162-76. (1956), 'al-Mar’a al-ghulämiyya fi ]-Islam’, al-Machrig, 50: 152-92. Zeller, E., and Mondolfo, R. (1932-8), La filosofia dei Greci nel suo sviluppo storico, 6 vols in 3 (Florence) [Italian translation, with revisions, of E. Zeller, Die Philosophie der Griechen in ihrer geschichtlichen Entwicklung, 6th edn. (Leipzig, 1919-20)]. Zimmermann, F. W. (1986), “The Origins of the so-called Theology of Aristotle’, in J. Kraye, W. F. Ryan, and C. B. Schmitt (eds.), Pseudo-Aristotle in the Middle Ages: The Theology and Other Texts (London), 111-240. Zonta, M. (1992), Fonti greche e orientali dell’Economia di Bar-Hebraeus nell’opera “La crema della scienza” {AION 52/1, Suppl. 70; Naples). —— (1995), Un interprete ebreo della filosofia di Galeno: GH scritti di Galeno nell’ opera di Shem Tob ibn Falaguera (Turin).