¹ The past fifteen years have seen the publication of Rufus of Ephesus, On jaundice; parts of his On melancholy; and some of his Case histories; and Galen, Commentary on Airs, Waters and Places (part). An edition of the Arabic version of the whole of this commentary is in active preparation, as is my own edition of Galen, On my own opinions.

² Galeni De optima medico cognoscendo libelli versionem arabicam primum edidit, in linguam anglicam vertit, commentatus est Albert Z. Iskandar, Corpus medicorum graecorum supplementum orientate 4 (Berlin, 1988).Google Scholar This will be cited throughout as Iskandar. Unless specific linguistic points are at issue, references within the text are given by chapter and subsection only.

³ Iskandar, A. Z., ‘Al-Rāzī wa mihnat al-tabīb’, Al-Machriq 54 (1960), 471–522.Google Scholar

⁴ Iskandar, A. Z., ‘Galen and Rhazes on Examining Physicians’, Bulletin of the History of Medicine 36 (1962), 362–5.Google Scholar

⁵ Dietrich, Albert, ‘Medicinalia arabica’, Abhandlungen der Akademie der Wissenschaften in Göttingen, phil.-hisl. Kl., 3 Folge, 66 (1966), pp. 190–206.Google Scholar In his description of the manuscript in his edition, p. 14, Iskandar omits the crucial library name (cf. Dietrich, , pp. 245, 258, for other libraries in Bursa).Google Scholar

⁶ Meyerhof, M., ‘Autobiographische Bruchstücke Galens aus arabischen Quellen’, Sudhoffs Archiv 22 (1929), 75–82.Google ScholarIskandar, , p. 16Google Scholar, adds a further citation from this author. In my edition of On prognosis, Corpus medicorum graecorum v.8.1 (Berlin, 1979), I was able to provide parallel passages only from the sections already in print.

⁷ Bergsträsser, G., Ḥunain ibn Isḥāq Über die syrischen and arabischen Galen-Übersetzungen, Abhandlung für die Kunde des Morgenlandes 17.2 (1925), p. 38 n. 112.Google Scholar The conjunction of Bakhtīshū’ ibn Jibrīl and Muhammad ibn Mūsā as the patrons of the respective translations gives at least a rough date and a place for the versions, c. 850 in Baghdad. Iskandar, , pp. 18–19, 25–6Google Scholar, rightly concludes that there is every reason to suppose that the present version is by Hunain.

⁸ See the comments by Ḥunain reproduced by Rosenthal, Franz, The Classical Heritage in Islam (Berkeley, 1975), pp. 19–21.Google Scholar

⁹ As-Safādī, translated by Rosenthal, ibid., p. 17.

¹⁰ Bergsträsser, G., Die bisher veröffentlichten arabischen Hippokrates- und Galen-Über-setzungen, Diss., Leipzig, 1912Google Scholar, later published in an enlarged form as Ḥunain ibn Isḥāk und seine Schule (Leiden, 1913)Google Scholar; Garofalo, I., Anatomicarum administrationum libri qui supersunt novem (Naples, 1986), esp. pp. xvii–xxviiGoogle Scholar, with a good bibliography of previous stylistic studies. I fail to understand why Iskandar, , p. 26Google Scholar, says that ‘nobody knows the method adopted by Ḥunayn… and his school… in rendering… Galen… into Arabic’, unless' method' is taken in the narrowest sense of discovering the precise meaning of a Greek word. Even so, Iskandar makes no reference to the work of any other scholar save Bergsträsser, and cites only parallel expressions found in other translations ascribed to Ḥunain and his school.

¹¹ The formulation is that of Bachmann, Peter, ‘Galens Abhandlung dazü, daβ der vorzügliche Arzt Philosoph sein muβ’, Nachr. Akad. Wiss. Göttingen 1965, p. 10.Google Scholar Cf. for his pupil Hubaish, Biesterfeldt, H. H., ‘Galens Traktat “Dass die Kräfte der Seele den Mischungen des Körpers folgen’, Abhandlungen für die Kunde des Morgenlandes 40.4, 1973, pp. 16–28.Google Scholar

¹² For similar warnings, note Diller, Hans, Kleine Schriften zur antiken Medizin (Berlin, New York, 1973), pp. 155–7CrossRefGoogle Scholar; and Alexanderson, Bengt, Galenas ΠΕΡΙ ΚΡΙΣΕΩΝ (Gothenburg, 1967), pp. 54–7.Google Scholar

¹³ Rosenthal, , The Classical Heritage, p. 19Google Scholar; note also Deichgräber, K., ‘Parabasenverse aus Thesmophoriazusen II des Aristophanes bei Galen’, Sitz. Akad. Wiss. Berlin, Kl. f. Sprachen, Lit. und Kunst 1956, 2.Google Scholar

¹⁴ Strohmaier, G., ‘Ḥunayn ibn Isḥāq et le serment hippocratique’, in Ḥunayn ibn Isḥāq (Leiden, 1975), pp. 321–2.Google Scholar

¹⁵ cf. Grignaschi, Mario, ‘La “Physiognomic” traduite par Ḥunayn ibn Isḥāq’, in Ḥunayn, pp. 285–91.Google Scholar This argument is employed by Toufic Fahd in his defence, against Manfred Ullmann, of the authenticity of Ḥunayn's translation of Artemidorus' Dream book, ‘Ḥunayn ibn Isḥāq est-il traducteur des Oneirocritica d'Artémidore d’Éphèse?’, ibid., pp. 270–84.

¹⁶ cf. Iskandar, , p. 36, for these two examples.Google Scholar

¹⁷ Dietrich, , op. cit. (n. 5), p. 194Google Scholar n. 1, had earlier ascribed this quotation to Posidonius.

¹⁸ Iskandar, , p. 95, 7Google Scholar, translates kalām as ‘words’, which might seem to make a contrast with ‘things’. However, its more usual meaning is ‘argument’, and that is how he translates it when the phrase is repeated at 94, 8–9 = 95, 11–12, and again in his summary of the Thucydidean comment at p. 141. The presence of ‘yaṣeḥḥu’, ‘made valid’ (‘authenticatedā Iskandar), might suggest that Ḥunain was thinking of a form of δικαιον, but it is more likely that he was simply aiming to give the sense of this gnomic statement.

¹⁹ Iskandar's painstaking English translation corresponds well to the Arabic, but, at times, his choice of words gives a meaning more remote than Ḥunain's Arabic from what Galen originally wrote.

²⁰ Iskandar, , pp. 12–13Google Scholar

²¹ Bergsträsser, G., op. cit (n. 10), nn. 118, 112.Google Scholar

²² The secondary tradition is reviewed by Iskandar, , pp. 18–25Google Scholar, but neither here nor on p. 12, where he cites the bibliographical entries in the late-tenth century Fihrist of Ibn an-Nadīm and in the thirteenth-century dictionary of Ibn al-ifṭī, are the details of the titles always given in full. The apparatus criticus at 1,1 cites only the variants in the headings of the two manuscripts, and hence the unwary reader may not realise the weight of authority against Iskandar's preferred title.

²³ . Bergsträsser, , op. cit. (n. 10), n. 112.Google Scholar Although he here expressly states that this book is not found in Galen's list of his own books, he makes no mention of it in the treatise he himself wrote on the books that Galen omitted from his catalogue, see Bergsträsser, G., Neue Materialien zu Ḥunain Ibn Isḥāq's Galen-bibliographie, Abhandlungen fur die Kitnde des Morgenlandes 19.2 (1932)Google Scholar

²⁴ Iskandar, , p. 30.Google Scholar His previous pages, 27–30, endeavour to establish the authority of Ḥunain, as a witness to authenticity, which is hardly at issue here.

²⁵ Iskandar, , pp. 30–4.Google Scholar The reference to Herodes Atticus as ‘the most able orator of our time’, 112, 11 = 113, 15, does not, pace Iskandar, , p. 31Google Scholar, place it at around the time of Herodes' death, c. 177, a date which he himself immediately rejects.

²⁶ Peterson, D. W., ‘Observations on the chronology of the Galenic Corpus’, Bulletin of the History of Medicine 51 (1977), 495.Google ScholarPubMed

²⁷ Ibid., pp. 492–3, accepted by Iskandar, , p. 32.Google Scholar

²⁸ Rightly, Iskandar, , p. 31Google Scholar, citing ix. 770–1 K.

²⁹ Iskandar, , p. 90, 15–17 = 91, 19–21.Google Scholar

³⁰ Bardong, K., ‘Beiträge zur Hippokrates- und Galenforschung’, Nachrichten Akad. Wiss. Göttingen 1941, p. 635.Google Scholar

³¹ Peterson, , op. cit. (n. 26), p. 491.Google Scholar

³² Bardong, , op. cit. (n. 30), pp. 605, 617.Google Scholar

³³ Iskandar, , p. 46, 13–48, 2 = 47, 19–48, 3.Google Scholar Galen's characterisation of On prognosis is found at In Hippocratis Epidemianm librum VI comment. VIII: CMG v 10, 2, 2, p. 495, 2–12.Google Scholar

³⁴ cf. Iskandar, , p. 52Google Scholar, 11 = 53, 19 with xiv. 602 K.; p. 62, 3–10 = 63, 3–14 with xiv.665–9 K.; and, for verbal parallels, pp. 100, 13–106, 2 = 101, 20–107, 2, with xiv. 600 K.

³⁵ Admittedly, this is far from being conclusive, for it is remarkable how Galen can take up, even after thirty years, problems discussed earlier, and continue his exploration often in similar language. He claimed consistency of doctrine, at least in essentials, throughout his career, but both tone and theme coincide in these two treatises even more than usual.

³⁶ For the date of On prognosis, see my edition, p. 49.

³⁷ cf. Iskandar, , p. 138Google Scholar, for Galen's references to this book as a written treatise.

³⁸ For On prognosis, see the discussion at CMG v.8,1, pp. 49–51.

³⁹ See On prognosis, passim.

⁴⁰ For the details of the system of tax-immune physicians, see my From Democedes to Harvey (London, 1988), chs. IV–VI.Google Scholar

⁴¹ Iskandar, , p. 165Google Scholar, relates the public ‘contest’ between doctors at 9.6 to the choice of public doctors. This is unlikely for several reasons. At this date, the choice was made by the town council, not by the population at large; the post Galen eventually obtained was that of doctor to the gladiators, and was in the gift of the High Priest of Asia, and, as such, was not a public appointment; what is described is less of a public examination than a challenge match, such as we know from other passages in Galen (e.g. viii.158, xi.188 K.). or the medical ‘contests’ held at Ephesus and possibly elsewhere in Asia Minor. The incident took place in Pergamum, and hence says nothing about conditions in Rome. Given that. Galen knew of public physicians (v.751 K.) and that some of his audience might well have served on a town council in their ‘native’ cities, his silence can hardly be accidental.

⁴² At 11.7 Galen specifically contrasts his own abilities with those of the average, but still competent, physician, and suggests that such a man ought not to be rejected for not being the ‘complete’ physician. The same adjective is often found on honorific decrees and tombstones. Interestingly, Soranus, , Gyn. 3Google Scholar, places the ‘complete’ physician at a lower level of competence than the ἄριστος, see Gourevitch, D., ‘Un thérapeute accompli. Note sur l'adjectif τ⋯λειος’, Rev. phil. 61 (1987), 95–9.Google Scholar

⁴³ Galen's plea is subtly different from the suggestion of Athenaeus of Attaleia, 150 years earlier, that for a layman to learn about medicine would provide him with intellectual stimulation as well as practical benefit, should he fall ill.

⁴⁴ Edelstein, Ludwig, Ancient Medicine (Baltimore, 1967), pp. 65–85.Google Scholar

⁴⁶ But cf. the comment of Philostratus, VS 536, that Timocrates of Pontus, fl. a.d. 140, as one who had devoted himself to the study of medicine, was fluent in the theories of Hippocrates and Democritus.

⁴⁶ cf. Edelstein, , Ancient Medicine, pp. 180, 190Google Scholar; Rawson, E. D., Intellectual Life in the Late Republic (London, 1985), pp. 170–80Google Scholar; Gourevitch, D., Soranos d'Éphèse, Maladies des femmes (Paris, 1988), pp. xix–xxii.Google Scholar

⁴⁷ cf. Garofalo, Ivan, Erasistrati fragmenta (Pisa, 1988), pp. 10–15.Google Scholar

⁴⁸ Galen, , On medical experience, 13Google Scholar; pp. –10 Walzer. This section uses these ‘dogmatist’ names to attack those dogmatists who reject experience and who rely solely on reason as a guide.

⁴⁹ von Staden, Heinrich, ‘Hairesis and Heresy: the case of the haireseis iatrikai’, in Meyer, Ben F., Sanders, E. P., edd., Jewish and Christian Self-Definition (London, 1982), iii, pp. 76–100, esp. pp. 81–3Google Scholar; this triple division of the sects may well have been canonical before Galen, , pace von Staden, p. 83.Google Scholar

⁵⁰ Fuchs, R., ‘Aus Themisons Werk Ueber die acuten und chronischen Krankheiten’, Rheinisches Museum 58 (1903), pp. 68–113.Google Scholar Fuchs' edition is greatly in need of improvement, not least because it was based only on two poor manuscripts. For the question of authorship and date, cf. Kollesch, J., Untersuchungen zu den pseudogalenischen Definitiones medicae (Berlin, 1973), pp. 24–8Google Scholar; and Garofalo, , Erasistrati fragmenta, pp. 9–10.Google Scholar

⁵¹ Kollesch, , UntersuchungenGoogle Scholar, is fundamental for any study of such introductory literature. Cf. also, Lawn, Brian, The Salernitan Questions (Oxford, 1963), pp. 3–5.Google Scholar

⁵² Notably in the case of Philistion, a young man from Pergamum who was led astray by following the teachings of his master Metrodorus, CMC v.1.1, pp. 401–3.

⁵³ Ḥunain's version, which I follow here, is somewhat banal, for obviously a man of intelligence and some learning is superior to a man who has neither. A better and more Galenic interpretation, which may have been clearer in the original Greek, would be that this man would be better than the unintelligent but experienced healer, and than the intelligent but inexperienced one. The latter proposition is obvious; in the former, Galen is ranking experience of cases derived from reading almost on a par with bedside experience.

⁵⁴ Kollesch, , op. cit. (n. 50), p. 14.Google Scholar

⁵⁵ Nutton, , From Democedes to Harvey, VIII, pp. 30–3.Google Scholar

⁵⁶ 9.21. Iskandar's translation of the original ⋯λπ⋯ζειν ‘hope’ gives a weaker impression of Galen's contempt.

⁵⁷ 1, 12; 8, 2; 8, 9. Cf. also Galen's complaint, xix.59–60 K., about those who have left off practising medicine and turned to politics.

⁵⁸ Philostratus, , Lives of the sophists 559–61.Google Scholar The statement of Iskandar, , p. 170Google Scholar, that Herodes himself heard Aristocles at Pergamum rests on a mistranslation of Philostratus' Greek by Wright, Loeb, ed., p. 185.Google Scholar Herodes sent his pupils to Pergamum but did not go himself.

⁶⁹ See Bowersock, G. W., Greek Sophists in the Roman Empire (Oxford, 1969), pp. 81–2.Google Scholar

⁶⁰ cf. Baldwin, Barry, Studies in Lucian (Toronto, 1973), pp. 38–40.Google Scholar The attempt by Scarborough, J., ‘The Galenic Question’, Sudhoffs Archiv 65 (1981), 1–31Google ScholarPubMed, to discredit Galen's evidence on the grounds of lack of confirmation from elsewhere is not convincing, see my From Democedes to Harvey, ch. III.

⁶¹ Galen's perspective is shared by MacMullen, Ramsay, Corruption and the Decline of Rome (New Haven and London, 1988), pp. 60–88.Google Scholar

⁶² Quintilian, , Decl. 268Google Scholar; cf. Garin, E., La disputa delle arti nel Quattrocento (Florence, 1947).Google Scholar

⁶³ Galen's Greek may originally have been ⋯φημερ⋯δες, with its hints of the diaries of Alexander and other generals. But it should be emphasised that Ḥunain may well have been merely interpreting Galen here, and the exact wording should not be pressed. For Charax, see Andrei, Osvalda, A. Claudius Charax di Pergamo: interessi antiquari e antichitè cittadine nell' etè degli Antonini (Bologna, 1984).Google Scholar

⁶⁴ cf. Momigliano, A. D., ‘History between Medicine and Rhetoric’, Ann. scuola norm. sup. Pisa, ser. 3, 15 (1985), 767–80, esp. 773Google Scholar, where he notes Galen's ‘attempt to distance himself from history and to side with philosophy’.

⁶⁵ See Kudlien, F., ‘Galens Urteil über die Thukydideische Pestbeschreibung’, Episteme 5 (1971), 132–3.Google Scholar

⁶⁶ xviiiA.450 K. = Thucydides, Hist. 1.63 and 118; xviiiB.849 K. = Hist. 3.23.

⁶⁷ cf. Silk, M. S., ‘LSJ and the Problem of Poetic Archaism: from Meanings to Iconyms’, CQ 33 (1983), 309–9CrossRefGoogle Scholar, who argues that, for much of his linguistic commentary, Galen will have been drawing on earlier Alexandrian grammarians.

⁶⁸ xix.60–1 K., where this (good) reason is contrasted with that of gaining a good Attic literary style. Iskandar, , p. 173Google Scholar, rightly draws attention to Galen's dislike of etymological argumentation but says nothing about Galen's own lexical studies.

⁶⁹ Jones, C. P., Culture and Society in Lucian (Cambridge and London), 1986CrossRefGoogle Scholar, is the most recent study; for the Latin side, see Champlin, Edward, Fronto and Antonine Rome (Cambridge and London, 1980).CrossRefGoogle Scholar

⁷⁰ See, in particular, Bowie, E. L., ‘Greeks and their Past in the Second Sophistic’, Past and Present 46 (1970) 3–41.CrossRefGoogle Scholar

⁷¹ For Cornelianus, see Bowersock, G. W., op. cit. (n. 59), p. 54.Google Scholar Cf. now also Argyle, S., ‘A New Greek Grammarian’, CQ 39 (1989), 524–35.CrossRefGoogle Scholar

⁷² Reitzenstein, 's Geschichte der griechischen Etymologika (Leipzig, 1897)Google Scholar, which formed the basis for his RE article ‘Etymologika’, RE vi.1, 807–17, is unhelpful. Better is Sandys, J. E., A History of Classical Scholarship (Cambridge, 1903), i.315–20Google Scholar, but he omits much. Some indication of etymological investigations in Galen's time can be gleaned from Latte, Kurt, Kleine Schriften (Munich, 1968), pp. 626–7, 649–66.Google Scholar It is perhaps significant that Galen does not attack here other physicians for their interest in etymologies, most notably Soranus, whose Περ⋯ ⋯τυμολιν το σώματος το ⋯νθρώπου is cited by later grammarians, see Kind, F. E., RE, ser. 2, 3, 1927, cols. 1117–18.Google Scholar

⁷³ Iskandar, , pp. 143–4Google Scholar, interprets ‘kings’ solely as Asclepius and the Asclepiads. But Galen never refers to Asclepius as a king, see, in particular, the fragments of his commentary on the Oath, ed. Rosenthal. Possible identifications would be with the Homeric heroes taught by Cheiron the centaur, Ptolemy I and II, Attalus III, and Ullmann, Mithridates V. Manfred, Rufus von Ephesos, Krankenjoumale (Wiesbaden, 1978), p. 120Google Scholar, notes that in Arabic ‘mulūk’ can stand for ‘nobles’ and ‘princes’ as well as ‘kings’, citing this passage from the translation literature. But here, as in the case in the Krankenjournale, an original βασιλες would fit nicely.

⁷⁴ Iskandar, , p. 142Google Scholar, identifies the ‘orators’ solely with his fellow townsman, the great orator Aelius Aristeides, whose hymns in praise of Asclepius are now lost. But this is far too restrictive; see the evidence for poems in honour of Asclepius and Hygieia (and plays, see Aelian, fr. 101 Hercher) collected by , E. J. and Edelstein, L., Asclepius (Baltimore, 1945), i, T.587–608 = pp. 326–37Google Scholar; MacMullen, R., Paganism and the Roman Empire (New Haven and London, 1981), pp. 15–20Google Scholar; and Bremmer, J. M., ‘Greek Hymns’, in Versnel, H., ed., Faith, Hope and Worship (Leiden, 1981), pp. 193–215.CrossRefGoogle Scholar Apuleius, Florid. 18 = Edelstein, T.608, is an example of a contemporary Latin orator who composed a carmen in honour of Asclepius.

⁷⁵ See Kudlien, F., ‘Galen's Religious Belief’, in Nutton, V., ed., Galen: Problems and Prospects (London, 1981), pp. 117–30.Google Scholar

⁷⁶ Ed. Rosenthal, , pp. 59–60Google Scholar, a passage confirming the Pergamene authorship of the commentary, which, despite Rosenthal's proper hesitations, can hardly have been written by anyone other than Galen. The comments, pp. 74–6, on the superiority of health to wealth, children or political power, which are similar to those in 1.1 and 9.14, are not noted by Iskandar. Kudlien, , op. cit. (n. 75), p. 119Google Scholar, appears to confuse the divine origin and status of healing in general (which Galen undoubtedly accepts) with a development of (Hippocratic) medicine out of temple medicine and temple records (which he would not). The longer passage from the commentary on the Oath, pp. 56–9Google Scholar Rosenthal, makes it clear that exactly what was implied in a divine origin for medicine was disputed, and Galen's own view would appear to be distinguished from that of ‘people in general’.

⁷⁷ So, rightly, Kudlien, , op. cit. (n. 75), pp. 118, 125.Google Scholar

⁷⁸ xiv.650; xix.18–19 K. For Galen's treatment of Marcus, see Hadot, Pierre, ‘Marc Aurèle, était-il opiomane?’, in Lucchesi, E., Saffrey, H. D., edd, Antiquité païenne et chrétienne; memorial André-Jean Festugière (Geneva, 1984), pp. 33–50Google Scholar, who shows that the idea that Marcus was a drug-addict is based on a gross misunderstanding of Galen's Greek.

⁷⁹ Aurelius, Marcus, Ep. ad Front. 3.9Google Scholar; Med. v.8; vi.43. His attitude towards the natural order of illness and death is evinced by Med. vi.33; x.2 and 36; iii.3.

⁸⁰ See above, n. 72, for the problem of the Greek word underlying the Arabic. At p. 102.1 = 103.1 (cited by Ullmann, , op. tit. (n. 73), p. 120Google Scholar, from Ibn Abī Uṣaibi'a), doctors are criticised for waiting for (rich) men at the doorsteps of ‘kings’ (so Iskandar). Here ‘kings’ can hardly be the right translation, for the emperors were to be found mainly in Rome, and the parallel passage in On prognosis, xiv.600 K., has the doctor making his morning call on ‘the rich and powerful’, and accompanying them everywhere. Either ‘mulūk’ represents another Greek word (οἱ δυν⋯μενοι) or, what is more likely, it forms part of Ḥunain's ‘explanatory expansion’ of some Greek words for social customs absent from the Muslim world. Given that Ḥunain elsewhere in his version refers to the habits of ‘the rich’ and ‘the powerful’ (see Iskandar's index), there was no obvious compulsion for him to translate either term by ‘mulūk’, but, equally, such a translation would easily convey the significance of the comment to a non-Greek audience, and may be what was intended in this passage on ‘kings’ and divine healing.

⁸¹ Lucian, Alexander. For the local background of the cult, see Robert, Louis, A trovers l'Asie Mineure (Paris, 1980), pp. 393–421Google Scholar, and for an attempt to peer behind Lucian's tendentious rhetoric, Lane-Fox, R., Pagans and Christians (Harmondsworth, 1988), pp. 241–50.Google Scholar

⁸² Galen, , In Epid. II comm. VI: CMC v. 1.1, pp. 402–3.Google Scholar See the discussion by Jones, C. P., op. cit. (n. 69), p. 19Google Scholar, who dates the incident to the 170s, which would perhaps strengthen the case for an allusion to Alexander and his shrine, but the Galenic context suits better a date in the late 150s in Egypt.

⁸³ , E. J. and Edelstein, L., op. cit. (n. 74), i, pp. viii, xiGoogle Scholar, for the omission; their list of sites is i, pp. 370–452 = T.707–861; Thraemer, K., RE 2, 1896, cols. 1662–77.Google Scholar Alessandria Semeria, ‘Per un censimento degli asklepieia della Grecia continentale e delle isole’, Ann. sc. norm. sup. Pisa, ser. 3, 16 (1986), 931–58Google Scholar, shows what can be done for a limited area.

⁸⁴ Robert, , A trovers l'Asie MineureGoogle Scholar; MacMullen, , PaganismGoogle Scholar; Lane-Fox, , Pagans and ChristiansGoogle Scholar, from different vantage points, illuminate the situation in the areas of the Roman world that Galen knew best. A comprehensive survey of the evidence for Asclepieia at this time is a desideratum.

⁸⁵ For question-and-answer handbooks, see Kollesch, , op. cit. (n. 50), pp. 35–46Google Scholar; and, for surgery, Marganne-Melard, M. H., ‘La chirurgie dans les papyrus grecs de medecine’, Proc. XXX Int. Congr. Hist. Med., 1986 (Dusseldorf, 1988), pp. 862–4.Google ScholarZalateo, G., ‘Papiri di argomento medico redatti in forma di dimanda e risposto’, Aegyptus 44 (1964), 52–7Google Scholar, sought a connection between such texts and the formal examination of civic physicians.

⁸⁶ See my From Democedes to Harvey, chs. VII and VIII, and Gourevitch, D., Le triangle hippocratique (Paris, 1984).Google Scholar

⁸⁷ Respectively, 1.9: 6.4, 8.5, 9.5, 13.6, 13.8: 4.2, 6.1–2, 11.3–4: 3.5–7, 3.9, 3.15, 11.2.

⁸⁸ e.g. 1.13, 9.14, 13.5, 13.9.

⁸⁹ See, as well as my commentary on On prognosis, Bowersock, , op. cit. (n. 59), pp. 59–75Google Scholar; Kollesch, Jutta, ‘Galen und die Zweite Sophistik’, in Galen: Problems and Prospects, pp. 1–11Google Scholar; Pearcy, Lee T., ‘Galen and Stoic Rhetoric’, Greek, Roman and Byz. Stud. 24 (1983), 259–72.Google ScholarPubMed

* I should like to thank my colleagues Amal-Abou Aly and Larry Conrad for their help with the Arabic.