The origin and transmission of the doxographical tradition placita philosophorum (Arius Didymus, ps. -Plutarch, Stobaeus,Theodoret, Nemesius, Porphyrius) Текст научной статьи по специальности «Языкознание и литературоведение»

Andrei Lebedev

THE ORIGIN AND TRANSMISSION OF THE DOXOGRAPHICAL TRADITION PLACITA PHILOSOPHORUM (Arius Didymus, Ps.-Plutarch, Stobaeus, Theodoret, Nemesius, Porphyrius)

А. В. Лебедев. Происхождение и филиация источников доксографи-ческой традиции «Placita philosophorum» (Арий Дидим, Псевдо-Плутарх, Стобей, Феодорит, Немесий, Порфирий).

В первой части статьи излагаются и дополняются наши аргументы против доксографической теории Дильса, прежде всего против атрибуции «Мнений философов» плохо засвидетельствованному автору Аэтию. Кроме того рассматривается жанровая принадлежность сохранившегося трактата Псевдо-Плутарха De placitis philosophorum и формулируется гипотеза, согласно которой в первоначальной форме этот текст представлял собой учебник физической философии (пара-Sraasiv!), в котором было не 6, а намного больше авторских определений ключевых понятий. Этот утраченный текст мы называем «более полным Плутархом» и обозначаем его как P+. Архетипом сохранившегося Плутарха (P) был, согласно нашей гипотезе, персональный список христианского апологета, которого интересовали исключительно «разногласия» именитых эллинских философов и их споры между собой, поэтому в своей копии он опустил как ненужные и даже нежелательные почти все авторские определения и другие «догматические» части, так как они служили примером общепризнанного знания, а не «разногласия». Во второй части рассматривается комплекс проблем, которые касаются как отношения между P+ и Стобеем, так и между двумя этими источниками и Арием Дидимом. К 8 известным авторским определениям, сохранившимся у Плутарха и Стобея, мы добавляем два неизвестных из Стобея. Первое из них (определение àvayKn, I,4,7b) мы связываем с леммой Ai5up<ou>, откуда следует, что компендий P+ был известен Стобею как сочинение Ария Дидима. Трактат Тертуллиана «О душе» содержит второе свидетельство об авторстве Ария Дидима (apud Arium). В последнем разделе статьи дается новая филиация доксографических источников, соотносимых с традицией Placita philosophorum, в виде древа (stemma), в котором центральная роль отводится и не гипотетическому «Теофрасту» Дильса, и не призрачному «Аэтию», а настоящему гиганту пост-эллинистической философии Арию Дидиму, учителю императора Августа.

Ключевые слова: пост-эллинистическая философия, стоицизм, аристотелизм, доксография, Арий Дидим, Аэтий, Феофраст, псевдо-Плутарх, Стобей, Феодорит, Немесий, Порфирий, христианские апологеты, Г. Дильс.

Table of Contents and Summary1

1. Preface. Status quaestionis.

2. The name and identity of AETIOS in Theodoret's triad of sources nA0YTAPX02-n0P$YPI02-AETI02. Four principal arguments against Diels' attribution of SP-Placita to a totally unknown writer «Aetius».

2. 1. Suspicious name, bad attestation. The personality of Theodoret: a bishop and preacher, not a scholar. 2. 2. The rhetorical character of triadic quotations in CAG. A close parallel from Cyril Contra Julianum.

2. 3. Analysis of the first two triadic quotations in CAG II and IV demonstrates that Plutarch and Porphyrius are Theodoret's real sources, the name of Aetius finds no match in the doxographical material quoted by the bishop.

2. 4. Analysis of the doxographical material in CAG V.16 ff. demonstrates that Theodoret indeed quotes from three different sources, but the name of Aetius does not correspond to the quotations from SP-Placita, and so even Theodoret himself does not ascribe SP-Placita to a writer called Aetius. In other words, the problem with Diels' attribution is not that evidence for Aetius as the compiler of SP-Placita is weak; the real problem is that such evidence does not exist. The confusion of names APEIOS and AETIOS is likely.

2.5 The meaning of Kai psvToi Kai: a linguistic argument against Diels' attribution.

3. The ®uaiKai So^ai of Theophrastus as the Urquelle of all Pre-Platonic physical doxography is a myth of the 19-th century Quellenforschung. Theopharstus is at best a marginal source in SP-Placita.

4. Authorial definitions in SP-Placita and the «better Plutarch» (P+). The extant P is a heavily abridged and truncated copy of P+. In P only 6 authorial definitions remain (and 4 in Stobaeus, two of them neglected) out of many more in P+. Possibly, P is a personal copy of a Christian

The first version of this paper was delivered on March 8th, 20014, at the Centre Léon Robin, Paris Sorbonne. The second, substantially expanded and revised version was delivered on December 3rd, 2015 at the International Colloquium «The Placita of Aëtius» organized by David Runia at Queens College, Melbourne, December 1-4, 2015. I have benefited from the discussions both in Paris and in Melbourne, and I am grateful to all participants for their comments, especially to Jean-Baptiste Gourinat, Jaap Mansfeld and David Runia. The present, third and final version essentially is based on the Melbourne paper, with some corrections and additions. I have added the analysis of the «Aëtius» quotations in Theodoret's CAG books II and IV (Section 2.3.) and four Appendices. I have also made some corrections in the Stemma (section 12) relating to the sources of Theodoret and Nemesius. The reason of this is explained in the Apendix 1 on Poprhyrius. Jaap Mansfeld and David Runia were kind enough to send me their comments in written form (dated February 16th, 2016) which I include in the bibliography as «Mansfeld and Runia, Critique of Lebedev's Melbourbe paper» and to which I reply in the Appendix 3.

apologist (Eusebius?) who was interested only in the divergent and conflicting views of the Hellenes (their diaphonia), not in the dogmatic (agreed) definitions that contradicted the contentious argument of the Christian apologists about diaphonia.

5. Is P+ identical with SP-Placita?

6. Two neglected authorial definitions in S.

7. The lost P+ was a «diaeretic» handbook of physics with apodictic definitions and/or authorial introductions or explanations in each chapter. Various doxai of other philosophers on the subject of the chapter were attached after the theoretical exposition presented as definitive authorial view.

8. The identity of Arius Didymus. Refutation of Goransson's arguments.

9. Evidence for Arius Didymus as the author (compiler) of the SP-Placita.

10. The so called Epitome of Arius Didymus as reconstructed by Diels in Dox.Gr. could never exist.

11. The relation between Arius' diachronic nspi aipsosrov and his diaeretic (problem-oriented) handbooks of physics (in Stobaeus book I) and ethics (in Stobaeus book II). Short (thematic) placita in Stobaeus as a rule come from the diaeretic handbooks, and «long» continuous doxography of each philosopher from the second (doxographical) section of the chapter dedicated to him in the grand scale nspi aipsosrov. This applies both to book I and II of Stobaeus. All three ethical doxographies in S II 7 are by the same author, Arius Didymus, but derive from his different works: (A) from the diaeretic introduction to ethics, (B) and (C) from the doxographical sections (smxop,ai xrov apsoKovxrov) in the general history of philosophy nspi aipsosrov.

12. The new stemma doxographicum.

Appendix 1. Porphyrius as a possible intermediate doxographical source. Appendix 2. The riddle of AIAYMOS ATHIOS (?) in Suda. Appendix 3. A reply to the critique of Mansfeld and Runia. Appendix 4. Some remarks on the Aetiana of Mansfeld and Runia. Appendix 5. Did "doxography" in Dielsian sense ever exist?

1. Preface. Status quaestionis

When my article titled «Did the doxographer Aetius ever exist?» was published in the Proceedings of the XVIIth World Congress of Philosophy in 1988 (held in Montreal in 1983), it was of a vox clamantis in deserto (Lebedev 1988). The authority of Hermann Diels' Doxographi Graeci was at that time unshaken and both his reconstruction of the supposed common source of Plutarch and Stobaeus, as well as his attribution of this doxographical compendium to a certain «Aetius» had never been seriously questioned by anyone in the previous 100 years. To-day I still cannot say that my views have won universal acceptance. My theory has been criticized by Jaap Mansfeld and David Runia who honored me with a special Appendix in the first volume of their Aetiana titled «Lebedev and the Rejection of Dielsian Hypothesis» (Mansfeld, Runia, Aetiana

1997: 333-338). I am grateful to my highly esteemed colleagues for this constructive criticism first of all because it gave me impetus to reinforce my theory with new arguments and additional evidence. Paradoxically, it was this criticism that contributed to the consequent and ongoing debate on the validity of Diels' reconstruction in which some serious scholars also expressed their scepticism about Diels' attribution as a result of which I do not feel myself in this debate to-day isolated. Robert Sharples admits that the name of Aëtios may be Theodoret's slip of pen for Areios and suggests that «if Lebedev is right», the middle Platonic doxographicum on three first principles in Plato (god, matter and ideas) and the conception of ideas as «thoughts» in divine mind «is pushed back to the latter part of the first century B.C.» (Sharples 1995: 78). In his review of the first volume of Aëtiana Michael Frede undertook an independent formal analysis of the passages with the name of Aëtius in T and came to the conclusion that these passage do not support Diels ' attribution of SP-Placita (which he designates as C, i. e. common source) to Aëtius (Frede 1999: 135-149). Jan Bremmer adduced solid prosopographic arguments against Diels' attribution pointing out that this name is uncommon before the 4th century AD (Bremmer 1998: 154-160)2. And Jean-Baptiste Gourinat published an important critical examination of Diels ' reconstructions of «Aëtius» and of the so called «Epitome» of Arius Didymus with special attention to Stoic doxography. Gourinat accepts the confusion of the names APEIOS and AETIOS in Theodoret CAG 5.16 and our explanation if it as lapsus calami or lapsus memoriae, and demonstrates how arbitrary is Diels' method of separation Arius from Aëtius thus providing additional support to one of our objections to Diels' reconstruction of the common source of S and P (Gourinat 2011: 173 ff., 177 ff.)3. As a matter of fact, although Mansfeld and Runia modestly state that they do not demolish the building erected by Diels, but only renovate and repair it here and

2 Ihor Shevcenko also pointed out to this fact during the discussion of my talk «A reconsideration of Diels' theory of doxography» at the Classics Department of Harvard University on February 2nd, 1990. Runia's reply to Bremmer (Runia 2009) is not convincing, cf. note 34 below.

3 According to Gourinat, the doxographical compendium attributed by Diels to Aëtius was known in antiquity only under the name of Plutarch. But the absence of Plutarch's name in Stobaeus, Nemesius, Achilles an some other sources speaks against this. In our view SP-Placita were cited in antiquity also as a work of Arius Didymus (by Stobaeus and, possibly by Tertullian).

their, their criticism of Diels' methodology occasionally goes beyond disagreement on minor points. There are quite a few pages in their Aetiana to which I can wholeheartedly subscribe and which, in my opinion, cast doubt on the basic principles of Diels' (and the whole 19th century) Quellenforschung and eo ipso provide support to my criticism of the Doxogrpahi Graeci: I have in mind e.g. their reflections on Diels' method, the impact of Lachmannian stemmato-logy and synoptic representations etc. (Mansfeld, Runia 1997: 87 ff. and especially Mansfeld's Doxographical studies in: Mansfeld, Runia 2009: 3-32). One gets an impression that in the second volume of Aetiana (II, 3) Mansfeld and Runia took a more tolerant stance towards the heterodox view about «Aetius» hypothesis. Very few scholars in these days regard Diels' Prolegomena to «Doxo-graphi Graeci» as impeccable and definitive solution of all doxographical problems4, and yet the name of «Aetius», despite the flimsiest (if any) evidence for the historical existence of such person, let alone for his authorship of the SP-Placita, is still being quoted by some as a name of a real ancient author allegedly "edited" by Diels in his Dox.Gr. I have no special anti-Dielsian agenda, on some important issues I side with Diels against his critics, e.g. against Goransson on the identity of Arius Didymus. And yet with all my respect for Diels I have to point out that when he prints on the title page of his edition of «Aetius» in the Dox.Gr., p. 268, in the left column the genuine title of the compendium of Ps.Plutarch and in the right column the invented ad hoc fictitious title IHANNOY STOBAIOY EKAOmN AnOO0ErMATON YnO0HK^N TA EH AETIOY (bold face is mine), this amounts to the falsification of evidence, or to Spielerei to use Diels' own favorit word. John Stobaeus never quotes any writer Aetius, period. I do not doubt Diels' integrity, but in this case he is apparently incapable to discriminate between his own (ill-founded) hypothesis and the manuscript tradition.

I will first summarize my main arguments against the Aetius hypothesis of Diels just to explain why I believe that they have survived the criticism of Mansfeld and Runia (1997), and also will supply some additional evidence that further undermines the reliability of Theodoret' «triads» of doxographical sources. I will also briefly discuss another hypothesis which constitutes the very

4 A notable exception is Zhmud' (2001), cf. idem (2013) who positions himself as an orthodox Neo-Dielsian and condemns as heresy any attempt to correct or to revise Diels. It is strange to read all this in the 21st century.

foundation on which the whole building of DG rests, namely Diels' reconstruction and evaluation of Theoprastus' lost work Oucixai So^ai. The discussion of these two topics will be brief and Ke^a^airoSnç, to use a doxographical term, intended only to clean the ground for the discussion of our main topic, the origin of the doxographical tradition of Placita phlosophorum. After this I will turn from the negative to the constructive part of my paper in which I will try to identify the real author of the original SP-Placita (as I coventionally designate the commom source of Ps.Plutarch and Stobaeus) and to prove my main thesis, namely that the real hero of the ancient doxographical epos was neither Theophrastus, nor an unknown «Aëtius», but one of the giants of the Post-Hellenistic philosophy, the teacher and friend of the emperor Augustus, Arius Didymus from Alexandria.

2. The name and identity of AETIOE in Theodoret's triad of sources nAOYTAPXOE-nOPOYPIOE-AETIOE.

Four principal arguments against Diels' attribution of SP-Placita to a totally unknown writer «Aëtius»

2.1 Suspicious name, bad attestation. The personality of Theodoret: a bishop and a passionate preacher, not a scholar.

No ancient Greek, Roman, Byzantine or Arabic source ever quotes by name a single philosophical doxa or placitum from a writer called Aéxioç. The prosopographical evidence collected by Bremmer (1998) leaves no doubt that this name became widespread only in late antiquity. The name of a philosophical writer Aéxioç occurs only in three passages of the Curatio Graecarum Affectionum (CAG) of Theodoret bishop of Cyrus composed in the 30-ies of the 5th century A.D. in a triad of doxographical sources nAOYTAPXOS - AETIOS - nOPOYPIOS as an alleged source of his information on the views of Greek philosophers, but it is not immediately clear which doxa(i) exactly is quoted from «Aëtius» and not from the other two sources. Theodoret by no means was a man of learning comparable in his knowledge of Greek literature and philosophy to Eusebius or Clement of Alexandria on whose writings (PE and Stromateis respectively) his CAG heavily depends5. He repeatedly distorts the names of Greek philosophers.

5 This is a commonly accepted view shared by Canivet (1958: 55, 58), Mansfeld and Runia I, 273: «Incontestably Theodoret's erudition is largely derivative, i.e. based on knowledge taken at second hand»; Scholten (2014: 36).

Thus, in CAG V.16 (a passage crucial for Diels' attribution) he confuses the philosopher A^K^alrov with the lyric poet AAx^av6, turns Tl^rov of Phleius into Ti^aio^, the Peripatetic AiKa(ap%o^ becomes K^eap%o^, and a fragment of Arius Didymus is falsely ascribed to Noumenius. Elsewhere he makes Pherecydes of Syros (Supio^) his compatriot from Syria, rather than a native of the island of Syros7 and sharing a widespread mistake relocates the birthplace of the atheist Diagoras from Melos to Miletus (CAG II.112). To quote a recent study of his life and work, «Theodoret was no academician given to single-minded research and the production of many books; he was a bishop and a pastor whose chief concern was the welfare of his diocese» (Clayton 2007: 3)8. Theodoret was consecrated bishop of Cyrus (Cyrrhus), the chief city of Cyrrhestica (a province of Euphratensis) in 423. From his letter 80 we learn that he had been bishop for 25 years «after passing his earlier life in monastery»9. It is therefore virtually certain that he wrote CAG in Cyrus, a provincial small town which he describes as wretched and solitary10. All these years he apparently had limited or no access to good libraries and cultural centers like Antioch. Isn't it a bit surprising that Theodoret was fortunate enough to find in such a solitary place in the middle of nowhere a precious complete copy of the original SP-Placita that was unknown both to Hellenic philosophers and Christian apologists far surpassing him in learning, a copy that could not be found in the greatest libraries of the ancient world?

6 Gourinat (2011) 173 n. 86 tries to acquit Theodoret of mistake citing Canivet's remark that Alcman is a dialect form of Alkmaeon. This would be true if Theodoret were living in preclassical times and writing in Doric dialect. But in classical and post-classical times the names of the lyric poet and of the philosopher from Croton were always clearly distinguished by litterati. Even Alcmaeon himself, although writing in Doric, retained the Ionian form A^Kpairov (B 1 DK). Alcman was a much more famous name, Alcmaeon of Croton was known only to specialists, hence Theodoret's mistake.

7 CAG 1.24, cf. also his Quaest. in libros Regnorum, MPG 80. 676 ®spsKuSn? o Supoq. This view is apparently due not only to the ignorance of the geography of Greece, but also to the popular thesis of the Christian apologists that the Greeks have stolen their wisdom from the Orient.

8 The practical, social aspect of Theodoret's activitity also is emphasized in Schor 2011.

9 Quoted by Clayton 2007: 10.

10 Epistle 138 quoted by Clayton 2007: 11.

2.2 The rhetorical character of the triadic quotations in CAG. A close parallel from Cyril Contra Julianum.

Theodoret quotes the triad of names Plutarch - Porphyry -Aetius in CAG three times: 1) in the Book 2 (sequence Plutarch -Aetius - Porphyrius) on the cosmogony and the origin of the world11. 2) In book 4 (sequence Aetius - Plutarch - Porphyrius) after a long series of philosophical doxai on the divine, the first principles, the cosmos and the stars12. 3) In Book 5 (sequence Plutarch -Porphyrius - Aetius) on the nature of psyche. In all three cases (especially in the first two) the triad of names is intended as a kind of «general bibliography» of sources used rather than a precise quotation of a certain view from a certain doxographer.

Both the triadic form of these quotations and the choice of authors are not casual (this was ignored by Diels when he proposed his attribution). The selection of the three names is determined by the fact that they all are famous and reputable among the Greeks themselves: this is a warranty that the (conflicting and sometimes impious) views of the Hellenes are not distorted or falsified by the Christian apologist who quotes them. This becomes obvious from Theodoret's remark in CAG IV.31 Ei Se Tig oiiexai Ka^e ouKoqtavxrfoai xoug avSpag, x^v na^no^nv auxrav Siaqiroviav e^ey^avxa, avayvroxro ^ev Aexiou x^v nepi apeoKovxrov ^uvayroynv, avayvroxro Se n^ouxap^ou x^v nepi xrav xofg 91^000901^ So^avxrov ernxo^nv' Kai nop^upiou Se ^ $1^00090^ ioxopia no^a xoiafixa SiSaoKei «If someone believes that I accuse falsely these men by exposing their total disagreement, let him read the collection of Aetius «On the opinions», let him read «On the views of the philosophers» by Plutarch, and the «History of Philosophy» by Porphyrius also provides a lot of such kind of reports». Cyril who in a similar apologetic context quotes various philosophical doxai from the same works of Plutarch and Porhyry repeatedly emphasizes that they are famous among the Greeks: Contra Julianum 2.14 n^otixap%og xoivuv, av^p xrov nap' atixoig ouk aon^og yeyovrog, ev xro Seuxepro PipMro x^g xrov 'OuoiKrov Soy^axrov' ouvayroy^g, ouxro

11 CAG 2.95 n^owapxoq Se Kai Ainoq Taq Trov ^ooo^rov sKnaiSsuouoi So^aq^ tov auTov Se Kai o nop^upioq avsSs^aTo novov, tov ¿KaoTou piov Taiq So^aiq npooTsOsiKroq.

12 CAG 4.31 Ei Si Tiq oisTai Kaps ouKo^avrqoai Touq avSpaq, T^v rcaprcoMnv auTrov Sia^roviav s^sy^avTa, avayvroTro psv AsTiou T^v nspi apsoKovTrov ^uvayroyqv, avayvroTro Ss n^ouTapxou T^v nspi Trov Toiq ^ooo^oiq So^avTrov snvrop^v Kai nop^upiou Ss ^ ^i^ooo^oq ioTopia rcoM,a ToiauTa SiSaoKsi.

ФП^г nepi xoü к0оцои «Plutarch, who was not an insignificant author among them (= Hellenes), in the second book of «Physical opinions» says the following about the cosmos...». Ойк ао^МО here is a litotes, the meaning is «very famous» among the Hellenes13. This rhetoric amounts to the following message: you, Hellenes, do not believe us, Christians, but then listen to your own wise, who provide unambiguous evidence on the disagreement of the Hellenes on things divine and the origin of the world». And here comes our first surprise. Two out of the three names chosen by Theodoret, Plutarch and Porphyrius (the archenemy of the Christians) are indeed famous and reputable Hellenic writers. But what about Aéxro^? The problem does not consists only in the fact that there is no famous Greek philosopher or doxographer of this name, an even more serious problem consists in the fact that such Greek philosopher is totally unknown. And - taking into account numerous distortions of the names of Greek authors in CAG - this raises the initial suspicion concerning the accuracy of Theodoret's report.

The criterium of celebrity in the choice of the Hellenic authors is reinforced by the triadic structure of quotation. Both in the Hellenic14, and in the Judeo-Christian15 legal traditions the evidence of three witnesses (xpeí¡; ^ápxupe^) was regarded incontestable. The imagery of «trial and witnesses» was suggetsed to apologists by the very concept of ano^oyia as apology in court against false accusations. If something is assured not by one reputable and famous witness, but at once by three, then it cannot be contested. Such triads of Hellenic authoritative writers providing confirmation of apologist's own thesis seem to be a rhetorical and dialectical cliché, a polemical device which we find also in Cyril of Alexandria. It is especially remarkable that Cyril uses it for the same puprose in the same context and citing virtually the same Greek names as Theodoret. In the first book of «Contra Julianum», after quoting divergent and conflicting views of Greek philosophers on the nature of god he adds:

13 cf. Cyrill. Contra Jul. 2.22 'Тфп yáp пои nspi aùxôv iayvoç юу ayav ó

П^оцхаруос év хф прюхю РфМю 'Фиагкюу 5oypáxrav' auvayraynç. 1.25

Порфйрю5 yoûv, éni nai5sía коарлк^ 5ó£av syrov пар' éKsivoiç qúk

áysvvn, év хф прюхю ^óyra x^ç ^i^0a090u_iax0piaç' xoioûxôv tí ф^аг nspi

xôv àvop,aap,évrav 'аофюу'.

15 Aelius Arist., Panathen. p. 146,8; 151, 31 Jebb.

15 Flav. Joseph., Antiqu. Iud. 4,129; NT. Epist. loan. 5,7.

Teypa9aoi Se nepi xotixrov n^otixap%og xe Kai exepoi xrov nap' atixoig ^oyaSrov, Kai o 0paoug Ka0' ^prov nop9tipiog16.

«On this subject [= opinions of Greek philosophers about god] have written Plutarch, as well as some other chosen among themselves [= pagans] authors, and Porphyrius who is insolent in his attacks on us [= Christians]».

The word ^oyaSeg «picked, chosen» writers here again points to the reputable and recognized authors who have written on the opinions of philosophers and whose testimony on the diaphonia of the Hellenes guarantees that Cyril is not lying or exaggerating. If the phrase Kai exepoi xrov nap' atixoig ^oyaSrov were positioned at the end, it would be of little importance, a kind of et cetera. But the fact that it is inserted between two names is unusual and calls for explanation. We get an impression that Cyril had in mind (or in some written source before his eyes) three names of the most famous Greek doxographers, but for some reason he dropped the second name and replaced it with a periphrastic expression «other chosen (e.i. highly esteemed) authors among the pagans». The first two possible explanations that come to my mind are the following.

1) Cyril knew the triadic quotations in Theodoret's CAG with the names of Plutarch, Porphyry and Aetius, and imitated the triadic formula of reliable (and self-defeating) pagan witnesses, but he was puzzled by the name of a doxographer Aetius he had never heard before and therefore replaced it with a reference without personal name, at the same time preserving the rhetorical force of a triad17.

2) Cyril knew from Eusebius PE that the most often quoted (and therefore ^oyaSeg) Greek doxographers were Plutarch, Areios and Porphyrius. But for a Christian bishop at that time to quote as a reliable source APEIOS without explanations was both inappropriate and hazardous: he could be accused by his opponents (Theo-doret among them) of supporting the condemned Arian heresy. The name of Areios at that time could only be quoted by a bishop with curses and indignation. That is why Cyril, presumably, prudently replaced the odious name by an opaque and innocent expression.

16 Cyrill. Contra Jul. I.39.

17 Theodoret and Cyril were contemporaries and opponents in Christological debates, the relative chronology of CAG and Contra Julianum is uncertain. Theodoret mentions Cyril Contra Julianum in a letter to John bishop of Antioch (died 442). Both may have known the apologetic works of each other. Terminus post quem for Cyril's work is 433, terminus ante quem is 442/1, see: Russell 2000: 190. Scholten (2014: 14) dates CAG to 437 AD (dubitanter).

2.3. Analysis of the first two triadic quotations in CAG II and IV demonstrates that Plutarch and and Porphyrius are Theodoret's real sources, the name of Aëtius finds no match in the doxo-graphical material quoted ny the bishop.

Diels' assumption that Theodoret's only real doxographical source was «Aetius» and that the names of Plutarch and Porphyrius he added «splendoris gratia» is higly implausible. To begin with, Theodoret never cites «Aetius» (his alleged main source!) alone, outside the triad of names «Plutarch - Porphyrius - Aetius», but he repeatedly cites elsewhere the Placita of Ps.Plutarch and Porphyrius' Historia Philosophos. Are we expected to believe with Diels that whenever Theodoret quotes Plutarch' s Placita and Porphyrius' Historia Philosophos in other passages he is also lying and that his real concealed source is the unknown «Aetius»? If Theodoret is lying even when he quotes well-known authors, exactly and with the number of book (quotations the authenticity of which is confirmed by the extant works or parallel tradition) how he can be trusted when he quotes a totally unknown work of an unheard author? Even on a priori grounds a Christian apologist is more likely to be acquainted with the works of his main theoretical opponent, Porphyrius, than with an unheard work of an unknown author. It seems that Diels was forced to make such implausible assumption on order to avoid another implausible assumption, i.e. to assume that Theodoret used both Ps.Plutarch and «Aetius», i.e. two hardly distinguishable doxographical compendia. And again, If Diels' assumption is valid, it must be valid in the two other Thedoret's quotations of the triad of doxographical sources as well. But a glance at the context of those passages will again reveal the impossibily of Diels' assumption. In CAG book II Пер! (on cosmogony and creation of the world)

Theodoret urges his pagan listeners/readers to compare Greek cosmogonical myths and philosophical theories on the origin of the world with the «apostolic and prophetic teaching» in order to realize how far the latter surpass the former. After enumerating the cosmogonies of Sanchuniathon, Manethon, Diodorus Siculus, Hesiod, Orpheus, Cadmus and theology of Cornutus, he refers to the three collections of philosophical views on this subject: (CAG II.95) П^оитар%о; Se ка! Аетю; та; xrov 9i^ooo9rov екяагйеиоиог So^o;-xov atixov Se ка! о Порфйрю; aveSe^axo novov, xov екаотои piov таг; йо^ац яроотебегкю;. «Plutarch and Aetius teach the opinions of philosophers. Porphyrius undertook the same labor having added a biography of each philospher to his opinions». Diels tried to throw out of the game not only Plutarch (as embarassing doublet of his

favorit Aetius), but also Porphyrius as an alleged pure biographer18. But Theodoret explicitly says that Porphyrius' Historia philosophos contained doxography as well as biography (the work of Diogenes Laertius is a close parallel). A glance at the extant fragments of Porphyrius' Historia philosophos will show that Diels is wrong and that Theodoret's report is true19. Theodoret does not specify in this passage which doxographical material comes from each of the three authors. And yet in the same book (CAG II.112) he quotes a list of atheists from Plutarch's Placita I.7 (Diagoras, Theodorus, Euheme-rus). An orthodox Dielsian will put himself in extremely uncomfortable position if he tries to persuade us that Theodoret mentions here Plutarch's name «splendoris gatia» and that his real source is «Aetius». Theodoret quotes Plutarch, period. He quotes nepi xröv apeoKovxrov, a work which is extant and confirms that his quotations are genuine and accurate. Additional confirmation is provided by the fact that he quotes this work repeatedly. Diels' denial of this obvious and undeniable fact (xo evapyeg) is motivated by what has been aptly described as «psicosi moderna della fraude antica» (Farinelli 2000), the obsession of some 19th century scholars with ancient «fraud» and their envie of Bentley's glory as demystifier. Diels was a child of his time, he likes to «expose» Spielerei of a supposed ancient Schwindler even when it exists only in his imagination. Classical scholar is not a lawyer, his work (his epyov) is to understand ancient minds and to interpret ancient texts bona fide, not to «expose», to condemn and to ridicule them with arrogant contempt and sense of superiority. In the very same book II Theodoret also quotes separately Porphyrius as his main source on oriental theogonies. The first quotation seems to derive from Contra Christianos (CAG II. 43-45), the second occurs in the same passage that quotes Historia philosohos. It is conceivable that Porphyrius discussed Sanchouniathon in both works. But what about «Aetius»? We do not find in book II (or anywhere else) a separate nominatim quotation from this author. We may conclude that if in his list of doxographical sources in book II Theodoret added some name «splendoris gratia», it was the name of Aetius, and not the names of

18 «quis igitur fuit? Porphyriusne qui Historia sua vitas philosophorum descripsit etc.» (Diels 1879: 46).

19 Vide e.g. Porphyr. fr. 197 (Egyptian cosmology), 205 (Milesian theories of matter), 220-221F (Plato's first principles, cf. ôo^aaai ... n^axrova) Smith. Note that Smith's collection of the fragments of Historia philosophos is incomplete. Diels' mistake has been corrected by Segonds in Des Places - Segonds (1982: 164).

Plutarch and Porphyrius, and if so, he did it for the sake of rhetorical effect: to quote треТ; цартире;. Theodoret is honest enough to acknowledge that he knows ancient cosmogonies from the great work of Eusebius: he gives only a brief summary (because, as he explains, he is afraid to be accused of telling too much nonsense, i.e. unholy things), the reader should consult Eusebius' exposition on his own risk. It is conceivable that in book II Eusebius is Theodoret's main source not only for the mythical and oriental theogonies, but for the philosophical doxai as well. A possibility cannot be excluded that Theodoret got his list of doxographical sources from Eusebius as well since his reference to Eusebius' Praeparatio Evangelica as a more detailed exposition of pagan cosmogonies than his own abridged version follows almost immediately after the triad of names «Plutarch - Aetius -Porphyrius» and the text of CAG II.97 leaves no doubt that Theodoret names Eusebius as his source not only for the mythical theogonies, but also for the theological and cosmogonical doxai of the philosophers20. In Eusebius' PE the three main doxographical sources regularly quoted are Plutarch, Arius Didymus (APEIOS) and Porphyrius. Theodoret, surprisingly, never mentions APEIOS, but instead combines with the names of Plutarch and Porphyrius an unknown AETIOS. Since he never quotes this author separately (unlike the other two), which means that he knows him only from second hand sources, even on the ground of CAG book II a conjecture can be made that he misred or misquoted Eusebius' APEIOS as AETIOS. Eusebius' triad of doxographical sources «Plutarch - Areios - Porphyrius», unlike Theodoret's strange and unparalleled triad with unknown Aetius, makes perfect sense and looks like a carefully selected list: Plutarch's Placita is the main source for Preplatonic opinions, Arius Didymus covers the main classical and Hellenistic schools (Plato - Aristotle - Stoics), Porphy-rius in his «Historia philosophos» dwells especially on Socrates (book 3) and Plato (book 4), and in his other works covers the later Platonic tradition from Numenius to Plotinus. Thus these three

20 CAG II. 95-97 Шои-rapxoç 5è каг AéTioç та; töv фЛоабфЮУ ÊKnaiSsuouai 56^aç- xov aùxov 5è каг о Порфирю; àvsSé^axo novov, xov Екаахои ßiov тац So^aiç лроат£0£ш»с. Toutoiç ùpaç, é avSpsç, à^irâ та ■qpiTspa napa0£ivai ... Ei 5é тю фí^ov каг aÙT^v Smyvravai T^v fuyKpioiv, sùp^asv TauT^v év Tovç Eüa£ßiou той na^aiaTÎvou ^uyypappaaiv, Eùayy£^iK^v 5è nponapaoK£u^v TovS£ tov novov ÊK£vvoç àvopaa£v syra 5è ^uvTopraç épô, àç Aiyurcuoi каг ФoívlK£ç каг töv 'EXXrçvœv oi noi^Tai каг фl^6g0ф0l ^ та орюрста ато^йа 0£oùç ùл£T6лaaav кт1.

doxographers cover the whole of Greek philosophy, and by quoting them a Christian apologist claims to demonstrate that he is not selective and that his refutation of the Hellenic wisdom is all-inclusive and complete. Note that Theodoret quotes three witnesses thrice.

Let us now turn to CAG book IV nepi u^ng Kai Koopou. After exposing the diaphonia of the Hellenes on the nature of the cosmos, matter and the celestial bodies in CAG IV. 5-29 Theodoret makes the following remark in IV.31 Ei Se Tig o'iexai rape ouKo9avx^oai xoug avSpag, x^v napno^nv atixrov Sia9roviav e^ey^avxa, avayvroxro pev Aexlou x^v nepi apeoKovxrov ^uvayroy^v, avayvroxro Se n^ouxapxou x^v nepi xrov xotg 9i^ooo9oig So^avxrov enixop^v Kai nop9upiou Se ^ $i^ooo9og ioxopla no^a xoiafixa SiSaoKei. «If someone thinks that I have falsely accused those men by exposing their numerous disagreements, let him read the «Collection of opinions» by Aetius, let him read the «Brief exposition of philosophers' opinions» by Plutarch, and Prophyrius' «History of philo-sophy» also provides a lot of similar cases». By this remark Theo-doret intends to emphasize his own objectivity and impartiality: he has not exaggerated the disagreement between Hellenic philosophers because all instances of disagreement he adduced are recognized by Hellenic philosophers themselves. The Hellenic wisdom is self-refuting. In this book, as in book II, it is not immediately clear what in the preceding text is quoted from Prophyrius and what is quoted from Plutarch, but the name of Aetius again presents a problem since there is nothing in Thedoret's excerpts that could be even tentatively identified as a quotation from this author. Porphyrius' Historia philosophos book 3 (on Socrates) is quoted in the very beginning (CAG IV.2 = Porphyr. fr. 216 F Smith) and virtually all short placita are found in the extant text of Ps.Plutarch's nepi xrov apeoKovxrov 9i^ooo9oig 9uoiKrov Soypaxrov emxop^. The title of Plutarch's work is quoted imprecisely (So^avxrov instead of apeoKovxrov). The title of the doxographical work of the mysterious Aetius looks suspicious. Both titles are shortened and probably quoted by Theodoret from memory, only the title of Porphyrius' work is quoted accurately. nepi apeoKovxrov ^uvayroy^ is impossible as original complete title, we would expect something like apeoKovxrov xotg 9i^ooo9oig, if it is a general doxographical compendium or, say, apeoKovxrov n^axrovi or SxroiKotg, if it is a specilized doxography of a single philosopher or a philosophical

school21. Our hypothesis that Theodoret' AETIOS is a misreading or a misquotation of Eusebius' APEIOS can easily explain the origin of the inaccurate title Euvayroy^ xrov apeoKovxrov from the quotations of Arius Didymus' doxographical works in Eusebius PE XV,20,8 xa x^g SxroiK^g 9i^ooo9(ag Soypaxa ano xrov enixoprov Apeiou AiStipou ouvei^eypeva, cf. Arius' own words apeoKei Se xotg npeoPuxaxoig kxL in the beginning of these excerpts 2. Su^eyro and ouvayro, ou^oyf and ouvayroyf in the sense of «collection» (of texts, excerpts etc.) are virtually synonymous, as are snonymous Soypaxa and apeoKovxa, and so Thodoret's ^uvayroy^ xrov apeoKovxrov may well be a paraphrase of Eusebius' ouvei^eypeva Soypaxa and Arius's own introductory words apeoKei kxL The word ouvayroyf in late sources is interchangeable with the word enixopf 3. It is noteworthy that in CAG V.25-26 Theodoret quotes verbatim two fragments of Arius Didymus exactly from these Stoic excerpts in Eusebius' PE, and the second of these Arius' fragments he misquotes as a text of Numenius (!)24. This fact proves beyond any doubt that Theodoret was either unable to read correctly the name of Arius in his copy of Eusebius' PE (e.g. unreadable or damaged marginal lemma) or otherwise tried to avoid the dangerous «heretic» name25.

2.4 Analysis of the doxographical material in CAG V.16 ff. demonstrates that Theodoret indeed quotes from three different sources, but the name of Aetius does not correspond to the quotations from SP-Placita, and so even Theodoret himself does not ascribe SP-Placita to a writer called Aetius. In other words, the problem with Diels' attribution is not that evidence for Aetius as the compiler of SP-Placita is weak; the real problem is that such evidence does not exist. The confusion of the names APEIOE and AETIOE seems very likely.

21 TLG gives not a single instance of ouvayroyf (^uvayroyq) Trov apsoKovTrov without dative, only (with dative) the title of Aristotle's lost ouvayroyf Trov apsoKovTrov Toiq nuOayopsioiq quoted by Simplicius, Cael., v.7, 386,23 = Aristot. fr. 164 Gigon (Zuvayroyf Trov nuOayopiKrov fr.167).

22 Cf. also Euseb. PE XI,23,2 sk Trov AiSuproi nspi Trov apsoKovTrov n^aTrovi ouvTsTaypsvrov.

23 Mansfeld, Runia 1997: 324 rightly point out that the title nspi apsoKovTrov ouvayroyf is awkward, but the original title nspi apsoKovTrov they propose ibid. 326 is also unlikely (a dative is obligatory in the complete title).

24 On this see below section 2.5.

25 On this see note 32 below.

Most important and crucial for the verification of the validity of Diels' hypothesis is the passage in Book V.16 since only in this case Theodoret says that he will quote from the authors he mentions, and only in this passage the doxographical excerpts start immediately after the reference to the sources, and only in this case we can identify with certainty the texts quoted from «Aetius» and to separate them from the quotations from Porphyrius and Plutarch:

CAG V.16 A Se ye 0еф ^е^ю, ¿к xrov П^оитар%ю ка! Порфир(ю ка! ц^тог ка! Ает(ю ^uyyeyраццevюv ерю. 0а^; тошот кек^^ке x^v акívnxov фйош кт^.

The doxographical material that follows covers many pages and is without interruption continued by extensive quotations from Plato's original works. Theodoret indicates that he has finished quoting from the three authors only at CAG V.44 Тооаит^ ка! ^иуурафец ка! фЛооофог ка! пог^та! ка! перг ка! оюцато;

ка! айт^; уе т^; той аубрюпои ^иотаоею; про; аМ^ои; ео%^каогу ерш ка! о! ^ev тайта, о! Se екеша

прeоPeтоvxe¡;...

The whole doxography and excerpts quoted between V.16 and V.44 can be on stylistical and other grounds divided into three sections, two major blocks and one short in between. The first major block: V.16 (Thales on soul) - V.25 (Aristotle on plant soul). These are short placita mostly with exact parallels in P, but with some additional placita of the same type (on these see Appendix 1).

The second major block starts with V.28 (Pythagoras and Plato on psyche as divine moira) and ends at V.44. From V.29 on it is interspersed with quotations from Plato's original dilogues cited as a confirmation of the proposed exposition of Plato's doctrine of the soul. The language of the doxography in the first and second sections is Hellenistic, the beginning of the second major block (third section) is marked by a sharp shift from Hellenistic to the Neoplatonic language and style. It is reasonable to suppose that from V.28 Theodoret starts quoting from Porphyrius. We believe that the following passage (possibly starting even earlier, from the Longinus quotation in V.27)2 is an almost verbatim quotation from Porphyry that should be added to the fragments of the book 4 of Philosophos historia (or to fragments from his other works, e.g. Пер!

26 Mansfeld and Runia (2016) rightly point to S 1,48,8 and admit (dubi-tanter) that the vouq 0upa0sv placitum may derive from SP-Placita. I agree that it may, but not directly. The strong admixture of Neoplatonic terminology in CAG V.28 (0sia poipa, аитократора, Moiprav v^paxa) points to Porphyrius as intermediate source between SP and T. See Appendix 1.

CAG V.28 'AvxiKpug Se xouxrov evavxia xa nu0ayopa Kai n^axrovi So^avxa- 0e(av yap Sf potpav xov vofiv £9aoav elvai. Hup9rovet Se Kai o NiKopaxou xroSe xro ^oyro, Kaixoi 0vnxfv elvai ano9nvapevog xfv yu%fv- a^o xi yap elvai xov vofiv napa xafixnv fine^apev. Kai oi nepi n^axrova Se Kai nuBayopav 0fipa0ev xofixov eioKp(veo0ai ^eyouoiv. na^iv S' au oi pev auxrov auxoKpaxopa xov vofiv £9aoav elvai Kai ayeiv Sfivao0ai, f av e0e^^, xa na0n' oi Se avayKng auxov Kai eipappevng ane9fvavxo Sofi^ov Kai npog xa xrov Moiprov ayeo0ai vf paxa, Kai xfg xofi oupavofi nepiSivfoerog Kai xfg xrov aoxprov ^uvoSou xag xofixou npa^eig Kai evepyelag fpxfo0ai. 0 Se ye n^axrov xfv evavxiav xofixoig nepi yu%fg e^enalSeuoe So^av-e^eu0epav yap auxfv elvai £9noe Kai Seonoivav xrov evo%^ofivxrov na0rov Kai ¿Kofioav f xfSe f eKetoe xropetv Kai e0e^ouo(rog ye 17 viKav f fxxao0ai. Kai xofixo oa9rog ev xotg Nopoig SeSf^roKe kxL2 Either Theodoret contaminates in V.28 the placitum on extracorporeal mind (vofig 0fipa0ev) from SP-Placita (cf. Stob. 1.48.7) with Pophyrian doxography or he quotes Porphyrius who already had rephrased the same placitum in Neoplatonic language28.

And so in the inetrmediate section 2 (V.25-V.27) between the preceding placita of SP-type in section 1 and Neoplatonic interpretation of Plato's doctrine of the soul in section 3 we are left with only three quotations: one of them (Zeno on sperm as anoonaopa xfg yu%fg) without indication of the source2 , another one misquoted from «Numenius», on the post mortem fate of the soul according to the Stoics30, and the third one from Longinus with his objection to the Stoic materialist conception of the soul31. The source of Theodoret for these three quotations is apparently

27 Not in Smith's valuable collection (Smith 1993: fr. 193-224). The introductory avTiKpuq in several passages is likely to be an addition of Theodoret intended to emphasize the contradictions between Hellenes.

28 On Pophyrius' use of SP-Placita see Appendix 1 below.

29 Theodoret. CAG V.25 Zfvrov Se o Kmsuc;, o rqoSs rqc; aipsosroc; fyqoapsvoc;, roiaSs nspi yuxfc; So^a^siv roue; otesiouc; ¿SiSa^s ^oiTHTaq-tov yap toi avOpronivov Oopov, uypov ovra Kai psrsxovra nvsuparoc;, rqc; yuxfc; s^nosv sivai pspoc; rs Kai anoonaopa Kai rou rrov npoyovrov onspparoc; Kspaopa rs Kai piypa, e^ anavrrov rrov rqq yuxfc; popirov lUvaOpoioOsv. Ou Sf xapiv aurfv Kai ^Oaprfv npoonyopsuosv.

Theodoret., CAG V.26-27 Noupfvioc; Se o nuOayopsioc; npoc; toutouc; anorsivopsvoq s^n «Tfv Se yuxfv ysv^rfv rs Kai ^Oaprfv Xsyouoiv, ouk suOuc; Se tou oroparoc; anaX^aysioav ^OsipsoOai, aXX enipsvsiv Tivag xpovouq KaO' saurfv, rfv pEv rrov onouSairov psxpi rqc; sk; nup avaXuosrog rrov navrrov, rfv Se rrov a^povrov npoc; noiouc; Tivag xpovoug.» __

31 CAG V.27 Kai o Aoyyivoc; Se aurrov avriKpuc; Karnyopsi, roSs ypayaq^ «Zfvrovi pEv yap Kai KXsavOsi vspsofosisv av rig SiKairoq ktX.

Eusebius PE book 15.20-21. The two Stoic placita in Eusebius are quoted from Arius Didymus: PE 15.20.8 Toiafixa Kai xa x^; Sxroix^; ^i^ooo^la; Soy^axa ano xrov 'Emxo^rov Apeiou AiSu^ou ouvei^ey^eva. And just as Theodoret immediately after these Stoic doxai on the nature and fate of the soul quotes Longinus' objection to the Stoics, so in the text of Eusebius they are followed by a much more extensive excerpt from Longinus with the same objection. It is obvious that «Numenius» is Theodoret's misquotation of the original name of Areios Didymus. It is also obvious that in Theodoret's triad of sources announced in CAG V.16 «Plutarch» corresponds to the short Placita of the first section, «Porphyry» to the third section from Historia philosophos book 4, and AETIOS to the intermediate section, namely to the two Stoic doxai on the soul from the «compendia» (emxo^ai) of Arius Didymus. Why does Theodoret misname Areios Didymus as Numenius? The confusion is probably due to the fact that one of the preceding chapters in Eusebius' PE, chapter 17 of the book 15 has a lemma: OTI OYK AN EIH S^MA TO ON KATA TOYS STOIKOYS- AnO TOY nPOTOY nEPI TArA0OY NOYMHNIOY. The subject of this extract from Noumenius is the same as the subject of Longinus' objection to the Stoics (PE, Book 15, chapter 21) quoted immediately after two Stoic doxai from Areios Didymos. Why does he misquote Areios Didymos as «Aetius»? We may never find a definitive answer to this question. It may be a lapsus calami or lapsus memoriae, it may be due to mechanical damage of one letter in the name A.EIOY in the manusript (or in a marginal lemma) which he misred as AETIOY, it may be due to a subconscious association of the two most prominent heresiarchs of his time APEIOS and his disciple AETIOS with whose followers he was passionately fighting32. Likewise we may never find an exact explanation of why

32 The names of the 4th century heresiarchs Areios and Aetios are very often quoted in patristic texts as a pair or together in a list of Arians, so the association of these two names at the time of Theodoret must have been similar to the modern association of Marx and Engels or Voltaire and Diderot etc. Some examples from the ecclesiastical writings of Theodoret: Hist. Eccles. p. 153, 25 Кат' ekevvov 5e xov Kaipov каг AExioc о Ewopiou SvSaoKa^oq, 5; x^v ApEiou Р^аафпршу таг; ¿rnvoiaiq кт1. p.

294, 20 Тайта ката xe x^; ApEiou каг Aexwd каг Ewopiou paviaq, каг рЕутог каг ката ХаРЕШои ... yEypафaaгv. also ibid. p. 70,17; 286,6. Aetios was also referred to as Аехюс; ApEio; in the sense of «Arian» (Greg. Nyss., Contra Eunomium, 3.9.55) or Аехюс; АрЕгауос;. Just as Theodoret confuses the names of two Peripatetics (for him they also belong to the same агрЕог;!), so he could easily confuse the name of doxographer ApEio;

he misquotes AiKaiap%o; as K^eap%o;. In a sense it does not matter. What matters, is the fact that he repeatedly distorts and misquotes Hellenic names, that he does not know himself whom he is quoting in CAG V. 25-27, and that he attributes these misquotations borrowed from Eusebius without reference to his factual source, to an totally unknown writer Aexio;. The real cause of these confusions is that he was not a scholar, but a passionate preacher and bishop whose primary concern was his diocese and the fight against pagans and heretics. He was not a liar as Diels claims. He just did not have time to cheque his references.

2.5. The meaning of Kai ^¿vtoi Kai: a linguistic argument against Diels' attribution of SP-Placita to «Aetius».

Theodoret CAG V.16 a Se ye ^uv 0eroi ^e^ro ¿k xrov n^ouxap%roi Kai nop^upiroi Kai ^evxoi Kai Aexiroi ^uyyeypa^evrov ¿pro. Here are the comments of Diels on this crucial for his attribution passage: «ubi cur Aetium tanta vi (Kai ^evxoi Kai) postposuerit, nullo modo dispicias, nisi ex hoc excerpsit, illos splendoris gratia nominavit. nam Aetio soli non cito Graeci credidissent, crediderunt illorum patrocinio commendato et sustentato» (Diels 1879: 48). «You can in no other way understand why in this passage he [= Theodoretus] with such force (Kai ^evxoi Kai) placed his name after those [of Plutarch and Porphyry], unless he made his excerpts from the latter [ = Aetius], and mentioned the other two for the sake of splendor. For the Greeks would not promptly believe Aetius alone, but they believed in the commendable and established authority (?) of the former [i.e. Plutarch and Porphyry]».

because of the subconscious association with identical names of two heresiarchs. Note that Theodoret does not mention the name of Areios and Arians without cursing them as condemned heretics (pavia, P^ao^npia etc.). So he would be psychologically unwilling to cite anyone called Areios as a reliable source, especially in a public lecture or sermon (Sra^s^iq) preaching orthodoxy to the pagans (see Appendix 3). Likewise in the Stalinist Soviet Union it was dangerous to mention Trotsky without official labels like «fascist», «Judas», «chief of the gang of murderers and spies» etc. Unlike Theodoret and Cyril, Eusebius of Caesaria had no reason to avoid the name of Areios because he was Arian himself. It must be emphasized that this is just one possible explanation of Thedoret's misquotation. Other explanations are possible as well (unreadable name, damaged copy of Eusebius; PE etc.). Our main arguments against Diels' attribution of SP-Placita to «Aetius» are primarily based on the analysis of the quotation in CAG 5.16 and its context.

So, according to Diels, Theodoret in this passage (which is crucial for Diels' attribution) tries to dupe his «Greek» readers, i.e. his pagan opponents, by putting the obscure name of his real source in the shadow of two glorious names. This is no more than a guess, and this guess is unlikely and ill-founded. In other passages Theodoret cites both Plutarch and Porphyry separately by individual name, but he never cites Aëtius alone. So it is the name of Aëtius that calls for suspicion rather than the names of Plutarch and Porphyry. In II.95 (doxai on the cosmogony) Theodoret does not cite Porphyry «splendoris gratia»: Porphyry's Historia philosophos is apparettly his source for the cosmogony of Sanchuniathon (and maybe of some other Oriental and mythical cosmogonies)33. As we have seen, the analysis of the doxographical material that follows after the announcement of three sources demonstrates that Theodoret is not lying: he indeed quotes from three different sources, but the name of Aëtius corresponds not to the quotations from P-style placita, but to the two fragments of Areios Didymos which he misquotes from Eusebius' PE. Diels' attribution of SP-Placita to Aëtius depends on his interpretation of the combination of particles ка! pevxoi Kai. What is the meaning of the phrase «with such force» (tanta vi) in Diels' Prolegomena? It seems that Diels takes ка! pevxoi ка( emphatically, i.e. as meaning something like «and especially», «last, but not least». It is by the «force» of this combination, according to Diels, that Theodoret marks his only real source. This combination occurs in CAG 80 times, and none of these instances is emphatic34. It is a colorless and somewhat pedantic expression which just adds an additional point or name, it is also often used to mark the termination of a list of names joined by ordinary ка(. Sometimes it introduces a second list of names35. In

33 CAG 2.94 ZayxrovraOrov pèv oûv o Bnpuxvoç xfv ®oivÎKrov Oso^oyiav ^uvéypays- psTfvsyKs 8è xaux^v siç xfv 'EM,aôa ^rovfv ®&rov, oùx o 'EPpavoç, àM,' o Bupkoç, xov ôè 2ayxrovia0rova Mav xsOaupaKsv o II °p9ùpv°ç.

Lebedev (1988). David Runia in Aëtiana III, 177 quotes in support of the emphatic use CAG i.12 Kai pévxoi Kai n^axrov ekevvoç. But the emphatic meaning of the reference to Plato in this passage is produced not by the combination of particles (which retains its ordinary meaning 'and also', 'as well as'), but by the word ekevvoç with its connotation of 'the well known' or 'glorious'. Below Theodoret adds EÙoxopia ôè nàvxaç ànoKpûyaç.

35 CAG 2.116 ouxroç ùpâç ôev xrôv Ava^ayôpou Kai nuOayôpou Kai n^axrovoç, Kai pévxov Kai Noup^viou Kai n^ouxapxou Kai n^roxivou Kai xrôv a^rov énaKouoavxaç ^ôyrov, énvZnxf°av xfv n^yfv. Here the combination introduces a second list of late Platonists distinguished from

our passage it is placed before the name of Aetius not because he is something special, but because it is the last name in a series. There are many passages in CAG where Kai ^evxoi Kai introduces the last of the three names without any emphasis36. A close parallel is found in CAG 1.14: (the most famous Greek philosophers, Pherecydes, Thales, Pythagoras, Solon and Plato received their knowledge of the true god from Egyptians and Hebrews) Kai xafixa SiSaoKei ^ev n^ouxap%o; o Boiroxio;, SiSaoKei Se Kai nop^upio; o Kaxa x^; a^nQela; Auxx^oa;, Kai ^evxoi Kai Nou^vio; o nuBayopeio; Kai exepoi n^eioxoi. Should we, following Diels' logic, conclude that the only factual source of Theodoret here is Noumenius whose name is quoted with «tanta vi» (Kai ^evxoi Kai) and that the names of Plutarch and Porphyry are added «splendoris gratia» to dupe his pagan readers? But this will not do: two of the 5 philosophers mentioned by Theodoret in this passage belong to 7 sages (Thales and Solon), and 7 sages were discussed in Porphyrius' Philosophos historia (201 F 4 - 203 Smith). Porphyrius was for the apologists an important source on Plato's theology, and Plutarch in his De Iside draws parallels between Egyptian religion and Greek philosophers, like Heraclitus and Empedocles.

Preplatonic philosophers and Plato. 5.82 toIIöv yap toutrav népi каг '^лократа каг Га^Уф ^uyypaфévтrav, каг pévToi каг nlàTravi каг H£voфövтl каг ApiaTOTéla каг ©шфраатю каг pupioiç ÉTépoiç. 6.75 каг à^p каг 0а!атта каг лотарог каг л^уаг каг v£фöv à5vv£ç каг 5poaoi каг aûpai, каг pévTOi каг фита rapm^d т£ каг arapna... 12.82 töv Çraypàфrav каг töv iaTpöv, каг pévTOi каг töv акитато^^ каг töv xalraTunrav каг töv xpuaoyoöv toùç pèv àpiaTOuç 0aup,àtpp,£v 12.95 éné5£i^a, TÎva pèv л£pг 0£oû каг uln; каг кла£га;, каг pévTOi каг àp£T^ç каг кaкíaç тог; töv 'Ell^vrav е5о^£ фlíoa6фOlç.

36 CAG 3.23 ÖTi npÖTOu; 0£ои; èvopiaav каг AiyunTwi каг ФoívlK£ç каг pévTOi каг 'Ell-qv^ ^li«v каг a£l^v^v каг yqv каг aaTpa каг Talla aTOix£Îa. 5.54 Outö 5è каг та alla та тошйта каг 'Орфои каг 'Наю5ои каг pévTOi каг Aiaxu^u. 6.55 0£panaivöv каг oiк£тöv, каг pévTOi каг y£öpyöv. 7.7 каг та; àyopàç каг та; àyuiàç, каг pévTOi каг töv £Ùn6prav та; огаа;. 7.45 Тайта pèv ош nàlai nap' 'Ell^vrav каг 'Prapairav каг Aiyurcuöv, каг pévTOi каг Eßpairav каг töv allrav Ето^ато ßapßdprav. 10.5 MapTuprî 5è 5iapp^5^v каг Пlouтapxoç ô Xaiprav£uç, àv^p о^х 'Eßpafo;, àll' 'Ell^v каг то yévoç каг T^v ylÖTTav, каг pévTOi каг таг; 'Ell^v^a!; 56^ai; 5£5oulraр,évoç... 10.67 ^стауои; 5è каг ло5пуой; каг pévTOi каг фгатаугауой; Еахпк6т£; тои; àлoaт6louç 10.70 oi по^рошта; .каг oi y£rapyoйvт£ç ... каг pévTOi каг oi all^v Tivà р£тax£lpíÇovт£ç Téxvnv 10.100 Каг ô ф5оро; 5è каг ô оу5оо;, каг pévTOi каг ô стато; каг nàvT£;, à; ело; rinrîv, oi yalp-ог... 6.37 Каг yàp ùyteia каг каНо; каг pöpn а0рато;, каг pévTOi каг Лойто; тои; pèv övqa£...

3. The OuoiKai 8o^ai of Theophrastus as Urquelle of all Preplatonic physical doxography is another myth of the 19-th century Quellenforschung

The extant doxography is based not on one classical, but on many Hellenistic (and Post-Hellenistic) sources: Stoic, Academic, Middle Platonic, Aristotelian, Epicurean. Theophrastus may be at best a marginal source. Diels seriously underestimated the direct use of Aristotle in SP-Placita (not mediated by Theophrastus) as well as the direct use of the original works of Preplatonic philosophers.

In his Prolegomena to Doxographi Graeci Diels compares the significance and impact of Theohrastus' Ouoixai So^ai37 on all subsequent physical doxography with the alleged fantastic success of Aristotle's emxopf p^toprov as described by Cicero in De inventione: Aristotle made such a pleasant and concise exposition of previous rhetorical theories that nobody after this event would consult the original handbooks: one concise compendium replaced many volumes of rhetorical tekhnai38. This comparison should be understood in the light of Diels' comments on the archaic obscurity of the early Greek philosophical prose in the Preface to the first edition of VS where he explains the purpose of his German translation of the fragments: «Denn abgesehen von der beabsichtigten oder unbeabsichtingten Unklarheit der Sprache, in der sich die aus der Tiefe zum ersten Male aufsteigenden Gedanke nur muesam durchdringen, steht diese archaische Rede weit ab von der periodisch gerundeten und semasiologisch abgeschlossenen Eleganz der Attiker des vierten Jahrhunderts.» (Diels, Vors. I, p. VI). What Diels wants to make clear by his comparison is that all those obscure and hard to read volumes of Preplatonic philosophers became obsolete after Theophrastus read them and made a concise and clear exposition of their contents in the «elegant» Attic prose of the 4th century. This is a bit surprising view given the fact there are hundreds extant Preplatonic fragments quoted from the original books during more the 1000 years, and on the other hand, there are only two attested quotations from Theophrastus' Phys. Opin. Diels seriously exaggerated both the obscurity of the Preplatonic books (especialy of the Ionian tradition Peri physeos) and the elegant

37 Diels himself used the incorrect title Фиагкюу 5ô^ai. I pointed out to his mistake in my 1984 paper on Democritus (Lebedev 1984: 14) and I am glad that Jaap Mansfeld arrived independently to the same conclusion (Mansfeld 1990: 3057).

38 Diels 1879: 119; Cicero, De inventione II.2.6.

clarity of Theophrastus, and at the same time he underestimated the linguistic competence of later Greek philosophers of different schools. Diogenes Laertius describes the style of Anaximenes' book as «simple and plain» (Ae^i; anAq ка! апергтто; - D.L. II,3.), and the preserved fragments of Anaxagoras confirm this chracterization of the Ionian scientific prose. On the contrary, the extant De sensibus of Theophrastus is one of the most obscure and hard to read classical philosophical texts that we possess. The language of Homer was much more archaic than that of the Ionian prose, but the Greeks were able to read and interpret it using extremely refined and sophisticated grammatical and hermeneutical techniques even in Byzantine times (e.g. Eustathius), and it never occured to anyone to replace the reading of the «obscure» original poems with lucid short summary expositions. If some archaic Greek philosophers were obscure (Pythagoreans, Heraclitus, Parmenides), they were intentionally so, not because they had difficulties in articulating their thought. I am not saying that Theophrastus' Фио1ка! So^ai was insignifcant work. No doubt it was a mine of information and for us it would be an invaluable source. But Hellenistic philosophers may have had a different opinion. I doubt that the Stoics were interested to know what Theophrastus said on Heraclitus, and Epicurus hardly got his knowledge of Democritus' atomic theory from summary expositions written in alien and obscure Peripatetic jargon by unfriendly critic. All Hellenistic schools were able to read the original Preplatonic books and to interpret them through the prism of their own philosophy, and they did so. It was Theophrastus' doxographical work that became obsolete in postclassical times, and not the works of early Greek philosophers some of which in Hellenistic times became foundational classics for the Stoics (Heraclitus), Epicureans (Democritus) and Sceptics (Xenophanes and Heraclitus).

Diels ' collection of the fragments of «Physical opinions» includes 25 texts (counting 5 a and 5b as two different fragments) (Diels 1879: 475-495). Three of these - fr. 2 (Anaximander and Anaximenes), 3 (Empedocles) and 8 (Atomists) do not cite and do not mention Theophrastus.

Of the remaining:

15 cite Theophrastus by name without specifying the work: 1, 4, 5, 9, 10, 12, 14, 15, 17, 18, 19, 20, 21, 22, 23.

Of the 7 that cite both name and work only 2 cite the doxographical work that may be identical with Фио1кюу РфМа in the catalogue of DL:

Fr. 5a (Galenus) веофраоод ...év тац xrov 9ugiköv So^rov énixo^mç xfv Eevoфdvouç So^av...

Фр. 8 0в6фраото<;..^ xroi nepi xrov фио1*ТО So^rov ката n^dxrova.

In other cases we have év xroi nproxroi nepi xrov фuоlкrov (fr. 6 Alex.), 0. év xfi émxopfi Avа^lцdvSpou ф-rçoi акойоаг (fr. 6a DL), év xroi nproxroi xfç фиогк^^ íотopíаç (fr. 7 Alex. Ap Simpl), év xoîç фиогкоц (13 Simplic), év xoîç фиогкоц (fr. 16 Placita II 2-3 on Xenophanes pyridia) - all these are quotations from Theophrastus' Physics (Та фиогга), not from Фuоlкаi 86^а1. Steinmetz rightly pointed out that «doxographical» quotations from Theophrastus in Simplicius derive from Theophrastus' Physics (Steinmetz 1964: 334-351).

FHS&G print in their edition of Theophrastus' fragments under «doxographica physica» both groups without separating them (Fr. 224-245; v. 1, pp. 402-434). In my opinion fragments from Фио1*ш 86^а1 and from Та фиогга should be printed under separate authentic titles.

Diels tried to prove the Theophastean origin of the extant doxography of the Imperial age in a synoptic table, the so caled «Theophasteorum apud excerptores conspectus.» (Diels 1879: 132144). This table may be of some practical value, but it fails to prove Diels' main thesis because it mostly postulates Theophrastus as a common source of any two or more authors none of whom actually quotes Theophrastus. Sometimes the similarity is too general, and the common source may be the original Preplatonic text. A Hellenistic doxographical compendium like Arius Didymus is also conceivable. The astronomical doxai of Heraclitus in SP-Placita (anathymiasis from the sea, celestial skaphai etc.) is paralleled by the doxographical section in Diogenes Laertius' chapter on Heraclitus. No other source mentions skaphai, so there may have been a common source. After Diels this common source has been commonly identified with Theophrastus' Phys. Opin. But neither SP-Placita, nor Diogenes Laertius quote Theophrastus as a source of the skaphai theory. Moreover, in the preceding context Diogenes Laertius quotes Theophrastus by name as saying that Heraclitus «left something unifinished, and made conflicting statements» in different parts of his book which amounts to assertion that Heracitus' book did not contain a consistent and complete physical theory39 .

39 D.L. IX, 6 ©EÔ9paaxoç 5é ф^сгу ùno ps^ayxoWaç та p,èv fpvxs^f, та 5è а^отЕ a^raç exovto ypaysiv.

However the «detailed» (та ¿п! церои;) exposition of his physical doctrine that Diogenes relates in IX.8-11 is complete, consistent and free of contradictions. Therefore Theophrastus as a main source is unlikely. The cosmogony may be based on the Stoic reconstruction, and the bizzare theory of celestial skaphai probably derives not from Heraclitus' original, but from a hemeneutical work on Heraclitus that «reconstructed» astronomical theories from Heraclitus' metaphors and analogies, e.g. analogies between cosmic processes and the torch race or sacrificial fire in the concavity of an altar (Lebedev 1987-88: 242-245). Another example of attested (quoted by name) doxographical report of Theophrastus that contradicts the hypothetical «Theophrastus» of Diels' table of «Excerptores» concerns Anaximander. I cannot reproduce here all my arguments against the authenticity of the term то апelрov (see Lebedev 1978 and 1988-2). In my view it is an Aristotelian term for what Anaximander himself called фйсц апе1ро;, ка! ау^рю;. Aristotle does not ascribe

the term то апelрov to Anaximander, he speaks about the infinite matter in Preplatonic cosmologies and states that «most of the physiologoi» (sic!) regard the infinite as a principle (ар%^)40. The doxographers extracted Anaximander's placitum on то апelрov' as ар%^ from this passage of Aristotle's Physics. Simplicius in Phys., 27,2 quotes Theophrastus as saying that Anaxagoras' theory of original mixture is similar to that of Anaximander41. I take the words екешо; as a reference to Anaximander, not to Anaxagoras (contra Diels). Thus we get a neglected fragment of Anaximander which contains an authentic analogy between the cosmogonical and the separation of gold particles from earth due to the rotational movement of a washing pan. That Anaximander conceived his primordial substance as a mixture and explained material change by apokrisis, diakrisis etc., not by qualitative transformation of a single substrate, is also attested by Aristotle42, and I cannot see how this consensus of Aristotle and Theophrastus can be neglected. Theophrastus compares the physical theories of Anaxagoras and Anaximander reducing Anaxagoras' first principles to two: xqv апeíрou фйсш ка! xov voйv, and concludes that Anaxagoras' theory of оюцат1ка сто^еТа is similar to that of Anaximander. Diels identified the source of the doxography of Anaximander in Simplicius Phys. 24,13 as Theophrastus and included it into his

40 Arist. Phys. 203b 6 ff. = 12 A15 DK.

41 Simpl. Phys, 27, 2 ff. = Anaxagoras A 41 DK.

42 Arist. Phys. 187a 20 = 12 A 9 DK.

collection of Phys. opin. (fr. 2 Diels). But Simplicius does not quote Theophrastus and apparently follows the doxographical cliché of aneipov as that was exctracted directly from Aristotle in

Imperial age. Moreover he ascribes to Anaximander a theory of cyclical change of the 4 elements which presupposes aristotelian alloiosis, and not the mechanistic apokrisis-diakrisis. This theory is incompatible with the theory of change and the authentic simile that real Theophrastus (quoted by name) ascribes to Anaximander in Phys. 27,2. It follow thart Simplicius' doxography of Anaximander in Phys. 24, 13 ff. is not based on Theophrastus.

Mansfeld and Runia have rightly pointed out to the parallels between SP-Placita and Aristotle's Meteorology and De Caelo (Mansfeld, Runia 2009: 110 ff., 135 ff.). It seems that the Pre-platonic doxography of the first principles in SP-Placita Book I heavily depends on Aistotle's Metaphysics Alpha, not on Theophrastus. Therefore Theophrastus should be acquitted of accusations like those of McDiarmid who compared in his study not Theophrastus with Aristotle, but Aristotle with Aristotle (McDiarmid 1953).

Here is yet another clear example of how Preplatonic doxai were digged out of Aristotle's works by the compiler of SP-Placita or his source.

[Plutarchus], De placitis philosophorum 1,13, 2. (Пер! ¿Aa%iGxrov).

'HpàK^eixoç ynYMàxià xiva ¿Aà%iGxa ка! ацер^ eioàyei.

Heraclitus introduces some kind of tiny scrapings which are minimal and have no parts.

Stob. 1,14,2 'HpàK^eixoç про xoù évoç 8океТ xiGi yrçy^axa KaxaAeineiv.

Heraclitus, according to some, admits tiny scrapings before the one.

Both texts ascribe to Heraclitus a kind of atomism, Heraclitus allegedly admits certain elementary particles which «precede» the one, i.e. the single substrate (fire). These strange placita are based on the misreading of the following pasage in Arstotle: (a) Aristoteles, De caelo 304a 9:

Oi ^èv yap aùxrov G%^a nepiànxouGi хф nupi, Kaôànep oi x^v nupa^iSa noioùvxeç, Ka! xouxrov oi ^èv ân^ouGxéproç Aéyovxeç oxi xrov ^èv Gxn^àxrov x^nxiKroxaxov ^ nupa^iç, xrov Sè Gro^àxrov xo nùp, oi Sè Ko^yoxéproç тф Aoyro npooàyovxeç oxi xa ^èv Gro^axa nàvxa GuyKeixai ¿к xoù Aenxo^epeGxàxou, xa Sè G%^axa xa Gxepea ¿к nupa^iSrov, roGx' ¿пе! xrov ^èv Gro^àxrov xo nùp Aenxoxaxov, xrov

Se oxnpaxrov f nupapig piKpopepeoxaxov Kai nproxov, xo Se nproxov oxfpa xofi nproxou oropaxog, nupapig av e'in xo nfip.

Oi Se nepi pev oxfpaxog ouSev ano9a(vovxai, ^enxopepeoxaxov Se povov noiofioiv, eneix' eK xouxou ouvxi0epevou 9aoi y(yveo0ai ihaah Ka0an£p av £i au^uara^Evou y^y^aTO«;.

Philosophers who ascribe to the particles of fire certain shape (pyramid) are Pythagoreans and Plato. Those who do not speak about figure, but simply regard the fire as thinnest element, should refer first of all to Heraclitus. The comparison that follows (in bold face) is a verbatim (or close to the original text) quotation from Heraclitus (Lebedev 1979)43. yfypa in the sense of gold dust is vox Ionica well attested in Herodotus, but alien to Aristotle and Attic usage. The quotation derives from the «technological» section of Heraclitus book in which he was drawing parallels between cosmic processes and technological practices of men to prove the basis thesis f xe%vn pipetxai xfv 9uoiv. In Ionian dialect yfypa «gold dust» was a word for «raw gold» in general, i.e. raw metal from which various figures and ornaments could be produced by melting and casting (oup9uoav). Heraclitus' original text did not imply any atomistic or corpuscular theory of fire. It is conceivable that the atomistic reading of the Aristotelian passage belongs to an Epicurean who in his polemics against the Stoics pointed out that their venerated authority, Heraclitus, admitted a kind of atomic theory and thus supported Epicureans, and not the Stoics. The Stoic compiler of Placita actually distances himself from this view by adding SoKet xioi.

43 All three texts are now included as fragment 116 in my edition of Heraclitus (Lebedev 2014: 202-203 with a commentary pp. 425-427). Of two possible interpretations of oup^uoropévou I discussed in my 1979 article only «melting» (with connotation «casting», «transforming») can be correct, not «smeltng» (separation of pure metal and ore). Aristotle probably has in mind this simile of Heraclitus also in De caelo 298b32.0i ôè xa p,èv a^a navxa yivsoOai ^aoi Kai pevv, Eivai ôè nayiroç où0év, ëv ôé xi pôvov ùnopévEiv, s£, où xafixa navxa pETao/np-axiCEoPai néyuKEv5flEp soÎKaoi Poû^EoOai ^éyEiv a^oi te rcoM,oi Kai 'HpaK^Eixoç o 'E^éoioç. Heraclitus' simile is imitated in De victu I,20 (xpuoiov spyaÇovxai ktL) and Plato, Timaeus 50a Ei yàp navxa xiç oxfpaxa n^aoaç ek xpuooû pnôèv pExan^axxrov nauoixo EKaoxa Eiç anavxa, ôeikvwtoç ôf xivoç aùxrôv ëv Kai spopévou xi nox eotî, paKprô npoç à^f0Eiav ào^a^éoxaxov EinEvv 5xi xpuoôç, xo ôè xpiyrovov 5oa te a^a oxfpaxa svEyiyvExo, p^ôénoxE ^éyEiv xaûxa àç ovxa ktL On the 17-th century reception of these placita in Olearius see: Primavesi 2015.

4. Authorial Definitions and the «better Plutarch» (P+)

The extant P in 5 books contains 128 thematic chapters. Apart from the proper doxai quoted with a preceding lemma (philosopher's name)44 P also contains 8 authorial definitions in the beginning of a chapter, 6 in the first book (definitions of matter, idea, cause, body, shape and color), one in the end of the second (Year) and one in the beginning of the third (Milky Way). All 8 authorial definitions in P are also preserved in S. The absence of the author's name and the oratio recta (with ¿oxi typical for definitions that answer the question xi eoxi x?) indicate that this is not someone's opinion, not a doxa, but an established truth. In other words all authorial definitions are intended as the correct and definitive view in contrast with various doxai that follow like endnotes.

1. YAH

P 1.9 'T^n ¿oxi xo unoKei^evov nproxov yeveoei Kai 90opa Kai xai; a^Aai; ^exapo^ai;.

S 1.11.1 T^n ¿oxi xo unoKei^evov nao^ yeveoei Kai 90opa Kai xai; a^Aai; ^exapo^ai;.

2. IAEA

P 1.10 ISea eoxiv otioia aoro^axo;, atix^ ^ev u^eoxrooa Ka0' auxqv, eiKoviZpuoa Se xa; a^op^ou; u^a; Kai aixia yivo^evn xq; xouxrov Sei^ero;.

S 1.12.1a ISea eoxiv otioia aoro^axo;, atixq ^ev u^eoxrooa Ka0' atixqv, eiKoviZpuoa Se xa; a^op^ou; u^a; Kai aixia yivo^evn xq; xouxrov Sei^ero;.

3. AITION

P 1.11 A'ixiov eoxi <Si' o> xo anoxe^eo^a q Si' o ou^Paivei xi.

S 1.13.1a A'ixiov eoxi Si' o xo anoxe^eo^a, q Si' o ou^paivei xi. apKei yap unoypa^iKro;.

4.SHMA

P 1.12 Sro^a eoxi xo xpi%q Siaoxaxov, n^axei, PaQei, ^qKei- q oyKo; avxixuno;, ooov ¿9' atixro- q xo Kaxe%ov xonov.

S 1.14.1a Sro^a eoxi xo xpi%q Siaoxaxov, n^axei, PaQei, ^qKei- q oyKo; avxixuno;, ooov ¿9' atixro.

5. SXHMA

P 1.14 eoxiv eni9aveia Kai nepiypa9q Kai nepa;

oro^axo;.

44 Mansfeld and Runia use the term lemma for the complete placitum (name-label plus tenet), I follow Diels in Dox. Gr. and Wachsmuth in his edition of Stobaeus who use the term lemma for the name-label only.

S 1.15.3b £%fpa eoxiv eni9aveia Kai nepiypa9f Kai nepag oropaxog (after lemma nop9upiou).

6. XP^MA

P 1.15 xpropa eoxi noioxng oropaxog opaxf

S 1.16.1 Xpropa eoxi noioxng oropaxog opaxf npo^youpevrog.

7. ENIAYTOS

P. 2.32 Eviauxog eoxi Kpovou pev eviauxrov nepioSog H, Aiog Se ip', 'Apeog Suetv, 'HMou iP' pfveg- oi S' auxoi Eppofi Kai A9poS(xng, iooSpopoi yap- oe^fv^g fpepai H- ouxog yap o xe^eiog pfv ano 9aoerog eig ouvoSov.

S 1.8.42c Eviauxog eoxi Kpovou pev eviauxrov nepioSog xpiaKovxa, Aiog SroSeKa, Apeog Suotv, 'HMou SroSeKa pfveg- oi S' auxoi Eppofi Kai A9poSixng, iooSpopoi yap- Se^fv^g kxL

8. rAAAHIAS

P 3.1 KuK^og eoxi ve9e^oeiSfg ev pev xro aepi Sia navxog 9aivopevog, Sia Se xfv ^euKo%poiav ya^a^lag ovopaZppevog.

S 1.27.1.1 KuK^og eoxi ve9e^oeiSfg, ev pev xro aepi Sia navxog 9aivopevog, Sia Se xfv ^euKo%poiav ovopaZppevog ya^a^lag

It is hard to imagine that the author of the original P included authorial definitions only in 8 out of 128 chapters. The choice of the preserved definitions seems to be accidental and must be due to the omission of many others. Six of the eight definitions come from the first book on principles, but why only 6 out of the expected 30 or so? Why shape and color are more important than god, time and fate? Why in the rest 4 books there are only two authorial definitions of the Milky Way and of the year? Are these topics more important than the fundamental concepts of the cosmos and Universe? All this looks erratic and illogical, and stands in flat contradiction with author's pedantic style and his promise in the preface to give a systematic account of the physical science (physikos logos). The conclusion seems inevitable that the extant text of P is a truncated and heavily abridged version (or rather a personal copy) of a complete handbook of physical philosophy from which more than 90% of the authorial definitions and/or other authorial explanations, introductions and remarks have been removed. There are two neglected authorial definitions (nepi xu%ng and nepi avayK^g) of which the second, in all probability, is quoted by Stobaeus from Arius Didymus (see below section 7). The authorial view on the topic of each chapter may have been stated not only in the form of a definition, but also in the form of an explanation/answer to a question of the type nrog ouveoxn o Koopog; (nrog opropev; etc.) or Sia xi yivexai; Authorial definitions are not to be expected in the

chapters with headings of the type ei to e^ppuov Zroiov; either. Such heading are typical for the genre of npop^q^axa. On the contrary lost authorial defintions are very plausible in chapters with headings that answer the question xi; otioia xoti...; It is also conceivable that some doxai of the Stoics and Aristotle (recognized as dogmatic authorities by the author of SP-Placita in the Preface) were intended as expressing the author's view. This is actually the case with the first chapter xi; eoxi 9tioi;; The archetype of all P's textual tradition may have been a personal copy of a Christian apologist who was interested only in the diaphonia of the Hellenes and used his «edited» copy to demonstrate the disagreement of pagan philosophers on all possible subjects. Apodictic authorial definitions were not just useless for his purposes (they provided no names), but could undermine his arguments since they looked like a commonly agreed knowledge. It is conceivable that he also ideologically edited the chapter on the origin of cosmos by removing creationist and teleo-logical cosmogonies (Plato, Pythagoreans, Anagaxoras, Empe-docles, Stoics) and leaving only the «atheistic» Epicurean atomistic version (Plut.1.4). We designate the original P as P+ («better Plutarch»). P+ was used by Stobaeus and the compiler of extant P (not by Theodoret and Nemesius!), possibly also by Porphyrius (see Appendix 1 below), traces of it are discernible in ps. Galen. The extant P became the source of later Patristic, Byzantine and Arabic doxography. Since P is not cited explicitly with Plutarch's name before Eusebius, there is a chance that extant P goes back to Eusebius' copy. Mansfeld's and Runia's observations on Eusebius' ideological editing of doxography apply here (Aetiana, vol. 1, 134 ff.). An obvious objection to this hypothesis is the existence of the 3rd century papyrus fragments of what looks like P from Antinoopolis (Mansfeld, Runia 1997: 126-129), but on the other hand these fragments are too scanty to allow firm conclusions and they do not attest Plutarch's name.

5. Is P+ identical with SP-Placita?

This is a tricky question. We arrive at the notion of P+ not by traditional methods of Quellenforschung, but by the method of internal reconstruction from remains (traces) or, in simple words, by the method of Sherlock Holmes or by hunter's common sense: if there is a skeleton of a bear, there must have been sometimes on the same spot a living bear. P is a «skeleton» of P+ stripped of almost all its apodictic/authorial flesh. Or, if you prefer an architectural

analogy, the extant P is like a 5 store apartment building with balconies in apparently ruined condition: only the first floor is preserved intact and it has a row of 4 balconies, some of the upper 4 stores are partially destroyed, and all balconies (except two preserved by chance in asymmetrical position) on this upper stores are missing. It is easy to conclude that originally there were 20 balconies in this building of which almost 80% have been lost. The existence of P+ is more certain than the existence of SP-Placita, i.e. of a hypothetical common source of P and S (identified by Diels with Aetios), because postulating a C as a common source of A and B on the ground of their similarity is a risky procedure: one of them may be the source of the other. And S indeed often looks as a (superior) source of the deteriorated or compressed P. So the question arises: is P+ an intermediate source between the original SP-Placita and extant P (possibility 1), or is it identical with SP-Placita (possibility 2)? This question can be also reformulated as follows: was the diairesis into chapters and thematic sections (problems) in SP-Placita the same (or at least almost the same) as in extant P, or alternatively, the extant diairesis was an innovation of P+ introduced by the compiler of P+ simultaneously with his abridgment of the anterior SP-Placita?

The principle «Placita praeter necessitatem not sunt multi-plicanda» advises us to prefer the second possibility, although the first possibly cannot be ruled out. Another important question is whether what we call P+ was from the start attributed to Plutarch. An important distinction must be made at this point. One thing is a better manuscript of P which had some additional lemmata missing from the extant P. This manuscript(s) may well have been already transformed from the handbook of physics (what we designate as P+) into a collection of doxai ad usum apologeticum, and stripped of most authorial definitions and other non-doxographic passages. And another thing is a diaeretic handbook of physics in its original form complete with 130 authorial explanations, definitions and other non-doxographical didactic materials. Better manuscripts of the extant P should not be confused with P+. The original P+ may have passed through several stages of deterioration, starting with «student's notes» of young Plutarch and ending with drastic expurgation by a Christian apologist.

6. Two additional authorial definitions from Stobaeus

It is commonly accepted that S provides numerous additional doxai omitted in P. But if so, S with equal success may be the source of additional authorial definitions. There are at least two such instances. The first is the definition of chance from the chapter oxi a^oyioxog f 9opa xfg xu%ng:

S 1.7.9b.1 Tuxn axaKxou evepyeiag eoxi npoo^yopia. «Chance is a name of an irregular event». Diels (Dox.Gr. 326, 12-13 ) prints it as a part of the preceding doxa of Anaxagoras and Stoics, but in the note ad. loc. agrees with Heeren that it is transmitted without lemma and in the Prolegomena (p. 46) attributes this definition to Democritus comparing Theodoret IV. 15. Wachsmuth disagrees with Diels in apparatus criticus ad loc. (vol. I, 92, 18) and prints it as anonymous separate doxa. Oratio recta without lemma is typical for authorial definitions, therefore in our view this definition of chance should be located in the very beginning of the chapter nepi xu%ng of the original SP-Placita. The second neglected authorial definiton is found in the chapter nepi avayK^g 0eiag kxL:

S 1.4.7b AvayK^ eoxi Kpioig PePaia Kai apexaxpenxog Suvapig npovoiag.

«Necessity is a firm decision (or «determination») and an irreversible force of the providence».

In Wachsmuth's edition this definition is printed with a lemma Eppofi eK xrov npog 'Approva, and on the basis of this the fragment has been included in the Corpus Hermeticum (Corp. Herm. fr. 13,1). But this is not what we find in the two main manuscripts (we have inspected both of them on photographs). Farnesinus (F) has in the text the following series of lemmata: ©a^xog. So9oK^f. ©ueoxn EupiniS. AiSupro. Eppou eK xrov n^axrovog, and in the margin aKprova nu0ayopou. Codex Parisinus (P) has in the text 0a^xog. oo9oK^fg. etipiniS. SiSup. eppofi eK xofi n^axrovog. aKprova nu0ayopou. Wachsmuth in his text of Stobaeus eliminates the transmitted name AiSupro / SiSup. and following a bold conjecture of Wackernagel makes it into the title of Euripides' play AiKupviog. As a result of this we have in Wachsmuth's text of S. 1.4.6 the following lemma with a tragic verse on fate:

EtipiniSng AiKupviroi- To xfg avayK^g oti ^eyeiv ooov ^uyov. 45

45 Accepted as a fragment of Euripides' Likymnios in Kannicht, TrGF vol. V, fr. 575. Euripides' Likymnios is quoted in Stobaeus once in 3.29.7 (3,628, 4 Hense) = TrGF V, 574. The proverbial phrase Zuyov avayKn?

There is no reason to doubt that the tragic verse comes from Euripides, but it is quoted without a reference to the play. The emendation of aK^rova to 'A^rova is plausible, but Hermes Trismegistus is quoted in the same chapter below (S 1.4.8), so the lemma 'Ep^ofi ¿k xrov npo; 'A^rova refers most probably to the same hermetic text and duplicates the lemma 'Ep^ofi npo; xov uiov: two different doxai of «Hermes» in the same chapter are unlikely. The name of AiSu^o; attested in two main manuscripts of S should be retained with a slight alteration AiSu^ou (proposed already by Heeren), and the lemma AiSu^ou in this case can only be attached to the definition of avayKn quoted after the verse of Euripides and the apophtegm of Thales (S 1.4.6-7a), and before the doxa of Pythagoras (S 1.6.7c) from which the excerpts from SP-Placita start. The location of this definition just before the excerpts from SP-Placita (cf. Diels, SP-Placita 1.25) is perfect for an authorial definition: all extant authorial definitions in SP-Placita always precede the first doxa of the relevant chapter. The language, style and terminology of the definition of ananke that we attribute to Didymus are Stoic, not Hermetic. And Arius Didymus was a Stoic. The style is concise with precision, not loquacious, not inflated and without mystical flavor typical for the hermetic texts. TLG searches in Corpus Hermeticum for a^exaxpenx-, for proximity of Kpioi; and pepaio; (in all cases) and for proximity of npovoia and Suva^i; (in all cases) yield no results46. On the contrary, both the doctrine and the language of the definition find exact parallels in Chrysippus:

cf. Chrysipp. ap Stob. Eclog. 1.5.15 Axponov Se, oxi a^exaxpen-xo; Kai a^exap^nxo; eoxiv o Ka0 eKaoxa Siopio^o; e^ aiSirov Xpovrov, Chrysipp. SVF II fr. 1092 = Plut. De commun. notit. adversus Stoicos 1061D napa xqv evvoiav eoxiv ayaBrov ^ev elvai ^eyioxov xo a^exanxroxov ev xai; Kpioeoi Kai pepaiov, SeioBai Se xouxou xov en' aKpov npoKonxovxa.

Chrysipp. fr. logica et physica, 30 npovoiav, xo ouve%eo0ai ^ia Suva^ei xov Koo^ov...

We have pointed to this definition as a possible quotation fom Arius Didymus already in our 1988 paper on «Aetius». Mansfeld and Runia in their Aetiana, I, 336 made the following objection to this: «The second part, with its clearly Stoicizing flavour, would not

occurs in Euripides' Orestes, 1330 and Aeschylus' Agamemnon, 1071. Surprisingly, it is very popular in church fathers.

46 Except one: the definition of ananke from Stobaeus under discussion erroneously included in Corpus Hermeticum on the basis of Wachsmuth's misplaced lemma 'Eppou.

be out of place at the beginning of P 1.25. The first half, however, has a logical-epistemological scope and could hardly fit there». The phrase Kpioig PePaia has a variety of different meanings in different contexts (medical, logical, psychological, legal, theological etc.). In Hippocrates it refers to the «definitive crisis», i.e. resolution of the disease47, in logical and epistemological contexts (when applied to the human faculty of judgment) it can refer to the «certainty» of judgment, in legal and political contexts (when applied to human will and intentions) it can refer to the «firm decisions» of the parties or allies involved. The genitive npovoiag qualifies not only the phrase apexaxpenxog Suvapig, but also the phrase Kpioig PePaia. We understand Kpioig PePaia not as «certain judgment» (that indeed would give the phrase logical-epistemological sense), but as «firm determination» or «decision» of the divine will, i.e. of pronoia. Unike the decision of human mind, the dicision of the divine mind has theological and metaphysical, and not epistemological meaning. Chrysippus' interpretation of the Moira's name Atropos uses exactly the same epithet apexaxpenxog, and the term Siopiopog «determination» is semantically close to Kpioig. The relevance and importance of the lemma AiSupou for the problem of the authorship of SP-Placta will be discussed below.

A promising source of further additional authorial definitions from the lost P+ may be Ps.Galenus' Historia philosophos48, Neme-sius' De natura hominis49 and the Definitions of Aquilius published by Marwan Rashed who has signalized parallels and convergences between the new collection of definitions on the one hand, and the Placita tradition, Arius Didymus and Alcinoos' Didaskalikos, on the other50. Rashed's general conclusion that the Definitions of Aquilius demonstrate especially affinity with the doxographical work of Arius Didymus seems plausibe enough (Rashed 2012: 170 ff.)51.

47 Hippocr., Coa praesagia, 147 'Tnvoi PaOssc;... PsPaiav Kpiaiv

anpaivouaiv.

49 Cf., e.g. definitions in Ps.Galen, ch. 4-14, 19-23 etc.

49 Cf., e.g. definitions of oxovxsiov (ch. 5), to ^avxaoxiKov (ch. 76), the senses (ch. 7-11), to SiavonTiKov (ch. 12), to pvnpovsuTiKov (ch. 13) etc.

50 Rashed (2012). The definition of body in Aquilius displays verbal coincidences with P 1.12: aropa saTiv ouaia Tpixfi SiaoTaTf psra avTi-Tuniaq (Def. 82), and the definition of time (Def. 81) would perfectly fit the chapter nspi xpovou in SP-Placita: xpovoq saTi psTpov Kivfasroc; Kai povfc; ff apiOpoc; tou sv Kivfasi npoTspou Kai uaTspou.

5 We agree with some reservations concerning the credibility of Diels' «reconstruction» of the «Epitome», of Goransson's claims about Alkinoos and Arius etc.

This remarkable discovery provides some additional and independent support to our hypothesis that in Arius' diaeretic handbooks definitions were used systematically, and not sporadically as in extant P.

Provided that the ethical doxography (A) in Stobaeus II 7 belongs to Arius Didymus, the passage which emphasizes the methodological importance of Aristotle's categories 2 can also provide some additional support to our thesis: Aristotelian categories play important role in the diairesis of chapters adopted in SP-Placita.

7. The compendium P+ as a handbook of physics

Mansfeld and Runia in their «Aetiana» emphasize the «dialectical» character and purpose of the SP-Placita («Aetius» in Diels' mistaken terminology) and connect it with the «diaeretic» method of presentation of conflicting tenets intended to demonstrate their diaphonia. We criticise this interpretation in Appendix 3. The text of the extant P, indeed, may look sometimes as a collection of divergent and contradicting opinions, but it is the result of epitomizing (see section 3-4 above). In its original form, i.e. P+, it was a theoretical handbook of physics in which various doxai were attached to the definitive apodictic definitions and explanations of the author. Its main purpose was teaching the basics of physical philosophy as approved by the most authoritative philosophers (predominantly Stoics and Aristotle as indicated in the preface). So the original Sitz im Leben of the lost P+ was different: it was a handbook, a school text that had to be learned by heart, hence the brevity and precision of the authorial definitions. A close parallel is provided by Ps.Galenus Historia philosophos, a compendium essentially based on P, but with some possible traces of the lost P+. In the Preface the author explains the purpose of his work as a kind of popular «teach youself» book for those 9i^o^a0ei; who wish to enrich their erudition and to be able to understand what the great

52 Stob. II 7. 2 qyoupsvoq 5' ёраитф npsnsiv про navxoq xqv ouoiav 5etv елюколету ou тц лрауратЕтетаг, Kansvra лоютпта xqv nspi auxqv каг посот^та, каг тоитог^ ЁфЕ^ то лро^ -ri, vopi^ra просЕлглоуптЕОУ [те] sivai каг та тюу a^rav елшколету, кабалЕр ой лаутюу, оитю^ тюу nspi тайта 5гЕУЕукаутюу. I understand this passage as a kind of acknowledgements addressed to Eudorus whose diairesis is quoted immediately afterwards: the diairesis of Philo (despite its usefulness) should be modified and enriched by the diairesis of Eudorus with its emphasis on Aristotle's categories. See also Viano 2005: 142.

philosophers are saying «without exegetes», i.e. without teachers53. After this he annonunces that he will «pass by» the «sophistic and excessive subtleties» of the dialecticians (xrov Sia^eKxiKrov) as worthless for the moral and intellectual progress (npooSog) of his readers54. By saying this we do not exclude that the collections of various doxai attached after authorial introductions and definitions (as a kind of endnotes) among other puproses could also serve for discussions and debates, i.e. for dialectical purpose. The authorial comment in the chapter «On vision» (SP-Placita 4.14 Diels) Suvaxai xig naoi xouxoig xoig Ke9a^a(oig %pfo0ai eni xofi nrog opropev «one may use all these basic summaries (kephalaia) with regard to the problem «ho do we see»55 is compatible with such use, although it does not specify exactly the occasions on which the collection of doxai relating to the problem of vision may be used. The purpose of the collection seems to be more general and the author rather refers to general learning and display of erudition: any encyclopedia or reference book can be used for debates, but it is not its primary purpose. Isn't it cool during a table-talk to cite, e.g. Pythagoras' view of catoptric images? Popular collections of philosophical doxai have something in common with the works of poikilographoi like Aelian or Deipnosophistai of Athenaeus. The «editor» of the extant P, probably a Christian apologist involved in the polemics against the Hellenic philosophers stripped P+ of most of its definitions and other authorial passages since they were useless for his polemical purposes. He needed only placita with lemmata (name labels), and especially with famous names in order to prove that even the most

53 Hist.Philos.2 npouOspsOa Ss Sia^xOfvai nspi TouTrov ouSsv psv i'Siov supnKOTsq, Ta Ss napa totc npoTspoiq anopaSnv sipnpsva auvayayovTsq, aayroc ts Kai auvToproc anouSaaavTsq nspi TouTrov Sia^xOfvai, iva oi yi^opaOsic toutoic nposvTuyyavovTsc pf SsoivTo Trov s&nyoupsvrov, a^a Si' sauTrov yiyvroaKoisv Ta KaO' EKaaTa Trov sipnpsvrov aa^saTspov. This recalls the modern genre of popular introductions of the type «difficult things made simple».

54 ibidem: aa^a^saTspov Ss nsipropsOa nspi TouTrov sinsiv Ta aoyiaTiKa Kai navu KsKoapnpsva Trov Sia^sKTiKrov naps^n^uOoTsc. ou yap sniSsi^sroc EvsKa npouaTnaapsOa Tfv nspi TouTrov anouSfv, a^a ypnaTa Ta fOn noifaai Pou^nOsvTsc toTc sk ^i^oao^iaq ^aPsiv Ti xpfaipov nposyvroKoai Kai vopiaavTsq ^i^oao^siv ou povov touc Ta TpiProvia nspinppsvouc, a^a Kai touc <Tfv> npooSov psv toTc no^oic opoiav svarqaapsvouc, Tfv Ss Siavoiav yiioaoyroTspav sxsiv anouSaaavTaq. He probably did not consider as «sophistic» the definitions of logical terms he quotes in 9-15 from Sia^sKTiKoi.

55 The importance of this remark is rightly emphasized by Mansfeld, Runia 2009: 183 et passim.

authoritative Greek philosophers disagree (diaphonia) with each other whereas all Christians adhere to one and the same eternal and unshakeable truth. The form of «short placita», precise and brachy-logical, not accompanied by any proof or argument supporting the summarized doctrine, typical for the SP-Placita, must have originated in the sceptical Academy (the Clitomachus tradition). Parallels between the SP-Placita and Cicero are easily explained by the common Academic heritage. And it is significant that the author of Stobaeus' ethical doxography (A) (Arius Didymus in our view) chooses his diairesis (to which he will «attach» areskonta) from two Academic philosophers: Philo and Eudorus. The very term xa apeoKovxa (which in Hellenistic times replaced the classical term So^ai) used in the title of P in all probability derives from the sceptical Academy. Just as xo KaBqKov is an abbreviated variant of the original phrase xo KaBqKov epyov, so xa apeoKovxa is a shortened version of the original phrase xa apeoKovxa Soy^axa which is preserved in its original form in the title of P. This terminology reflects the sceptical attitude towards dogmatic doctrines that are «dear» to or «favored» by dogmatic philosophers, whereas the sceptic is equally indifferent to all. Greek sceptics invented this powerful polemical weapon (collection of conflicting views on the same subject) targeting only the dogmatics, the Christian apologists adopted it to discredit and reject Greek philosophy as a whole. However the compiler of the P+ had a completely different task. He clearly describes the aim of his compendium in the Preface to Book 1:

MeAAovxe; xov 9uoiKov napaSrooeiv Aoyov avayKaiov qyou^eBa etiBti; ev ap%ai; SieAeoBai xqv xq; 9iAooo9ia; npay^axelav, iv' eiSro^ev xi eoxi Kai nooxov ^epo; atixq; q 9uoiKq Sie^oSo;.

«As we intend to teach the physical science (or «theory») we believe that it is necessary from the very start to define the subject matter of philosophy in order to understand what is philosophy, how many parts it has and which of these parts is the detailed exposition of physics». (Follows the tripartite division of philosophy by the Stoics and Aristotelian division into theoretical and practical part).

The word napaSrooeiv is extremely important for the understanding of the purpose and genre of the compendium: it is used here in the rather late technical sense 'to teach' (still extant in modern Greek napaSoon 'lecture')56, and not in the classical sense

56 LSJ s.v. napaSiSropa I,4,b «teach doctrine», cf. especially Arrian., Epictet. 2.14.2 oxav xsyvn napaSiSroxai

'to hand down' by tradition (legends, opinions etc.)57. We hear the voice of a professor who in the beginning of his course of lectures defines the subject matter of the discipline he is going to teach58. napaSrooeiv in this context has nothing to do with doxography, with 'handing down' opinions, and it is not a term specifically connected with debates, it means to teach physical theory to the students, so the compendium is presented as a handbook. This is reaffirmed in the beginning of the first chapter on physis.

EneiSq npoKeixai q^iv xa 9uoiKa Beropqoai, avayKaiov qyofi^ai Sn^rooai, xi nox' eoxiv q 9uoi;- axonov yap neipaoBai ^ev 9uoio^oyeiv, ayvoeiv S' atixo xofixo, xqv xq; 9uoero; Suva^iv.

Since we intend to study physics, I believe it is necessary to explain what is nature, for it is inept to try to theorize about nature and at the same time to ignore the very subject of the theory, i.e. what nature means.

What follows is the Aristotelian defintion of nature as the principle of motion and rest, and the whole passage is phrased and formulated in Aristotelian terms. In Aristotle dialectics may be useful for the philosohy, but strictly speaking Qeropia stands above dialectis, and Beropqoai in the quoted passage has little to do with dialectics in Aristotelian sense. Theoretical science according to Aristotle relies on apodeictic, not on dialectical arguments.

8. The identity of Arius Didymus. Objections to Goransson

In his 1995 monograph Goransson makes some claims which, if correct, could result in complications for some of our conclusions. I mean first of all his denial of the identity of Arius the doxographer and Arius the court philosopher, the friend of Augustus, contrary to the widely accepted (and argued by Hahm) view that goes back to Meineke and Diels. This thesis is closely connected with an attempt

57 LSJ napaSiSropi I,4. Contra Mansfeld-Runia, Aetiana, II, 62. 18. Correct explanation is given by Lachenaud (1993: 191, note 1).

58 In Lebedev (1988: 6) a hypothesis was proposed that the extant P may be based on student's notes made ano ^rovqq and that this student may have been young Plutarch himself. Because it was written by Plutarch's hand, after his death the text of notes was included into Lamprias' catalogue. Charlotte Schubert develops this thesis with more detail and precision in Schubert (to appear). Ps.Galenus in the Preface to his compendium makes it clear that he used two classes of sources: what he has «heard» from his teachers (i.e. lectures) and what he has «learned from readings», i.e. from handbooks: Hist.Philos. 2 oca napa xrov nsnaiSsuKoxrov qpac aK^Koapsv Kai Si' avayvroapaxrov spaOopsv, vuv auvaysiv snixsipqaopsv.

to reconsider the traditional view (since Diels) that the chapter 12 of the Didaskalikos is based on Arius Didymus (fr. 1 Diels): according to Goransson it is Arius who depends on the text of Didaskalikos, and since the date of the latter is uncertain, the doxographer Arius cannot be dated more precisely than between the second half of the 1st century B.C. and third century A.D. (Goransson 1995: 196 ff.). Both these claims are ill-founded, and the bold assertion that Arius Didymus is «beyond doubt» the borrower can be easily refuted by the linguistic analysis and the statistics of usage of divergent keywords in the passage on Platonic ideas that the Swedish scholar ignores. The philosophical language of Arius Didymus is Hellenistic with classical antecedents, precise and technical, the language of Didaskalikos is coloured by mysticism and comes close to the Neoplatonic usage. On the other hand Goransson's criticism of the Diels' reconstruction of the hypothetical Epitome of Arius Didymus is wholly justified.

One of the main arguments of Goransson against the identity of Arius the doxographer and Arius the court philosopher is that (allegedly) the court philosopher is always called Arius (but not Didymus), and the doxographer either Didymus or Arius Didymus, but never just Arius (Goransson 1996: 211 ff.). This is incorrect: Tertullian calls the doxographer just Arius without addition of Didymus59. One should also take into account the different types of sources. Reports on Arius' friendship with Augustus come from historical sources: no doubt in private life Augustus called his friend by his personal name. But the official full double name was used as author's name in the writings of Areios Didymus to avoid confusion with other writers called Areios.

The first and general objection that can be made against Goransson's second claim (i. e. that Arius borrows from Alkinoos, not vice versa) is that of two similar texts a grammatically superior, philosophically more sophisticated and longer text (i. e. the text of Arius) cannot derive from an inferior (corrupt), less sophisticated and a shorter one.

1) In Didaskalikos 12.166.39 xrôv rata ^ûciv aioB^trov Kai Kaxa pépoç rôpicpéva Tiva Seî napaSeiypaxa eîvai xàç iSéaç the

59 Tertullian, De anima 54.2, 73. 1-2 Itaque apud illum (scil. Platonem) in aetherem sublimantur animae sapientes, apud Arium in aerem, apud Stoicos sub linam. 5.4, 74.-4-6 Sed in aetherem dormitio nostra cum puerariis Platonis aut in aere cum Ario aut circa lunam cum Endymionibus Stoicorum? This has been rightly pointed out by Brad Inwood (1995) in his review of Goransson.

position of Kaxa ^epo; creates a syntactical ambiguity, Arius Didymus has a better reading Kaxa yevo; and syntactically smooth text fr. 1D xrov Kaxa 9uoiv aioBnxrov Kaxa yevo; ropio^eva xiva napaSeiy^axa 9a^evo; elvai xa; iSea;.

2) Alcinoos 12.166.25-26 has a corrupt text xq; iSea; otion; aixia; ap%q (PV) xofi elvai eKaoxov xoiofixov oiov atixq unap%ei. Goransson follows the implausible emendation apxqv (adverb) of Hermann and Louis (Goransson 1995: 107, n. 7), Whittaker corrects to aixia; <Kai> ap%q; in conformity with Arius who has a perfect and clear text Arius Didymus fr. 1 Diels elvai Se xqv iSeav ai'Siov otioiav, aixiav Kai apxqv xofi eKaoxov elvai xoiofixov oi'a eoxiv atixq. The words ai'Siov otioiav, essential for the definition of Platonic idea, are apparently omitted by Alkinoos, and not «inserted» by Arius contrary to Goransson's claim (Goransson 1995: 201).

3) In Arius the same definition of idea is followed by the explanation omitted in Didaskalikos: KaBanep otiv xa; Kaxa ^epo; roonep apxexuna xrov aioBnxrov nponyeioBai oro^axrov, ouxro; ktA. The term nponyeioBai is used here in the same technical ontological sense of antecedent, first order being which is also attested in the Pythagorean formula of moral telos quoted by Stobaeus in the ethical doxography (A) in Book 2 and which also is paralleled in Didaskalikos. The comparison of these two passages is revealing as to how Alkinoos rewrites Hellenistic doxography in late Platonic language:

Stob. 2.7 (p. 49, 16-18 Hense) nuBayopav Se nap'atixov (sc. "O^npov) eineiv 'Enou Beroi- SqAov ro; ofi% opaxroi Kai nponyou^evroi, vonxroi Se Kai xq; Koopxq; etixa^ia; ap^oviKroi.

Didaskalikos 28 (p. 181, 43-45 Whittaker) xo xeAo; elvai av xo e^o^oiroBqvai Beroi, Beroi Sn^ovoxi xroi enoupaviroi, ^q xroi ^a Aia unepoupaviroi, o; otiK apexqv e%ei, a^eivrov S'eoxi xauxn;. Gorans-son wrongly denies that these two texts should be taken as close parallels60 and seriously mistranslates nponyou^evo; as «leading the way» (Goransson 1995: 193) which is by no means the «basic sense» of the term in logical, ontological and methodological contexts. The «antecedent» or the «preceding» god is the supreme god, the first order god of the Middle Platonists distinguished from the inferior demiourgos that «follows» him. Moral action cannot be modelled on this supreme god because he does not partake in any praxis or activity and therefore «surpasses virtue». In Stobaeus' text

60 And for no good reason reproaches Guista for «careless reading» of which he is guilty himself: Goransson 1995: 191.

it is tempting to read oti vo^xroi Kai npo^youpevroi, opaxroi Se kxL Alkinoos grasps the basic thought of the original correctly, but he replaces the technical term npo^youpevroi with a more poetic and popular unepoupaviroi. Although this is not a corruption, but an intentional rephrasing, in a sense Stobaeus' texts gives a lectio difficilior. Before the Christian era unepoupaviog is only once attested in Plato's Phaedrus 247c, there are some instances in Philo Alexandrinus and Middle Platonists, but the bulk of evidence (629 instances in TLG online) comes from the Neoplatonists (102 instances only in Proclus!) and Church fathers (especially Origenes, Gregory of Nyssa, Didymus Caecus et al.). Although of Plato's coinage, by the frequency of use it is predominantly a Neoplatonic and Christian term. On the contrary npo^youpevog (and adverb npo^youpevrog) as a logical and ontological term for «antecedent» or «prior», i.e. for «first» causes etc. is Hellenistic (forshadowed already in Theophrastus) and restricted to highly specialized philosophical and medical contexts61. It is worth noting that in his compressed version of Arius Didymus fr. 1 Diels (on Platonic ideas) Alkinoos also omits the term nponyeio0ai in the sense of transcendental reality. The omission of an earlier and more technical term and its replacement by a later and more mystical one leaves no doubt as to who is the borrower. This can be used both as additional evidence in support of the attribution of Stobaeus' ethical doxo-graphy (A) to Arius Didymus and as additional refutation of Goransson's claims about the relation between Arius Didymus and Alkinoos.

9. Evidence for Arius Didymus as the author (compiler) of the SP-Placita

As we have pointed out from the start no ancient, Byzantine, Arabic or any other source ever quotes any certain passage or a concrete doxa from «Aetius». In all three occurences of this name in Theodoret's CAG «Aetius» is only included in a kind of general «bibliography» of doxographical sources (allegedly) used by the

61 Theophr. De igne, 14 nponyoupsvroq opp. KaTa aupPsPnKoq. nponyou-psvroq explained as nproTroq by Galenus, De dignos. pulsibus v. 8, 846, 16 K. There are 9 instances of nponyoupsvov, nponyoupsvroc in Didaskalikos, opposed to snopsvroc and used as value term (syn. Tipiov 2.2, syn. apiaTa 2.3, syn. ouairoSsq 32.7) or something of higher order (virtues of the rational part of the soul are nponyoupsvai compared with the virtues of the irrational part: 30.3) or more elementary (aTspsrov nponysvrai Ta sninsSa, syn. npoyova 13.2), see also 3.2, 4.1, 4.4, 17.1 (in temporal sense).

Christian apologist, but it is not clear what exactly is quoted from «Aëtius». In the case of Arius Didymus we have at least two direct quotations from a doxographical work under his name that clearly and unambiguosly cite him by individual name as a source of identifiable doxai from SP-Placita.

1) The lemma ArSu^ou (SrSu^. P : ArSu^ro F) in Stobaeus' chapter on necessity has been discussed above in section 6. One might object to this that the authorial definitions in SP-Placita as a rule have no lemma (name-label), exactly because they are authorial. But in this case we have a lemma added by Stobaeus to indicate the source of his excerpt, and not a lemma attached to a doxa that stood in the original text of SP-Placita quoted by Stobaeus. It indicates the source not only of the authorial definition, but also of the placita of philosophers (Pythagoras, Parmenides, Democritus, Leucippus etc.) that follow after the authorial definition. Even if the lemma indicates the source of Stobaeus' excerpt from SP-Placita, the author of the authorial definition will be the same as the author of SP-Placita. This lemma is very similar to the well Known lemma 'Ek xqç ArSu^ou émxo^qç in Stob. IV (p. 918, 15 Hense) which Stobaeus (and not his source) attaches to his excerpt from Arius' ethical doxography on the origin of happiness. It is the famous lemma that Wachmuth, following the suggestion of Meineke, restored in the beginning of the doxography (A) in Stobaeus' second book.

A second possible quotation from SP-Placita with Arius as author's name is preserved in Tertullian's De anima. In chapter 54 he quotes a collection of different opinions of philosophers on the immortality of the soul and afterlife. He mentions Pythagoras, Plato, Empedocles, the Stoics and Arius62. It is possible that in his source «apud Arium» was a reference to the doxographical compendium in which the opinions of Pythagoras, Plato, Empedcles and the Stoics were cited, but he took Arius as one these philosophers, and not as a source of their opinions. We may compare Tertullian's passage with

62 Tertullian. De anima 54.1 Omnes ferme philosophi, qui immortalitatem animae, qualiterqualiter volunt, tamen vindicant, ut Pythagoras, ut Empe-docles, ut Plato, quique aliquod illi tempus indulgent ab excessu usque in conflagrationem universitatis, ut Stoici, suas solas, id est sapientium, animas in supernis mansionibus collocant. 2. Plato quidem non temere philosophorum animabus haec praestat, sed eorum qui philosophiam scilicet exornaverint amore puerorum. Adeo etiam in philosophos magnum habet privilegium impuritas. Itaque apud illum in aetherem sublimantur animae sapientes, apud Arium in aerem, apud Stoicos sub lunam.

the chapter «On the immortality of soul» in Ps.Plutarch where we have the same sequence of names nu0ayopag n^axrov SxroiKol.

Cf. Ps.Plut. De Placitis 899C (p. 149 Lachenaud) = SP-Placita IV.7 Diels

nepi a90apoiag yu%fg.

nu0ayopag n^axrov a90apxov elvai xfv yu%fv... <'HpaK^eixog Kai> oi SxroiKoi e^iofioav eK xrov oropaxrov xfv pev ao0eveoxepav 90eipeo0ai ** apa xotg ouyKpipaoi <Kai> oKiSvao0ai ***, xaux^v S' elvai xrov anaiSeuxrov xfv S' io%upoxepav, oi'a eoxi nepi xoug oo9oug, <emSiapeveiv> Kai pe%pi xfg eKnuprooerog63.

<'HpaK^eixog Kai> supplevi || xfv pev ao0eveoxepav 90eipeo0ai scripsi : uno9epeo0ai xfv pev ao0eveoxepav codd. || <Kai> oKiSvao0ai scripsi : y(veo0ai codd.

The passage of Tertullian also displays verbal coincidences64 with the doxography of Stoic view of the soul quoted by Eusebius ano xrov Enixoprov Apeiou AiSupou

Elvai Se yu%fv ev xro o^ro 9aoiv, o Ka^ofioiv ai0epa Kai aepa KuK^ro fnepi xfv yfv Kai 0a^aooav Kai eK xouxrov ava0upiaoeig-xag Se ^oinag yu%ag npoone9uKevai xaux^, ooai xe ev Z®oig eioi Kai ooai ev xro nepie%ovxv Siapeveiv yap eKet xag xrov ano0avovxrov yu%ag. evioi Se xfv pev xofi o^ou aiSiov, xag Se ^oinag oup-piyvuo0ai eni xe^euxf eig eKeiv^v. e%eiv Se naoav yu%fv fyepoviKov xi ev auxf, o Sf Zrof Kai a'io0noig eoxi Kai oppf. Kai exi pex' o^iya-Tfv Se yu%fv yevnxfv xe Kai 90apxfv ^eyouoiv. otiK eti0ug Se xofi oropaxog ana^ayeioav 90eipeo0ai, aW enipeveiv xivag Xpovoug Ka0' eauxfv xfv pev xrov onouSairov pe%pi xfg eig nfip avaAuoerog xrov navxrov, xfv Se xrov a9povrov npog noooug xivag Xpovoug. xo Se Siapeveiv xag yu%ag ouxrog Aeyouoiv, a9povrov npog noooug xivag xpovoug. xo Se Siapeveiv xag yu%ag ouxrog Aeyouoiv, oxi Siapevopev fpetg yu%ai yevopevoi xofi oropaxog %ropio0evxeg Kai eig e^axxro pexaPa^ovxeg otioiav xfv xfg yu%fg, xag Se xrov a9povrov Kai a^oyrov Zrorov yu%ag ouvano^uo0ai xotg oropaoi65.

63 (a) Olympiodorus, Comm. in Plat. Phaedonem, 10.2; p. 139 Westerink. Tpirq ôô^a f Aéyouaa xfv pèv ànaiÔEUxov yuyfv é^voûaav xoû arapaxoç EÙ0ÙÇ 90EÎpEa0av, xfv 5è nEnaiÔEupévnv axopra0EÎaav xaîç àpExaîç émpévEiv <péxpt> xfv ÊKnupraaiv xoû navxoç коарои, fç ôô^nç fv Kai 'HpaK^Evroç [<péxpi> addidi].

64 Cf. aetherem ... aerem in Tertullian with о Ka^oûaw ai0épa Kai àépa in Arius.

65 Arius Didymus fr. 40 Diels.

Apart from these quotations additional evidence for Arius as the compiler of SP-Placita is provided by the striking similarities in method, purpose, style, philosophical language and phraseology between physical doxography (SP-Placita) and ethical doxography in Stobaeus Book II, chapter 7. We cannot discuss here at length the controversy about the authorship of three doxographies; our investigation compells us to agree with those scholars (like Giusta and Hahm) who regard Arius Didymus as the single author of all three ethical doxographies: the first (thematic) on diairesis of the ethical topos (A), the second on Stoic ethics (B) and the third on Peripatetic ethics (C). There are some similar features even between the extant remains of SP-Placita and Stobaeus' ethical doxography (hence quoted as SED-A, SED-B and SED-C), but the similarity becomes more striking if we compare SED-A with P+ . The language and style of the authorial definitions, of diairesis of subjects, of the basic summaries (kephalaia), of the transitions is very similar. The most striking common feature, however, is the emphasis laid on the Aristotelian categories as methodological basis of the adopted diairesis.

10. The so called Epitome of Arius Didymus as reconstructed by Diels in Dox.Gr. could never exist

Only the magic of the «great name» of Diels can explain the fact that so many scholars took seriously Diels' reconstruction of the alleged doxographical work by Arius Didymus with a title Epitome. The existence of such original book title is no more plausible than the existence of a book title pip^iov or ouyypa^a. Epitome is not a title of a book, but a designation of type or genre of text, it can be used instead of the complete title only as an abbreviated reference to the complete title mentioned before, like 'Emxo^q xrov apeoKovxrov (e.g. nAaxrovi). The attested pluralis emxo^ai refers to the doxographical sections of different chapters in nepi aipeoerov dedicated to individual philosophers or schools. Diels' reconstruction of the so called Epitome has been subjected to critical exmination and revision by Mansfeld and Runia, and Jean-Baptiste Gourinat66.

Diels wishfully rejected the attested title of Arius' opus magnum nepi aipeoerov. Instead he printed the selected fragments of physical doxography from Stobaeus relating to to Plato, Aristotle and the Stoics under another fictitious title invented ad hoc: EK THN

66 Mansfeld, Runia 1997: 238-265; Runia 1996; Gourinat 2011. On ethical doxography (B) see also Viano 2005.

APEIOY AIAYMOY EniTOMHS OYSIK^N (Dox. Gr. 447). This title is nowhere attested, it is another Spielerei. And what is worse, it is composed in faulty Greek: to be meaningful and intelligible as a title of a doxographical work it requires something like émxo^q xrov ^uoiKrôv Soy^axrov or 'Emxo^q xôv apeoKovxrov n^axrovi, Apioxoxé^ei Kai SxroïKoîç ^uoiKrôv Soy^axrov. As it stands it could only mean emxo^q of a work titled (Ta) OuoiKa, i.e a brief exposition of physics, but this does not square with Diels' own reconstruction of the so called Epitome.

11. The relation between Arius' diachronic nspi aipéasrav and his diaeretic (problem-oriented) handbooks of physics and ethics

with areskonta

If our attribution of P+ (= SP-Placita) quoted in Stobaeus' book I to Arius Didymys is correct (as we believe it is since it is based on direct quotations) and if we accept that the author of all ethical doxography (and not only of the C which is attested by direct quotation) in Stobaeus book II also derives from Arius Didymus, it turns out that Stobaeus used in Eclogae I—II as his main source of philosophical placita two «diaeretic» handbooks by Arius Didymus written not as diachronic histories of philosophy, but as problem-oriented expositions of physical scince (P+ in Stobaeus book 1), and of ethics in book 2 (hence forward referred to as SED = Stobaeus' Ethical Doxography). The subject of the first handbook is Aoyoç ^uoiKoç, that of the second is Aoyoç q0iKÔç (or xo q0iKov eiSoç xqç ^i^ooo^iaç), both are parts of a Siaipeoiç xoù Kaxa ^i^ooo^iav Aoyou, a kind of general philosophical encyclopedia. Stobaeus cites in the beginning of book II also a third work of Arius Didymus titled nepi aipeoerov. This is a title of a regular diachronic history of philosophy also known as AiaSo%ai. The question of the relation between Arius' history of philosophy and two theoretical introductions to physics and ethics is complicated by the fact that the evidence on the first is scanty. The nepi aipeoerov, like the ten books of Diogenes Laertius, no doubt combined biography with doxography. The doxography in the historical work, however, was presented not in the form of concise basic summaries (kephalaia) dispersed through many chapters according to the subject, but as a continuous exposition of the physical and of the ethical theory either of the whole school (like the exposition of the Stoic doctrine in DL VII) or of a certain philosopher. In all probability this continuos and coherent exposition of the doctrine followed after the introductory

biographical section, as is the case in the work of Diogenes Laertius. The most plausible scenario seems to be this. The two handbooks were relatively compact works. The nepi aipeoerov was a grand-scale work, presumably the Opus Magnum of Arius. Stobaeus quoted doxography both from the big historial work (continuous) and from the compact handbooks (short placita). This explains the existence of the long and short versions of similar doxography. It follows that the so called «coalescence» of several placita into one continuous text in Stobaeus (as Diels and his followers assume) in fact preserves the original text quoted by Stobaeus from the grand-scale work nepi aipeoerov. Whether the continuous physical doxo-graphy was split into short kephalaia (either verbatim quotations or basic summaries) by someone else or by Arius Didymus himself depends on how we solve the problem of the authorship of P+. If we accept the authorship of Arius Didymus (and there is no alternative to this based on direct evidence), then Arius did the job himself.

12. The new Stemma Doxographicum

The stemma doxographicum attached to this section is essential one, it includes only the key-texts and its purpose is to delineate the main lines of the origin and transmission of the Placita philoso-phorum tradition. Note that it does not take into account the relative chronology of different sources. One should compare it with the stemma in Mansfeld and Runia, Aetiana, I, 328 which essentially follows Diels (ibidem, 81).

The following sigla are used in our stemma:

SP-Placita - the hypothetical common source of Stobaeus and Ps.Plutarch (corresponds to Diels' «Aetius», but Diels' synoptic text in Dox.Gr. is cited as «Plac. Diels»). P+ - The «better Plutarch», the original diaeretic handbook of physical

phlosophy that combined didactic authorial material with doxography. In the stemma I assume that SP-Placita and P+ are identical, but some

doubts remain (see section 5 above). P - the extant text of Ps.Plutarch's Placita philosophorum. AD - Arius Didymus, C - Cyril, E - Eusebius, G - Ps.Galen, Hist.Philos., N - Nemesius, Q - Qusta Ibn Luca (Arabic translator of P, 9th century), S - Stobaeus, T - Theodoret. SED - Stobaeus' Ethical Doxography in Anthol. II 7. This includes SED-A (diaeretic handbook of ethics), SED-B (Stoic ethics, continuous doxography) and SED-C (Peripatetic ethics, continuous doxography). SPhD - Stobaeus' Physical doxography in Anthol. I (Plato, Aristotle and Stoics) - roughly equivalent to Diels' Epitome of Arius.

Broken line denotes a secondary source, dotted line a contamination, a line marked by a small circle means «pertains to a group» rather than «derives from a source».

Academic Stoic dox. Original works of Plato,

Placita (Posidonius) Aristotle & Stoics

AD AD

Diaeretic Itcpi aipÉOEtdv

Lebedev. Stemma dox. June 2016.

Appendix 1. Porphyrius as intermediate doxographical source between SP-Placita (=AD), on the one hand, and N, T (occasionally) and S (occasionally), on the other

Apart from the definition of ananke discussed above, there is only one (to our knowledge) another instance of a name-label attached to an authorial definition from SP-Placita. This time it is the name of Porphyrius, not of Arius Didymus. In Wachsmuth's edition this text is printed like follows: Stob. I.15.3 (p. 145, 3-6 W.)

3a. nop^upiou.

3b. eoxiv em^aveia Kai nepiypa^q Kai nepa^ oro^axo^.

Both P (Parisinus 2119) and codex Farnesinus attach the lemma npo9upiou to this definition of shape67. Wachsmuth, as always following Diels, regards this as «empty» lemma followed by alleged lacuna, and therefore separates the lemma (S 1.15.3) from the definition (S 1.15.3b). But let us believe our evidence, i.e. the consenus of the two best manuscripts of Stobaeus. Rather than postulating a lacuna or a confusion of lemmata by Stobaeus it is safer to admit that Stobaeus in this particular case quotes SP-Placita second-hand from Porphyrius. Which means that Porphyrius may well have been a neglected intermediate source between SP-Placita (Arius Didymus in our view) on the one hand, and some late doxographical sources, on the other. Nemesius and Theodoret first come into consideration. In the case of Nemesius this is virtually certain since he does not quote Plutarch or «Aetius» as his source of placita, but he does quote in proximity with P-style placita Porphyrius who is in general his main source for the philosophical doctrines of the Hellenes. From the 4th century on Porphyrius was the main opponent of the Christian apologists, so they were so to speak obliged ex professo to know his works in order to refute him as the main defender of paganism. Dorrie has convincingly argued that Porphyrius' Symmikta zetemata contained doxography and was one of the main sources of Nemesius' De natura hominis (Dorrie 1959: 18 ff., 111 ff.). Another work of Pophyrius that contained doxography was Historiaphilosophos. Contrary to Diels' implausibe hypothesis, Theodoret is not lying when he quotes Porphyrius as one of his three doxographical sources along with Plutarch and «Aetius» (misquoted Areios). A long series of various placita on the nature of the soul in Theodoret quoted from SP-Placita is supplemented in CAG V.18 by two «additional» doxai of Empedocles (mixture of aetherial and aerial substances) and Critias (blood). Diels thought they come from «Aetius'' (Dox.Gr. 389; Placit. IV, 3, 13-14). Since they are not found in S and P, but are found in Nemesius (only Critias) whose source is Porphyrius, as well as in Macrobius (both Empedocles and Critias), Porphyrius' Historia philosophos (explicitly quoted by Theodoret as one of his doxographical sources for this passage) seems to be a more likely source68.

67 See the page 42v at: http://gallica.bnf.fr/ark:/12148/btv1b52501705k/ f8 8.item.r=stobaeus.zoom

68 Nemesius, NH V.18 (p. 127, 12-13 Morani); Macrob. In Somn. Scip. I, 14, 19-20; On this see Mansfeld 1990: 3073 ff.

Sometimes T interweaves with standrad P-type placita additional doxographical material from a source of SiaSo%ai type69 . A remarkable example is provided by the doxography on the nature of the cosmos and matter in CAG IV. 5-1470. Any attempt to identify T's source in this case has to start from Porphyrius as the most plausible candidate. His Historia philosophos fits the bill: 1) it is a work of SiaSo%ai type combining doxography with biography; 2) virtually all additional material71 in CAG IV relates to Preplatonic authors and thus complies with the chronological limits of Hist.Phil. 3) Porphyrius' Hist.Phil. is explicitly quoted both in the preceding paragraphs CAG IV, 1-2 and after the doxographical excerpts in IV, 31 together with the names of Plutarch and «Aetius» as a source of the preceding placita. Incidentally, this provides additional confirmation that Theodoret is not lying about Pophyrius as one of his doxographical sources and eo ipso once again undermines and refutes Diels' Aetius hypothesis.

Appendix 2. The riddle of AIAYMOE ATHIOE (?) in Suda.

In Adler's edition of the lexicon «Suda» A 871 we read:

AiSu^o^ Axqio^ q 'Аххю^ %pn^axioa<;, фг^ооофо^ AKaSn^aüKO^. nröavrov Kai оoфшцaxюv Айоец ev ßrßAior^ ß' Kai aAAa noAAa.

Nothing is known about this Academic philosopher from any other source. One poossible explanation is the following. Although Diogenes Laertius affiliates Arius Didymus with the Stoics, the author of Stobaeus' Ethical Doxography (A) in Stob. II.7, most probably Arius Didymus, quotes as his predecessors with approval two academic philosophers, Philo of Larissa and Eudorus of Alexandria. So, on the basis of his own writings Arius Didymus could be regarded by some as Academic philosopher. The works of Theodoret were held in high esteem in Byzantium, the compilers of

69 Mansfeld, Runia 1997: 281. For M&R «additional» in this case means «added to SP-Placita», for us it means «added to P».

70 In a recent stimulating discussion of this passage Gérard Journée (2015) opts for an intermediate stage of the SP-Placita tradition accessible to T, but not to P and S. This violates the principle of economy «Placita praeter necessitatem...».

71 The only exception is Epicurus. The complete name 'EnÎKoupoç Nso-kAeouç AOnvaioç is found in P, the words né^nx^v ysvsâi psxa AqpoKpixov remain a puzzle. It is conceivable that Porphyrius mentioned Epicurus in the chapter on Democritus as one his famous historical successors, just as D.L. concludes his chapters with ysyôvaai paOnxai aùxoû kxA. In CAG IV. 11-14 the complete names from Porhyrius are mixed with P-Placita.

the lexicon «Suda» quote his works many times. The «Index auctorum» in Adler's edition lists more that 200 loci from Theodoret 's works72. No doubt they also knew Eusebius' PE. Suppose that they noticed a remarkable discrepancy between Eusebius' PE and Theodoret's Curatio: both apparently quote the same or similar doxographical excerpts, but in Eusebius the names of the doxographers are Plutarch, Porphyrius and APEIOS AIAYMOS, and in Theodoret they are Plutarch, Porphyrius and AETIOS. They may have marked this discrepancy in the entry on Didymos, which, one may guess, in its original form before corrpution could read like this:

AiSupog Apfiog (= 'Apeiog) f Aexiog... 73

The combination of names AiSupog Apeiog could sound disturbing if not scandalous to the ears of devout since Apeiog was used sometimes in the sense of «Arian heretic» and AiSupog (the Blind) was the name the most ardent opponent of the Arian heresy.

Appendix 3. A reply to M&R's criticism

Jaap Mansfeld and David Runia have produced two sets of criticism of my work on «Aetius» and Diels' theory of doxography. The first was published in the «Appendix» to the Aetiana volume I (1997) and deals with my 1988 paper (written in 1983). I agree with the following objections and remarks in this set: 1) the reading of all MSS in Eusebius PE XIV, 16,1 (p. 299, 19 Mras) is Aiayopag o MiAfoiog; 2) I overlooked that in his 1881 paper Diels recognized a limited contamination of S from P established by Elter (Diels 1881: 343-350)74. But these corrections concern minor points and do not affect my main arguments at all; unfortunately M&R keep silence on the statistics of the use of Kai pevxoi Kai in T which plays decisive role in the rejection of Diels' Aetius hypothesis. Most of the other objections are answered, I hope, in the present paper. Faced with the hard to refute statictics, M&R admit that my intetrpretation of Kai pevxoi Kai as meaning simply «and also», «as

72 Suidae Lexicon. ed. Ada Adler, Pars V: Praefationem Indices disseratitionem continens. München; Leipzig: 2001, 85-86.

73 Even if the transmitted text of Suda is correct, this does not affect our «hard» arguments against Diels' Aëtius' hypothesis.

74 In fact I meant something more important than limited textual contamination during medieval transmission. I meant the possibility that Stobaeus himself used selectively both P and its source (AD) in which case SP-Placita similar to P may have never existed. But this subject requires a special treatment and I cannot address it here.

well as» rather than «and especially» (as Diels claimed) is correct, they recur to an argument that looks like a dialectical tour de force: they propose to interpret the non-emphatic meaning of Kai pevxoi Kai as Theodoret's intentional attempt to «conceal rather than draw attention to his main doxographical source» (p. 335)! In support of such interesting conjecture they quote a (what they believe to be) parallel from Clemens Alexandrinus Strom. 2.78-100 where Clemens quotes his source (Philo Alex.) after a long excerpt only at 100.3 to conceal his plagiarism. We do not find this example particularly «illuminating». To begin with, «parallels» of this type from another writer prove nothing: Theodoret is not Clemens Alexandrinus. But this particular parallel is not a parallel at all because we have here not only two different writers, but also two very different contexts. Clemens does not emphasize the name of his main source because he tries to conceal plagiarism (according to M&R) and because his aim is to emphasize his own originality. Theodoret in this passage does not claim originality at all: on the contrary, he emphasizes that to avoid suspicions that he exaggerates the diaphonia between Hellenes on the nature of man, i.e. falsifies the evidence, he will quote pagan authorities thus proving his objectivity and impartiality.

From the second set of M&R's critical comments75 I also accept with gratitude some helpful remarks and corrections on minor points, but I do not find in them convincing alternative answers to the hard questions about Diels' Aetius hypothesis raised in my Melbourne paper (2015). Below I reply to those comments that have not been already answered in the preceding pages76.

On section 2.1: In this work T is an apologist..., not a preacher] In the very title of Theodoret's work @ЕражЕтгщ r&v 'ЕХХцук&у жавщахюл> the «Hellenic diseases» stand for different pagan «heresies» (both mythologies and philosophical schools) that inflict the souls of the pagans, Theodoret presents himself as a doctor, and the «therapeia» consists in the conversion of hellenizontes to Christianity and the salvation of their souls77. The purpose of T's work is not theoretical, but practical, the йрец constantly addressed by him, are not intellectuals, but simple people of his diocese (of low class and low culture as it seems) who still remain unconverted.

75 Mansfeld, Runia (2016), Critique of Lebedev's Melbourne paper (2015).

76 The verbatim quotations from M&R's comments are set in italic.

77 CAG Proem, p. 421-22 Raeder Бую psv ouv xfq xe xrav voaouvxrav svsKa 0EpanEiaq Kai xfq xrav uyiaivovxrav npopnOoupEvoq юфЕ^а^ xouxov avESE^apnv xov novov kxL

He feels compassion for these les misérables and calls them ф1Аог, but he is adamant in his hatred of Porphyrius. What we call the 12 «books» of the CAG, Theodoret himself calls SiaAé^eiç, a term which means «speech, conversation»78 and comes close to a homilia, exhortation or sermon. It seems likely that the 12 dialexeis were originally delivered as sermons or public lectures79.

On 2.3: What are the additional placita: they are of course what is not covered by the SP-Placita precisely because they are ST-Placita] By «additional» placita I mean the placita that Theodoret adds to quotations from the extant P from a different source, most probably Porphyrius (see Appendix 1). An alternative to this could only be a better (than extant P) manuscript of P.

On 4: No attempt is made to understand what the role of such definitions might be in a Placita type work. No account is taken of what Mansfeld writes in his discussions in Aëtiana vol. 2.1.] This remark takes me aback. Authorial definitions are not discussed at all in Aëtiana vol.2/1, this important topic has been overlooked or rather omitted by M&R since I pointed out to the importance of definitions already in my 1988 paper. And it is easy to see why: any elementary handbook, whether ancient or modern, contains apodictic definitions of the key notions and summary expositions of main theories of a science that it teaches. Our hypothesis that the original SP-Placita was a handbook of physical science explains «the role of such definitions», but for M&R's hypothesis of the SP-Placita as a dialectical collection of opinions the presence of such definitions in the text is apparently a stumbling block and a challenge to their theory.

ibidem: ...a work beginning each chapter with a definition is simply a chimera or a pure speculation. Would it be even possible to commence every chapter with a definition?] I agree that it would be risky to postulate a lost definition in every chapter, my thesis is that there were much more authorial definitions, as well as other expositions of authorial views (see the section 4 above). But the rhetorical question of M&R can be answered in affirmative way. M&R overlook the case of Stobaeus' ethical doxography in II.7: each chapter of this handbook of ethics starts with authorial definitions of the key terms and only then proceeds to the exposition of different doxai with name-labels. This similarity is not accidental:

78 Lampe, A Greek Patristic Lexicon, p. 356, s.v. SiaAs^i; 1. In modern Greek this word means lecture, whether for students or general public.

79 Initial oral performance is admitted by Scholten (2014: 53-54).

the author of both the handbook of ethics in Stobaeus' book II and handbook of physics in book I is one and the same, Arius Didymus. A similar procedure is followed by Nemesius' in NH: chapters 5, 6, 7, 12, 13 (first comes definition and then doxai), 3, 4, 5 (first comes theoretical discussion of the problem and then doxai) etc.

On 6: It is very possible that the origin of these compact definitions was indeed in a hermetic work] It is impossible and ruled out with 100% certainty for the following reasons: 1) Stobaeus never quotes two doxai of Hermes in the same chapter; Hermetic text (from a dialogue) is quoted in the same chapter «On necessity» as the krisis bebaia definition. 2) Excerpts from Hermes in S normally are quoted together with Platonic/Neoplatonic excerpts after short Placita, the krisis bebaia definition precedes Placita quoations and is a part of them. 3) The terminology is Chrysippean and the theory of identity ofpronoia and ananke, is a typically Stoic doctrine.

On 8: This section on Goransson is largely superfluous. One paragraph or even a footnote would be sufficient]. In their Aetiana M&R took the arguments of Goransson on the identity of Arius very seriously and pointed to me that these arguments allegedly invalidate my attribution of SP-Placita to Arius the court philosopher (M&R, Aetiana, 241 ff., 336). Now that I have fulfilled their demand and removed this particular objection, M&R still express their dissatisfaction! The section 8 is not superfluous since scholars still continue to quote Goransson's ill-founded claims about the relation between Arius the doxographer and Alcinoos.

On 11: nspi aipsasmv and AiaSoxai are generally regarded as separate Hellenistic genres] Both in Aetiana and at the discussion of my paper in Melbourne it was pointed to me that nepi aipeoerov, unlike AiaSo%ai, allegedly discussed only Post-Socratic schools. Arius Didymus had a different opinion: in his nepi aipeoerov he discussed the Seven sages, the Pythagoreans and Xenophanes, as well as Plato, Aristotle and the Stoics, in other words this was a general history of Greek philosophy that covered all philosophical schools. Since this fact was a stumbling block to Diels' forced separation of Arius Didymus from Aetius in Stobaeus, Diels wishfully denied that the quotation from Didymus in Clemens come from Arius Didymus. But let us believe our evidence: Xenophanes (quoted by Srobaeus, not by Clemens!) in any case cannot be thrown out of the game, and this fact alone is sufficient to reject Diels' reconstruction of the so called Epitome allegedly treating only Plato, Aristotle and the Stoics.

On 12: In fact the PST-Placita cannot be anything else than A(etius), whom M&R are attempting to reconstruct and edit.] In their comments M&R repeatedly use the uncommon designation PST-Placita (and once ST-Placita). I find these designations potentially misleading. A convergence of any two (or more) doxographical sources (say, A and T) should not be quoted as ST-Placita (remember the rule «Placita ptaer necessitatem...»), it should be quoted just as a convergence of S and T. The source of virtually all short placita in T is P, not SP-Placita. A couple of «additional» doxai not found in S or P are quoted from Porphyrius (see Appendix 1 above).

Appendix 4. Some remarks on Mansfeld and Runia's Aëtiana

There is no doubt that the 3 volumes of the Aëtiana by Jaap Mansfeld and David Runia constitute a remarkable scholarly achievement and that anyone interested in the subject will find their work indispensable for years to come. I have already signalized some of its virtues in the preface. But now let me also point to what I find questionable and misleading in this work.

1) Uncritical acceptance of Diels' Aëtius hypothesis and failure to convincingly answer the criticism of Lebedev, Frede, Bremmer and other critics of Diels. An attempt to downpaly the importance of doxographer's name is hardly convincing8 . The answer to the question «What's in a name?» depends on what kind of name it is. The name of Aëtius is indeed just an empty name, it gives us no cue to the time, location and the affiliation of the author. If we replace the name of Aëtius ad libitum with Manetius, Pupetius or Dilsetius nothing will change, one empty name is as worthless as another. But if the name belongs to a well known philosopher, as is the case with Arius Didymus, the mere knowledge of his name allows us to date exactly his work to the second half of the first century A.D.81, to determine its historical-cultural context - the melting pot of philosophical ideas in the 1st century B.C. Alexandria - and to

80 cf. Runia, «Aetius, or what's in a name» (2009); Mansfeld and Runia, Aetiana II/1, p.3 «Whether or not he [=Diels] was also right in attributing it to Aetius (we believe he was) is irrelevant from the point of view of this successful restoration». It is not irrelevant: when the real name of the author and the purpose of his work are ignored, the restoration cannot be successful. M&R themselves correctly criticize Diels for the elimination of authorial remarks , chapter numbering etc.

81 Diels' date of SP-Placita (Aetius) c. 100 AD lacks any serious ground. Mansfeld and Runia, Aetiana, I, 328 follow Diels without explaining their reasons. Dilesius dixit.

establish his philosophical affiliation. Arius Didymus was a Stoic, not a dogmatic orthodox Stoic of the Hellenistic era, but a new-style Stoic of the Post-Hellenistic period, with a deep respect for all «ancients» (Stoics, Plato and Aristotle), working in tradition of Platonizing Stoicism of Antiochus and having close relations with contemporary Academics of the Neo-Pythagorizing trend (Eudorus). This portrait matches almost exactly the general tendencies of the SP-Placita and thus provides additional indirect confirmation of our attribution.

2) No less serious and even more detrimental to MR's project is the erroneous interpretation of the SP-Placita as a «dialectical» collection of placita and failure to understand that it is a handbook of physical philosophy. If someone sets as his goal to reconstruct a work of an ancient author, but from the start erroneously defines its genre (e.g. mistakes tragedy for comedy), such reconstruction is doomed to failure: the structure of any work is determined by its genre. The imposition on the extant remains of an alien form will result in the serious distortion of its original structure. M&R have seriously misunderstood the meaning of the introductory sentence pe^ovxeg xov 9uoiKov napaSrooeiv Aoyov kxL in the Preface to P which reveals the philosophical genre of the SP-Placita: the author's goal is to teach physical science, i.e. to write a handbook of physics, and not to present a collection of various doxai useful for dialectical debates. In philological and historical-philosophical hermeneutics82 alike we adhere to the following fundamental principle: any text has to be first interpreted on its own, on the ground of internal structural and semantic analysis, no (supposed) external «parallels» are allowed at this stage. Only after the «immanent» intetrpretation has been completed, external «loci paralleli» can be adduced as confirmatory evidence. Analogical arguments are generally invalid: analogy and similarity, like beauty, often lie in the eye of the beholder. M&P violate this principle when they base their interpretation of a complete and complex work, the SP-Placita, on an arbitrarily selected and rather irrelevant passage from Aristotelian «Topics»83. For the genuine and historical «intellectual context» of

I use the term hermeneutics in its original Greek sense of the art of interpretation with no relation to the homonymous 20th century philosophical trend.

3 M&R, Aetiana, II/1, 158 ff. I do not deny that this Arstotelian passage may throw some light on the origin of Peripatetic collections of doxai like Theophrastus' Physiakai doxai. But the context of this recommendation in the «Topics» it totally different from the purpose of SP-Placita, a hand-

our doxographer one should look not into this technical advice on how to build a database of quotations, but into modern literature on Antiochus of Ascalon, the «Platonic Stoicism» of the 1st century BC, the Post-Hellenistic philosophy in general and the Aristotelian revival of the 1st century BC8 . Another serious mistake is the forced «dialectical» interpretation of the term Sialpeoi^. The use of Siapeiv in the Preface of P has nothing to do with diaphonia, debates, antithesis etc.85 It is applied to the Stoic «division» of philosophy into three disciplines and Peripatetic division of philosophy into theoretical and practical parts. The closest parallel to this use is found in the title of Eudorus' work Aialpeoi^ xoti Kaxa ^iAooo^lav Aoyou quoted by the author of SED (A), i.e. by Arius Didymus, who quotes it in the discussion of the best diairesis of the ethical topos86. The ethical part (meros) of philosophy is further divided into three species (eidos), and each of these is «subdivided» (unoSiaipeixai p. 42, 24). In the end of the exposition of Eudorus' «division» the author says that he will start the discussion of the ethical «problems» (npopA^axa) following his own arrangement of topics (Siaxa^iv) because he is «sure that his own division is more precise»8 . Both in the handbook of physics quoted by Stobaeus in book I (the SP-Placita), and in the handbook of ethics quoted in book II Siaipeiv/Siaipeoi^ is used in the same sense and denotes the arrangement of topics or «problems», i.e. refers to structure of science and has no connection whatsoever with diaphonia or debates. The lost Diairesis of philosophical science by Eudorus was a kind of philosophical encyclopedia comprising three handbooks (on logic, ethics, physics). Arius' diaeretic handbooks of physics (= SP-Placita) and ethics (SED-A) were partly modelled on it. Both in Eudorus' and Arius' works Aristotle's categories played important role in the formulation and sequence of problems. According to the author of SED-A this Aristotelian diairesis is superior to the older

book of theoretical physics. And we doubt that the Peripatetic collections of doxai professed «to teach physical science» and contained authorial definitions and explanations. If they did, they would have been theoretical pragmateiai, like Aristotle's Physics, and not collections of doxai. 184 Sedley (2012), Bonazzi (2007), Boys-Stones (2001), Sharples-Sorabji (2007), Karamanolis (2006) etc. Figures like Philo of Larissa, Posidonius, Eudorus and Philo Alexandrinus also come into consideration.

85 Ps.Plut. Plac. p. 51, 12; 51, 24 Mau.

87 Stob. II,7; p. 42, 7-11 W.

87 Stob. II,7; p. 45, 8-10 apKxsov Ss xrov npoP^npaxrov, npoxaxxovxa xa ysvn Kaxa xqv spot ^aivopsvqv Siaxa^iv, qv xiva nsiOopai npo; xo aa^saxspov Siqip^Ksvai.

one of Philo of Larissa (the hexameres), but his own will be even better.

3. I am not going to pass judgement on the volume II/2, the Specimen reconstructions of SP-Placita Book II, as a whole. One has to live with it, consulting it on many occasions, in order to appreciate it. It is a mine of meticulously collected information, surpassing Diels and invaluable for further research. I would restrict myself to one critical remark. The commentary to each chapter contains a final section titled «Dialectical-doxogaphic parallels». These collections of parallels are useful as such, but the texts quoted under this heading with few exceptions have nothing to do with «dialectics». This is just a restatement of authors' mistaken interpretation of the genre and purpose of the SP-Placita. The interpretation of tradition should always be presented separately from the tradition itself and not implanted in it.

Appendix 5. Did "doxography" in Dielsian sense ever exist?

The term "doxography" is used primarily in two senses: in the original narrow sense of a (supposed) special ancient genre of philosophical literature originated (according to Diels) by Theophrastus' Фиткт So^ai and in much broader sense as applied to any exposition or interpretation of philosophical views of any philosopher A by another philosopher B. In this second sense one can speak about Aristotelian doxography of Eleatics or atomists without implying that Aristotle was a doxographer in the first sense. I will call the first type pure or genuine doxography, and the second type "embedded doxography" since philosophical theories of a philosopher A in this case are "embedded" in the theoretical context of a philosopher B. It is possibe that Theophrastus' Physical Opinions was indeed genuine collection of various doxai, but we know virtually nothing about its structure and contents (see section 3 above). It is conceivable that some other works of genuine doxography, i.e. doxography for its own sake, existed in antiquity (e.g. the compendium of Celsus described by Augustinus), but where can we find a single specimen of this genre among the extant philosophical texts? Which of the 9 authors/texts included by Diels in the second part of his Dox.Gr. pertain to this genre of genuine doxography? The texts of Christian Apologists (Hippolytus, Epiphanius, Hermias) are apparently embedded in the refutation of heresies and and exposition of "diaphonia" of the Hellenes. Cicero's and Philodemus' theological doxography is embedded in the context

of the Hellenistic philosophy of religion. Theophrastus' De sensibus is anything but a collection of opinions for its own sake. As we have seen, the doxography of Arius Didymus printed by Diels under the title "Epitome" comes from a work of different philosophical genre, nepi aipeoerov. Only the Stromata of Plutarch look like a collection of doxai on its own, but this text pertains to the SiaSo%ai style continuous doxography, not to the diaeretic placita. The most important surviving "specimen" of the genuine problem-oriented doxography was for Diels (and still remains for many) the compendium SP-Placita (Diels' Aëtius). We have seen that SP-Placita was a handbook of physical philosophy, consequently the doxography quoted in this work was also embedded in the teaching the basics of physical science and was not a "genuine" collection of placita. Ps.Galen in the preface to a closely related text explicitly states that his work is an elementary handbook of philosophy that will help his reader to read and understand great philosophers. The exceprts from SP-Placita in Stobaeus are part of anthology composed by John of Stobi with eductional pupose for his son Septimius. The Didaskalikos of Alkinoos, as its title indicates, is a handbook of Platonic philosophy, and it is at least partly based on Arius Didymus. We may conclude that the most important texts on which Diels's reconstruction of the supposed genre of genuine "doxography" was based, are in fact handbooks.

Bibliography

Ayres, L. (ed.) (1995): The passionate imtellect: essays on the transformation of classical traditions presented to I. G. Kidd. New Brunswick.

Algra, K.A., van der Horst, P.W., Runia, D.T. (edd.) (1996): Polyhistor: Studies in the History and Historiography of Ancient Philosophy: Presented to Jaap Mansfeld on his Sixtieth Birthday, Leiden. Baltussen, H. 2000: Theophrastus against the Presocratics and Plato.

Peripatetic Dialectic in De Sensibus. Leiden. Baltussen, H. 2005: The Presocratics in the Doxographical Tradition. Sources, Controversies, and Current Research. Sudia Humaniora Tartuensia 6.A.6, 1-26. Bonazzi, M., Helmig, Chr. (edd.). 2007: Platonic Stoicism - Stoic Plato-nism. The Dialogue between Platonism and Stoicism in Atiquity. Leuven.

Boys-Stones, G.R. 2001: Post-Hellenistic philosophy. Oxford. Bremmer, J.N. (1998): Aëtius, Arius Didymus and the Transmission of

Doxography. Mnemosyne 51, 154-160. Canivet, P. (ed.). 1958: Théodoret de Cyre. Thérapeutique des maladies Hélleniques. Paris.

Clayton, P. B., Jr. 2007: The Christology of Theodoret of Cyrus.

Antiochene Christology from the Council of Ephesus (431) to the Council of Chalcedon (451). Oxford.

Des Places, E., Segonds, A.-Ph. (ed.) (1982): Pophyre. Vie de Pythahgore. Lettre a Marcella. CUF, Paris.

Diels, H. 1879: Doxographi Graeci. Berolini.

Dörrie, H. 1959: Porphyrios' «Symmikta Zetemata». München.

Farinelli, C. 2000: Lobone di Argo ovvero la psicosi moderna del falso antico. In: Cerri G. (ed.). La letteratura pseuepigrafa nella cutira greca e romana. Napoli, 367-379.

Fortenbaugh, W. (ed.) 1983: On Stoic and Peripatetic Ethics: the Work of Arius Didymus. New Brunswick.

Fortenbaugh W., Huby P., Sharples R., Gutas D. (edd.) 1992: Theophrastus of Eresus. Sources for his Life, Writings, Thought and Influence. Pt. I. Leiden; New-York.

Frede, M. 1999: Aëtiana. In: Phronesis 44, 135-149 (review of Mansfeld-Rinia, 1997).

Giusta, M. 1964: I dossografi di etica. Vol. 1-2, Torino

Göransson, T. 1995: Albinus, Alcinous, Arius Didymus. Göteborg.

Gourinat, J.-B. 2011: Aëtius et Arius Didyme sources de Stobée. In: Reydams-Shils, G. (Ed.), Thinking through Excerpts. Studies in Stobaeus. Brepols, 143-201.

Hahm, D.E. 1990: The Ethical doxography of Arius Didymus. In: Haase 1990: 2935-3055, 3234-3243.

Haase, W. H. (ed.) 1990: ANRW II 36.4: Philosophie (Epikureismus, Skeptizismus, Kynismus, Orphica; Doxographica). Berlin; N.-Y.

Inwood, B. 1995: Review of Göransson (1995). In: BMCR 95. 12.08

Journée, G. 2015: Thedoretus on Hippasus and Heraclitus (CAG IV.12). Paper delivered at the Melbourne colloquium «The Placita of Aëtius» 3.12.2015, published at «academia.edu».

Karamanolis, G. 2006: Plato and Aristotle in Agreement? Platonists on Aristotle from Antiochus to Porphyry. Oxford.

Lachenaud, G. 1993: Plutarque, Oeuvres morales, tome XII/2: Opinions des philosophes. Paris.

Lebedev, A.V. 1978: TO APEIRON: Ne Anaximandr, no Platon i Aristotel [TO AnEIPON: not Anaximander, but Plato and Aristotle]. Vestnik drevney istorii [Journal of Ancient History], 1, 39-54; 2, 43-58. Лебедев, А.В. 1978: TO APEIRON: Не Анаксимандр, а Платон и Аристотель. Вестник древней истории 1, 39-54; 2, 43-58.

Lebedev A.V. 1979-1980: ¥НГМА ZYMФYZQMENON. Novyy fragment Geraclita [¥НГМА ZYMФYZQMENON. A new fragment of Heraclitus]. Vestnik drevney istorii [Journal of Ancient History], 2, 3-23; 1, 29-48.].

Лебедев, А.В. 1979-1980: ¥НГМА ZYMФYZQMENON. Новый фрагмент Гераклита. Вестник древней истории, 1979 № 2, 323; 1980, № 1, 29-48.

Lebedev A.V. (1984): "ФУЕЕ TAЛANTEYOYZA": Neglected Fragments of Democritus and Metrodorus of Chios. In: Proceedings of the First International Congress on Democritus II. Xanthi, Greece, 13-25.

Lebedev, A.V. 1988: Did the doxographer Aëtius ever exist?. In: Philosophie et Culture. Acts/Proceedings. XVII Congres mondiale de philosophie. Montreal 1983 Edited by Venant Cauchy, Ed. du Beffroi / Ed.Montmorency, v. 3 (1988 ) 813-817 (microfilm).

Lebedev, A.V. 1988-2: Anaximander the originator of the mixis theory of matter and the founder of mechanistic physics. in: First International Conference on Greek Philosophy. Ionian Philosophy. Samos, 27-31 August 1988 - Pythagoreion, Abstracts of papers, 56-57.

Lebedev, A.V. 2014: Logos Geraclita: rekonstruktsiya mysli i slova. S novym kriticheskim izdaniem fragmentov [The Logos of Heraclitus: A Reconstruction of his Thought and Word. With n New Critical Edition of the Fragments]. St.Petersburg, «Nauka», 2014. Lebedev, A.V. 2014: Логос Гераклита: реконструкция мысли и слова. С новым критическим изданием фрагментов. Санкт-Петербург, «Наука», 2014.

Mansfeld, J. 1990: Doxography and Dialectic: the Sitz im Leben of the Placita. In: Haase, W.H. (ed.), 3056-3229.

Mansfeld, J. 2010-1: Doxographical Studies, Quellenforschung, Tabular Presentation and Other Varieties of Comparativim. In: Aëtiana III, 3-32.

Mansfeld, J. 2010-2: Physikai Doxai and problemata physika in Philosophy and Rhetoric from Aristotle to Aëtius. In: Aëtiana III, 33-98.

Mansfeld, J., Runia, D. 1997: Aëtiana. The Method and Intellectual Context of a Doxographer. Vol. 1: The Sources. Leiden; New-York.

Mansfeld, J., Runia, D. 2009: Aëtiana. The Method and Intellectual Context of a Doxographer. Vol. 2: The Compendium. Part 1-2, Leiden; Boston.

Mansfeld, J., Runia, D. 2010: Aëtiana. The Method and Intellectual Context of a Doxographer, vol. 3: Studies in the Doxographical Traditions of Ancient Philosophy, Leiden; Boston.

Mansfeld, J., Runia, D. 2016: Comments of Lebedev's Melbourne paper (unpublished).

McDiarmid, J.B. 1953: Theophrastus on Presocratic Causes. In: Harvard Studies in Classical Philology, 1-156.

Primavesi, O. 2015: Olearius on Atomism and Theism in Heraclitus: A Presocratic in late 17th century Germany. Rhizomata 3(1), 94-123.

Runia, D. 2009: Aëtius, or What's in a Name?. Mnemosyne 62, 464-470 (= Mansfeld, Runia 2010: 173-182_.

Runia, D. 1996: Additional Fragments of Arius Didymus on Physics. In: Algra, van der Horst, Runia 1996, 364-370.

Rashed, M. 2012: Les définitions d' Aquilius. BICS 55 (1), 131-172.

Russell, N. 2000: Cyril of Alexandria. London; New York.

Sharples, R. 1995: Counting Plato's Principles. In: Ayres, 67-82

Sharples, R., Sorabji, R. (edd.) 2007: Greek and Roman Philosophy 100 BC - 200 AD. Vol. I-II. London.

Schofield, M. (ed.) 2013: Aristotle, Plato and Pythagoreanism in the first century B.C. London.

Schölten, C. 2014: Theodoret. De graecarum affectionum curatione. Leiden; Boston.

Schor, A. M. 2011: Theodoret's People. Social Networks and Religious

Conflict in Late Roman Syria. Berkeley; Los Angeles. Schubert, Ch. (to appear): Die Arbeitsweise Plutarchs: Notizen, Zitate und

Placita (to appear in Rheinisches Museum für Philologie). Sedley, D. (ed.) 2012: The Philosophy of Antiochus. Oxford. Smith, A. 1993: Porphyrii Philosophi Fragmenta. Stutgardiae; Lipsiae. Steinmetz, P. 1964: Die Physik des Theophrastos von Eresos. Bad

Homburg; Berlin; Zürich. Viano, C. 2005: L'Epitomê de l'éthique stoïcienne d'Arius Didyme (Stobée, Eclog. II, 7, 57, 13-116, 18). In: G. Romeyer-Dherbey, J.B. Gourinat (edd.). Les Stoïciens, Paris. Zhmud', L.Ya. 2001: Revising Doxography: Hermann Doels and his

Critics. Philologus 145, 219-243. Zhmud', L.Ya. 2013: Die doxographische Tradtion. In: von Flashar, H., Bremer, D., Rechenauer, G. (Hrsg.). Grundriss der Geschichte der Philosophie. Bd. 1: Die Philosophie der Antike. Frühgriechische Philosophie. Basel, 150-174.

Abstract. In the frst part we will recuperate and reinforce our arguments againsts Diels' Aëtius hypothesis. We will also discuss the nature and purpose of the extant Ps.Plutarch's De placitis philosophorum and will argue that in its present form it is a truncated copy of the original handbook of physics (napaöroasiv!) with only 6 authorial definitions remaining from many more in the original. We assign to this «better Plutarch» the siglum P+ . The archetype of the P tradition was probably a personal copy of a Christian apologist who was interested only in dia-phonia of the Hellenes and therefore dropped most of the apodictic definitions and other authorial remarks and explanations as worthess. The second main complex of problems we will address is the relation between P+ and Stobaeus (S) on the one hand, and between both of them and Arius Didymus, on the other. We will refute Göransson's claim that Arius the doxographer should be distinguished from Arius the court philosopher. We add to 8 known authorial definitions in P and S two neglected ones from Stobaeus. One of them (on avayKq, I,4,7b) we connect with the lemma AiSup<ou>. Which means that P+ was known to Stobaeus as a work o Arius Didymus. Tertlullian is the second independent source who quotes P+ as «apud Arium». Finally we will present the new stemma of doxographical sources that emerges from our research in which the central role is accorded neither to Dielsian hypothetical «Theophrastus», not to the ficticious writer «Aëtius», but to the real giant of post-hellenistic philosophy Arius Didymus from Alexandria, the teacher of emperor Augustus.

Keywords: Ancient Greek philosophy, Post-Hellenistic philosophy, Stoicism, Aristoteliesm, doxography, Arius Didymus, Aëtius, Theo-phrastus, Pseudo-Plutarch, Stobaeus, Theodoret of Cyrus, Nemesius of Emesa, Porphyrius, Christian apologists, Hermann Diels.