The Cambridge Companion to Early Greek Philosophy

by A. A. Long

  Aetius’ Placita too is lost, but Diels provided a reconstruction that, though not without major flaws, is basically correct.3 He magisterially showed (1) that the extant topic-oriented Placita ascribed to Plutarch (but in fact by pseudo Plutarch) and dated to the second century A.D. is a (rather drastic) abridgement of Aetius (and that the greater part of the shoddy Historia philosopha ascribed to Galen is a further abridgement of a version of ps.-Plutarch); (2) that Ioannes Stobaeus (fifth century A.D.) in the first book of his gigantic and only partly preserved anthology, the so-called Eclogae physicae, had incorporated large portions of Aetius and preserved important material abridged away by ps.-Plutarch; and (3) that the Church Father Theodoret (also fifth century) in his Cure for the diseases of the Greeks, the only source to mention the name of Aetius (three times), had also used Aetius’ work on an important scale.

  Accordingly, Diels argued that the information on the early Greek philosophers contained in his reconstructed Aetius,4 though debased and modified in the course of transmission, is linked to Theophrastus’ great work in a direct and vertical line of descent. This lends an air of historical reliability to what we find there. A similar conditional reliability is postulated for the authors who used the Vetusta placita (a work Diels, perhaps wisely, did not attempt to reconstruct). Diels further argued that the following texts largely go back to Theophrastus himself: most of the doxographical passages in the first book of the ecclesiastical author Hippolytus’ Refutation of all heresies (early third century A.D.), in the Stromateis of another ps.-Plutarch preserved by Eusebius, in several chapters dealing with the early Greek philosophers in the work of the otherwise unknown Diogenes Laertius (also early third century) entitled Lives and maxims of those who have distinguished themselves in philosophy and the doctrines of each sect,5 and finally in a few other works of minor importance.

  This reconstruction of the secondary tradition forms the backbone of Diels’ splendid edition of the Fragmente der Vorsokratiker (1903), which he revised and expanded three times in his own lifetime, and which was further revised by Walther Kranz, who added an indispensable index volume.6 This work too is continuously being reprinted, and it still is the basic edition of the texts of the early Greek philosophers. Fragments, both verbatim and secondary, are habitually cited with the numbering of Diels and Kranz (abbreviated DK). All other editions of the so-called Presocratics or of individual Presocratics, even though further material is occasionally added or verbatim fragments Diels believed to be spurious are authenticated, are entirely indebted to DK and so to the hypothesis concerning the genealogy of the secondary sources which underlies this work.7

  Diels firmly believed that verbatim fragments (designated B-fragments) cannot be understood apart from the testimonia (designated A-fragments).8 In spite of this, however, his format is designed to highlight the importance of the verbatim fragments. Hence, Diels gave each individual (or in the case of the Pythagoreans an individual group) its own numbered chapter, in chronological order and even in an order according to “succession,” instead of following the supposedly systematic lay-out of Theophrastus, or that of the extant ps.-Plutarch. This procedure unfortunately often entailed the cutting up and distribution of the testimonia, which in the majority of our ancient sources tend to assemble and oppose to each other the views of several philosophers rather than discussing those of individuals. In this process Diels tended to overlook some details, or put them in a chapter where one would not suspect them to be. In our sources verbatim fragments too are sometimes quoted in clusters to illustrate an issue in natural philosophy or theology or ethics.

  Diels’ quasi-biographical mode of presentation, though based on a (too) clear hypothesis concerning the transmission, effectively obscures its own foundations and also inhibits access to the original sources themselves. The reign of the individual Presocratic fragment became firmly established, and the relative reliability of an A-fragment was believed to have been securely ascertained by the place assigned to its source in the tradition as reconstructed, that is, its counting as good or less good. The verbatim fragments on the other hand were viewed in the way works of art found in the course of a premodern excavation were appreciated, and so as having a value not dependent on the ruins that happened to preserve them.

  This view, indeed, is not entirely false, and certainly not always. Such fragments often travel from one source to another, and the context in which we find them is by no means always decisive for their interpretation – even in those cases where we can be relatively or even entirely certain that what was copied out is the original work. Even here we should realize that quotation need not necessarily be exact; errors are unavoidable, and texts that are quoted may be adapted to their context.9

  Before Diels’ reconstruction of Aetius, scholars in Germany believed that all the above mentioned later authors had used, or revised, a common source already available in the age of Cicero. The Vetusta placita is what remains when Diels’ Aetius has been subtracted: a nice example of a shrunken hypothesis. It is, therefore, not at all surprising that the section of Diels’ overview dealing with the Vetusta placita is far from satisfactory, and that the nearer we get to Theophrastus, the more hazardous the route becomes. Following in the footsteps of Usener, Diels was not at all bothered by the fact that the majority of the larger fragments (those dealing with the principles) he ascribed to Theophrastus’ doxographical work are cited from the Physics.10 He also failed, apart from a remark tucked away in a later article,11 to take Aristotle’s influence into account, although Zeller had pointed out the similarities between Theophrastus’ and Aristotle’s accounts of the early Greek and Platonic principles.12 To be sure, most Aristotelian passages dealing with the early Greek philosophers are to be found in DK, but Aristotle’s role in shaping the tradition had fallen by the wayside.

  Diels also failed to ask himself for what purpose placita had been collected in the first place, and why it was that they continued to be added to, abridged, or revised in other ways. He did not take into account the possibility, that before Aetius more than a single tradition may have existed, or that mutually diverging witnesses to the same tradition may have been available.13 Those who contributed to the tradition(s) were in no way obliged to preserve their predecessors’ material unchanged. But Diels’ main purpose was to get as close as he possibly could to the undefiled Theophrastean origin of the doxographical tradition by unmasking what he saw as fraudulent practice, and so to come nearer to the pure fount of early Greek philosophy itself. It was a sort of rescue operation, which in itself of course is not at all a bad idea. But as already intimated, his hypothesis is currently being revised and in need of still further revision, so the account which follows, though still preliminary in the sense that this revision is not yet completed, will in part go beyond Diels.

  TWO SOPHISTS AND PLATO

  Collections of views were already composed by two sophists, Hippias and Gorgias. Plato and Aristotle among others presumably used them, and were influenced by them.14 Hippias put together a topic-oriented anthology of related views in both prose and verse, culling the poets as well as what came to be called the philosophers.15 This will have had the purpose of providing easy access, perhaps mainly for rhetorical purposes, to what must have been an already bewildering variety of ideas. By assembling related views from the old poets down to just before his own times, Hippias in effect emphasized agreement and continuity. Important echoes of his approach are to be found in Plato and Aristotle.16

  Gorgias, on the other hand, stressed what he took to be the philosophers’ insoluble disagreements. We still have a short paraphrase of a part of his original argument and a significant remark in one of his extant declamations.17 In addition, his work is echoed in two early Hippocratic works, in Xenophon, Isocrates, and even in Plato.18 The philosophers, so Gorgias stated, could not agree whether the things that are were one or (infinitely) many, whether they were generated or ungenerated, and whether motion exists or does not exist. He amusingly
went on to argue that all were wrong. Both Plato and Isocrates provide lists arranged according to the number and nature of things that were assumed, a feature that we shall also find in Aristotle and others.

  Plato and Aristotle combined the approaches of Hippias and Gorgias and added to the material they had collected. Indeed, a presentation according to similarity (e.g., a list of the views concerned with how many and what things there are) may be combined with one according to disagreement. In his later dialogues Plato, who had begun his career as a sort of Socratic sophist, turned more and more to the great masters of the past, discussing and adapting their ideas in order to go beyond them, and we may well believe that he had studied the original works of, for example, Anaxagoras, Parmenides, Heraclitus, Zeno and Empedocles. Still, his approach to these past masters was coloured by their reception in the sophistic works mentioned above, and also by the way the ancient thinkers had been interpreted by lesser followers.19 This, for instance, is why Plato emphasizes Heraclitus’ doctrine of flux and diversity and tends to neglect what he has to say about unity and stability, and why when speaking of Parmenides, he emphasizes his idea of the Oneness and immobility of all there is, though he is by no means blind to the question of Being (e.g., Soph. 24Id).20 Above all note that what we have in Plato is not doxography but a form of dialectic (see the following section on Aristotle), and that the more or less rigid schemes which underlie his expositions are presented in the course of imaginary conversations among civilised people, not as ingredients of a systematic treatise.

  ARISTOTLE, THEOPHRASTUS, AND THE LATER Placita

  Discussion of the tenets of his predecessors, often including Plato and Plato’s immediate pupils, is a standard feature of Aristotle’s systematic treatises (pragmateiai).21 He prepared himself quite thoroughly by assembling a library, and presumably included abstracts and quotations in the critical monographs he wrote about Melissus, Alcmaeon, the Pythagoreans, Gorgias, and Zeno (D. L. V.25), a two-book treatise Problems from the (writings) of Democritus (D. L. V.26), and one in three books On the philosophy of Archytas (D. L. V.25). Only a few fragments of these works, still available to the later Aristotelian commentators, are extant. There can be no doubt that he also read and excerpted other major figures such as Parmenides and Empedocles, from whom he quotes individual lines and even a few longer passages. That he also used and was influenced by the anthology of Hippias already has been noted, and, as the author of a monograph on Gorgias, he had of course firsthand knowledge of the latter’s argument. In addition, he was also influenced by the way Plato cited and used his predecessors. But Aristotle converted Plato’s urbane approaches into a discipline, namely dialectic, which follows a set of specific rules set out explicitly both in the Posterior analytics and in the Topics.22

  It is part of Aristotle’s method, when engaged in the dialectical discussion of a problem (defined at Topics I.11 104b 1-8), to divide a genus into its species in order to review the relevant doxai, and to set out the disagreements and the views which are held in common, so as to evaluate and criticize them in the most apposite way, and to go on from there. Probably the best known example of this procedure is the discussion of the antecedents, from Thales to Plato, of his own theory of the four causes that takes up much of the first book of the Metaphysics.

  One who embarks on the discussion of a question or problem (which may be put in the form of a statement) should proceed in an orderly way. He should establish what is the genus of the matter, for example, whether it is a question in one of the theoretical disciplines, such as physics (and then, of course, what is the species, for example, zoology), or ethics. Furthermore, four types of questions must be distinguished and treated separately – whether or not the object of the inquiry has a certain attribute or not, the reason why it has this attribute, the existence or nonexistence of the object of the inquiry, and its substance or definition (APo. II.1 89b24-35).

  The categories play a crucial part in this connection, because it is of major importance to establish to what category (substance, quality, quantity, place, and so forth) the object of inquiry and its attributes belong (e.g., De an. I.1 402a7-10, 402a23-b3). Again, the four types of questions may be formulated for each category.

  At Topics I.14 we are told how to select and classify statements (protaseis) or problems (problêmata) that are to be discussed; I quote parts of the text:

  Statements should be selected in as many ways as we drew distinctions in regard to the statement. Thus one may select the tenets [doxai] held by all or by the majority or by the experts …. We should also make selections from the existing literature and put these in separate lists concerned with every genus, putting them down under separate headings, for instance about the good, or about the living being, and about the good as a whole, beginning with the question What is it? One should indicate separately the tenets (doxai) of individuals, for example, that Empedocles [representing expert opinion] said that the elements of bodies are four …. Of statements and problems there are, roughly speaking, three sorts: for some are ethical, others physical, and others logical. Ethical are such as, for example, whether one should rather obey one’s parents or the law, if they disagree, logical, for example, whether the knowledge of opposites is the same or not, physical, for example, whether the cosmos is eternal or not. The same holds for problems (105a34-b25).

  Statements (or propositions) and problems may be exemplified by tenets, doxai; accordingly, as there are three classes of statements, so there are three classes of doxai: ethical, physical, and logical. This explains the title of Theophrastus’ treatise, Physikai doxai, and also makes clear to what kind of context this work belongs.

  A fundamental Aristotelian example of such a division of a (sub-) genus is to be found at the beginning of the Physics. It is concerned with three categories, namely the quantity, substance, and motion of the principles or elements, and, true to the precept of the Topics, names are added in some cases (Phys. I.1 184b15-21). Numerous other examples could be cited from Aristotle’s technical treatises.

  One can prove that Aristotle’s method profoundly influenced the Placita literature because in numerous cases the types of questions and the categories determine the layout of individual chapters and indeed entire sequences of chapters in ps.-Plutarch. For example, Chapter I.7, “About the gods,” first discusses the issue of existence and then goes on to list the various views (name labels added) about the substance and shape (i.e., the quality) of the gods. Chapters IV.2-7 are concerned with what the soul is, the number of its parts, the substance and location (category of place) of its ruling part, its motion, and the issue of its immortality (name labels added throughout). The placita on the earth (ps.-Plut. IV.9-15) ultimately depend on Aristotle’s discussion at De caelo II.13, even as to part of their contents, and so on.23

  My working hypothesis for Theophrastus’ Physical tenets is that it was a systematic collection of the problematic tenets of the physicists (and presumably of some doctors) according to genera and species, and that he applied the method of division and availed himself of the types of questions and the arrangement according to categories. We have an explicit testimony that he also added the required objections (enstaseis).24

  In his topic-oriented extant work On the senses, Theophrastus applies the method of division throughout. The main and explicitly stated division is between those who believe cognition is “by like” and those who assume it is “by unlike.” But another division also plays a part, namely between those who believe there is a difference between sense perception and thought, and those who do not. Furthermore, within each group the members are arranged according to the number of senses postulated. The last philosopher to be discussed is Democritus. This is because, according to Theophrastus, he argues that cognition is both by like and by unlike and so fails to fit the main division. This structure, involving a division of representatives on either side of an issue followed by one or more exceptional tenets, is not typical of Aristotle’s dialectical overviews, but
it is entirely similar to numerous chapters in ps.-Plutarch.25 Diels believed that On the senses is a large fragment of the Physical tenets, but this is by no means certain.26 The predecessors of Aetius presumably used not only the Physical tenets but also other works by Theophrastus. In fact, they may have used works by Aristotle himself or even, on occasion, the original sources or available epitomes of such originals. We may call this practice retrograde contamination. Still, it is clearly Aristotle’s methodology as revised by Theophrastus that determines the layout of the Placita.

  That a collection of this nature, including tenets of post-Theophrastean provenance, was already available in the time of the Stoic Chrysippus is proved by a verbatim fragment of the latter concerned with the soul’s ruling part, quoted by Galen.27 This collection already went beyond Aristotle and Theophrastus in that, like Aetius, it clearly brought out the profound disagreement (antilogia, or diaphonia) among the experts. In a sense this is a return to the manner of a Gorgias, but, in actual fact, it is a symptom of the impact of Hellenistic scepticism.

  The relation of Seneca’s Natural questions to the Placita traditions needs a fresh inquiry that cannot be provided here.28 It is clear that he must have used material prior to ps.-Plutarch, and it is also clear that this can hardly have been Aetius, or Aetius alone: the differences are simply too substantial, and Seneca provides much more information on individual doctrines than Aetius, who moreover may have to be dated a bit later than Seneca. Presumably, Seneca had also studied original treatises in the field of meteorology. Yet, the Natural questions as to its topic-oriented contents corresponds – with omissions, and differences of order, differences that are complicated by the uncertainty about the original order of the books of Seneca’s treatise – with Aristotle’s Meteorology as well as with the third book of ps.-Plutarch, which is also about meteorology (including IV.1, on the Nile). The last Greek philosopher to be cited is Posidonius, just as in Aetius. There is a certain emphasis on the early Greek philosophers just as in Aetius, though Seneca is far more selective as to names (though fond of citing anonymi). With some hesitation, I would plump for the suggestion that among Seneca’s numerous sources were one or more versions of Vetusta placita, in which he found the rather rich information about the early Greek philosophers – including perhaps even Aristotle, Theophrastus, and Posidonius – that he wanted to use. He is an independent and creative writer, so he used the information in an independent way. What above all tempts me to accept this suggestion is Seneca’s procedure: he cites doctrines which he subjects to a dialectical scrutiny, stating objections, making the proper choice, and occasionally even coming up with a solution of his own. This is exactly the way many ancient authors used placita material.