The Cambridge Companion to Early Greek Philosophy

by A. A. Long

  This cluster of features is best read against the backdrop of Empedocles’ theory, whose story is largely teleological (in as much as Love, in the person of Aphrodite, acts as a craftsman) and even eschatological. For Empedocles, perception and thought are occasions of excellence: they allow distinctions and hierarchies. This is the reason why the section of his poem dealing with thinking led to a treatment of intellectual gifts and craft knowledge43 which seems to have been without parallel in other thinkers. The main theme that organizes his analysis of cognitive capacities, however, is that of fragmentation and synthesis. Travel is not only necessary for contact and thus perception to take place; rather the other way round: perception is one way for the elements to come together. In the perceptual process, the encounter of the elements is only temporary. When water finds water (this is what the perception of dark objects consists of), or fire finds fire (when the perception is of bright objects), air and earth are left aside.44 From this point of view, thought, in which all the four elements are at work (for blood is an harmonious composition of all four elements), only intensifies a movement of unification that is already discernible in sense perception. Whether sensation or thought, every cognitive act is the anticipation, within the limits of human life, of the ultimate fusion of the elements in the unity of the divine Sphere. These are acts of love, and this is why they are linked to pleasure.45

  The beautiful correlation between Anaxagoras’ and Empedocles’ views on thought and the senses on the one hand, and their overall philosophical world-view on the other hand, is also to be found, or at least suspected, in other thinkers, such as Diogenes of Apollonia46 or Democritus, even if, in the latter’s case, the bulk of his systematic interest was on the sensibles (the atomic forms) rather than on the senses themselves.47 This is less true of other thinkers because of the state of our information, or because of the kind of philosophy they practised (Alcmaeon for instance, though a physicist, must have had strong medical interests48). The balance between scientific programme and systematic interest, even in the most skilful thinkers, is a delicate one. One could even argue that there is in any given author, and from one author to another, a certain tension between his systematic project and the obligation to comply with some kind of scientific programme – a programme implicit in the set of relatively closed data and questions I have alluded to before.49 But on the whole, one can say that early Greek philosophers succeeded in integrating cognitive processes, no less than cosmological phenomena, with their physiological approach. This may have been at the cost of an epistemological consciousness that would be more congenial to post-Socratic, or more precisely post-Platonic, interests.

  NOTES

  1 As was shown by Solmsen [497] 160–64.

  2 Descartes, Second meditation: “What is a thing that thinks? A thing that doubts, understands, affirms, denies, is willing, is unwilling, and also imagines and has sensory perceptions.” (“Qu’est-ce qu’une chose qui pense? C’est-à-dire une chose qui doute, qui conçoit, qui affirme, qui nie, qui veut, qui ne veut pas, qui imagine aussi, et qui sent.”)

  3 For the semantic aspects of the history of the word psychê, see Jouanna [493].

  4 See below, p. 261.

  5 See Snell [128], with critique by T. Jahn, Zum Wortfeld ‘Seele-Geist’ in der Sprache Homers (Munich, 1987).

  6 Claus [486], to whom I owe the title of this section.

  7 Jarcho [492].

  8 Burkert [201] 120–65. See Huffman in this volume p. 70.

  9 Huffman [198] 330.

  10 See Riedweg [367].

  11 See Furley [489].

  12 De an. 1.2 405a19–21 = DK 11 A22. The assimilation of “All is full of gods” to “Soul is everywhere” is explicitly presented by Aristotle as a conjecture (De an. 1.5 411a7).

  13 The fragment is sometimes considered inauthentic, as betraying later influence (Diogenes of Apollonia or the Stoics). For a discussion, see Whorle [188] 63–64.

  14 As Huffman [198] 307–14 suggests it is.

  15 Aetius, IV.3.5 (= DK 68 A102) and IV. 8 (cf. IV.2), cf. Aristotle, De an. 1.2 404a5.

  16 The information about Diogenes is in Theophrastus (Sens. 44), who is not helpful, however, as far as Democritus is concerned (58); see below p. 259. On the function of the brain for Democritus, see Sassi [421] 73ff. as against Bicknell [410].

  17 As translated by Kahn [416].

  18 DK 22 B12, 36, 45, 67a, 77, 85, 98, 107, 115, 117, 118, as well as a group of fragments where the term seems implied with various degrees of probability (B26, 88, 136, A16). Cf. Nussbaum [256].

  19 See Claus’ [486] and Schofield’s [261] critiques of cosmo-physiological interpretations of the psyche-fragments, as represented for example by Mansfeld [255]. For further discussion of Heraclitus’ account of psychê, see Hussey in this volume p. 101.

  20 1 take it in the reverse order of Aristotle’s presentation, leaving aside the group of anonymous thinkers, which raises problems too intricate to be treated here. For a full study (including the question of order), see Mansfeld [40].

  21 The text and the meaning of the fragment are much disputed. For references, and an attempt to understand to pleon in the last line as saying that “the full” (not “the more,” or “what prevails”) “is thought,” see Laks [301].

  22 See below, p. 263, 266.

  23 Sedley [378] 26–31, thinks that the principle “like knows like” in Empedocles applies only to thinking (cf. DK 31 B109). Although On the senses does make a scholastic use of the traditional two principles “like by like”/“different by different” (for which see Müller [496]), I do not think that Theophrastus misrepresents Empedocles’ doctrine here.

  24 Contrast Theophrastus’ use of hôs with phasi in the De anima passage and hypolambanein in the Metaphysics passage.

  25 On Xenophanes and Heraclitus, see Lesher in this volume pp. 228–36, and Lesher [494] 13, 20–23. On Democritus, see Taylor in this volume pp. 196–7.

  26 We find the same kind of assesment in Theophrastus, fr. 227 FHSG.

  27 See DK 31 B3, p. 262 below.

  28 For a defence of the transmitted text (meta tên kinêsin), see Sassi [421] 187ff. With the usual correction (kata tên krasin), the text reads: “when soul is balanced in its mixture.”

  29 Lesher [494].

  30 See Long [366] 268.

  31 Lesher has argued forcefully for Xenophanes’ commitment to empirical observation (see p. 230 in this volume), but this is a matter of reconstruction, and the question of empiricism reaches far beyond that of the relationship between sensation and thought (although both questions are obviously related).

  32 Tr. Lesher [494] 24.

  33 For doubts about this, see Mansfeld [40].

  34 This interpretation may gain some support if one takes logôi not as “reason,” but simply as argument (Lesher [494] 24, n. 46, and in this volume p. 239).

  35 As the word barbaros in Greek means “who does not speak Greek,” “barbaric” in Heraclitus’ fragment is often taken to imply that the souls in question “do not understand the language of the senses.” This seems doubtful, but see Hussey in this volume p. 90.

  36 Adopting the construal of the sentence by Mansfeld [12] vol. I, 254, which must be right.

  37 There are reasons to think that Empedocles was the one who set up the agenda.

  38 This is at least one possible explanation of why cognition requires some symmetria (Theophrastus, Sens. 3, to be compared to the criticism of Empedocles’ theory in 15).

  39 Democritus’ account of vision is much discussed. See K. von Fritz, “Demokritos Theorie des Sehens” in his Grundprobleme der Geschichte der antiken Wissenschaft (Berlin, 1971) 594–622, and O’Brien [419].

  40 Cherniss [34] 314–16.

  41 Theophrastus’ classification of previous theories of sense perception relies on an opposition between those who explain sensation in terms of “similarity” between the perceiver and the object perceived, and those who assume some “dissimilarity” or “opposition.�
�� On this, see Mansfeld in this volume p. 30.

  42 On this, see Laks [394].

  43 This is the implication of Theophrastus’ report in Sens. 10–11.

  44 Theophrastus Sens. 10–11.

  45 See Bollack [356] vol. 1, 263–64.

  46 On whom see Laks [425].

  47 This aspect of the problem has been omitted here. The reader may refer to the second part of Theophrastus’ treatise.

  48 The problem of Alcmaeon’s intellectual profile is treated in Mansfeld [495].

  49 This, of course, concerns cosmology no less than physiology, in the restricted sense of the term.

  MARIO VEGETTI

  13 Culpability, responsibility, cause: Philosophy, historiography, and medicine in the fifth century

  “The idea of nature as implying a universal nexus of cause and effect comes to be made explicit in the course of the development of Presocratic philosophy”: G.E.R. Lloyd.1 “The conception of cause is borrowed from the language of medicine, as is clear from the word prophasis which Thucydides uses”: W. Jaeger.2 “The word aition is, from the Hippocratic writings on, a standard word for ‘cause’, and its relative aitia … meant a complaint or an accusation, but already by the time of Herodotus’s book it can mean simply ‘cause’ or ‘explanation’”: B. Williams.3

  These three distinguised scholars, distant though they are from one another in their intellectual orientations, seem to agree on the opinion that a precise and well-defined conception of causality is present in fifth-century philosophy, history, and medicine. This judgement is widely shared, but it needs to be corrected, or at least clarified and formulated, from two different but complementary perspectives.

  First, as we shall see, lexical investigation of causality (aitia, aitios, to aition, prophasis) shows that explicit theoretical reflection on causal connections and forms of explanation based upon them emerged only gradually and with considerable uncertainty from the fuzziness of moral, political, and judicial language to do with culpability, responsibility, and imputability of facts and actions. Interestingly, the conceptualization of causality developed in medical contexts rather than in early Greek philosophy (judging from the fragments of the latter and setting aside the causal formulations provided by Aristotle and Peripatetic doxography).

  Second, there is a need to clarify the relationship between the development of theoretical reflection on causality and the kind of causal connections it describes. For example, Aristotle’s treatment of “cause” in book two of his Physics does not include a Humean conception of causal connection, according to which the cause is the necessary antecedent of the effect. In this respect, he is faithful to the complexity of fifth-century thought; he tends to reproduce, albeit in the context of a rigorous theory, the diverse dimensions of causality that were beginning to emerge there more or less vaguely. On the other hand, a conception of a cause as that which is necessary and sufficient to bring about the effect is found in a part of the medical testimonies, and in this respect, it prefigures Stoicism rather than Aristotle.

  Fifth-century thought was largely lacking in any explicit theoretical reflection on the problem of causality and in a “strict” conception, in the Humean sense, of causal connections. But it was quite capable of conceiving (more or less spontaneously) relationships between things and events that later theory would have included in the general context of causality. What we need to recognize is that these relationships came to be described in terms that are different from the language of causality that I will analyse in this chapter.

  There are, for example, phenomena that occur “by nature” (physei), that depend on the regularity of the world’s natural order. This dependence is often described, both in philosophy and medicine, as “necessity” (anankê). Sometimes this necessity can be connected, not to natural regularity, but to the decrees of destiny and divinity, as in Agamemnon’s famous statement: “Not I am culpable (aitios) but Zeus, Moira and the Erinyes” (Il. XIX.86). The necessary dependence of events on the plan of destiny occurs frequently in Herodotus with the expression, “as had to happen” (edei). If the regularity on which the events depend is not divine or natural but human, the connection is often expressed, especially in political and judicial contexts, with the weaker term eikos (plausible, probable, likely).

  Yet, these connections of dependence between things, events and forms of order can only be brought within the context of causality and causal explanation by using later thought patterns. To demonstrate this, let us briefly survey some clear examples from early Greek philosophy, starting with the famous fragment of Anaximander, reported by Simplicius (DK 12 B1):

  The source of coming-to-be for existing things is that into which destruction, too, happens “according to necessity; for they pay penalty and retribution (dikên kai tisin) to one another for their injustice according to the assessment of time.” (tr. KRS, 108)

  The universal and necessary connection that binds things in the cosmic cycle is evidently conceived here in the moral/juridical terms of guilt and punishment rather than those of causal explanation.

  In the immediately succeeding philosophers, we find the widespread idea of a dependence of things and processes on the “power” of an originating principle (archê). Thus, Parmenides (DK 28 B12): “The daimôn that governs and rules (archei) all things.” There appear, especially in Empedocles and Anaxagoras, principles that much later would be interpreted as prefigurations of (efficient) causality: in the first case, Love and Strife (philia and neikos), in the second case Intelligence (nous). These principles exert their actions on other originating principles of a biological kind, such as the “roots” (rizômata) of Empedocles and the “seeds” (spermata) of Anaxagoras.

  Now in Empedocles’ text (see, for instance, DK 31 B26), Love and Strife seem to be somewhat anthropomorphic metaphors for the cosmic elements’ aggregation and separation, and not separate from the elements themselves. In any case their action is expressed in the political language of power (e.g., DK 31 B17.28: “They prevail (krateousi) in turn as time goes round”). In contrast with Empedocles, Anaxagoras’ principle Nous is explicitly conceived as separate from the things on which it exerts its own action. This is probably why Plato in the well-known passage of the Phaedo (97b ff.) refers to Anaxagoras as the initiator of the idea of final causation. Yet, the way Anaxagoras expresses Nous’ separation and its action on the world is once again the language of political and military power: Nous is “authoritative” (autokrates), exerts its own force (kratein, ischeuei), and has the power (archê) of initiating the rotation of the world (DK 59 B12). Thanks to this power that it has, Nous imposed order on all things (panta diekosmêse). This conception of a regulating activity has probably influenced the way Plato in the Timaeus thinks of the Demiurge’s action on the spatiotemporal world. It seems clear that this embryonic form of causal thinking is still completely clad in metaphorical language derived from the political sphere. The need to explain the beginnings of the cosmic order does not imply a theoretical reflection on the concept of cause, but rather it is forced to express itself in terms of the power that the gods exercise in the world or that men exercise in society, just as in Anaximander the language remains juridical and ethical.

  To approach the question of how causal thinking began, we cannot improve on the words of Michael Frede:

  When the use of ‘aition’ was extended such that we could ask of anything ‘What is the aition?’, this extension of the use of ‘aition’ must have taken place on the assumption that for everything to be explained there is something which plays with reference to it a role analogous to that which the person responsible plays with reference to what has gone wrong; i.e., the extension of the use of ‘aition’ across the board is only intelligible on the assumption that with reference to everything there is something which by doing something or other is responsible for it.4

  My aim in this chapter is precisely to verify, by correcting the widely shared opinions mentioned at the beginning, just when and ho
w this extension, conceptualization, and generalization of causal thinking took place. In other words, when did the transition occur from the personal language of culpability and moral, political, and legal responsibility to the abstract and “neutralized” language of cause? (This does not necessarily imply, as we shall see, the substitution of the neuter substantive to aition for the forms aitia and aitios.) We shall be dealing with a lengthy and complex process, one that spans all of fifth-century thought and that left a profound trace even on the more developed theorizing of Plato and Aristotle.

  THE PHILOSOPHERS

  The surprising result that emerges from lexical investigation of causality in the early Greek philosophers is the virtually total absence of any reflection on the problem of causal explanation. This is surprising because, of course, the evidence on them includes abundant references to the language of cause. Yet, from our perspective in this chapter, that evidence has no value whatsoever because it depends entirely on Aristotle’s interpretation, found in book one of his Metaphysics and book two of his Physics. There Aristotle looks at these thinkers as imperfect predecessors of the research into causality carried out by himself. When instead we focus only on the fragments that, to some extent, reflect the early Greek philosophers’ original language, the terminology of causality proper is virtually absent. What we find is terminology that conforms completely to the traditional moral and juridical connotations.