PARMENIDES ON THOUGHT AND BEING. Those who make a study of Parmenides are today sharply divided into two camps over

PARMENIDES ON THOUGHT AND BEING Those who make a study of Parmenides are today sharply divided into two camps over the relationship between thought and Being in his poem. One camp argues for the identity of thought and its object Being (or the identity of mind and Being); the other for 1 their non-identity. The identity thesis opens the door to idealism, which in its most robust definition reduces reality to mental entities, minds and their thoughts and their 2 other mental states. Or in Vlastos' more modest statement of the idealism he attributes to Parmenides and to any idealist, `the dogma that thought and reality are 3 coextensive'. There are good reasons for supposing that Parmenides holds no variety of idealism and that he does not sanction explicitly or implicitly the identity of mentality and Being. THE IDENTITY THESIS The identity thesis has often found supporters and the support of formidable scholars. Vlastos is a modern advocate. Long seconded him a few years back in a detailed 4 argument for identity, and he in turn receives the firm support of Sedley in a briefer 5 commentary. Kahn too argues for identity, but observes the lack of attention `standard 6 modern scholarship' gives to the thesis. Antiquity, however, had to wait a long time, until the neoplatonist Plotinus, for an advocate of identity (Ennead 3.8.8, 5.1.8). The temptation to take up the identity thesis is strong for three main reasons. (1) Some fragments may be read as expressing or implying identity of thought and Being. Two in particular have been the focus of attention. The clearer or the more striking of these is the important fragment B3,, which lacks a context, as it has come down to us, but which most scholars think to be 7 central to Parmenides' speculation. Its foundational role is indicated by its being expressed in the form of a premise by means of the inferential particle. A popular presumption held by Diels/Kranz is that the incomplete verse B3 completes the 1

8 incomplete verse at the end of B2, where it fits in metrically. This position makes B3 the premise of an argument for the unthinkability and inexpressibility of non-being:... / (B2.7-8) (B3). The fragment is most naturally and simply translated as a statement of identity, 9 as even the opponents of the identity thesis, such as Tarán, are willing to concede. In the spirit of the identity thesis B3 may be translated as, `For to think and to be is one 10 and the same thing' or, more simply, `The same are thinking and Being'. B3 provides the pillar of the identity thesis, and Plotinus is the first to appeal to it in his own argument for the identity of Being and thought in Parmenides' metaphysics. 11 The other passage is embedded in lengthy B8, verses 34-36: ',,, 12 which Long, after the translation of Sedley, renders as, `Thinking and that which prompts thought are the same. For in what has been said [i.e., the preceding 13 arguments] you will not find thinking separate from being'. In particular, Long puts stress on the passage's opening verse as another clear statement of Parmenides' commitment to the identity of thought and Being. As he puts it, the verse identifies thinking with the `source and object' of thought, namely, Being. The similarity in wording between B3 and the opening of B8.34 provides a good reason, as Coxon points out, 14 for the similarity in translation of these two passages. 15 Long contends with Kahn over his identification of Being with, which Kahn 16 translates as `knowing'. Kahn recognizes what he calls the `curious asymmetry' between mind and Being. The identity lies in only one direction, and it is only or that is `identified with' or `reduced to' `its object', which is Being, but `never 2

conversely'. Kahn observes that we find the same asymmetry in Aristotle's psychology, when it comes to Aristotle's identification of thought with its object (e.g. DA 431b17), where we never find the converse identification. Long maintains, in disagreement with Kahn, that identity between Being and thought holds in both directions in an equal 17 fashion and that Being is mind just as much as mind is Being, that there is a 18 `reciprocal identity of mind and Being'. Sedley shows that he is of the same mind 19 when he says `that what thinks is, and that what is thinks'. Nor does Long believe that there is a reduction of mind to Being. They both retain their own integrity by their possessing distinct essences and thereby distinct definitions. There is no identity, as Long puts it, in type between mind and Being, but only the identification of their tokens so that, just as Vlastos also maintains, thought and Being share the same extension. 20 If we take `Being' and `mind' to be distinct characterizations of the same subject, Long would seem to commit Parmenides to a version of property dualism, in which the same subject displays a double nature. When Long treats Being as a characterization, he indicates that he may believe that Parmenides treats `being' or `existence' in a qualitative fashion or, as some would put it, as a `predicate'. This, however, is reasonable, since every philosopher before Kant in all probability regards `existence' as a quality. An object's existence, just as its color and shape, would contribute to its description. Property dualism entered the repertory of the philosophy of mind as a replacement for Cartesian substance dualism, and as a position that is friendly to physicalism and yet avoids a reductionist analysis of mentality. Typically, advocates of property dualism hold that every entity is a physical entity, but that some of these display a dual nature by also being mental entities, the mentality of which is not defined in physical terms. Accordingly, these entities are both physical and mental, and they possess these features as essential properties. Two difficulties emerge for Long from his attributing property dualism to 3

Parmenides. First, it becomes difficult to see how he can avoid ascribing idealism to 21 Parmenides, an ascription he appears to prefer to avoid. Whatever counts essentially as an entity also counts essentially as a mental entity on Long's view; reality just is what is mental. Despite the predicative role Being assumes in the framework of property dualism, mentality must be attributed to Being essentially. There is no other subject available to receive mentality, which functions as an essential property within the dictates of property dualism. Second, this dual nature of Being puts in jeopardy its homogeneity, its being of a 22 single sort, (B8.4), or its being `all alike' qualitatively (B8.22). Long recognizes the threat to the unity and homogeneity of Being posed by his interpretation of identity. He tries to disarm this threat by maintaining that his identification of Being as thought is no more a problem for him than the attribution of `thinkable' to Being, or for that matter the attribution of any of the many properties, such as `ungenerated' or 23 `indestructible', Parmenides is willing to apply to Being. In other words, it is just a problem for Parmenides. Yet these sorts of properties, `ungenerated' and the like, seem to be of a different order from those that go into the essence or nature of an object. `Thinking' is not on a par with `ungenerated', certainly not under Long's characterization of it as `the life or heart of Being', which under such a description would seem to be the very nature of Being. Those properties Parmenides applies to Being as necessary properties, `ungenerated', `one', `continuous', and the like, he probably would not regard as telling us what Being is, which in its uniformity is presumably nothing but Being. The so-called `Cornford fragment', if genuine, indicates that Being just amounts to Being when the fragment affirms that the `name' of what is 24 `sole, unchanging' is, `to be', and thus presumably no other `name' (description) would illuminate its true nature. It is perhaps pertinent to recall that Aristotle allows for necessary but not essential properties of an object in at least the form of the proprium, 4

a property that is necessarily convertible with its subject (Top. 1.5.102a18-30). Aristotle's example of such a property for humankind is `grammatical', whereas 25 scholastic Aristotelians preferred `risible'. It is the post-aristotelian understanding of essentialism that insists that every necessary property of a subject be an essential one. The homogeneity of Being is also jeopardized by another consequence of the 26 identity thesis. Long acknowledges this consequence, which the advocates of identity do not always recognize, at least not openly. The identity of Being and thought entails the odd consequence that Being would have itself for its object of thought, since the only object available for thought is Being. Accordingly, Being may be characterized in a reflexive way as `thought thinking thought', which is strikingly like Aristotle's characterization of divinity, the unmoved mover (Metaph..9.1074b34-35). A selfreflexive Being would seem to introduce a complexity in its nature that undermines the alleged homogeneity of Being. There would be at least a conceptual heterogeneity within the essence of Being because Being by its very nature is both thought and the object of the thought. Again, we end up with a version of property dualism: Being is essentially thought and the object of that thought. (2) There is the strong motive of identifying Being with divinity, and Long for 27 instance feels its pressure. Since divinity could hardly be deprived of life and mentality, divine Being should then at least think, and on the identity thesis is even identical with its thinking. The attribution of divinity is not unreasonable, especially upon a mystical interpretation of Parmenides, although Long does not evoke such an interpretation on behalf of his analysis. Parmenides, through the words of the unnamed goddess who tutors a nameless youth about reality, gives evidence, as West affirms, 28 of having undergone a mystical experience, of an ordinary enough variety, in which there is an awareness of the unity of all things. In Parmenides' case Being constitutes an eternal unity that makes no allowance for any distinction between things. Long finds 5

it implausible that Parmenides' successors, the pluralists Anaxagoras and Empedocles, could consider themselves `relatively Parmenidean' and could endow their fundamental principles with thought -- Anaxagoras' and Empedocles' divinities -- unless they assumed that Parmenides did the same with his Being and believed it to be `divine 29 mind'. Presumably, Long takes Empedocles' divine `roots' of air, earth, fire, water, and the divine powers of `love' and `strife', to be endowed with thought just because they are divine. Accordingly, if Being is divine, it should be a thinking thing. Being is fundamental to Parmenides' metaphysics, to be sure, just as the of Anaxagoras and the divinities of Empedocles perform fundamental roles in their cosmologies. But the divinity of Being is difficult to maintain in the face of Parmenides' 30 never designating it as `divine' in the surviving fragments, and Mourelatos has pointed out that Parmenides makes no use of any of the traditional epithets for divinity in his 31 description of Being. Even though the poem's proem is full of superhuman features and the activities of divinities, it displays no pious attitude towards any deity, and the rest of the poem betrays none towards Being. Notwithstanding West's suggestive remarks, the youth of Parmenides' poem in his encounter with the goddess undergoes no mystical experience in which he comes in direct contact with Being. Even though Parmenides' doctrine condones a conception of Being that unites all things into a homogeneous whole, the identity of the youth is not submerged within Being in a mystical union, for example, of the typical sort we find, for instance, the English poet Tennyson recounting when he records the subsumption of his personality under a greater reality, where `individuality itself seemed to dissolve and fade away into 32 boundless being'. Instead, the youth in the goddess' presence becomes the auditor of the rigorous argumentation she articulates at length in her lecture. (3) The major proponents of the identity thesis also marshal additional `indirect', as they call them, considerations. The most compelling of these indirect considerations 6

is the obvious suggestion that, as Vlastos puts it, `the thought which knows being could 33 hardly be denied existence'. It is simply difficult to appreciate how thought could not be a part of existence, and since all that exists is Being, `for nothing other either exists 34 or will exist besides Being' (B8.36-37), and since Being is homogeneous in nature 35 (B8.22), thought and Being should be the same. Sedley contributes to this indirect consideration by maintaining that the non-identity thesis would `undermine' Parmenides' 36 `monism', since it would divorce the thinking subject from its object of thought. Kahn argues that Parmenides would not wish to identify `knowledge and truth' ( ) with non-being, and thus he intends them to be identical with Being, since his exclusive disjunction, `it is or it is not', allows for no other alternative, and thus no other place for 37 thought to reside. This indirect consideration is an inference readers of the poem draw based on their understanding of Parmenides and on the presumption that he could not have neglected the ontological status of thought and language. Unfortunately, Parmenides and his goddess may simply have not faced up to the issue of how to accommodate thought and language within their metaphysics, and Owen, for instance, finds reasons for suspecting that Plato regards Parmenides as having failed to address 38 the ontological status of thought and language. Parmenides' oversight would doubtless be a telling criticism of him, and a problem he possibly ignored or never even recognized. But it should be noted that there is no end to the `indirect' considerations of the sort proposed by Vlastos and adopted by Long. Thought, to be sure, plays a prominent role in Parmenides' argument. Surely it should be included in the circle of Being. But the goddess and the youth play a big role in the poem. Surely they too, unless they are fictional, should have a claim on existence. By the same token, the author of the poem should certainly have a claim and the members of his audience. After all, must not the argument, this `much contested refutation' (B7.5), have an author and an audience whom the author undertakes to persuade? Yet the introduction of 7

these agents within the embrace of Being would undermine its homogeneous unity, 39 unless, of course, they too are identical with Being. Parmenides simply has nothing to say about these subjects, however philosophically reprehensible these oversights may be on his part. A reason is available, however, that may help to explain Parmenides' neglect of thought and language. It is plausible that the early philosophers would not from the outset of metaphysics have hit upon the vexed question of the status of thought and language within reality. In his reflection on the nature of thought and language Parmenides may have been able to appreciate them only as guides in the development of his metaphysics and not as genuine parts of reality. Thought and language provide guides about reality because what thought cannot think and speech cannot name cannot be real. The conception of generation, for example, involves the unthinkable and the unnamable. Thus it too cannot be real and Being is ingenerable. By his focus on thought and language as items that guide him to the nature of Being, Parmenides may have been directed away from considering where they should fit into his metaphysics. THE NON-IDENTITY THESIS Those who hold the non-identity thesis typically locate the basis of Parmenides' thinking in his position, which Guthrie paraphrases as, `it is impossible to think what is not: there 40 is no thought without an existing object'. Guthrie seems to take these two statements to be equivalent. The requirement of an object by thought and language is often drawn from the difficult B6.1-2, which in Tarán's translation reads, `it is necessary to speak 41 and to think Being; for there is Being, / but nothing is not...'. This rendering of B6.1-2 42 allows it to be the `exact complement', as Tarán points out, of B2.7-8, which again in his words reads `for neither could you think (or know) non-being... nor say it'. Accordingly, Being is the only thing thinkable and the only thing thought requires. Consequently, non-being in its unthinkability is absurd, and thus is dismissed from the 8

roster of reality, along with anything the conception of which presupposes non-being, such as generation or destruction. Long readily admits the advantage of the non-identity thesis. It is a perfectly `clear and intelligible idea' to tie together Being and the thinkable, since `we cannot think what is not'. The `inconceivability of the non-real' furnishes Parmenides with his `starting-point', which as a first premise has the advantage of being 43 `attractive to common sense'. Tarán, who advocates non-identity, in contrast with Long, is among those who believe that the absurdity of non-being is basic and unargued for in the goddess' thinking and that it provides one of Parmenides' `points of 44 departure'. But, despite what Tarán affirms, Parmenides' goddess argues for the absurdity of non-being. The proponents of the non-identity thesis devise their own plausible translations of B3 and B8.34 which would avoid committing Parmenides to the identity thesis. Guthrie offers a translation of B3,, which is in keeping with the aspirations of the non-identity theorists: `the same thing is for thinking and for being'. This `literal' rendition of B3, as Guthrie calls it, brings out what he and 45 Burnet take to be the `original dative force' of the infinitives. Guthrie and Coxon point out the syntactical parallel with... in B2.2, `which... are for thinking', and Coxon who provides references maintains that tenses of with a transitive 46 infinitive are `idiomatic in the fifth century and later'. Guthrie prefers over the `literal' translation, which relies on the dative meaning of the infinitives, a rendering that displays the `potential' nature of when it is coupled with an infinitive, which is 47 related to the dative reading of the infinitives, as Barnes maintains, `for it is the same thing that can be thought and can be'. This translation gives B3 a modal nature, and it states the identity of what can be thought and what can exist. On either of these renderings of B3, the dative or the potential, thought and Being are not identical. 48 The assumption that the similarity in wording between B3 and the opening of 9

B8.34 requires a similar translation of both passages yields then a translation of verse 34 of the sort proposed, for example, by Coxon, `The same thing is for conceiving as is cause of the thought conceived', or that proposed by Cornford, `Thinking and the 49 thought that it is are one and the same'. In the one case, the object of thought and the cause of the thought are the same; in the other, thinking has only Being for its object and no other. In neither translation are thought and Being identified. Most of those who adhere to the non-identity thesis translate the of the second half of verse 34, 50 like Cornford, as if it meant instead of. Yet a number of scholars 51 construe along the lines of Coxon: Schofield, Burnet, Simplicius. Burnet provides the most promising rendering in English: `The thing that can be thought and that for the sake of which thought exists are the same'. Those scholars who shun reading as usually read it causally or explanatorily: for instance, `cause' 52 (Coxon), `why' (Schofield). But the causal reading is not limited to the champions of non-identity. Long too reads it causally by proposing `which prompts thought' for his translation of, and when he glosses the object of thought as the `source' of thought. 53 There is, however, a good philosophical reason for not reading causally, or not merely causally in the modern sense, but instead `teleologically', and 54 thus for adopting Burnet's rendition, `for the sake of which'. This rendering makes possible a reading of Parmenides' conception of thought that permits the basis of his argument to be even more appealing to common sense than what the non-identity theorists seem to count on when they argue that non-being is inconceivable simply because thought always requires an object. The teleological nature of thought endows it with an end and thus with its capacity for being directed towards its end, the object of the thought, which is Being for Parmenides and the only object available for thinking in his metaphysics. The end of thought is Being because thought by its nature makes 10

Being its content. This directionality of thought may be plausibly construed in terms of the intentionality of thought, of its aboutness, or of its being by its nature `of an object'. Intentionality Brentano judged to be essential to the whole range of mentality -- thought, perception, belief, desire, fear, love and the like -- and he took it to be the defining 55 property of mentality that marks the mental off from the physical. The object of intentionality provides the mental with its content, and the intentional object may be a particular thing, a property, or a state of affairs. The intentional object may exist or not exist, but the intentionality of thought in Parmenides' analysis has for its object what invariably exists and its only object is that of Being. Accordingly, the teleological directionality of mind is stronger than intentional directionality, because thought for Parmenides cannot help but be drawn inexorably towards what is real without qualification, namely, Being. Therefore, in B8.34 the goddess ties together the thinkable and the object thought is directed towards. The inconceivability of non-being, its unthinkability or absurdity, follows readily from the intentionality of thought, if it is assumed in addition that its intentional object must have actual existence. This assumption is perhaps easy for ancient Greeks to make, as Plato bears witness to, when in his comments on the nature of thought, where he treats it as possessing intentionality, he has the young Socrates answer without argument and in few words when the old Parmenides questions him about the nature of thought, that thought cannot be of nothing, but only of something, of something that exists (Prm. 56 132b-c). The intentionality of thought has clearly the appeal of the approval of common sense: thinking is always thinking of something, the possession of some content: so how could you think of nothing, a mere blank, the thought of which is devoid of all content? Those who argue that Parmenides is committed to the view that thought merely requires an existing object do not have the philosophical advantage of the common- 11

sense appeal the intentionality of thought brings with it. Yet the non-identity theorist might protest that it is obvious that Parmenides' statement of `requirement' is a statement of intentionality. After all, what else could it mean? Thought's requirement of an object is just its being directed towards an object as its content. But this is not obvious. Thought's requirement of an object might be nothing more than a causal requirement, that thought requires an object for its cause, where `cause' has no more character than what its contemporary analysis credits it with, and not with the rich range of Aristotle's analysis that includes final causality. 57 Long too is perhaps in a position to agree that thought requires an object, as the non-identity theorists do, but in his interpretation thought requires no object other than itself, which would rob the requirement of its attraction. The identity thesis may also be compatible with the intentionality of thought. Even the reflexive nature of thought on Long's interpretation of Parmenides may not be incompatible with the intentionality of thought: thought would have itself as its intentional object. Yet the identity of mind and Being in itself has no obvious appeal, certainly none for common sense, as a startingpoint of argumentation, and Long confesses that the `key challenge' for the defenders 58 of identity is its clarity and philosophical justification. Thus B3 as a statement of identity could have no appeal if it serves as a basic premise in Parmenides' argument, as it surely would if it completes the fragment of verse at the end of B2, where it would serve as the premise from which the unknowability or unthinkability and inexpressibility of non-being follows. These features of non-being in turn provide the reasons the way of `is not', as opposed to the way of `is', cannot be pursued as a way of inquiry, as that which is an `utterly unlearnable path', (B2.5-6). On the identity thesis these consequences would have to follow from the mere identity of Being and thought, and thus the argument from identity would look something like this. It is impossible to think of non-being. Why? Because there is no thought without Being. 12

59 Why? Because thought is the same as Being. (But why in the world think this!) Yet if identity is not among the primary premises, but a conclusion, how should it be argued for? And why do we find no trace of any such argument in the fragments or in the testimonia? One might object that common-sense appeal is nothing that Parmenides ever worried about, since surely he is a philosopher, if ever there was one, who had no fear of paradox. But philosophers always back up their puzzling reasoning or their counterintuitive conclusions, with an intuitively appealing starting-point in their argumentation, as Russell bears witness to, when he observes that `the point of philosophy is to start with something so simple as not to seem worth stating, and to end with something so 60 paradoxical that no one will believe it'. The paradox typically lies in the conclusions, not in the first premise, or among the primary premises, the place B3 may be located in Parmenides' thinking. Accordingly, B3 should be and is susceptible to an intentionalist reading, for example, in the translation of Barnes: `The same thing is both for thinking 61 of and for being'. Since thinking is always thinking of Being, thought's requirement of an object in the form of Being stems from thought's being directed towards an actual object. B3 then can provide Parmenides with the foundation of his argumentation in thought's intentionality, and it finds reinforcement in the statement of thought's intentionality in B8.34 in thought's being for the sake of Being. these lines: A credible translation of the whole passage B8.34-36 may be constructed along The same is for thinking and for the sake of which thought is. For not without Being, in which it is expressed, will you find thought. However we may gloss the troublesome relative clause,, of verse 35, it is clear from the passage that Parmenides speaks in an asymmetrical 13

fashion of thought and Being. Parmenides characterizes thought as dependent upon Being for its occurrence, since it is thought, not Being, that he treats as accompanying something else when he says of it that it cannot be found apart from Being. This, of course, does not rule out a mutual dependence, where Being, in turn, always accompanies thought and is never found without thought. Yet Parmenides never speaks in the surviving fragments of Being in such dependent terms. Instead, we have clear reason to suppose that it is independent in every way, `for it is not in need', (B8.33). The asymmetry between thought and Being, which Kahn is sensitive to, but which Long is not, and which amounts to thought's dependence upon Being, or its requirement of Being, fits in with the intentionality of thought. The occurrence of thought depends on its possession of an object, of its direction towards an object that serves as its intentional object. And this consideration may contribute to our understanding of the meaning of the difficult relative clause,, which has baffled so many of Parmenides' students. Tarán is exasperated with those many commentators who insist in translating the relative clause in such a way that thought must be `expressed in Being', since he finds that this could make no sense. Only by giving up this translation can there be any hope 62 of making sense of this difficult clause. He proposes as a sensible translation of the clause, `in what has been expressed', namely, what was expressed in the preceding verse 34,, `thought that it is', and thus thought is not found apart from Being. Long and Sedley offer a rendition in keeping with Tarán's translation with their rendering, `in what has been said', but they avoid the awkwardness of Tarán's assumption, that the reference goes back only to the preceding line, by their proposing that the reference is more general in its scope and takes in the whole range of the 63 `preceding arguments'. Schofield, who is a champion of non-identity, also renders the 14

clause in the same terms: `in all that has been said'. Yet what may count as the most natural reading of the relative clause, which Mourelatos labels the `usual translation', 64 may yet have a reasonable defense. The verb can signify nothing other than ways of speaking: `to tell of', `speak of', `make a statement' in court, or `to promise' and `to engage' oneself in marriage. In the passive voice it may mean `to be called' or `named', and LSJ cites Parmenides' use of the passive participle in the relative clause 65 of B8.35 as an instance of its meaning `to be expressed'. In the one other place the verb shows up in the surviving fragments of Parmenides, at B8.60, it clearly means `tell'. None the less, it is arguable that since thought cannot be found apart from Being, thought may be presumed to display or manifest itself in Being when it is thought of Being, or when it is directed in an intentional fashion towards Being. Only when there is thinking of Being is there at least genuine or veridical thought, and can such thought then have the occasion of making itself evident. In this way thought may be construed as `expressed in Being'. 66 The issue of identity cannot be settled by merely a debate over translations of the relevant passages. Each side can offer plausible translations, some more natural than others, but all perfectly credible, and the debate comes then to a stalemate over translations. The decision turns upon a philosophical consideration, not a philological: which of the lines of thought proposed for Parmenides by each of the contending parties provides the more philosophically satisfying result? This falls squarely on the side of the non-identity theorists. The identity theorists compel the goddess to argue 67 from the identity of thought and Being, which is thoroughly counter-intuitive, whereas their adversaries are in a position to allow her to argue from the intentionality of thought. Their argument would go something like this. It is impossible to think of non-being or what is not. Why? Because there is no thought without Being. Why? Because thought requires by its nature an object. It requires an object because it is always a thought of 15

something. From this plausible-sounding beginning, which yields the unthinkability of non-being, the goddess goes on in her bold speculation to draw her incredible conclusions, which leave her audience dumbfounded: reality is reduced to a single, eternal object, indivisible, uniform, and unchanging in every way. The identity theorists may counter with their own claim to philosophical plausibility when they argue that identity allows a clear place in Parmenides' metaphysics for thought, which he surely could not suppose lies outside the bounds of reality. This is true enough, but the decision turns on the philosophical cogency of the first steps in Parmenides' thinking, and this surely is provided by the non-identity theorists who avoid identity and are in a position to call upon the intentionality of thought as the foundation of Parmenides' argument. 68 16

BIBLIOGRAPHY Barnes, J. 1979a. The Presocratic Philosophers. vol. 1. London.. 1979b. `Parmenides and the Eleatic One,' Archiv für Geschichte der Philosophie 61. 1-21. Brentano, F. 1973 (German original 1874). Psychology from an Empirical Standpoint. English ed. L. L. McAlister. Trans. A. C. Rancurello, D. B. Terrell, L. L. McAlister. New York. Originally Psychologie vom empirischen Standpunkt. 2 vols. O. Kraus (ed.). Leipzig. Burnet, J. 1930 (original publication 1892). Early Greek Philosophy. 4th ed. London. Cornford, F. M. 1935. `A New Fragment of Parmenides', Classical Review 49. 122-123.. 1939. Plato and Parmenides. London. Burnyeat, M. 1982. `Idealism and Greek Philosophy: What Descartes Saw and Berkeley Missed', Philosophical Review 91. 3-40. Caston, V. 2001. `Concerning Traditions: Augustine and the Greeks on Intentionality'. In D. Perler (ed.), Ancient and Medieval Theories of Intentionality. Leiden. 23-48. Coxon, A. H. 1986. The Fragments of Parmenides. Phronesis, suppl. vol. 3. Assen. Curd, P. 1991. `Parmenidean Monism', Phronesis 36. 241-264.. 1998. The Legacy of Parmenides: Eleatic Monism and Later Presocratic Thought. Princeton. Diels, H. 1951. Die Fragmente der Vorsokratiker. 6th ed. rev. W. Kranz. 3 vols. Berlin. Fränkel, H. 1975a (German original 1951). Early Greek Poetry and Philosophy. Trans. M. Hadas and J. Willis. Oxford.. 1975b (German original 1955). `Studies in Parmenides'. In R. E. Allen and D. Furley (eds.), Studies in Presocratic Philosophy, vol. 2. London. 1-47. Originally `Parmenidesstudien' in Wege und Formen frühgriechischen Denkens. Gallop, D. 1979. ` Is or Is Not?', Monist 62. 61-80. 17

. 1984. Parmenides of Elea, Fragments: A Text, with an Introduction. Phoenix, suppl. vol. 18. Toronto. Graham, D. 1999. `Empedocles and Anaxagoras: Responses to Parmenides'. In A. A. Long (ed.), The Cambridge Companion to Early Greek Philosophy. Cambridge. 159-180. Guthrie, W. K. C. 1965. A History of Greek Philosophy, Vol. 2: The Presocratic Tradition from Parmenides to Democritus. Cambridge. Hussey, E. 1972. The Presocratics. London. Jaeger, W. 1947. The Theology of the Early Greek Philosophers. Trans. E. S. Robinson. Oxford. James, W. 1985 (original publication 1902). Varieties of Religious Experience: A Study in Human Nature. Cambridge, Mass. Kahn, C. H. 1968. `Review of Tarán, Gnomon 40. 123-133.. 1968/9. `The Thesis of Parmenides', Review of Metaphysics 22. 700-724. Kirk, G. S. and J. E. Raven. 1966. The Presocratic Philosophers: A Critical History with a Selection of Texts. 1st ed. Cambridge. Kirk, G. S., J. E. Raven, and M. Schofield. 1983. The Presocratic Philosophers: A Critical History with a Selection of Texts. 2nd ed. Cambridge. Long, A. A. 1996. `Parmenides on Thinking Being', Proceedings of the Boston Area Colloquium in Ancient Philosophy 12. 125-151. Mansfeld, J. 1964. Die Offenbarung des Parmenides und die menschliche Welt. Assen. Mourelatos, A. P. D. 1970. The Route of Parmenides: A Study of Word, Image, and Argument in the Fragments. New Haven. Owen, G. E. L. 1960. `Eleatic Questions', Classical Quarterly NS 10. 84-102. Russell, B. 1985 (original publication 1918). The Philosophy of Logical Atomism, D. Pears (ed.). LaSalle. 18

Phillips, E. D. 1955. `Parmenides on Thought and Being', Philosophical Review 64. 546-560. Searle, J. R. 1983. Intentionality. Cambridge. Sedley, D. 1999. `Parmenides and Melissus'. In A. A. Long (ed.), The Cambridge Companion to Early Greek Philosophy. Cambridge. 113-133. Tarán, L. 1965. Parmenides: A Text with Translation, and Critical Essays. Princeton. Vlastos, G. 1953. `Review of Zafiropoulo, L'école éléate', Gnomon 25. 166-169. West, M. L. 1971. Early Greek Philosophy and the Orient. Oxford. Woodbury, L. 1958. `Parmenides on Names', Harvard Studies in Classical Philology 63. 145-160. Williams, B. A. O. 1981. `Philosophy'. In M. I. Finley (ed.), The Legacy of Greece: A New Appraisal. Oxford. 202-255. Zeller, E. and W. Nestle. 1963 (Original publication 1919). Die Philosophie der Griechen in ihrer Geschichtlichen Entwicklung, 1. Tiel., 1. Abteilung, Vorsokratische Philosophie. Leipzig. Reprint by Georg Olms. 19

NOTES 1. The existential reading of is the most plausible of the possible renderings when it and the various forms from it are used for the subject of the goddess' lecture; for arguments against non-existential renderings, see Barnes 1979a: 158-161. Texts of and references to Parmenides are those of Diels/Kranz. For ease of reading iota subscripts replace iota adscripts. Unattributed translations are the author's. 2. Williams 1981: 204 offers this sort of definition of `idealism', which he maintains is a modern philosophical position never held by any ancient philosopher. For an elaboration upon Williams' observation, see Burnyeat 1982. 3. Vlastos 1953: 168. 4. Long 1996. 5. Sedley 1999: 120, whose brief comments are embedded in a general exposition of Parmenides and Melissus. German scholars, as Long 1996: 132 and n. 14 observes, have often translated B3 as a statement of identity, e.g. in the translation of Diels/Kranz, `Dasselbe ist Denken und Sein', even if their commentary states otherwise, of which Mansfeld 1964: 67-68 provides an example. 6. Kahn 1968/69: 720-724; 1968: 132-133. Others who may be ranked among identity theorists include Phillips 1955 and Hussey 1972: 94. Barnes 1979a: 170-171 does not identify thought and Being, but he does place Parmenides among the intellectual ancestors of Berkeley, since Parmenides believes that what can be thought of exists. 7. As Tarán 1965: 41 says, many scholars consider B3 `to contain the essence of Parmenides' philosophy'. Mourelatos 1970: xv refrains, however, from offering an interpretation of B3, as well as B6.1-2, because of their indisputable `syntactic ambiguity'. 8. Also Burnet 1930: 173 and Coxon 1986: 180. See Coxon for additional reasons for placing B3 at the end of B2. 20

9. Tarán 1965: 41. 10. Translation by Tarán 1965: 41. 11. Clement also preserves B3, which he evidently takes, like Plotinus, to be a statement of identity, Strom. 6.23. Proclus provides an inaccurate rendering of the line, in Prm. 1152. 12. Long 1996: n. 21. 13. Long 1996: 136. The bracketed material is Long's. 14. Coxon 1986: 209. 15. Kahn 1968/9: 722 also appeals for his interpretation of identity to B6.1, which he translates as `Cognition and statement must be what-is', '. 16. Kahn 1968/9: 723-724. Barnes 1979a: 158-159 questions the translation of as `know' by Kahn and others. 17. E.g. Long 1996: n. 14 18. Long 1996: 138. 19. Sedley 1999: 120. 20. Just as Vlastos 1953: 168 maintains. Long 1996: n. 13 and p. 146. 21. Long 1996: 146. 22. It is becoming more fashionable to deny the qualitative homogeneity of Being and to maintain that it is qualitatively heterogeneous and only existentially homogeneous. Nowadays probably the most prominent spokesman for this view is Curd: 1991 and 1998. Graham 1999 too has lately taken up a similar position, although cautiously he argues that Parmenides may be read as a pluralist and that the pluralists who succeeded him did so read him and could claim him in good faith as their predecessor. Mourelatos 1970: 130-133 also argues against the homogeneity of Being. Barnes 21

1979a and 1979b most likely initiated this revisionist interpretation of Parmenides, which attempts to overthrow the monism that has been the keystone of the traditional interpretation of Parmenides since the days of Plato and Aristotle. 23. Long 1996: 146. 24. Cornford 1933 and Woodbury 1958: 148-149, 153-155; see also Mourelatos 1970: 185-188. The dispute over the Greek of the fragment,, ' ', is usually limited to the first word: whether it should be, as it is here, or. 25. Plato perhaps provides another example of a necessary, but non-essential, property when in the Meno he has Socrates give as the definition of `shape', `alone of existing things that happens always to follow color' (75b). This may belong necessarily to shape, but it hardly seems to be essential to or to illuminate the nature of shape. 26. Long 1996: n. 22. 27. Long: 1996: 129-131, 141, 143. 28. West 1971: 218-226 who even maintains that the arguments of Parmenides are often `contrived and artificial' and that they merely lend `a show of logic' to beliefs that he largely came to through `direct perception, a mystical experience'. Fränkel 1975a, who also holds to the identity thesis (357, cf. 1975b: 34), believes that Parmenides `personally experienced the unio mystica with true Being' (366). 29. Long 1996: 131, 141. 30. Jaeger 1947: 107. Anaxagoras never describes as `divine', even though it is a thinking, living thing, which is responsible for the state of the cosmos (B12). 22

31. Mourelatos 1970: 44 and n. 108. He finds `conspicuously absent' the adjectives,, for `deathless', for `eternal', for `unaging', and or for `great' or `greatest'. Mourelatos allows that, `inviolate', is the only word Parmenides uses (B8.48) to describe Being which might carry a religious significance. 32. From Tennyson's letter to Mr. B. P. Blood, a portion of which James 1985 [1902]: 304 n. 3 records. 33. Vlastos 1953: 168. 34. Tarán 1965: 59 points out against those who have denied Parmenides' adherence to the so-called `tautology', `Being exists', that, as an implication of B8.36-37, he does allow for the construction. 35. Long 1996: 141-143 agrees with Vlastos, but he tries to strengthen the case by an appeal to the meaning of at B1.29, where he exploits and elaborates upon an analysis of the word by Coxon 1986: 168. On the basis of Coxon's reasoning, Long contends that Parmenides does not use metaphorically, but intends it to bear its literal traditional Homeric meaning of the `seat' of vital and psychological powers. Accordingly, Long takes to be a description of Being that designates it as a living, thinking thing. Sedley 1999: 120 adopts Long's interpretation of and gives it the strong translation of `mind' in B1.29. Most scholars, as Long acknowledges, take to be used metaphorically in verse 29, and he argues that Coxon's research shows that if Parmenides used the word metaphorically, it would be `completely anomalous' on his part. But if one considers Parmenides' clear claim to originality it would hardly be surprising if he were the first to use `heart' as a metaphor for the `central features' of a subject, as it has clearly come to be used across many languages. Although Coxon adheres to the non-identity thesis, Long is right to think that some of his comments reveal him to be highly sympathetic to the identity thesis: e.g. Coxon 1986: 181. 23

36. Sedley 1999: 120. 37. Kahn 1968/9: 722. 38. Owen 1960: n. 55 tries to discredit Vlastos' presumption that Parmenides must have found a place for thought within Being by appealing to Soph. 244c-d, 248d-249a, in which, Owen thinks, the Eleatic Stranger's comments imply that Parmenides neglects placing thought within reality by his failing to place language, the correlate of thought, within reality. Williams 1981: 224 concurs in Owen's judgement. See also Tarán 1965: 200 who holds that on a reading `without prejudice' of the Sophist it should be acknowledged that Plato was the first to pose the problem of the existence of language and thought. Although Kahn 1968: 133 argues for the identity of v and, he observes a further asymmetry between them in that the one `seems to take place in time' when the other is `eternally self-identical'. 39. Tarán 1965: 199-200 offers similar observations in his criticism of Vlastos; cf. Williams 1981: 224. Long 1996: 139 is well aware of this line of criticism. 40. Guthrie 1965: 40. 41. Perhaps a better translation of the opening of B6.1 is `It is necessary that that which is for speaking about and thinking of exists'. 42. Tarán 1965: 58. 43. Long 1996: 135-136. 44. Tarán 1965: 43. 45. Guthrie 1965: 14 and : 173 n. 2. Guthrie's translations are after those of Zeller 1963: 558 n. 1, `Dasselbe kann gedacht werden und sein', and Burnet 1930: 173 and n. 2, `for it is the same thing that can be thought and that can be'. Zeller is possibly responsible for the popularity of translating B3 so that it does not express the identity of thought and Being; see Tarán 1965: 43. Phillips 1955: 548-549, who ranks among the identity theorists, seems to allow that a datival rendering of the infinitives in B3 yields a 24

`literal rendering'. He rejects this rendering for one that identifies thought and Being because it gives the `natural sense', and he holds that no one would have taken the grammar of B3 to yield anything but a statement of identity if the identification had been between an `unexceptionable pair'. 46. Coxon 1986: 174, 180. Long 1996: n. 24 acknowledges the parallel between B3 and B2.2, but demurs without elaboration over Guthrie's insistence upon its exactitude. 47. Barnes 1979: 159. 48. A few examples of the translations of B3 by those who hold the non-identity thesis: Burnet 1930: 173: `for it is the same thing that can be thought and that can be'; Barnes 1979a: 157, `The same thing is both for thinking of and for being'; Gallop 1984: 57, `... because the same thing is there [ ] for thinking and for being'; Coxon 1986: 54: `for the same thing is for conceiving as is for being' (Coxon's frag. 4). Burnyeat 1982: 15-16 supports the non-identity reading of B3 and appeals to it to support his view that Parmenides does not identify Being and thought. Burnyeat argues that no philosopher from antiquity adopted idealism, not even Plotinus. Williams 1981: holds the same view, and also argues that Parmenides does not identify thought and Being. 49. Coxon 1986: 70 and Cornford 1939: 43. 50. E.g. Guthrie 1965: 39, Mourelatos 1970: 165 (`wherefore'), Barnes 1979a: 179, Kirk and Raven 1966: 277, Gallop 1984: 71, Diels/Kranz 1951. 51. Kirk, Raven, and Schofield 1983: 252. Since Schofield is responsible for the chapter on Parmenides in this edition, I refer to its comments on Parmenides as merely `Schofield'. Burnet 1930: 176 and n. 2; Simplicius Phys. 87.17. 52. Long 1996: n. 23 characterizes these sorts of translations as bringing out the `causal force' of the basis of thought. 53. Long 1996: 136. 25

54. Similarly, Kahn 1968/9: 721 with his translation of `the goal [or aim or motive]'. Guthrie 1965: 39 does not adopt this sort of causal reading, but allows for its possibility, which he makes clear when he explicates it as `the purpose or foundation of thought'. 55. Brentano 1973: 88-89. Searle 1983 provides one important contemporary account of intentionality. Caston 2001: 26 groups Parmenides with Gorgias and other sophists as having similar concerns over the intentionality of thought. 56. See Burnyeat 1982: 21-22. Williams 1981: 219-220 reads the esti of B3 existentially so that he translates B3 as `The same thing is there to be thought, and to be', and he appreciates B6.1 in the same fashion, `What is there to be spoken of and thought, must be'. Behind the phrase `is there to be', Williams finds the `primitive conception' that thought and language possess their content only because they touch and see what is. Accordingly, if there is nothing there, then there is nothing there to see or to touch and thus for Parmenides nothing to think or to name. 57. Coxon 1986: 209 very likely intends to limit `cause' to something like Aristotle's final or teleological cause. 58. Long 1996: 136. 59. Gallop 1979: 69 with n. 36 also points out that reading B3 as a statement of identity makes it difficult to appreciate it as a premise for B2.7-8. 60. Russell 1985: 53. 61. These are the words of Barnes 1979: 157 but my italics; cf. the rendering of B3 by Mourelatos 1970: 75 n. 4, and for a paraphrase of B3 along similar lines, see Burnyeat 1982: 15-16, where the italics are his: `... it is one and the same thing which is there for us to think of and is there to be: thought requires an object, distinct from itself, and that object... must actually exist'. 62. Tarán 1965: 128. 26

63. Long 1996: 136; Sedley 1999: 120. Tarán 1965: 128 acknowledges that he is following the lead of Albertelli. Mourelatos 1970: 171 n. 19 finds this translation to be `not very plausible'. 64. Mourelatos 1970: 170. 65. LSJ s. v. I. See Mourelatos 1970: 170-174 for his interpretation of the meaning of the verb in verse 35 and his justification for his translating the relative clause as `to which it stands committed'. 66. The translation of Burnet 1930: 176, `For you cannot find thought without something that is, as to which it is uttered', and the similar translation of Cornford 1939: 43 maintain that thought is expressed only with respect to Being. Mourelatos 1970: 170 offers a similar translation with his rendition, `to which it stands committed'. These translations are on the right track, and Mourelatos comes close to expressing an interpretation that treats thought as intentional when in his commentary on the relative clause he roughs out its sense as `to which (scil. the ) it (scil. ) refers' or `to which it is addressed'. In this context the language of `reference' is virtually that of `intentionality'. 67. Long 1996: 138 appears to maintain this position. 68. The author wishes to express his gratitude to... for helpful comments on a draft of this paper. 27