Printer Friendly

Anaxagoras as probability theorist.

A serious sticking point in Anaxagoras' cosmogony is a general scholarly consensus that it grinds to a halt at the initial cosmic event, and thus cannot actually begin. While his hypothesis indicates the difference between cosmogony (the fact of Motion) and its pre-existing condition (the state of Rest), it does not explain how it was possible for Mind (nous) to begin moving. This is especially odd given his explicit endorsement of Parmenides' absolute prohibition on Change, itself conceived as a transition between is and is-not. The transition internal to nous would then be something both required and impossible; to adapt Parmenides on the point, if nothing is happening then nothing will ever happen, and, by parity of reasoning, if something is happening then something has always been happening. (1) But then it is unclear how any cosmogony is possible, particularly if Mind has the cosmogonic function Anaxagoras ascribes to it, namely of somehow (2) bringing about the motion which in turn results in the phenomenal world. McKirahan, for instance, implies that Anaxagoras was quite deeply confused and ought to have been somewhat embarrassed by his having ultimately dispensed with the principle of sufficient reason, (3) and it seems to be rather unlikely that Anaxagoras would not so much as have heard about Anaximander's work in general and of this central cosmological postulate in particular - but even if the unlikely should somehow turn out to be true, there can be little question, I think, that Anaxagoras' consciously working within Parmenidean constraints implies that in fact he deeply respected the principle of sufficient reason in its characteristically Parmenidean formulation, namely as 'Principle(s) of Reason Sufficient for Reality and Thought' which quite explicitly contradicted ordinary common sense. The view that Anaximander made "a brilliant leap into the realms of the mathematical and the a priori" (KRS 134) by means of his cosmological postulate (4) describes an intellectual feat which is in its way fully consistent with Parmenidean logic, and Kahn's judgement that Anaximander's successors reverted from his bold "use of such speculative reasons in radical contrast with the evidence of common sense ... to more solid considerations to keep the earth in its place" (1960: 78) (5) extends to Anaxagoras only insofar as Mind somehow or other initially effects the cosmic whirl at the centre of which lies Earth.

My present aim is to argue that the available primary evidence (6) is consistent with Anaxagoras' having committed no such rational vice. After a preliminary discussion of the general philosophical context within which Anaxagoras' thought is to be comprehended, I begin by distinguishing between noetic stuff in general (nous, or 'something') and nous as a definite something (hereafter Nous, or 'some thing'); this distinction reasonably supports the hypothesis that it is the latter rather than the former which is cosmogonically efficacious. In the second major section I argue that the mathematics of probability provides a scientifically respectable explanatory mechanism for the transition from nous to Nous. The final section argues that the transition between nous and Nous clearly violates Parmenides denial of the possibility of change only if nous has not always already been in motion, a consequent which is with much plausibility denied given the fragmentary indications of pre-cosmic, variously sized noetic agglomerations.

[alpha]: The Philosophical context for Anaxagoras

A brief, preliminary discussion of the major philosophical influences upon Anaxagoras will be quite helpful for the sake of the general aim of this paper. That the towering figure of the time was Parmenides is clear from Anaxagoras F19/B17 and T2/B10. In F19 Anaxagoras provides a summary, if Waterfield's ordering is even close to being accurate, of what has been stated previously (F1/B1-F18/B19): "Greek usage of the words 'generation' and 'destruction' is incorrect. Nothing is generated or destroyed; things are combined from already existing things and dispersed [into them]"; (7) T2 conveys a similar thought whether or not it is a genuine fragment: "For how... could hair come from not-hair or flesh from not-flesh?" (8) The whole of Parmenides B8 is relevant to these two Anaxagorean remains, but B8.20-21 are especially so, (9) and among the particularly significant precursors to Parmenides in this regard are Heraclitus and Xenophanes. Heraclitus appeals to the Logos common to all things (B1, B2, B50, B89), but this logos is especially relevant to nous (B104), which alone can comprehend it (B2), and furthermore logos amounts to a single rule governing all cosmic change; qua rule, the 'harmony of opposites' (B51) does not change, evidently, but what is ruled no less evidently does (B84a); finally, his observation that, in the case of a circle, beginning and end coincide (B103) finds similar expression in Parmenides B5. (10) Xenophanes' remarks on the completeness (B24) and motionlessness (B26) of the god is echoed in Parmenides B8.4 and B8.29-30. (11) If, finally, Pythagoras is to be included among these precursors, we may observe that the conceptions of the kinship of all living nature (14A8a; in Parmenides B8.34-37 'living' is refined to 'rational') and those of sameness and unity (evident from Pythagorean pebble arithmetic, 58B28) recur in Parmenides B8.5-6 and B8.22-25 respectively. (12) In general, it is not difficult to comprehend Parmenides as arguing about what-is from a stable, logically coherent point of view, one from which all previous cosmic speculation could be shown to have been deficient precisely in having uncritically accepted the reality of change: we might reasonably say that, as far as Parmenides is concerned, while harmony is ultimate, the ultimate harmony is unity.

The question of influence between Anaxagoras and Empedocles, while important, (13) is not immediately relevant to my present concern, and so I shall note only that the same thought expressed in Anaxagoras F19/B17 and T2/B10 is present in Empedocles B17.30, B12, and B8. (14) Democritus B9 shows a similar sort of response to Parmenides. (15) Each of these three 'pluralist' responses to Parmenidean 'monism' sought, as is generally accepted among scholars, to take account of Parmenides constraints upon reality while preserving, in their own ways, the phenomenon of change: on one hand they invoke somatic things (Empedoclean 'roots', Anaxagorean 'portions' and 'seeds', Democritean 'atoms'), on the other hand they appeal to something causally responsible for the assembly and disassembly of these fundamental units of changeless reality (Empedoclean Love/Strife, Anaxagorean Mind, Democritean weight/collision). This causally efficacious factor is in each case to be construed most basically in a mechanical rather than a teleological manner, and consequently, while the evident phenomenon of change in general is of course not a 'thing' in the way that 'roots', 'portions' or 'seeds', and 'atoms' are, it hardly follows from this that the physical conditions responsible for change were themselves simply nothing at all.

It might be thought, as it was by Guthrie (1965: 277) over forty years ago, that Anaxagoras was, unlike Empedocles and Democritus, pressing toward immaterialism but lacked the vocabulary to do so. This is at odds with some important surviving evidence, however: Anaxagoras' explanation concerning the requirement for noetic purity is that if nous were impure then it would be like the 'everything in everything' in the sense that it would include among its ingredients portions of everything else, and "the things mixed with it would stop it ruling anything in the way it does by being on its own and by itself" (F10/B12.6-8). This suggests that unless X and Y are significantly different, X cannot rule or act upon Y; but since this difference cannot be generic (for if it were there could be no causal relationship between them), it must be specific. X and Y are generically identical in that both are material, then, but specifically different in that X is capable of acting upon Y but not vice versa. (16) Incidentally, this provides an excellent example of reasoning from effect to cause, as illustrated by "one is bound to suppose" (chre dokein) in F4/B4a and F5/B4b. (17)

In addition, however, there are two sorts of extrinsic reasons for the implausibility of the Guthrie-type view, both of which are independent of the natural consonance between materialism and dysteleology. One kind of reason is a matter of the philosophical record: firstly, Plato has 'Socrates' complain that Anaxagoras' nous explains the 'how' rather than the 'why' things (Phaedo 97b ff.); secondly, in Metaphysics 1.4, 985a18-23, Aristotle rejects Anaxagoras' nous as no more than "a deus ex machina for his cosmogony, and when he finds it impossible to explain why something necessarily is as it is, he drags mind in, while elsewhere he uses anything other than mind to explain how things happen" (W. 322); and thirdly, in Parts of Animals 4.10, 687a8 ff., Aristotle criticizes Anaxagoras' view that we see and grasp things because we have eyes and hands (rather than, as Aristotle himself holds, that we have eyes and hands in order to see and grasp them). We cannot reasonably ask for a clearer report of Anaxagorean dysteleology--and, I would add, the strictly emergent character of mentality from physicality, an extension easily inferred from Plato's antimaterialist argument in Laws 10 that soul pre-exists body.

The other sort of reason I would adduce is rooted in common human experience: even if Anaxagoras' own parents had decided to have a child, it is hardly credible that they decided to have just him, and even less so that he himself decided to be born; we may add to these considerations, if we wish, that exactly like anyone else he would not have been able to recall the very first decision he made. The natural inference from ordinary observation of human life, as opposed to metaphysical speculation, then, is that mental things--e.g., wishes, intentions, decisions, and the like--are ultimately dependent upon something somatic, even if immediately dependent upon something noetic. Appealing to Mind's 'decision' to begin cosmogony (18) does not, therefore, solve the problem at hand but merely shifts it elsewhere: now we will need to find some sufficient reason for Mind's decision at time T rather than at either [T.sub.-1] or [T.sub.+1] (cf. McKirahan's criticism above). It is better, I conclude, to concentrate our attention upon the initiation of motion without invoking any decision of nous to do so, (19) since noetic physical motion is necessary, as I shall argue in greater detail below, for its analogous motion in terms of higher cognition (wishing, intending, deciding, and the like--including, crucially, the apparent purposiveness of nature in general).

That the physical conditions responsible for change were themselves not nothing but something (even if not 'some thing') is not hard to see if we recall the four basic axioms of Parmenidean thought: (1) Being is, (2) Nonbeing is not, (3) Something cannot become (or come from) nothing, and (4) Nothing cannot become (or come from) something. Of these, the first two are correlated as general metaphysical principles; the second two are correlated as different ways of expressing the inviolability of the metaphysical chasm separating generic Being (to einai) from generic Non-being (to me einai). (20) In B2.3-5, Parmenides distinguishes between two general ways of approaching things: "There is the way that it is and it cannot not be ... Then there is the way that it is not and that it must not be"; (21) the former is the Way of Truth whereas the latter is "an altogether misguided route" (B2.6) taken by "mortals [who], knowing nothing,/Stray two-headed" (B3.4-5). The result of conjoining these four principles is that whatever is, is always (or, more strongly put, necessarily) existent, and that whatever is not, is never (or, more strongly put, impossibly) so; (22) from this Parmenides deduces that change is ruled out, metaphysically speaking, as a non-starter. This much is generally uncontentious as a description of his fundamental position regarding the nature of reality. (23)

With this philosophical background in place, we may now proceed to the Anaxagorean distinction between something in general and some particular thing. This distinction respects Parmenides' ontological constraints, themselves indicative of the principle of sufficient reason if anything is. Parmenides' set of requirements for the actual existence of anything clearly applies both to 'something' and to 'some thing', but while the actual existence of any given thing can conceivably arise only given the antecedent actual existence of something, the actual existence of something is entirely compatible with the actual non-existence of any given thing, (24) although the actual existence of something is incompatible with the potential existence of any given thing. (25) There is, on this analysis, a one-way entailment from 'some thing' to 'something': the actual existence of a thing presumes that of something, but not vice versa. This distinction, which maps with ease onto nous-without-article ('something') and nous-with-article ('some thing') respectively, has to the best of my knowledge not yet been considered relative to the problem of cosmogony in Anaxagoras, and suggests a way of resolving that problem satisfactorily.

[beta]: From nous to Nous

I turn now to Anaxagoras' cosmogony. The fundamental problem facing him in this regard has generally been construed as that of deriving motion from not-motion, or rest. (26) Since any such derivation is plainly inconceivable, as Empedocles and Leucippus/Democritus understood, (27) it appears that Anaxagoras' respect for Parmenides' rules was incomplete, for even if motion is not 'some thing' but only 'something' (rather than 'nothing'), he cannot at all reasonably have thought motion to have simply begun just like that without immediately having abandoned Parmenides. The overall task of this section, accordingly, is to establish the proposition that there is an important ontological distinction between nous ('something') and Nous ('a discrete, definite thing in its own right'). This difference is marked by the absence (F9/B11 and F10/B12) or presence (F13/B14 and F14/B13), respectively, of the definite article accompanying nous. Especially illuminating in this respect, given W.'s ordering of the remains available to us, is the resulting cosmic narrative: Anaxagoras opens with general remarks about the physical properties and processes of pre-cosmic somatic stuff, including an initial statement concerning the separation of related opposites from the original 'everything in everything else' (F1-F8), then turns to a variety of general remarks on mind (F9 and F10), (28) after which for the first time he links nous with an article (F13 and F14); the consequence of a nous thus stabilized somewhere or other in the 'everything in everything else' is that the actual phenomenal world can begin to be articulated. (29) An especially important phase within this cosmic narrative is F10, which records the situation where nous, while present, is not yet capable of actually having an effect upon the pre-cosmic 'everything in everything'; in this regard, W.'s placement of F10 between F9 and F13 seems highly plausible as indicating the cosmogonically required transition from nous-in-general to the nous in particular, i.e. Nous, a transition whose completion is recorded in F13. Put another way, F1-F8 constitute Anaxagoras' general description of pre-cosmic somatic reality, F9 describes nous as the exception such that, for instance, if X has among its attributes being pure, being somewhere but not everywhere, being limitless, being independent, and so forth, then X is noetic rather than somatic (and, by contrast, anything impure, everywhere, limited, dependent, and so forth is somatic rather than noetic--but we must keep in mind that nous is ultimately a particular sort of body), F10 subsequently synthesizes F9 with F1-F8 such that the nous in F13 is rendered possible, and, finally, F13-F14 constitute his actually held cosmogonic theory: until nous somehow becomes Nous, cosmogony cannot actually begin.

As mentioned above, F10 sufficiently indicates that nous is to be taken as conceptually and ontologically sui generis: "Everything else has a portion of everything, but Mind is limitless and independent; it is mixed with nothing, but is on its own and by itself." (30) Anything other than nous is a mixture of different kinds of things; (31) a particular thing--a given bone, for instance--is, in other words, actually a combination of 'the seeds of all things'. (32) The bone's preponderant ingredient is 'bone', of course, since otherwise the particular bone would not even exist (and hence could never be identified by anyone as a bone), but it also includes ingredients of all things besides 'bone', for instance 'cloud', 'stone', 'hair', 'wood', and 'flesh'. The same point evidently holds for the unperceived (because dominated) ingredient 'hair' within the perceived (because dominant) ingredient 'bone' in the perceived bone: the 'submerged' or dominated ingredient 'hair' has as its predominant (but not exclusive) ingredient 'hair', but any one of the ingredients, in turn, of 'hair'--for instance 'cloud', 'flesh', 'wood', even 'bone' (since nothing clearly precludes its inclusion at a level ontologically inferior to its superior, perceived level--will be doubly 'submerged' or dominated in the perceived bone. In principle this analysis has no limit, (33) and thus we see not only that one of the reasons there are bones in the first place is that the ingredient 'bone' ends up dominating every one of its constitutive subordinates, but in addition that, since each of the perceived bone's infinity of constitutive subordinates has some --perhaps, at its own level, even dominant--portion of 'bone', and since there is no smallest unit of ontological analysis to be reached in the case of these emergent 'mixtures', the ingredient of 'bone' is itself present to infinity within any given bone.

The case is quite different with any given portion of nous, not because nous has some smallest unit of existence or analysis, but because however deeply one cares to analyse it, the 'dominating' and the 'dominated' elements are identical in terms of their fundamental nature, namely as being the ultimate cause of motion and order in things. Since, however, the charge of irrationality with which this paper opened concerns nous's being the source of motion in things rather than with the order present among them, what is needed is an account of how nous is the source of motion, not in 'every thing' but rather in 'everything' insofar as the latter is, as indicated above, necessary for the existence of any identifiable physical object in the first place. Given an original 'everything', there is clearly 'no [definite] thing' yet, and hence what Anaxagoras needs to be able to explain, if he is not to have violated the principle of sufficient reason, is how nous can mix with 'everything' so as to produce 'every thing' that actually exists, has ever existed, and will ever exist. The operation of 'mixing' will be impossible, or at least extraordinarily difficult to conceive, if nous is 'something' but not yet 'some thing'; on the other hand, if nous has always been the nous, i.e. without a prehistory of its own simply as nous, then, given that nous is said to be "on its own and by itself" (B12.2-3), any reasonable account for the specific timing of a genuine, historically original cosmogonic act would indeed be doomed. Whether or not (and, if so, how) 'something' might become 'some thing', then, is crucial for Mind's own becoming 'some thing' from having previously been no more than 'something'; without its having somehow or other actually become 'some thing, I cannot conceive how it might have any effect upon on anything at all so as to have managed to produce even a single 'thing' other than itself.

The claim needing justification at this juncture is that it is only when nous has become Nous that it has become cosmogonically efficacious; F10/B12.8 is relevant to this point, (34) but this line does not, as far as I can see, rule out a causal kind of relationship between generically identical but specifically different kinds of thing. Such a relationship can be regarded as holding between unorganized mind-stuff (nous) and organized mind (Nous, or nous with article), specifically different from each other in terms of the absence or presence of organization (or possibly the degree of organization) rather than in terms of whatever it is that is organized. If this hypothesis is cogent, any requirement that nous 'decided' or 'wished' to begin rotating all of a sudden is without sufficient warrant, and furthermore there will be no justification at all for supposing that it did so in order to bring anything else about in the antecedently existing original 'everything in everything else'. (35)

The most serious difficulty for this general interpretation is unquestionably A45.27-29, the only surviving testimonial evidence suggesting a teleological dimension to Anaxagorean cosmological thought: "Anaxagoras seemed to say that when all things had been together and at rest for an unlimited time until then, the Cosmos-making Mind, wishing to separate off the forms which he calls homeomeries [sic], produced motion in them" (tr. Lesher 129, emphasis mine). (36) The whiff of goal-directedness detectable here arises precisely from Mind's "wishing" (bouletheis) to do something, an obvious impossibility if Mind is itself the product of the same mechanical motion it is said to induce in the initial 'everything other than Mind'. However, "seemed" (edokei), to judge from its placement at the beginning of a new clause, seems to govern also the participle 'wishing', a judgement which finds support in two related respects: first, every other element of this testimonial is independently confirmed by independent fragmentary evidence; second, this specific element alone is not similarly confirmed. Simplicius' use of the imperfect tense here is highly significant precisely because it is relatively rare in his reporting of Anaxagorean thought, and it can therefore plausibly be taken as his means of establishing a distance between himself and what he is reporting; that is, the imperfect in this case carries no implication that he himself thought that what he was reporting was any more than an accurate report--if he did, he would have used the present active indicative dokei ('it seems [to me]') instead, as he regularly does. The implication is rather that someone else thought Mind "wished" to do something, and we do not have to look far at all to identify him: it is the teleologist upon whose Physics Simplicius is here commenting. (37) But as we have seen, there is no reason beyond this particular report that Anaxagoras was a closet teleologist--if anything, the available evidence strongly supports the view that he was a confirmed dystelelogist.

If, then, the organizing principle of the world is not itself eternal, something must have occurred within whatever is ultimately responsible for cosmic order. Taking our cue from Anaxagoras' theory of the articulation of specific medium-sized objects--that X is predominantly but not exclusively X--out of the original 'everything in everything' by means of the nous, this latter was itself previously nous-in-general. If, moreover, the basic characteristic of nous, whether as an unconstituted 'something' or as a constituted 'some thing', is to be able to produce motion, then its ontological status as 'something' or 'some thing' is strictly irrelevant; whatever motion it can produce will be the same in kind regardless of its extent, the extent of motion mapping directly onto the extent of nous itself. If so, then there will have been motion, ineffectual as it was for cosmogonic purposes, wherever nous existed in the pre-cosmic 'everything'; that is, the question concerning which sort of nous is causally responsible for the order of all things and which is not, or which is directly and which is only indirectly so, can be plausibly answered by noting that there is no fundamental inconsistency in supposing that nous as 'something' produces order in all things indirectly by first producing order within nous itself such that only subsequently does it become 'some thing' capable of producing the order evident in all things other than nous. This view is supported by Anaxagoras' having used the aorist diekosmese in B12.18, denoting a completed action but without reference to how long the action took to complete. Greater precision regarding Anaxagoras' use of the aorist in this case is not easy, but I would suggest as especially sensible candidates 'ingressively', 'resultatively', or possibly both. (38) If the former alone was his intention, then we can say no more than that at some definite point in time ('the beginning') nous began to govern; if the latter alone, then we are directed to the effects of that governance, namely the rotation. I do not see how a combination of these senses can definitively be ruled out in this ambiguous case, but in any event, Mind's ruling anything other than itself implies a previous rule of itself, and if its self-rule was not intentional, as suggested above, then all that we can confidently infer on the point is that the self-rule of 'the nous' has as its chronological antecedent 'something' which was capable of resulting in such self-rule without itself already being self-ruled. The establishment of self-rule, if so, would coincide exactly with the initial (but blind) formation of 'the Mind' from generic 'Mind-stuff'. That the ordering capacity of nous had a temporal beginning, then, would indicate that it has not done so from eternity even though it has by 'the beginning' been completed.

This is not the first time the aorist was used in this fragment. Five lines previously (B12.13) we read that Mind ruled or governed (ekratesen) the rotation at the cosmic beginning. The significance of this, it seems to me, is that while Mind as 'something' eventually does attain success, it does not do so merely qua 'something', since in that case cosmogony will clearly have regressed infinitely into the past, any 'beginning' of an infinitely remote past being a contradiction in terms. Thus, if there is to be a beginning, nous as 'something' must somehow become nous as 'some thing' which will thereby have acquired the very possibility of setting the 'everything' in motion, meaning that if it is the nature of the mind to produce motion in that 'everything' other than itself, it is in effect the nature of Nous as 'some thing' emergent from its only and pure ingredient, generic mind-stuff. It therefore goes beyond the available evidence to hold that the resulting organized cosmic system was foreseen (let alone intended) by nous before that system was actually brought about without reference to that standardly presumed 'foresight' on the part of mind; all we can safely say is that the cosmic result was sufficiently stable for its variety of 'separated' component 'things' to somehow persevere and, in some cases, even to reproduce. For B12.11-12 states that "Mind rules every animate creature", (39) but since ge seems in this clause to be either restrictive or emphatic, and since according to B12.22-25 "the dense is being separated off from the rare and the hot from the cold and the bright from the dark and the wet from the dry", (40) and further since it stretches credulity considerably to suppose that such things as dense, rare, and so forth are alive (they might, however, be 'powers' (41)), it is preferable to select that interpretation which is very easily grounded in ordinary human experience, i.e. to construe ge as emphatic. Finally, Anaxagoras says in B12.21-22 that "this rotation caused the separating off"; (42) but immediately prior to this clause he refers to future results as well as present and past ones; (43) we might have expected to find the imperfect here, but instead Anaxagoras again uses the aorist. The general conclusion to be drawn from the foregoing, I suggest, is that it was the motion of the nous, first clearly identified as such in F13/B14, which had successfully been transmitted to the non-noetic 'everything in everything else', and which thereby resulted in the cosmic 'separating out' of all distinct physical things.

This is an opportune occasion to examine the suspiciously corrupt-looking first five words of F13/B14. (44) As the first three phrases of this fragment were printed by DK, [TEXT NOT REPRODUCIBLE IN ASCII]. (45) These phrases are rendered by McKirahan as follows: "Mind, which always is, is very much even now where all other things are too." (46) It is noteworthy that no significant variants are recorded in the manuscripts of F13; Simplicius read [TEXT NOT REPRODUCIBLE IN ASCII]..." (47) McKirahan regards Simplicius' reading as "[one that] does not make good sense here" (199 n.12), and, provided that hosa is to be construed with nous, his judgement is not to be disputed. (48) But the emendation proposed by DK supposes that all known manuscripts are corrupt in precisely the same way at precisely the same textual location. While not impossible, of course, it might very reasonably give one pause, as it has recently Sider. I would suggest, on the basis of what has been argued so far, that the text as it has been transmitted to us might with at least some plausibility be retained. For in the first place, since hosa is plural and neuter, it would have to agree with its related noun, but this need not be--and as the text stands it cannot be--nous: it does agree, however, in precisely these respects with ta alla panta, suggesting that hosa, in every bit as jarring a grammatical disagreement with nous as Anaxagoras held the original state of things to have been, hangs in suspension until ta alla panta eight words later (immediately preceded by hina kai) recalls and completes it. Putting these results together, we get something intelligible as the sentence's syntactic skeleton: "and the Mind ... also exists wherever all other things are." Proceeding to fill in the remainder of this part of the fragment (including hina kai), and presuming Anaxagorean respect for Parmenides' requirements, esti te karta kai nun estin is not unintelligibly rendered as 'it really (49) is, i.e. it is now'. Second, deleting DK's commas after panta and periechonti, we get a sensible result: "wherever all other things are in the surrounding mass" (after KRS) and "both in the things joined together and the things that have been separated". (50) Third, by virtue of the careful placement of hosa right after nous, and supposing that ta alla panta completes the thought launched by hosa, we may now observe the manner in which ta alla panta completes hosa: it answers the originally implied question 'How many of these distinct mind-things exist?' Anaxagoras' answer, as recorded by Simplicius, is: 'As many of them as there are distinct physical things.' This deliberately vague answer is actually recorded in the previously cited F3/B7: "it is impossible to know ... the number of [physical] things that are being separated out." Thus, assembling and repunctuating the foregoing results, we get the following: 'And the Mind (how many of them?) --it really is, i.e. it is now--also exists wherever all other things are in the surrounding mass, (51) both in the things joined together and the things that have been separated." Whatever demerits this suggestion may have, it seems to have some not insignificant strengths: in the first place, it allows us to retain Simplicius' text as it stands in all the manuscripts, thereby nullifying any requirement to emend it; and in the second place, it avoids attributing to Anaxagoras the unintelligibility noted above by McKirahan, which should be avoided unless the cost of doing so is simply too high. If his Greek was originally mangled so as to best reflect his understanding of the pre-cosmic ('mangled'?) situation, then any objection that this suggestion 'mangles the Greek' loses whatever force it might entirely reasonably otherwise have had.

[gamma]: The Probability of Cosmogony

If it was not nous-in-general but the nous in particular which was cosmogonically efficacious, then what we now need is a plausible account of how the latter might conceivably have arisen from the former. This particular 'arising', like any other, can be explained in the way that scientists typically account for things, namely by reference to general laws of mechanistic and descriptive physics which are framed as economically as possible, economy in explanation being a virtue for anyone seeking to explain anything with any hope of actually being explained. We have already seen evidence of Anaxagoras' skepticism regarding human knowledge in F3 above; this skepticism would reasonably apply also to the general laws of physics, in consequence of which those very laws cannot be framed in absolutely necessary terms but rather in the more limited terms of likelihood or probability--in effect, exactly what Aristotle would later call 'hypothetical necessity': if X will exist, then Y must first exist, there being no necessity that X actually exist. (52) By parity of reasoning, if X now exists, then Y must first have existed, although there is no necessity that X now exist. Applied to the general laws of physics, we can similarly reason 'backwards', as it were, to what must have been so given that what is so is in fact so. Thus, if the world is organized then there must be some principle of organization, and if it has been organized then that principle itself cannot have been eternally efficacious. The alternative, that it has always been organized by means of an eternally efficacious principle presumes the sort of knowledge ruled out by F3, which explains why Anaxagoras can consistently suppose no more than that it has not always been organized. This result, however, entails a view that the world is fundamentally contingent, and the best way to express this ontological contingency, it seems to me, is probabilistically. Such a theoretical possibility in the case of Anaxagoras may initially seem surprising, but it will seem less so if, as I have previously argued, the concept of probability as plausibility ('the likelihood of a proposition's being true') was in circulation by his time: the more a particular claim was thought to 'make sense' (however 'making sense' is to be more specifically interpreted), (53) the more it was thought likely to be true. (54) What Anaxagoras does, apparently for the first time in the history of western philosophy, is to extend the notion of probability from the subjective realm of beliefs concerning the facts to the objective realm of the facts themselves.

That Anaxagoras actually made use of probability theory can be seen by inspecting his theory of matter as a mixture of seeds and portions. Any medium-sized perceptible object X is not exclusively but predominantly X, we read in F4/B4a, F5/B4b, and especially F10/B12.28-30: "each [perceptible] item is and was most distinctly those ingredients which predominate in its mixture." (55) Conspicuously absent from this account of physical objects is anything remotely resembling a teleological explanation, as we have already seen Plato's Socrates and Aristotle complaining about: Anaxagoras explains how, but not why, there are any such objects. If the general answer to the 'why' question is some variation on 'it's better that way', then Anaxagoras cannot be said to have held that it's better (without therefore implying that it's worse), but only that it is, in fact, the way it is. A teleologically minded natural philosopher will of course object, but Anaxagoras will legitimately be able to counter that, strictly speaking, the inference from existence to goodness goes beyond any readily available evidence. The only plausible exception would be knowledge itself, which is reasonably enough construed as something good; but this is undermined by the view that human knowledge is limited (F3) and by human sensation's being so weak as to be "incapable of discerning the truth" (F20/B21). This does not imply that we are capable of discerning the truth (provided that we use something other than the senses), but rather that the truth of things is actually beyond our ability to confirm definitively. (56) The right sequence, then, as far as Anaxagoras is concerned, is: motion-order-knowledge-goodness; on this analysis teleology is apparent rather than real, the result of rather than the reason why there is mechanical motion in the first place.

This has in no way explained any supposed derivation of motion from rest, however; at best it can explain how, once moving, the nous is able to begin causing motion in anything other than itself. But if an account similar to that concerning perceptible items applies to nous itself, i.e. both that there is no 'why' for Nous but instead only a 'that' and therefore a 'how', it remains to show how this explanatory 'how' in the case of Nous does not violate the principle of sufficient reason. It will, I take it, be unreasonable to demand that a non-teleologist include final causes in order to satisfy the criterion for legitimate acceptance of this principle. If the foregoing account accurately represents Anaxagoras' thinking, he recognized that genuine cosmogony was ruled out by any appeal to mechanical inevitability, but this left open the middle course of non-teleological probability: arguing 'forwards' from explanans to explanandum, assuming a specific ontological possibility (the nous) and as much time as is needed, it will have seemed a reasonable prediction (even if not one with absolute certainty) that nous will eventually become established as Nous, thereby having acquired the power necessary to begin actually moving the non-noetic 'everything in everything else'; arguing 'backwards', on the other hand, from the observed facts to the precondition of those facts, or the 'hypothetical necessity' given them, no other possibility will satisfy a non-teleologist cognitively. If so, then the world as we know it, according to Anaxagoras, was neither necessary nor merely possible (bare possibility itself being inherently insufficient for actual generation) but instead probable. Everything probable is possible, of course, but not everything possible is probable; the probable is that subset of the possible in which the conditions for change (in the broadest sense of the term) are likely to obtain. Since nothing can properly be said to be likely to happen if its chance of occurrence is counterbalanced by its chance of non-occurrence, X can be likely to happen only if the chance of occurrence outweigh the chance of non-occurrence, however slightly. That cosmogony was always more likely than not is suggested by Anaxagoras' hypothesis that nous was present--not everywhere but here and there, not all portions having the same sizes but in greater and smaller amounts--from everlasting in the original 'everything in everything'. All that would be needed to tip the cosmogonic scale is for a sufficiently large agglomeration of nous to constitute Nous and thereby to begin moving the pre-existing cosmic body. The possibility of such an agglomeration, however, clearly requires that nous be in motion already.

This study opened with the sticking point involving the impossible requirement that motion be derived from rest. That such a derivation is impossible is clear, but that Anaxagoras required it has generally seemed considerably clearer than it actually is--more boldly put, the likelihood is that, as far has the evidence goes, there has always been motion. If successful, then the characterization of him as a post-Parmenidean presocratic philosopher who ultimately failed by having violated the principle of sufficient reason constitutes a significant misrepresentation in need of correction. As has been standard operating procedure above, we will need, for the sake of establishing the central thesis of this section, to provisionally grant him his cosmogonic thesis, then to determine what that thesis presupposes, ensuring that the evidence reasonably supports attributing to him the relevant presupposition(s). The present case is complicated by the fact that the relevant evidence is indirect rather than direct: we cannot point to the linguistic device of the presence or absence of the definite article to support a distinction between nous as 'something' and Nous as 'some thing', but instead must be content to begin with F10/B12.27-28, which prima facie does not seem to be relevant to the proposition concerning the everlastingness of motion. In this passage we read that "wherever it is found, in larger or smaller amounts, mind is always identical." (57) Since in this case nous takes no article, we may infer that precosmically it existed in different sizes (or volumes), but that not one of these sizes was in fact sufficient for cosmogonic purposes.

If Anaxagoras was serious about cosmogony, the question concerning how any pre-cosmic size of nous might become sufficient to constitute Nous must be answered. But if the pre-cosmic 'everything' (including nous) was at rest, not only would no 'becoming' be possible, even the mention of differences in noetic size would have been pointless. For either some such precosmic size would have sufficed for cosmogony or it would not have, in consequence of which cosmogony would either have occurred an eternity ago or have been effectively impossible; but neither of these possibilities is consistent with a seriously held cosmogonic hypothesis that describes an expansion outwards from an initial small area (apo tou smikrou, B12.14). I conclude that it must have been possible for nous to grow, if cosmogony is to be preserved. While it might be thought that growth, as one kind of 'becoming', is ruled out by F19/B17, this text indicates that X cannot become Y rather than that X cannot become X*, but it is this latter (from nous to Nous) which I am here supposing that Anaxagoras thought of as 'growing'. (58)

The only way for anything to grow, however, is by way of its incorporating something similar to itself, as we have seen in the case of hair and flesh in T10 above. But the incorporation of something by something else - generically identical but specifically different--will be impossible unless they somehow come into contact, and their coming into contact will similarly be impossible unless one or the other (or both) is in motion, either rotationally or rectilinearly (or both). (59) Now if cause can be inferred from effect, as seems clear from Anaxagoras theory of the nature of matter as 'everything in everything else', it seems quite reasonable to suppose, given the cosmic rotatory effect of Nous, that a stable expanse (60) of Nous would be neither utterly motionless (like Xenophanes' god), nor moving rectilinearly, but instead simply rotating somewhere or other--at first in some small area, then increasing in scope. But this does not in turn imply that nous also rotates, for if the larger and smaller expanses of eternally existing nous were all rotating naturally, and if Nous were produced as a result of noetic rotation alone, then cosmogony would once again have retreated to the infinitely remote past, and in consequence Anaxagoras would in effect have postulated a thoroughly useless postulate. On the other hand, however, noetic rotatory motion might conceivably arise from previous rectilinear motion alone, if only a given expanse of nous chanced to have collided obliquely with another expanse of nous; (61) and while the possibility of rectilinear motion evidently presupposes an infinite universe, such an infinite expanse finds support in Anaxagoras' mention of the limitlessness of the surrounding vault (F2/B2) and the possibility of other cosmoi within the single universe (F5/B4b). But even if noetic motion should turn out to be naturally rotatory rather than rectilinear, movement of one expanse of nous from one place to another (and hence the possibility of contact between discrete expanses of nous) might conceivably result from their whirling or tumbling about, in any number of directions at any number of times, through the pre-existing and thicker (62) 'everything other than nous'. That none of these are inherently impossible explanations for the relative sizes of pre-cosmic nous requires only that, given enough time, they are entirely likely to occur sooner or later. But this, in turn, indicates nothing more than that cosmogony is probable given certain specifiable antecedent conditions, the character of which it has been my overall aim to describe here.

Given the foregoing extended description of Anaxagoras' cosmic thought, it seems quite reasonable to suppose that eventually some expanse of nous adequate for establishing an emergent, larger and more powerful expanse of Nous capable of initiating cosmogony will accumulate somewhere --if only by sheer accident. But even if the probability here envisioned should approach necessity so nearly as to be in effect indistinguishable from it, nonetheless they remain conceptually distinct because only the former preserves the genuine possibility of cosmogony. I can, for my part, think of no better way of expressing this indistinguishability than that the beginning of our cosmos was bound to happen to occur. If Anaxagoras' thought on this specific topic has been fairly reconstructed from the evidence remaining to us, then we will need to give credit to him for having made deliberate use of the probability as 'the chance of an event's occurring', and unless someone prior to him can be shown to have used it, we ought also to credit him with its invention. (63)


BARNES, J.--The Presocratic Philosophers, Volume 1. London: Routledge and Kegan Paul, 1979.

BARGRAVE-WEAVER, D.--"The Cosmogony of Anaxagoras." Phronesis 4:2 (1959) 77-91.

BURNET, J.--Early Greek Philosophy, Fourth edition. London: Adam and Charles Black, 1930.

CHERNISS, H.--"The Characteristics and Effects of Presocratic Philosophy." Journal of the History of Ideas 12: 3 (June 1951) 319-45.

CURD, P.--The Legacy of Parmenides. Princeton: Princeton University Press, 1998.

DICKS, D. R.--Early Greek Astronomy to Aristotle. London: Thames and Hudson, 1970.

DIELS, H. and KRANZ, W.--Die Fragmente der Vorsokratiker, Volume 2, Sixth edition. Berlin: Weidmannsche Verlagsbuchhandlung, 1960.

GALLOP, D.--Parmenides of Elea. Fragments: A Text and Translation with an Introduction. Toronto: University of Toronto Press, 1984.

GOMPERZ, T., tr. L. Magnus--The Greek Thinkers, Volume 1. London: John Murray, 1901.

GUETTER, D. L.--"The Cognitive Sense of nPEn- and nPOEHK-." Phoenix 60 (2006) 329-45.

GUTHRIE, W. K. C.--A History of Greek Philosophy. Volume II: The Presocratic tradition from Parmenides to Democritus. Cambridge: Cambridge University Press, 1965.

HUFFMAN, C.--"Review [of D. Sider, The Fragments of Anaxagoras]." CJ 79 (1983) 79-81.

HUSSEY, E.--The Presocratics. New York: Charles Scribner's Sons, 1972.

JAEGER, W.--The Theology of the Early Greek Philosophers. London: Oxford University Press, 1967

KAHN, C. H.--Anaximander and the Origins of Greek Cosmology. New York: Columbia University Press, 1960

KERFORD, G. S.--"Review: The Fragments of Anaxagoras." CR, N.S.18 (1968) 279-81

KIRK, G. S., RAVEN, J. E. and SCHOFIELD, M., eds.--The Presocratic Philosophers, Second Edition. Cambridge: Cambridge University Press, 1983.

KINGSLEY, P.--In the Dark Places of Wisdom. Inverness, CA: Golden Sufi Press, 1999.

LANZA, D.--Anassagora, Testimonianze e Frammenti. Florence: La Nuova Italia, 1966.

LAKS, A.--"Mind's Crisis: On Anaxagoras' NOTE." The Southern Journal of Philosophy 31 (Supplementary Volume, 1993) 19-38.

LESHER, J. H.--"Mind's Knowledge and Powers of Control in Anaxagoras 'DK' B12." Phronesis 40 (1995) 125-142.

LLOYD, A. M.--"The Poetry of Anaxagoras' Metaphysics." Journal of Philosophy 4 (1907) 85-94.

LOBEL, E. and PAGE, D. L., eds.--Poetarum Lesbiorum fragmenta. Oxford: Clarendon Press, 1955 (corrected edition 1968).

Mckirahan, R. D.--Philosophy Before Socrates. Indianapolis: Hackett, 1994.

MOORHOUSE, A. C.--"AEN in Classical Greek." CQ, N.S. 12 (1962) 235-238.

O'BRIEN, D.--"The Relation of Anaxagoras and Empedocles." JHS 88 (1968) 93-113.

REESER, M. E.--"The Meaning of Anaxagoras." CPh 55 (1960) 1-8.

ROBINSON, J.--"Anaximander and the Problem of the Earth's Immobility," in ANTON, J. P. and KUSTAS, G. L., eds.--Essays in Ancient Greek Philosophy. Albany: SUNY Press, 1971: 111-18.

ROBINSON, T. M., tr.--Heraclitus. Fragments: A Text and Translation with a Commentary. Toronto: University of Toronto Press, 1987

ROSS, W. D.--Aristotle, Physics. A Revised Text with Introduction and Commentary. Oxford: Clarendon Press, 1956

SCHOFIELD, M.--An Essay on Anaxagoras. Cambridge: Cambridge University Press, 1980.

--"Review: Anaxagoras." CR, N.S. 32 (1982) 189-91.

SIDER, D.--The Fragments of Anaxagoras, Second Edition. Sankt Augustin: Academia Verlag, 2005.

SILVESTRE, M. L.--"Significato e Ruolo del Nous nella Filosofia di Anassagora." Il Contributo 12 (1988) 29-52

SMYTH, H. W., rev. G. M. Lessing--Greek Grammar. Cambridge, MA: Harvard University Press, 1956.

STANNARD, J.--"The Presocratic Origin of Explanatory Method." PhQ 15 (1965) 193-206.

WATERFIELD, R., tr. and comm.--The First Philosophers: The Presocratics and the Sophists. Oxford: Oxford University Press, 2000.


University of Windsor

Windsor, Ontario, Canada

(1) While Parmenides' 'Way of Truth' would not regard an event as a 'thing' in its own right, the same cannot obviously be said for his 'Way of Opinion', although in this latter case an event would need to be regarded more generally as 'something' (as opposed to nothing). I shall return to this distinction presently.

(2) Cosmogony begins, according to Anaxagoras, only when nous somehow brings about circular somatic motion. But "how Mind imparted the first rotatory movement [to Body] is by no means obvious; it may be that even Anaxagoras himself had no clear mental picture of the process" (Kirk, Raven, and Schofield [hereafter KRS] 1983: 362).

(3) McKirahan remarks that "if Anaxagoras is not to violate the Principle of Sufficient Reason, he needs to specify some feature of Mind itself or the state of the mixture of all things that accounts for why the rotation [which gets cosmic generation going in the first place] began at one time rather than another, and there is no reason to think that he did so" (1994: 224, emphasis mine). This should have been particularly embarrassing since it was well before Anaxagoras that Anaximander had made "the first known application of the Principle of Sufficient Reason" in explaining why Earth was at rest in the centre of the cosmos (40; for a sample of the history of this position, see Barnes 1979: 24, Hussey 1972: 26, Jaeger 1967: 36, Burnet 1930: 65-66, and Gomperz 1901: 51).

(4) According to Cherniss 1951: 324, "The details of his [= Anaximander's] theories aside, the startling and important features of his thought are its universality of scope, its freedom from anthropomorphic orientation, and the strictly impersonal causal nexus which is assumed to hold together all objects and events."

(5) Dicks 1970: 45 claims that Kahn painted a "vastly distorted picture of ... [Anaximander which] implies a familiarity with the concept of the celestial sphere and its main circles which is entirely anachronistic for his time, and for which there is no good evidence before the latter part of the fifth and the beginning of the fourth century B.C."; Robinson 1971: 111-118 similarly argues that this view is wrongly ascribed to Anaximander by Aristotle.

(6) The order of fragments given in Diels-Kranz 1960 (hereafter DK) is haphazard compared to the coherent re-ordering provided by Waterfield 2000 (hereafter W.), and accordingly I shall cite Anaxagoras by W.'s designations (F = fragment, T = testimonial) but include the standard DK numbers for ease of reference, especially in cases where specific lines of DK's Greek are at issue. I use Sider's text (2005), which on occasion updates DK, and any references to his commentary on Anaxagoras are indicated by "on X.y", where X and Y respectively denote the fragment/line numbers in DK. Authors other than Anaxagoras will be referred to by DK numbers alone. Finally, the translations below are W.'s unless indicated otherwise.

(7) Anaxagoras F19/B17: to [TEXT NOT REPRODUCIBLE IN ASCII].

(8) T2/B10: [TEXT NOT REPRODUCIBLE IN ASCII]; McKirahan, in a brief discussion of the disputed authenticity of this fragment, notes that "the ultimate source (Aristotle) of the interpretation is quite likely to be correct, since he is thought to have access to Anaxagoras' book and is unlikely to have misunderstood an obvious point like the one in question here" (198n.9). Unlike Schofield (1980: 135 ff.), Sider (121f.) argues that B10 is authentic.

(9) Parmenides B8.20-21: "If it came to be, it is not, and likewise if it will be some time in the future./Thus birth has been extinguished and perishing made inconceivable."

(10) Heraclitus B1: "everything happens in accordance with this principle"; B2: "although the principle is common, the majority of people live as though they had private understanding"; B50: "it is wise for those who listen not to me but to the principle to agree in principle that everything is one"; B89: "the universe for those who are awake is single and common"; B104: "What intelligence or insight do they have? They trust the people's bards and take for their teacher the mob"; B51: "They are ignorant of how while tending away it agrees with itself--a back-turning harmony, like a bow or lyre"; B84a: "While changing it rests" (tr. Robinson 1987); and B103: "In the case of a circle['s circumference] beginning and end are common (ibid.); Parmenides B5: "The point from which I start/Is common; for there shall I return again".

(11) Xenophanes B24: "Complete he sees, complete he thinks, complete he hears"; B26: "He remains for ever in the same place, entirely motionless,/Nor is it proper for him to move from one place to another"; Parmenides B8.4: "Entire, alone of its kind, unshaken, and complete"; B8.29-30: "It stays in the same state and in the same place, lying by itself,/And so it stays firmly as it is.". Incidentally, W. suggests an intriguing alternative to the standard translation of Parmenides B3 ("For the same thing both can be thought and can be"): "Thinking and being are the same", noting that "if this [alternative] translation is correct, and mind and being are identical for Parmenides, a whole new light is shed on his poem. Its subject would not so much be being per se as thinking about being; and what-is would be a living, sensible entity, somewhat akin to Xenophanes' god" (319).

(12) Pythagoras 14A8a: "we should regard all ensouled creatures as akin"; 58B28: "when gnomons are placed around the one, and apart, in the one case the shape is always different, and in the other it is always one" (tr. McKirahan; for its representation see 94-95); Parmenides B8.34-37: "The same thing both can be thought and is that which enables thinking./For you will not find thinking apart from what-is, on which it depends/For its expression. For apart from what-is nothing else/Either is or will be..."; B8.5-6: "It was not once nor will it be, since it is now, all together,/Single, and continuous"; B8.22-25: "Nor can it be divided, since all alike it is. Nor is there/More of it here and an inferior amount of it elsewhere,/Which would restrain it from cohering, but it is all full of what-is."

(13) For an excellent discussion of this question O'Brien 1968.

(14) Empedocles B17.30: "Nothing comes into existence or ceases to exist; there is only them [i.e., the four 'roots': Fire, Air, Water, and Earth]"; B12: "For there is no way for what-is-not to be born,/And for what-is to perish is impossible and inconceivable,/Since wherever it is planted at any time, there it will always be"; B8: "Listen now to a further point: no mortal thing/Has a beginning, nor does it end in death and obliteration;/There is only a mixing and then a separating of what was mixed,/But by mortal men these processes are named 'beginnings'."

(15) Democritus B9: "Sweet exists by convention, and so does bitter, warm, cold, and colour; in reality there are atoms and void." On this fragment W. includes a useful citation from Democritus' Criteria by Sextus, Adv. Math. 9.139: "There are two forms of knowledge, one genuine, the other bastard. To the bastard kind belong all the following: sight, hearing, smell, taste, touch. But the other kind is genuine and is far removed from the bastard kind." Democritus A14, finally, indicates that medium-sized perceptible objects are the consequence of entanglements of (imperceptible) atoms.

(16) This point gains support from Anaxagoras' theory of perception: "[he] says that perception occurs thanks to opposites [tois enantiois], because similars are unaffected by one another" (T16/A92). Applied to the case of ordinary vision, it would be an error to suppose that objects are visible because we see them; rather, we see them because they are visible. That visible things can be seen but not heard, or audible things heard but not seen, shows not only that tois enantiois are both generically similar and specifically different, but also that it is the object of sense which 'acts' upon the 'passive' sense organ.

(17) For features of scientific explanation related to this one, see Stannard 1965: 201-202, and esp. 206.

(18) For this view, see Lesher 1995: 128; for the emphasis upon Mind's knowledge, see Laks 1993: 19-38.

(19) For this view, see Silvestre 1988: 29-52.

(20) More ontologically expressed, 'whatever is, is, and whatever is not, is not; if something is, it cannot cease to be, and if something is not, it cannot begin to be.'

(21) A somewhat looser translation of Parmenides B2.3-5 than W.'s may be warranted if the distinction between the negation of thought or will (me) and that of fact or actuality (ou) merits preservation, namely, "One route is: 'that which in any way exists is also that which one cannot in fact [ou] just think [or 'will': me] out of existence'.; the alternative route is: 'that which does not in fact [ou] exist is also that which it is imperative not to think [or 'will': me] into existence'."

(22) Curd 1998 makes a compelling case for the view that Parmenides is better understood as a reformer within the tradition of natural philosophy than as metaphysical supplanter of it, arguing that the 'singularities' Parmenides has in mind apply to any discretely existing member of a plurality of things. Kingsley 1999 argues forcefully for the very deep kinship in thought between Parmenides and Pythagoras, concluding that metaphysics was due to a very influential neo-Pythagorean named Plato; but it is not easy to see how Parmenides' reflections on the criteria of generic Being do not qualify as metaphysics, even if it was only subsequently that Plato invented transcendental particulars.

(23) It is on the basis of his application of these metaphysical principles that the cosmology presented in Parmenides' 'Way of Opinion' (B8.50-B19) is fundamentally misleading if taken as a description of 'The Real'; the "deceitful ordering" of the goddess' words in 8.53 (tr. Gallop; W.s "lies" is too strong, implying as it does a wilfulness about this deception, rather than a caution concerning it) is, although linguistically not only intelligible but even "plausible" (eoikota, 8.60, tr. Gallop) as far as it goes, not therefore objectively accurate, i.e. intelligible the whole way, relative to whatever is actually the case independently of human cognitive capabilities or proclivities.

(24) Not only is this fully consistent with the systems of Empedocles and Democritus, but in the case of Democritus the void, while not a thing in its own right (only atoms are), is not therefore nothing; rather, "[it is] 'something' in the sense of matter or substance" (Moorhouse 1962: 236; McKirahan's "hing" [305], by contrast, misses Democritus' point). For a previous instance of this stark separation of ouden into ou and den (Simpl., in Cael. 295.4-5) to make what Moorhouse 238 cautiously regards as a generally similar point see Alcaeus 320 (Lobel-Page 1955). The compound ouden indicates, then, either 'nothing' or 'no thing', and since the motion-enabling void is clearly not nothing, it follows that void is 'something' but not 'some thing'.

(25) Even if Aristotle invented the words entelecheia and dunamis as names for 'actuality' and 'potentiality' respectively, it in no way follows that he also invented the concepts related to these names; on the contrary, the conceptions themselves seem clear enough given the actual work of the three post-Parmenidean materialists here considered.

(26) Bargrave-Weaver, for instance, argues that a single comprehensive stage of cosmogonic differentiation succeeds a pre-cosmic era during which everything was static and motionless (1959: 88-89), but does not speak to how Anaxagoras might have thought such a transition to be reasonably possible.

(27) The everlastingness of motion is implied by Empedocles' cosmic cycle. B16: "For they are as they were and will be, and never, I think,/Will boundless time be emptied of the two of them [sc. Love and Strife]"; B17.6, 13: "The roots never cease from continuous alteration ... They are for ever unchanging in a cycle"; B26.5-6: "Now being brought together by love into a single orderly arrangement,/Now being borne asunder by the hostility of strife"; Democritus A16: "Leucippus and Democritus. claim that the primary bodies are in constant motion in the infinite void"; A18: "they say that there is always motion" (McKirahan).

(28) Specifically, in F9/B11 nous constitutes the sole exception to the preceding narrative by virtue of being something pure and in some things only, and in F10/B12 it is something limitless, independent, pure (again), self-identical throughout, decisive, supremely powerful, and capable of governing, initiating, and ordering.

(29) For the singularity of the universe in Anaxagoras see F6/B8 and T1/A63; for the multiplicity of worlds within the universe see F5/B4b.


(31) The proper sense of 'things' is given in F4a/B4a.2-3: [TEXT NOT REPRODUCIBLE IN ASCII], the seeds of all identifiable 'things'. Any specific sort of 'seed' (or 'ingredient'), if predominant in the emergent mixture, would as an entirely natural consequence result in a corresponding 'thing'.

(32) Since 'everything' is clearly presupposed by 'every thing', it seems that the 'seeds' themselves also exist in the 'everything' from which all discrete 'things' will eventually be articulated. Anaxagorean 'seeds' are, according to Sider, indestructible (96f.) and identical with their powers (dunameis, 171), but these powers can make their presence felt only after they have "collected in sufficient strength" (104, emphasis mine); now since "each dynamis [is] present in and uniformly dispersed throughout [the seed]" (171), and it seems entirely reasonable for us to suppose that the only way nothing at all was happening prior to Mind's 'prompting' would be for these powers to be precisely balanced, Sider's subsequent remark that "the dynameis are joined inextricably together in varying proportions into seeds, some of which can be characterized as aer, bone, etc." (139, on B12.22-25) is perplexing, entailing as it does an imbalance within seeds--and therefore, I would say, almost certainly also between them, in which case there would always have been movement rather than absolute rest, even if that movement was below the threshold of human perception, as is at the very least consistent with Anaxagoras' view of the weakness of human sensation in F20/B21.

(33) Reeser 1960: esp. 1 and 7. She argues that Anaxagoras' conceptions of the larger and the smaller, unlike Zeno's, do not violate Parmenidean ontological requirements and prohibitions; the same respect for Parmenides can be seen in other aspects of Anaxagoras' philosophy.

(34) F10/B12.6-8: "at first it [the cosmos] began to rotate out from a small area, but now it is rotating over a wider area, and it will rotate over a wider area still." Cf. F6/B8.5-6: "Mind also controlled the whole rotation, in the sense that it was responsible for initiating the rotation."

(35) Mind qua 'something' is not yet ontologically ready to so much as even form intentions in the preliminary stage of its own existence, as will be discussed in some further detail presently. For the moment, intentions require the emergence of 'some thing' capable of them from a chronologically prior 'something' which is not similarly capable.

(36) A45.27-29 = Simpl. in Phys. 10.1121.21-24 (not included in W. T6).

(37) Lanza 1966 argued that Simplicius did not have direct access to Anaxagoras' book, but instead got his information from Theophrastus, one of Aristotle's most well-known pupils. If Lanza is right, then the odds are low, it seems to me, that Theophrastus was not swayed by his teacher's teleological view of things, at least in general terms. This would help to explain why Simplicius wrote "seemed" rather than "seems". Kerford (1968: 279) disputes Lanza's thesis in this connection: conceding that Lanza's view may not be impossible, he notes that "the manner in which [the relations between the fragments of Anaxagoras in Simplicius] are given suggests otherwise ... [Simplicius' arguments with Anaxagoras] certainly do not derive from Theophrastus" (emphasis in original).

(38) For the general use of the aorist to denote a completed action without reference to the time required to complete it, see Smyth 1956: 436, #1923; the more particular 'ingressive' and 'resultative' possibilities are at 430: #1924 and #1926 respectively.



(41) Although Kerford notes that "the processes [in F11/B9 and F14/B13] ... seem to refer to the mechanical processes such as in a modern centrifuge rather than to anything biological or 'parabiological' " (291), the point is also relevant to this case.


(43) B12.17-18: [TEXT NOT REPRODUCIBLE IN ASCII]. We may note that where emellen ('would') implies an expenditure of effort, that effort in turn implies a predicted future eventuality, and that 'will' is ambiguous between these alternatives; on the other hand, a predicted future eventuality does not imply an expenditure of effort, as is clear from any strictly natural process, e.g. of salt's lowering the freezing temperature of water. If, as I shall argue in greater detail presently, nous has always existed but Nous has not, these lines cannot confidently be taken to indicate any more than what will eventually happen to occur.

(44) Schofield (1982: 190) criticizes Sider's not having obelized this corrupt passage; Huffman comments, on Sider's printing (n44), that this is "an attractive possibility but the passage is so confused that confidence is not possible" (1983: 69).


(46) In this, McKirahan follows KRS (who in turn follow DK): "But Mind, which ever is, is assuredly even now where everything else is too, etc."

(47) Simpl. in Phys. 35.157.7; the only variation is D (Laurentianus 82.2, s. xii-xiii), which reads ECTTai instead of ectti.

(48) Taking hosa as a pronoun relative to the immediately preceding nous, DK divided hosa into hos and a, then searched for a plausible word beginning a, theorizing that aei is the complete original word: thus they print "Mind, which always is", etc. DK's emended text thus plainly implies that some definite, already formed Mind is everlasting, and it is clear enough that this emendation, if accurate as a representation of Anaxagoras' original, is fatal to the cogency of what I have been suggesting above, namely that not only is there an important difference between Mind without an article ('Mind-stuff-in-general'), and Mind with the definite article ('some thing identifiable as a stable unit of Mind), but also that the latter did not exist in a fully formed state from everlasting but instead somehow succeeded in achieving a stability that it did not have previously.

(49) LSJ suggest "very" or "extremely", but this suggests, I think implausibly, that Mind has a greater degree of existence than do 'the other things'. By contrast, "really" is not only as emphatic as 'very' or 'extremely' are, it is also evocative of specifically Parmenidean thought.

(50) I have in mind here the two sorts of things already familiar in Anaxagoras--those which have been separated out from the 'everything' (hot/cold, wet/dry, and so on) and those particular things which result from that previous separation (flesh, bone, hair, and so on).

(51) F13/B14.2: [TEXT NOT REPRODUCIBLE IN ASCII] (DK, KRS), another disputed text; Sider prints [TEXT NOT REPRODUCIBLE IN ASCII] instead of [TEXT NOT REPRODUCIBLE IN ASCII]. It should be noted that the plausibility of pollai as an emendation of Simplicius' polla (ibid., 157.8) depends upon this particular word's having been intended by Anaxagoras to be understood within its immediately surrounding dative construction. But as before, this emendation requires that all the mss. of Simplicius have it wrong in just the same way on precisely the same point. Now, Anaxagoras does not yet have a reputation for poetry, and perhaps he never will--but only if this text does not turn out to be a case in which he is illustrating his more general cosmic point by his very linguistic choices: in the same way that the many discrete 'things' are literally within the surrounding general 'everything' but nevertheless ontologically distinct from it, so here polla is emphatically surrounded by a dissonant dative construction and thus linguistically distinct from it. If so, then, insofar as polla agrees with ta alla panta and insofar as this linguistic association would suggest a connection between polla and the plural 'things' rather than with the singular (surrounding) 'everything', the resulting sense would be this: "Mind exists. wherever all the other things do within that which surrounds [them]." For the suggestion that the content of Anaxagora's metaphysics--homoeomeries and nous--was significantly (whether or not intentionally) poetic, i.e. absurd and paradoxical, but only for prosaically minded thinkers, see Lloyd 1907, esp. 93.

(52) See, e.g., Arist. Phys. 2.9 and GC 2.11.


(54) Guetter 2006: 330-31.


(56) For a similar view in Xenophanes see F17/B35 ("Let these things be believed as approximations to the truth") and F19/B18 ("The gods did not intimate all things to men straight away, /But in time, through seeking, their discoveries improve").

(57) F10/B12.27-28: [TEXT NOT REPRODUCIBLE IN ASCII]. This passage refers to the expanses of nous itself rather than, as at B12.11-12, to the larger and smaller animate creatures in which the presence of nous can be discerned.

(58) For the idea that krateo, generally construed in terms of 'ruling', may also signify digesting and assimilating, see LSJ (s.v. [TEXT NOT REPRODUCIBLE IN ASCII] II.2.b). Although nous might be naturally fit to govern, it does not follow that it actually governs (only Nous does); what seems to me to be required is that nous somehow was able to 'feed' upon nous so as to have 'grown' into Nous. Once established, Nous will be quite capable of further growth and governance.

(59) Cf. T6/A45: "Moreover, this source [for all coming into being] must be a single principle, of the kind which Anaxagoras calls 'mind', and there is always a starting-point [arches tinos] at which our minds stop thinking and set to work." The hypothesis advanced in this paper--that Nous causes motion by virtue of being in motion, and that the rotatory motion of Nous is due to the inherent (rotatory, rectilinear, or both) motion of nous--is, it may be noted, consistent with Ross's third of three interpretive possibilities for this 'starting-point', namely as indicating a reference to a principle intrinsic to nous itself (1955: 546).

(60) A stable 'expanse' of Nous may be functionally equivalent to a particle, but these are ontologically distinct. Any grounds upon which one might attribute to Anaxagoras a particulate theory of nous are even less secure than those seeking to attribute to him a particulate theory of matter. Now if a volume of nous is something but not particulate, then it can expand or contract only by the addition or subtraction of other volumes of nous.

(61) A head-on collision between larger and smaller 'expanses', as I imagine here, might result in either the absorption of the smaller by the larger, or, supposing that the larger were simultaneously struck by a sufficient mass of smaller ones (whatever their number), a disintegration of the larger--but this would not by itself produce the required noetic rotation.

(62) The relative thickness of 'everything other than nous' is easily inferred from the description of nous itself as leptotaton (extremely fine) at F10/B12.9.

(63) This paper has benefited greatly from considerable critical feedback from Brad Inwood of the University of Toronto.
COPYRIGHT 2009 Universidade de Lisboa. Centro de Estudos Classicos da Faculdade de Letras
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Guetter, David L.
Publication:Euphrosyne. Revista de Filologia Classica
Date:Jan 1, 2009
Previous Article:On the origins of the sciences.
Next Article:O Etna em Pindaro e em Esquilo: duas perspectivas de um fenomeno fisico e de um mito.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters