‘Antisthenes the Cynic, unable to answer [Zeno’s arguments against motion], got up and walked, deeming a proof by action more potent than any logical confutation’ (Elias, 29 A 15). Zeno’s paradoxical conclusions disagree outrageously with what we like to call ‘common sense’; and if common sense has no part to play in the serious dramas of science, in philosophy it often assumes a leading role. Moreover, in its antagonism to Elea, common sense has a powerful ally: perception. We perceive, everyday, the falsity of Eleatic metaphysics andZenonian immobility; and our common sense is trustworthy just because it is securely backed by those quotidian perceptions.
The Eleatics, naturally enough, became enrolled in the sceptical army: in the crude words of Aëtius, ‘Parmenides says that the senses are false’ (28 A 49). Sextus, our chief quarry for ancient scepticism, numbers Parmenides among his tribe (adv Math VII. 114); and Timon, the sceptical satirist, praised Parmenides, ‘who turned his thoughts from the delusion of fantasy’ (fr. 44=A 1). Aristotle sums it up: the Eleatics ‘pass over perception and disregard it, thinking that one should follow reason…. In the light of their arguments this seems to follow; in the light of the facts it is near to madness to hold such opinions’ (GC 325a13–18=A 25).
No Eleatic could be unaware of the way in which his conclusions disregard the data of perception; and we might expect some little argument from Elea to excuse or justify its high-handed treatment of the chief instrument of Ionian science. Epistemology, after all, was in existence, a young discipline but not an infant; and if Parmenides had a sceptical predecessor in Xenophanes, he had an opponent in Heraclitus. A philosopher of the fifth century could not simply shrug off his epistemological commitments.
In fact we find little in the Eleatic fragments; and the little we find is probably all there ever was. Nothing suggests that Zeno wrote in an epistemological vein; for Melissus we possess one substantial fragment, but no hint that his work contained anything further of that sort; and for Parmenides we have a few brief lines.
In support of his claim that Parmenides ‘makes it clear that one should not attend to the senses but to reason’, Sextus quotes text 153; and Diogenes quotes the same lines in the same connexion (IX. 22= A 1). Plainly, the later tradition knew no other sceptical utterances from Parmenides. (The quatrain on thought, B 16, is irrelevant here: it will be considered in chapter XXII.) I transcribe 153 again (see above, p. 170):
Do not let much-experienced habit force you along this road, to let run an aimless eye and an echoing ear and a tongue; but judge by argument (logôi) the much-contending refutation uttered by me.
The lines do not argue for scepticism: they enjoin, they do not reason. But many scholars find in them a wholesale rejection of sense-perception. I think that the lines say both more and less than that.
First, the lines mention the tongue; and the tongue is the organ of speech as well as of taste. I agree with those scholars who attend to the former function and suppose that Parmenides has in mind not gustatory illusions (never a very rich source of sceptical argument) but rather the perils inherent in ordinary language. The empty words of mortals, which Parmenides lists at 156. 40–1, habitually trip off our tongue; and if we let them do so, they will lure us, like a fatal ignis fatuus, along the marshy path of Opinion.1 It seems probable that the ‘echoing ear’ is to be understood in the same fashion. Parmenides is not thinking of auditory illusions; he is warning us against listening to the foolish words of other mortals which perennially seduce us from the narrow path of Truth. Compare Heraclitus’ advice to disregard other men’s flowers and ‘seek for ourselves’ (above, p. 145). If I am right, then two of the three organs mentioned in 153 feature not as instruments of disreputable sense-experience but as channels for the subtle and semi-conscious insinuation of mortal opinion. The lines in 153 do more than warn against the errors of the senses.
Second, 153 does less than utter a general warning against perception. The lines occur in a specific context and their moral has a specific point: it is not that the senses are in general to be distrusted; it is that the senses are not to be used against Parmenides’ ‘much-contending refutation’. Parmenides’ request, as I have already remarked, is entirely just: when we turn to the backgammon board, we may find it impossible to believe that the Way of Truth is the way to metaphysical bliss; but that is no disproof of Parmenides’ contentions. If we are to reject Parmenides’ conclusion, then we must match reason with reason: we must show where his argument goes wrong.
In sum, 153 has very little to do with scepticism. Parmenides is saying no more than this: ‘If you think my argument wrong, then prove it wrong; don’t fall back into the lazy habits of common sense.’ I do not deny that Parmenides was an enemy of the senses and that he ‘hurled the senses out of truth’ (pseudo-Plutarch, A 22). But that enmity is left implicit in Parmenides’ poem: we have no formal argument for scepticism in the text, and no explicit statement of scepticism. Parmenides made no contribution to the history of Pyrrhonism.
The case is otherwise with Melissus. Fragment 30 B 8 is long, but it merits a complete translation:
[i] Now this argument is the greatest sign that there is only one thing; but there are the following signs too.
[ii] If there were many things, they would have to be such as I say the one thing is. For if there is earth and water and air and fire and iron and gold and living and dead and black and white and the other things which men say are real—if there are these things, and we see and hear correctly, then each of them must be such as it first seemed to us, and must not change or become different (heteroion), but each thing must always be just as it is.
[iii] But now we are saying that we see and hear and grasp (sunienai) correctly; but what is hot seems to us to become cold, and what is cold hot, and what is hard soft, and what is soft hard, and living creatures seem to die and to come into being from what is not alive, and all these seem to alter (heteroiousthai), and what was and is now seem to be in no way homogeneous; but the iron which is hard seems to be rubbed away by the finger…,2 and so do gold and stone and everything else that seems to be strong, {Hence it follows that we neither see nor know what is the case3} and earth and stone seem to come into being from water.
[iv] Now these things are not in agreement with one another. For although we say that the many things both are eternal and have forms and strength, they all seem to us to alter and to change from the state in which they are at any time seen.
[v] It is clear, then, that we do not see correctly, and that those many things do not correctly seem to be. For they would not change if they were real, but they would be just as each seemed to be; for nothing is greater than what is real. But if they change, what is has perished and what is not has come into being. Now in this way if there were many things they would have to be such as the one thing is (191).
As paragraph [i] explicitly states, the burden of 191 is to provide additional support for monism: the ‘argument’ to which [i] refers is presumably that of 164; whether 191 contains all the additional ‘signs’ that Melissus promises is unknown. Aristocles, who quotes part of [ii] and [iii], says that Melissus ‘wants to show that none of the phenomena and things we see exists in reality’ (A 14); and Simplicius introduces 191 to illustrate Melissus’ attitude to sense-perception. And those ancient critics were at least half correct; for one of the things Melissus attempts to do in 191 is to argue that ‘we do not see correctly’.
The argument is in reductio form. Paragraph [ii] sets out the hypothesis to be reduced; it is in fact the conjunction of:
(1a) There exists several things, a1, a2,…an, and
(1b) Our senses are veridical. Melissus’ illustrations for (1a) sound strange to modern ears; but they doubtless reflect the various pluralisms, lay and professional, that Melissus met with in his philosophical conversations.4 The conjunct (1b) is usually stated for the cases of sight and hearing; but I assume that Melissus has the general case in mind, and I believe that by ‘sunienai (grasp)’ in [iii] he means ‘perceive’.
Paragraph [ii] also begins the reductio: asserting, reasonably, that each member of the plurality in (1a) would have to have the properties which the Eleatic deduction has shown to be essential to all existents, Melissus infers in particular that:
(2) No ai ever changes. He also argues, with needless ingenuity, that:
(3) If ai seems φ at t, then ai is always φ.
Paragraph [iii] looks more seriously at (1b): any number of changes seem to take place; the hardest things are rubbed away, and the most different things emerge from one another. Our senses tell us that everything changes; hence, by (1b), we get to:
(4) Every ai constantly changes. (The word ‘constantly’ is not in Melissus’ text; but it is, I think, implicit in the last phrase of paragraph [iv].)
The argument is now over; for, as [iv] points out, (2) and (4) ‘are not in agreement with one another’. Thus [v] concludes to the negations of (1a) and (1b): ‘we do not see correctly, and…those many things do not correctly seem to be’.5 The remainder of [v] merely repeats the assertion made in [ii], that from (1a) an Eleatic can properly derive (2) and (3): if the many things are real, then the predicates of Eleatic metaphysics must hold of them.
What are the merits of this ingenious piece of reasoning? First, it is worth noting that, despite [i], it does not purport to be entirely independent of Eleatic metaphysics; on the contrary, the move from (1a) to (2) explicitly applies familiar Eleatic properties to a putatively plural world. Such an application might seem wholly trivial: apply monism itself, and (1a) leads at once to a contradiction. But that would be unbearably jejune; and Melissus does not intend it. In his move to (2) he applies (T 7), the thesis that what exists does not alter. Now (T 7) was indeed inferred from (T 5), the thesis of monism; but it was also inferred directly from (T 1), the thesis that what exists cannot be generated (above, p. 215), and paragraph [v] serves the important function of indicating that it is that second inference that Melissus means to call upon. In short, Melissus argues that any pluralist must accept at least (T 1), and hence (T 7); and that then pluralism collapses. The ‘neo-Ionians’, to whom the next several chapters are devoted, did, some of them, attempt to hold both (T 1) and pluralism.
Aristocles introduces his quotation of [ii]–[iii] with a scathing criticism:
Now this is most absurd: showing by argument that [the senses] are useless, in fact they continually rely heavily upon them—Melissus, who wants to show that none of the phenomena and the things we see exists in reality, proves it by means of the phenomena themselves.
Having quoted Melissus’ own words he proceeds:
When he says this, and much more in the same vein, one might well ask him: ‘Is it not by perception that you know that what is now hot later becomes cold?’—and similarly in the other cases. For as I said, he will be found to be doing away with and refuting the senses by means of a peculiar trust in them (192: A 14).
Aristocles here initiates a longstanding objection to scepticism of the senses: in order to construct their arguments, the sceptics have to start from the data of perception; so based, their arguments are bound to be self-refuting. Melissus relies heavily on the senses; for he sets down (1b) as a premiss, and construes it in the strongest way possible, as saying that every sense report is true. Having said that, he proceeds to infer the falsity of all sense reports. What could be more absurd?
The charge of self-refutation may stand against some sceptics, but it has no hold on Melissus. Aristocles has misread Melissus’ argument: Melissus does not assert that (1b) is true—he presents it as a hypothesis which he will show to be false. Nor does he assert that (2) is true; for (2) depends upon (1b). The only fact about the senses to which Melissus does commit himself is this: that, according to our senses, things change. To believe that is not to show ‘a peculiar trust’ in the senses.
Aristocles’ charge of self-refutation fails, but there is a sound point in his criticism. Melissus construes (1b) in a strong fashion; and no self-respecting partisan of the senses would maintain that every sense report is true. In fact, Melissus does not need the strong construe of (1b); without it he will get neither (3) nor (4)—but he does not really use (3), and he does not require (4). To discover a contradiction he need only establish the contradictory of (2), which is not (4) but:
(5) Some ais sometimes change.
However weak we care to make (1b), it will surely remain powerful enough to give (5): the man who pretends to place some trust in his senses and yet believes that the world is an unchanging place can hardly be taken seriously. Partisans of the senses must not believe everything their favourites tell them; but their partisanship is empty if they deny such propositions as (5).
A more tolerable objection remains: ‘Melissus’ handling of the reductio is poor: the conjunction of (2) and (5) is certainly a contradiction; but Melissus cannot infer that both (1a) and (1b) are false: at most he can infer the negation of their conjunction, “Not both (1a) and (1b)”.’ That objection is sound; and it makes a fatal breach in Melissus’ argument as he states it. But it is possible to repair the wall and restore the argument.
Since he cannot retain both (1a) and (1b), Melissus’ opponent has two positions open to him: he may abandon (1a) and hold to (1b); or he may maintain (1a) and give up (1b). The first of those positions is quickly demolished by an argument closely parallel to that of 191: if our senses are veridical, then we live in a plural world. Just as any serious advocate of (1b), however weakly he construes it, must admit the truth of (5), so any serious advocate of (1b) must allow that the world exhibits diversity and is not a monolithic whole. It is absurd to support the senses and be a monist. Melissus’ opponent must therefore retreat to the second position; and that too must be abandoned if Melissus can prove:
(6) If there exist several things, a1, a2,…an, then our senses are veridical.
Now Melissus will certainly have reflected that it is only sense perception which suggests a plural world: reason, as the Eleatic deduction shows, leads inexorably to monism. That reflexion will not yield (6) but it will yield:
(7) If there is reason to believe that there exist several objects, a1, a2, …an, then our senses are veridical.
I imagine that even pluralists who deny Melissus’ rational path to monism will be prepared to accept (7). In itself (7) does not suffice to demolish the second position; but it does show that any occupant of that position is committed to:
(8) There exist several things and there is no reason to believe that there exist several things.
Propositions of the same form as (8)—‘P and there is no reason to believe that P’— are not self-contradictory; but anyone who holds to (8) is thereby acting irrationally, in one clear sense of that term. Many philosophers will maintain that some irrationalities of that sort are acceptable: there are some things we may or even must believe in the absence of reasons; but even if that is true, it seems unlikely that (8) can be numbered among such favoured propositions. I conclude that in (8) Melissus has a final answer to his opponent: the argument of 191 cannot be rationally defeated. Of course, Melissus has not proved scepticism; he has argued that, given the fundamental thesis of Eleatic metaphysics, (T 1), scepticism can be securely established by way of (T 7). The argument is not general: it is tied to Eleatic theory; but it is, for all that, ingenious and powerful.