Bett the Sophists and Relativism

The Sophists and RelativismAuthor(s): Richard BettSource: Phronesis, Vol. 34, No. 2 (1989), pp. 139-169Published by: BRILLStable URL: http://www.jstor.org/stable/4182328 .Accessed: 04/08/2011 18:49

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Phronesis.

http://www.jstor.org

The Sophists and Relativism RICHARD BETI'

It is frequently alleged that the Sophists were relativists. Claims to this effect can be found in general histories of philosophy, in histories of Greek philosophy, in studies of the Sophists, in studies of relativism, and elsewhere. Sometimes it is moral or ethical relativism specifically that is ascribed to them, at other times a broader relativism concerning knowledge, truth or reality in general; but that the Sophists were some species of relativists is something of a commonplace.' In fact, I am not aware that it has ever been explicitly contested in print.

My contention, however, is that this view of the Sophists is largely erroneous. There is but one Sophist, Protagoras, whom we have reason to regard as a relativist in any deep or interesting sense. It is not entirely clear whether even he deserves this label. But if he does, it is solely on the basis of his famous doctrine that "Man is the measure of all things" (DK 80B1) - a doctrine which he is never said to have shared with any of the other Sophists. The tendency to describe the Sophists as a group as relativists derives, I think, from at least two sources; first, from a tendency to regard Protagoras as representative of, and indeed authoritative for, the whole movement, and second, from a too hasty examination of the relation between the Sophists and Plato. On the first point, it is no doubt true that

I Attributions of relativism to the Sophists as a group include Frank Thilly, A History of Philosophy (3rd ed., revised by Ledger Wood. New York, 1957), p. 58; Samuel Stumpf, Socrates to Sartre: A History of Philosophy (4th ed. New York, 1988), p. 31; Eduard Zeller, Outlines of the History of Greek Philosophy (13th ed., revised by William Nestle, translated by L.R. Palmer. London, 1931), p. 93; W.K.C. Guthrie, History of Greek Philosophy, Vol. III (Cambridge, 1969), pp. 47, 50-1; G.B. Kerferd, The Sophistic Movement (Cambridge, 1981), ch. 9; Jack W. Meiland and Michael Krausz, eds., Relativism: Cognitive and Moral (Notre Dame, 1982), p. 6. For attributions of ethical relativism specifically, see also James L. Jarrett, ed., The Educational Theories of the Sophists (New York, 1969), p. 15; H.D. Rankin, Plato and the Individual (New York, 1964), p. 106; Alasdair Maclntyre, After Virtue (Notre Dame, 1981), pp. 130-1. Of course, this list makes no pretence to completeness.

Phronesis 1989. Vol. XXXIV12 (Accepted October 1988) 139

Protagoras, given his stature and his historical position, influenced other Sophists in many ways. But this does not, of course, imply that the Sophists in general imitated him or agreed with him in every respect. Whether or not other Sophists subscribed to some form of relativism - even assuming, for the moment, that Protagoras did so - has to be determined by looking at the evidence concerning them, not him; and, as I shall argue, the evidence does not support any such conclusion. As for the Sophists and Plato, it is clear that Plato was opposed to much of what the Sophists stood for, and that his mature philosophy includes a view of truth, including moral truth, as robustly objective, rather than relative. One may be tempted to think, therefore, that objectivism versus relativism must have been the issue around which the dispute between them revolved. However, this would be a needless, as well as a groundless, hypothesis. There is plenty for Plato to have objected to in the Sophists' attitudes and activities, quite apart from any supposed relativism on their part; as I shall suggest in closing, his antipathy to them can be uncontroversially accounted for on grounds which have nothing to do with relativism.

Clearly, the plausibility of my claim depends in part on what relativism itself is taken to be. The term "relativism" is often used very loosely; and the resulting confusion is only increased when "relativist" is employed, as it frequently is, as a term of philosophical abuse.2 This unclarity is very probably a further factor responsible for the generic labeling of the Sophists as relativists; certainly, those who perpetuate this view of the Sophists do not usually explain what they mean by the term. In any case, it is essential that I spend a little time explaining exactly what I mean by the term "relativism", when I deny that the Sophists in general adhered to relativism. There is, I shall concede, a weak sense of "relativism" in which many of the Sophists may plausibly be viewed as ethical relativists. Howev- er, I shall be concerned to distinguish this from relativism in the deep and interesting sense - the sense in which, as I said, it is only Protagoras who has any serious claim to be regarded as a relativist.

My argument will begin, then, with a characterization of the deep form of relativism. I shall not, at this point, refer directly to the Sophists; I shall simply define relativism, using contemporary terminology, as precisely as I can. This approach may perhaps seem perverse or incongruous in a paper purporting to deal with ancient thinkers. But the term "relativism" is, after all, a modern philosophical coinage3; and it was not coined, as far as I know, 2 The confusion over the use of the term, and its pejorative connotations, are well discussed by Mark B. Okrent, "Relativism, Context and Truth", in Is Relativism Defen- sible?, The Monist 67, no. 3 (1984), pp. 341-58.

140

with Greek philosophy specifically in mind. While the applicability of the term to ancient thinkers is a matter for historical scholarship - to which we will shift as soon as the first section is over - an acceptable definition of the term cannot but be a reflection of its usage in modern philosophy.4

Following these preliminaries, I shall briefly examine the weak, unin- teresting "relativism" which many Sophists (and many other thinkers) do admittedly display. I shall then tackle my central question; are there any grounds, other than Protagoras' "Man the Measure" doctrine, for ascribing relativism, in the deep sense, to any of the Sophists? Finally, I shall quickly touch on, but not resolve, the question whether even the "Man the Mea- sure" doctrine is a form of relativism. I assume throughout that the Sophists deserve to be treated as being of philosophical interest - whatever else we may wish to call them besides philosophers. I hope it is no longer necessary to justify or apologise for this.

Relativism, in what I am calling the deep and interesting sense, may be stated in the broadest terms as follows. It is the thesis that statements in a certain domain can be deemed correct or incorrect only relative to some framework. Every part of this definition calls for further explanation.

One can be a relativist about various different types of subject-matter. As I noted earlier, the Sophists are sometimes said to have been relativists about ethics in particular. One could also be a relativist about aesthetic judgements, about judgements in the social sciences, and so on. Then 3In fact, it is little more than a century old; the earliest usages attested by the OED are 1863 for "relativist" and 1885 for "relativism".

I I admit that those who describe the Sophists as relativists may not all intend the term in precisely the same sense as I shall propose. This is a difficult matter to judge since, as I observed, explicit definitions of the term, and even perspicuity in its use, are rare in this context. However, I would defend my definition on the grounds that it would be thoroughly orthodox among people who do define the term really carefully - namely, those contemporary philosophers whose subject-matter is relativism itself (on this, see further note 9). My impression, in any case, is that interpreters of the Sophists do generally have in mind a definition akin to mine (at least as one among others); this seems to me most clearly true of the passages (cited in note 1 above) from Stumpf, Guthrie and Meiland/Krausz. If there are some who do not, then my thesis is not, of course, the direct negation of theirs. But negating someone else's thesis is not the only way of supplanting it; and my hope is that, whether or not writers on the Sophists mean the same by "relativism" as I do, my position will be judged superior to theirs by the uncontroversial standards of clarity, precision in the use of terms, and cogency in the treatment of the ancient evidence.

141

again, more global types of relativism are possible, according to which claims to know things, about any subject whatever, or claims concerning the existence of something, can be deemed correct or incorrect only relative to some framework; the above two positions might be labeled "cognitive relativism" and "ontological relativism" respectively. Finally, one could be a relativist about truth in general, in which case, presumably, any sentence with a truth-value would fall within the scope of the thesis. Some of these forms of relativism would naturally be combined with others; but I shall not consider the possible relations among the different types. In any case, the relativist must be prepared to specify which categories of statements his or her version of relativism applies to; and this is the point of the phrase "statements in a certain domain", in the definition above.

Standardly, for a statement to be "correct" is for it to be true, and for a statement to be "incorrect" is for it to be false. However, one might hold that sentences concerning a certain subject-matter do not admit of truth or falsehood, yet insist that they can be judged as warranted, or assertible (or their converses), in some other sense; value judgements, in particular, have often been viewed in this way. The terms "correct" and "incorrect" are intended to encompass these non-standard notions, as well as truth and falsehood; a statement is "correct" if it measures up to the demands for assertibility, whatever they are, that are appropriate to statements within the subject-matter in question.

Next, we have the phrase "some framework". The word "framework" is intended loosely. What the relativist has in mind is some kind of basic commitments, on the part of some group or individual, concerning the subject-matter in question. Other terms besides "framework" have been used to describe such commitments - "belief system", "conceptual scheme", "background" and "perspective" are common examples - and the different terms carry different implications as to the character of what they refer to. Given our primary concern with the Sophists, we cannot take up any of the deep and difficult issues in this area. In addition, various different views are possible concerning whose framework is the appropriate one against which to assess any given statement. On one common view, the relevant framework is the one prevalent in the speaker's culture; alterna- tively, and more radically, it might be taken to be a framework specific to the individual speaker. This latter type of view is often called Subjectivism, in one of the many senses of that word; but it may also be regarded as a limiting case of relativism. Besides these, there are forms of ethical relativism, at least, which take as criterial the framework embraced not by the speaker, but by the agent referred to in the sentence.5 Again, we cannot

142

consider these permutations in any detail. The essential point is that, for any form of relativism, there must be some framework, shared or not, relative to which the correctness of statements in some domain is held to be alone open to assessment.

This leads us to the final, crucial question; what is meant by the phrase "only relative to"? To say that a statement can be deemed correct or incorrect only relative to a framework means two things. First, it means that the correctness of the statement is a matter of the consistency between the statement and the framework; and second, it means that there is no further sense, aside from this issue of consistency, in which the statement can be assessed for correctness. If, in discussing the habits of some remote tribe, I say "this tribe's cannibalism is an abomination", and there is a moral relativist in earshot who takes the moral framework of the speaker's culture to be the standard for assessment, then that relativist will judge my statement correct, because consistent with my moral framework; however, if the same statement is made by a member of the tribe itself, the same moral relativist will judge it incorrect, because inconsistent with the tribe's moral framework.6 There is nothing absurd or contradictory here - any more than it is absurd or contradictory to think that "I am hot" might be true as spoken by A and false as spoken by B; indeed, one might say that the effect of relativism is precisely to assimilate a certain class of statements to the category of indexical sentences.7 The same kind of account will apply to statements such as "There is no such thing as Phlogiston", if one is a relativist about science, or statements such as "We now know that the earth is not flat", if one is a cognitive relativist. In each case, the relativist will

5 For a version of this type of relativism, see Gilbert Harman, "Moral Relativism Defended", Philosophical Review 84 (1975), pp. 3-22. It is this type, which takes the agents' framework as criterial, that seems to lie behind the popular view that one cannot, or must not, criticize the ideas or behavior of people from other cultures. 6 If "correctness" is taken to be something other than truth (which we earlier admitted as a possibility), an analogous non-standard concept of consistency would have to be developed. I assume that this difficulty is not insuperable. In addition, the notion of "consistency with a framework" seems to presuppose that "frameworks" have a specifia- ble propositional content - something which might well be denied by some, especially by those of a Wittgensteinian persuasion. But neither of these difficulties will affect our examination of the Sophists. 7 Some versions of relativism draw an even closer analogy with indexicals, by claiming that an implicit reference to some specific set of standards is part of the semantic content of the statements in question. See, in particular, the version developed by Gilbert Harman, op.cit. But Philip E. Devine, "Relativism", Is Relativism Defensible?, pp. 405- 18, argues that relativism need not claim anything of the kind; and in fact, many treatments of relativism do not even address these semantic issues.

143

specify whose framework is the appropriate one against which to assess the statement, and will judge the statement correct if it is consistent with that framework.

My second point about the phrase "only relative to" was that the relativist refuses to allow that correctness or incorrectness, for the sentences falling within the scope of the relativism, can be construed in any other way than the one just described. For a statement in this category to be correct is for it to be consistent with the framework, whichever it is, that is taken to be relevant in the given case; and beyond this, there is no sense in which one can talk of such statements being correct or incorrect. In particular, there is no independent perspective, external to all the various frameworks embraced by cultures or individuals, from which the statements can be assessed; nor, therefore, is there any way of assessing these frameworks themselves. Or rather, the only possible perspective from which to assess one framework is the perspective of another framework. Thus, from within my framework, I can bemoan your framework, and vice versa. But this is mere name-calling; there is no neutral perspective from which the two frameworks can be compared, and their relative merits determined.

Relativism, then, is in one sense a denial of objectivity, for the domain to which it applies. It denies that any kind of "God's-eye view" is available in this domain; and it denies that judgements in this domain can be correct or incorrect in an absolute, unqualified sense.8 Instead, they can only be correct or incorrect in relation to some framework or other. On the other hand, the relativist may well maintain that this kind of correctness is quite robust enough for all our needs - though others may disagree.

This interpretation of relativism is, I believe, fairly cautious and un- contentious.9 It is also deliberately broad; a great many different positions 8 There is another sense, though, in which at least some relativists can perfectly well admit objectivity. The judgement that a certain statement is correct, or incorrect, relative to a certain framework (that is, is or is not consistent with that framework) may itself be objectively true - provided that the relativism is not about truth itself. Indeed, it might be argued that the relativist must understand consistency on objectivist lines; for, it might be said, if consistency is itself a relative matter, this will place the relativist in danger of an infinite regress. If so, then an entirely global relativism about truth would be untenable. By contrast, there is no difficulty in "local" relativisms admitting objectivity in domains other than the ones to which they themselves apply. I Characterizations of relativism which I take to be consistent in essentials with my own (even if they are stated in somewhat different terminology) are those of R.B. Brandt, "Relativism Refuted?", pp. 297-307, and C. Behan McCullagh, "The Intelligibility of Cognitive Relativism", pp. 327-40 in Is Relativism Defensible?; the Introduction to Relativism: Cognitive and Moral; and, from Martin Hollis and Steven Lukes, eds., Rationality and Relativism (Cambridge, Mass., 1982), Barry Barnes and David Bloor,

144

would qualify as forms of relativism under my rubric. Finally, while much more could be said by way of clarification (and would be demanded, in a full treatment of the subject), I have tried to present the doctrine as neither senseless nor self-refuting - as it is often claimed to be. If any of these things were not so, the interest of my thesis would, of course, be considerably lessened. But now that they have been as far as possible established, we may leave aside many of the niceties of the previous discussion. The vital points, for our purposes, are the ones having to do with the phrase "only relative to". Our question now is whether any of the Sophists were relativists in the sense we have outlined.

II

My first step, in answering that question, is to observe that the view I have called "relativism" is quite different from a view, also called "relativism" by some, which is with good reason ascribed to at least some of the Sophists. The view I have in mind has to do specifically with evaluative concepts, and may be stated in outline as follows. What is good, or just, or virtuous, or desirable, depends on the circumstances; there is no one set of things, or actions, which are good, just, virtuous or desirable in all circumstances for all people. The same action (for example, helping a friend) may be just in one situation and unjust in another; what is desirable for a sick person to eat may be undesirable for healthy people to eat; and so on.

There is evidence that an emphasis on this kind of relativity was routine among the Sophists. Protagoras, questioned by Socrates as to whether he equates "good" with "beneficial to humans", is made by Plato to reply with a lengthy harangue on the variability of what is good or beneficial, depending on who one is talking about (Protagoras 334a3-c6). It is clear from the context that this is intended as a parody of Sophistic thinking (as well as of Sophistic rhetorical display); and I see no difficulty in ascribing these kinds of ideas to the historical Protagoras. But even if this would be to take the less than sympathetic Plato too much on trust, it can hardly be doubted that the speech is at least meant as an allusion to ideas generally current among the Sophists.

"Relativism, Rationalism and the Sociology of Knowledge", pp. 21-47, W.Newton- Smith, "Relativism and the Problem of Interpretation", pp. 106-22, and Ernest Gellner, "Relativism and Universals", pp. 181-200. One well-known characterization of relativism which is harder to relate to mine is that of Bernard Williams, "The Truth in Relativism", Proceedings of the Aristotelian Society 75 (1974-5), pp. 215-28.

145

It appears, too, from a reference in Aristotle (Politics 1260a26ff.), that Gorgias held a similar view about the variability of virtue. Different kinds of activity constitute virtue for different kinds of people - men, women, children, slaves and so on. Virtue is not any one thing, the same for everyone; what is virtuous for any given person depends on that person's station in life. (Incidentally, the view receives Aristotle's own approval.) And Plato puts the same view in the mouth of Meno, an enthusiastic admirer of Gorgias (Meno 71e1-72a5), in answer to Socrates' request for a definition of virtue.

Another bountiful source for such arguments is the treatise known as the Dissoi Logoi (DK 90). The first three chapters of this deal with good and bad, decent (kalos) and indecent (aischros), and just and unjust respectively. In each case, the first part consists of arguments to the effect that the seemingly opposed members of these pairs are "the same"; and the arguments pursue in relentless detail the theme we have been considering.10 What is bad for one person (say, the wearing out of one's shoes) will be good for another (in this case, the cobbler) (1,5); what is indecent in some places (women bathing, for example, in public places) is decent in others (in this case, at home) (2,3).11 Again, the piece does not allow us to attribute these ideas to any specific Sophist; but it does show us that such ideas were commonplace among the Sophists in general.12

Now, these arguments admittedly have to do with a kind of relativity; they assert that one cannot say what is good, just or virtuous without qualification, but only in relation to specific circumstances. However, this is

10 There is one section (2,9-20) which does not appear to conform to this pattern. On this, see note 14 below. '" The same style of argument is even extended to the concepts of truth and falsity, madness and sanity, wisdom and ignorance, what is and what is not (chs. 4 & 5). Here the reasoning becomes even more blatantly trivial than in the ethical cases; e.g., truth and falsehood are said to be "the same" on the grounds that the same statement may be either true or false, depending on the circumstances - that is, depending on whether or not things actually are as the statement says they are (4,2). If this is relativism, "relativism" about truth is true by definition. 12 Despite our ignorance of the author's identity, the connection between the Dissoi Logoi and Sophistic thinking seems clear from two points; first, the apparent function of the work as a piece of rhetorical pedagogy, and second, the date of around 400 B.C., established by an allusion to the Peloponnesian War as "most recent" (1,8). This dating has been disputed by Kerferd (The Sophistic Movement, p. 54), on the grounds that the author merely means that the Peloponnesian War is the most recent of the wars he is going to mention. I agree that this is what the words literally mean. However, it seems to me that it would be strange to make a point of saying "I begin with the most recent", unless the war was in fact shortly before the date of composition.

146

only superficially similar to the view outlined in the previous section, and only superficially relativistic. The relativism we looked at earlier denied the possibility of objective judgements - that is, judgements whose correctness was independent of the framework of any particular group or individual- in the domain to which it applied. The present view, by contrast, is quite compatible with objectivity; what it denies is not the possibility of objective judgements, but the possibility of broad and invariable universal rules of conduct"3 - which is by no means the same thing. For, firstly, it does not deny that there might be some objectively correct general definition of goodness, justice, or whatever the evaluative concept under scrutiny. It might be, for example, that justice could be defined quite generally as "Giving people their due" (as Polemarchus tried to define it at the beginning of the Republic). There is no conflict between saying this, and saying that the specific actions which justice, as thus defined, demands vary widely from situation to situation, and are not formulable in any simple rules. (Meno appears to take his view about the variability of virtue as conflicting with the Socratic search for a unitary definition of virtue; he is reluctant to admit that virtue has any unitary character (73a4-5). But there need be no conflict, if the definition is on a sufficiently general level.) Secondly, the view we are examining does not deny that whatever, in some specific set of circumstances, turns out to be good, or just, is so objectively speaking. For instance, there is no difficulty in saying that some things, such as bedrest and medicines, are, as an objective matter, good for the sick but, also as an objective matter, not good for the healthy. None of the passages we have looked at suggest that what counts as good or just depends on one's evaluative framework; rather, they are suggesting that what is, as a matter of fact, good or just depends on who or what one is dealing with.

Even when an author explicitly cites differences in ethical practice between different communities, it may be that no more than this superficial kind of relativity is at issue. Consider the beginning of Thrasymachus' speech in Book I of the Republic (338c-339a). Thrasymachus' claim here is that the specific requirements of justice vary from society to society, but that these superficial variations are due to differences (themselves the result of differing political systems) in what is in the interest of the ruling party. In a deeper sense, then, the nature of justice is the same everywhere. 13 This is not to say that it denies what, in modem moral philosophy, has been called the "universalizability" of moral judgements (for discussion of this concept, see especially R.M. Hare, Freedom and Reason (Oxford, 1963), chs. 2 and 3). That is, it does not deny that what is good in one set of circumstances is also good in all relevantly similar circumstances; it merely insists that the circumstances may often be relevantly dissimilar.

147

Justice is always whatever is in the interests of the ruling party; what, in particular, this turns out to be depends on the political circumstances in which the ruling party finds itself. There is no hint that what is in the rulers' interests might itself be relative to some framework; Thrasymachus appears throughout to be confident that questions about people's interests admit of definite, objective answers. By the same token, questions about justice will admit of definite, objective answers - even though the answers will be different from community to community. What is just in a democracy is different from what is just in an oligarchy, because the rulers, and hence the rulers' interests, are different in the two cases; but given that people's interests, in any particular set of circumstances, can be specified objectively, justice-in-a-democracy and justice-in-an-oligarchy can also be specified objectively. As in the cases we examined above, what is responsible for the variation is not relativity to different ethical frameworks, but only differences in circumstances. There are not,)in Thrasymachus' view, competing basic principles of justice, between which no neutral judgement can be made; rather, there are different power relations in different communities, which give rise to local differences in what specific types of behavior are (really, objectively) just. At least as portrayed by Plato, then (and none of the other evidence about him suggests otherwise), Thrasymachus is not a relativist in the deep sense. His other central assertion - that justice, being merely the interest of the stronger, has no genuine claim on our allegiance - does not affect this verdict in any way.14

In summary, we have good reason to ascribe to some of the Sophists an interest in one kind of relativity about value; but this is not of any great philosophical significance, and does not warrant our calling them relativists, if relativism includes a denial of the possibility of objective judgements in some domain. I cannot, of course, forbid people from calling the

14 Precisely the same pattem is apparent in Dissoi Logoi 2,9-20. Here the author cites a number of variations in what is considered decent in different communities (2,9-17). This sounds like the makings of a more serious form of relativism about decency; and indeed, one might well think this was a case which cried out for an analysis in terms of brute differences of attitude in different communities. But even in this case, the author resorts, much less plausibly, to an analysis in terms of differences in circumstances. Variations in what passes for decent behavior, he says, are due to variations in kairos, "opportunity"; "to put it generally, everything is decent when it is opportune, and indecent when it is inopportune" (2,20). Thus, what is "opportune" for one community will not be "opportune" for another, because of the different circumstances in which the two communities find themselves; and concomitant with these variations will be variations in what constitutes decent behavior. This is the same kind of variability, compatible with objectivity, which we saw in Thrasymachus' view, and throughout this section.

148

kinds of observations we have looked at in this section "relativism" as well. But there are two strong reasons not to do so. First, this would result in a serious ambiguity in the term; for, as we have seen, the views examined in this section are far from equivalent to relativism in the sense examined earlier. Second, views of this kind are widespread among thinkers who would not normally be considered relativists.'5 Aristotle, for example, would qualify as an ethical relativist in this superficial sense, and so would any act-utilitarian. Both Aristotle and the act-utilitarian will hold that the right thing to do must be determined in the individual case, and cannot be specified in advance by exceptionless general rules.16 But both may never- theless insist that there is a single, non-relative determinant of what is right; what is right is determined, in Aristotle's case, by the judgement of the phronimos - the person who sees the world aright, ethically speaking - and in the act-utilitarian's case, by which action, among those possible for the agent, promotes the greatest utility. Neither is remotely committed to holding that the very same individual action could be judged both right and wrong (relative to different sets of standards), there being no neutral way of determining which judgement is correct. 17 One could even include Socrates on the list of superficial "relativists" - which surely constitutes something close to a reductio of this use of the term. Socrates' interlocutors, in Plato's early dialogues, will frequently offer initial definitions of virtues in behav- ioral terms - Laches says that courage is facing the enemy and not running away (Laches 190e4-6), Charmides says that s6phrosune is a kind of quiet- ness (Charmides 159b5-6), and so on - and Socrates will typically object that the kind of behavior described would be virtuous only in some circumstances, not in all.'8

For these reasons, it seems preferable to restrict the term "relativism" to the more exciting and controversial view described in section I. In terms of this nomenclature, therefore, neither Thrasymachus nor the author of the Dissoi Logoi is a relativist. We have not yet finished with Gorgias or

" Indeed, on some of the topics we have looked at (such as the question of what is good for the sick and for the healthy), it would be insane not to adopt a measure of this kind of "relativism". 16 For Aristotle see, e.g., NE 1094bl4-16, 1109b14-23; compare also Aristotle's endorse- ment of Gorgias in the Politics passage cited above. 17 At least, the utilitarian is not committed to this view qua utilitarian. A utilitarian could also be a relativist; but if so, it would be for reasons that were independent of utilitaria- nism. (Aristotle, of course, holds no such view.) 18 For discussion of this point, see Terence Irwin, Plato's Moral Theory (Oxford, 1977), pp. 42-47.

149

Protagoras; but so far, we have no grounds for regarding either of them as relativists.

IlI

What, then, of relativism in the deep sense? I now come to the heart of my argument. In this section I shall consider the claim of Gorgias to be called a relativist in this sense; in the following two sections, I shall look at a number of points relating to the Sophists in general.

Gorgias is often associated with relativism; usually this is because some form of relativism is taken to be a corollary of the ideas expressed in his treatise Peri Tou Me Ontos *19 As far as I can see, this has absolutely no basis. I say this not because I take the still widespread view that the treatise is no more than a joke, affording no evidence for Gorgias' actual views - though if this were so, it would, of course, undermine any attempt to foist relativism on him; my reason is simply that the text itself, taken entirely seriously, has nothing whatever to do with relativism. As is well known, the treatise, as reported both by Sextus Empiricus (M. VII.65-86) and in the third section of the pseudo-Aristotelian De Melisso Xenophane Gorgia, contained three main conclusions;20 first, that nothing is, second, that even if anything is, it cannot be known about, and third, that even if anything can be known about, it cannot be communicated to someone else. Of these, the second and the third are forms not of relativism, but of what would now be called scepticism.21 The arguments for both trade, in a manner reminiscent of epistemological scepticism since Descartes, on the idea of an unbridgeable gap between thought, and speech, on the one hand, and that which thought and speech purport to be about, on the other. What these arguments are supposed to show is not that knowledge, or successful reference, is possible only within some framework, shared or otherwise. The point is

19 Writers who draw this connection include Guthrie, History of Greek Philosophy, Vol. III, pp. 272-3, and James N. Jordan, Western Philosophy (New York, 1987), p. 55. ' The differences between the two reports, while far from negligible, need not concern us here. For a very full account of these differences (together with an argument for their essential compatibility), see G. B. Kerferd, "Gorgias on Nature or that which is not", Phronesis 1 (1955), pp. 3-25. 21 This is not, of course, the same as scepticism in the ancient sense, the hallmark of which is the refusal to assert anything in one's own person; in the ancient sceptic's terms, Gorgias "dogmatises" as much as anyone. Modern epistemological scepticism, by contrast, consists in a denial of the possibility of knowledge or certainty, and it is this which Gorgias' conclusions resemble.

150

much simpler; there can be no such thing as knowledge, or successful reference. Knowledge and reference are not relativized, but absolutely denied.

Nor does the first conclusion, that "Nothing is", involve any kind of relativity. It is not suggested that "what is" is relative to a scheme, to a set of perceptions, or anything of the kind; again, there is simply no such thing as "what is". It makes no difference, for the present point, whether the claim is understood (as traditionally) to be that nothing exists, or (as has found favor in recent years) that nothing is, any more than it is not, the bearer of any predicate.22 Either way, and whatever detailed interpretation of the very difficult argument one adopts, a relativizing move is nowhere in sight. Again, the position is sceptical (in the modern sense) rather than relativistic, in that it negates a very basic, pre-reflective assumption - the assumption that things "are".

All of this seems so plain that one may well wonder why anyone would ever have seen Peri Tou Me Ontos as implying relativism. One reason may simply be the careless, yet unfortunately all too common, conflation of relativism and scepticism; but I think there is another factor responsible. Consider what Gorgias' three conclusions imply about the status of human discourse. It follows from these three conclusions that no statement will have any more claim to correctness than any other; hence debate will be a free-for-all, in that no one assertion can be judged superior, on grounds of truth, to the rest. Now, this may well sound like an extreme form of relativism. However, if we examine more closely why it is that no statement will be any more correct than any other, it becomes clear that we are dealing with a different view. All statements are on a par, as far as truth is concerned, not because all statements are true relative to some framework or other, and frameworks cannot be neutrally ranked for correctness, but because all statements are false. They are all "equally true" in the sense that none of them is the slightest bit true. All of them purport to be about onta, "things which are"' (this is so whichever interpretation of esti one adopts); ' For the traditional reading, see, e.g., Guthrie, op.cit., pp. 192-200; for the other interpretation, see Kerferd, "Gorgias on Nature or that which is not", and The Sophistic Movement, pp. 93-100. As the latter passage amply documents, this second interpretation is associated with a much broader shift in the understanding of "being" in early Greek philosophy. 3 It might be objected that this is not the case with avowedly fictional sentences. However, it is doubtful whether Gorgias, or anyone else at the time, had a clear concept of fictional discourse. Apparently he referred to tragedy as a "deception" (apaten, DK 82B23); for a useful discussion of the significance of this, see W.J. Verdenius, "Gorgias' Doctrine of Deception", in G.B. Kerferd (ed.), The Sophists and Their

151

but there are no onta, and even if there were, and they could be known about, discourse about them would be impossible. Speech, therefore, is mere speech, having no connection with anything beyond itself; thus speech as such is deceptive. On this nihilistic and extraordinary view, truth becomes not relative but non-existent.

Could Gorgias seriously have accepted these arguments, and this consequence of them? This is a difficult matter to judge. It may seem that no one could accept them, and that the "parody" interpretation of Peri Tou Me Ontos must be correct. But one has to be careful, when dealing with Greek philosophy, about claiming that some doctrine is "too bizarre to be be- lieved". Gorgias' position, after all, is scarcely any more of an affront to common sense than those of the Eleatics (with which, on any interpretation, it obviously has much in common); one might also mention Xeniades of Corinth who, according to Sextus (M. VII.53) actually did maintain that "every impression and opinion is false". Gorgias' pupil Isocrates, too, apparently takes him at his word; in two passages he reports Gorgias' assertion that "Nothing is", and compares this, in a matter-of-fact tone, with the views of certain other thinkers (Helen 3 and Antidosis 268). Isocrates' own point, on both occasions, is that such speculations are a pointless waste of time; but the implication is that the people concerned, including Gorgias, thought otherwise. As for Gorgias' other writings, there are some echoes of the position sketched here, but also some apparent conflicts with it. However, since most of the surviving material comes from rhetorical set-pieces, one could hardly expect to be able to lean very heavily on it as evidence for its author's philosophical views.24

Legacy, Hermes, Einzelschriften 44, 1981, pp. 116-28. In any case, fictional sentences are not, presumably, candidates for regular truth; so the central point, that Gorgias' arguments imply that no statements are true, still remains. ' The Helen speech remarks on the impossibility of knowledge, and on the deceptive- ness of speech itself (DK 82B11, paras. 11 & 10-14 respectively); the Palamedes refers to the inability of words to convey the truth plainly to their audience (DK 82B1 la, para.35). See also DK 82B26, t6 Xv rtvat &(pavig x 6Xov roe boxriv, r6 bi boxeCv do0evig ji^ rvx6v roe revat. (It is unclear, however, whether >& riX6v should be rendered "since it does not attain" or "if it does not attain". Either way, the fragment refers to a gulf between seeming and being; but on the latter reading, the gulf would not be explicitly stated to be unbridgeable.) But against these passages must be set other parts of the Helen and the Palamedes, which do seem to presuppose the possibility of discovering, and speaking, the truth; see Bli, para.1, and Blla, paras. 4,24,33,34. Kerferd, The Sophistic Movement, pp. 81-2, attempts to reconcile both strands of thought in "a common conceptual model". But I am not sure either that I understand the composite position he sketches, or that this position coheres with all the relevant texts.

152

In sum, there seems to me no strong reason to doubt that Peri Tou Me Ontos was intended to express a serious position. To return, though, to my main point, that position was not, nor did it imply, any form of relativism. It will indeed follow from this position that, as the Helen asserts (DK 82B11, paras. 11 & 13), people's opinions are changeable and insecure; for none of them will be anchored, as it were, to anything outside themselves. This in turn will place the orator, especially one who, like Gorgias himself, possess- es a rhetorical techne, in a very powerful position; with his skills, he will be able to induce in his audience whatever opinions he wants. However, none of this suggests that the correctness of opinions is relative to some framework; to repeat, the malleability of opinions derives from the fact that all opinions are (equally) false, not from the fact that they are all true relative to some framework or other.?5

IV

Gorgias was not, of course, alone in developing a technique of persuasive speech - even though his particular theory of persuasive language may have been idiosyncratic, and more fully worked out than most. The teaching of rhetoric, and an interest in methods for effective public speaking, were the main things the Sophists had in common; it is this shared interest, above all, which justifies their being grouped together, despite considerable differences in what else they taught or thought, as members of a single "movement". Now, it is sometimes alleged that this very activity, teaching and theorising about the art of persuasive speaking, is an indication that the Sophists, as a group, were relativists.' For, it is argued, if one devotes a great deal of one's energy to the question of how best to persuade people of whatever position one wishes - or, as it was often expressed, how to be equally persuasive on either side of a case - one can hardly fail to drift into the view that truth is merely relative to the speaker (or, perhaps, relative to the bewitched listener), that the truth about any given subject simply is whatever position is most persuasively presented at the time. But this line of

I On the connections between Gorgias' philosophical position and his theory and practice of rhetoric, see Verdenius, op.cit. ' See, e.g., the passages from Thilly and (especially) Guthrie cited in note 1; also Carl Joachim Classen, "The Study of Language Among Socrates' Contemporaries", in Carl Joachim Classen (ed.), Sophistik (Wege der Forschung 187: Darmstadt, 1976), p. 228, where relativism about values is held to be a necessary condition of developing a flexible method of persuasion.

153

thinking is very shaky indeed. It is true that a rhetorician might also be a relativist; but the teaching of rhetoric by no means commits one, or even necessarily encourages one, to relativism. First of all, one could very well teach the art of persuasive speech without holding any theory about the nature of truth. But even if one did hold some such theory, there is no reason why that theory should be relativism. One could certainly understand a lifelong teacher of rhetoric ranking persuasion as more important than truth, as truth is traditionally conceived. A dispassionate concern for what is really, objectively the case about some issue is not easily compatible with a concern for being able to argue equally persuasively on any side of it; one could hardly, at any rate, act upon both concerns at the same time. But it by no means follows that a teacher of rhetoric would have to deny that there is any such thing as the objective truth, and hold, instead, that the truth just is whatever one may be persuaded of. Cigarette advertisers presumably rank persuading people to buy cigarettes as more important than the truth about cigarettes; it does not follow that cigarette advertisers are relativists. There is no conflict whatever in holding a standard, objectivist conception of truth, but in having a less than consuming interest in the truth, as thus conceived; to believe that truth is objective is not necessarily to believe that truth is of overriding importance. The tendency to equate the two seems to be strong among philosophers. But it is a misguided tendency, nonetheless. One may hold the two beliefs together, but one is not required to do so. The Sophists' enthusiasm about rhetoric is by itself, then, no evidence of relativism on their part; if any of them were relativists, this has to be determined on other grounds. With this issue in mind, let us now consider some other aspects of the Sophists' concentration on language.

I begin with that interest in linguistic precision which was variously known as orthoepeia, "correct diction", and orthotes onomat6n, "correctness of names".27 Socrates tells us in Plato's Cratylus that the Sophists in general were experts on orthotes onomat6n (391b9-c4); but it was two Sophists, Protagoras and Prodicus, who were especially associated with this subject-8 Of these, Protagoras appears to have been particularly interested

" There is some question whether these two terms were interchangeable. For discussion of this issue, see Guthrie, op.cit., pp. 204-6. 1 I assume that the orthotes grammaton, in which Hippias is said to have been interested at Hippiss Minor 368d4-5, was something quite different. I also leave out of account Democritus, who is said to have written a work nQ'4 'O 'LQoV 6LAoOErEsLT xaQl y)Awatwv (DK 68B20a); my reason is simply that he does not count as a Sophist, as the term is usually understood.

154

in the mechanics of language - in distinctions among grammatical genders, and among what would now be called speech-acts.29 On the topic of gender, moreover, he seems to have been interested in more than mere descriptive classification; a passage from Aristotle (Soph. El. 173bl7-25) suggests that he wanted to alter the existing grammatical genders of words. The well known satirical passage in Aristophanes' Clouds (658ff.) is also evidence of this kind of activity during the period, though Protagoras is not specifically named there. Prodicus, on the other hand, appears to have been especially interested in the precise definitions of words, and particularly in precise distinctions of meaning between nearly synonymous words; reports, and parodies, of this are plentiful in Plato and elsewhere.'

It may be wondered why I have introduced this topic. What possible connection could there be between such researches and Sophistic relativism? My answer is that the study of "correctness of names" seems to have carried with it an interest in the relation between words and the world, and that the prevalent views about this relation, as far as we can uncover them, bear significantly on the question whether the Sophists were relativists. Of course, an interest in the details of language need not by itself indicate any kind of metaphysical theory, relativistic or not. These studies may well have been pursued for their value as part of a rhetorical technique; to the extent that this was so, they would not suggest any particular philosophical views. Even the readiness to reform language need not have been based on anything more than a desire to make language itself, for whatever purpose, neater than one found it.3" On the other hand, Plato's Cratylus, the only extended treatment of "correctness of names" which survives from the period, suggests overwhelmingly that more than merely linguistic consid- erations were involved.

The debate in that dialogue centers around the question whether there is a natural "correctness of names"; and the dialogue opens with a statement of two opposing positions. Cratylus, who answers in the affirmative, holds that, for each segment of reality, there is some one correct term, uniquely appropriate to what it refers to; Hermogenes, on the other hand, holds that there is no more to "correctness" than convention, or an agreement to use

2 On grammatical gender, see Aristotle, Rhetoric 1407b6; on distinctions among speech-acts, Diogenes Laertius IX.53-4. `' E.g., Cratylus 384b, Protagoras 358a, and the passages collected in DK 84A13-19. 31 More would presumably be at stake if the motivation for reform was to bring grammatical gender into line with natural gender. The Clouds passage suggests that this was at least one motivation; again, though, we do not know who (if anyone in particular) is being satirised.

155

words in a certain way. This latter view, as quickly becomes clear, is taken to entail that any use of words is as good as any other - that any individual is at liberty to use words in any way whatever (384d-385a); to say that correct usage is "merely a matter of convention" is, then, apparently tantamount to saying that there is no genuine distinction between "correct" and "incorrect" usage. Now, Hermogenes was the brother of Callias (391cl), the man renowned above all others for his patronage of the Sophists; one might well expect, then, that Hermogenes' position would reflect a position held by one or more of the Sophists. On the other hand, since Hermogenes' position in effect dismisses the whole notion of "correct" versus "incorrect" usage as a fiction, it is hard to see how a serious student of orthotes onomaton could adopt it; for clearly, the study of orthotes onomat6n depends on the assumption that there is some distinction between correct and incorrect usage. It is Cratylus' view32 which, of the two, seems by far the more congenial to such study.

But now, if the study of orthotes onomaton presupposes a view akin to Cratylus', those Sophists who studied the subject were not relativists; quite the reverse. To insist that certain forms of words are "naturally correct" is to imply that the world is of a determinate character, and that there is just one objectively right way of describing any particular portion of it; this is as far from relativism as one could conceive. However, if I were simply to conclude, at this point, that the study of orthotes onomaton was inherently anti-relativistic, there would certainly be objections. It would be replied, first, that I am putting far too much credence in Plato's way of framing the issue; the two positions presented in the Cratylus need not have been the only options thought to be available at the time. Second, it might be said that Hermogenes' view, even if it is inimical to the detailed study of orthotes onomat6n, is at least a view about the topic; and, as I suggested, it could easily reflect a view held by actual Sophists. So even if we do accept Plato's presentation of the alternatives, we have evidence of a radical Sophistic relativism, emerging from reflections about language. Let me take up these two objections, beginning with the second.

Suppose that a view resembling Hermogenes' was indeed held by one or more of the Sophists. Is Hermogenes' view actually a form of relativism? The answer, perhaps surprisingly, is that it is not. Hermogenes explicitly denies allegiance to Protagoras' "Man the Measure" doctrine (386a5- 7,391cS-7) - a doctrine which, as I have admitted, may be a form of

32 I am referring to the Cratylus of the dialogue; the question of the historical Cratylus' views need not concern us here.

156

relativism; instead, he readily agrees with Socrates that things have "some firm essence of their own" (airat ai'rrdv oi'raav Exovtd -Ttva ,fatL6v O= Ta n Q6y,guaa, 386e1-2 - cf.a3-4). His view appears, then, to be this.

Reality is fixed and determinate, but each determinate real entity can just as well receive one name as another; no name is any more "fitting" to its object than any other. Socrates argues that if one repudiates Protagorean- ism, one is committed to Cratylus' view about language (386a-391a). But this is not Hermogenes' original view, and he is reluctant (390e5-391a3) to accept Socrates' argument; he wants to combine a view of reality as determinate with a view of language as arbitrary. Now, this composite view is not relativism. It is true that, according to this view, a sentence such as "the cat is on the mat" may be either true or false, depending on whether "cat" is used in the normal way, or to refer (for example) to what we usually refer to by "dog"; it will also follow that "the cat is on the mat" will be just as true as "le chat est sur le tapis". But these banalities do not amount to a relativism about truth or reality. In Hermogenes' picture, language is not thought of as in any way shaping the reality it describes. Contrary to relativism, the true facts about the world, in this picture, are objective and independent of us; it is merely that these true facts are equally expressible in any number of different, but mutually translatable, notations. To develop any serious kind of relativism out of reflection on language, one would need something like the notion of a conceptual scheme, embodied in a language and imposed on the world through the use of that language; nothing like that is hinted at by Hermogenes, or by any other source of the period. This is not surprising; such a notion would require a degree of sophistication and self-conscious- ness inconceivable in the fifth century, when reflection about language had only just begun. As many have noted, and as the term orthotes onomaton itself suggests, the whole of language, in this period, was generally regarded on the model of proper names; words were thought of as labels for things in the world. This type of view is presupposed as late as Plato's Theaetetus, and is responsible for the inconclusiveness in one part of that dialogue;33 it is not until the Sophist that we see a clear advance beyond it, in the distinction between onomata and rhemata (261e-262e). But the view that words are merely labels for things in the world presupposes that things in the world are the way they are prior to, and independent of, the act of labeling.

It turns out, then, that both of the opposing positions presented in the Cratylus imply an anti-relativist metaphysics. But what of the other ob-

3 In the "dream theory" (201d-202d). On this, and the ensuing criticism (202d-206c), see the commentary of John McDowell (Oxford, 1973), pp. 231-50.

157

jection raised earlier - that the Cratylus provides an artificially narrow conception of the alternatives? Might there not have been a third view, which held language to be conventional, but which still insisted on a valid distinction between correct and incorrect usage? Certainly, this is possible. The idea that language is a human creation is common to Protagoras' myth in Plato's Protagoras (322a6) and to the anthropological account in Diodo- rus Siculus (1.7ff.), which is generally thought to derive from a fifth-century source (traditionally, though not definitely, Democritus); but, to judge from what we saw earlier, Protagoras, at any rate, would not have agreed that any usage is as good as any other. However, these ideas no more suggest relativism than do either of the positions in the Cratylus. The Protagoras passage merely says that primitive humanity "articulated speech and names" (qpwviv xa v6&6ga-ra. . . 6LiQ6QWaaTo), which does not suggest any particular view about the relation of language to the world. And the Diodorus passage seems to suggest the same naive, non-relativist view of the relation as we just mentioned; creating a language is a matter of fixing which series of sounds will serve as the name, or "token" (sumbolon), for which object in the world (I.8,3). The names may be quite arbitrary, as the author observes (kb tuxE, I.8,4); but the objects are simply "out there" waiting to be named, and unaffected by the process of naming.

The upshot is this. The Sophists' researches into the "correctness of names", and various associated ideas about the nature of language, tell against, rather than in favor of, an interpretation of them as relativists about truth, or about the nature of the world. It is, of course, possible that if the evidence about this matter were less fragmentary, it would support a different verdict. But such speculations are necessarily idle; any discussion of the Sophists is bound to be subject to the qualification "as far as we can tell". I now want to glance at one more topic from the realm of language. Two related theses are said to have been held by numerous Sophists; first, that contradiction is impossible, and second, that false statement is impossible. While the ancient sources do not always mention them together, the connection between the two is evident. Briefly, and in modem terminology, contradictory propositions, by definition, have opposing truth-values; if there is no such thing as falsehood, there is no possibility of statements being opposed to one another in truth-value, and hence no possibility of contradiction.' We are told that one or other of these theses was held by I Conversely, if there can be no contradiction, there can be no opposition in truth- values, and hence only one possible truth-value. This does not, however, entail the impossibility of falsehood without the additional assumption that there are some true statements. Presumably most people would allow this - but not everyone, as we saw

158

Protagoras and his associates, by Prodicus, by Antisthenes, and by others not named.35

These positions have sometimes been linked with relativism; 36 and it is true that they would be plausible corollaries of a position which makes truth relative to the framework of each individual speaker. For, if every statement is "true for" the person who delivers it, then two people who deliver what we would normally think of as opposing statements (say, "Astrology is utter mumbo-jumbo" and "Astrology is not utter mumbo-jumbo") are not really contradicting each other; instead of one statement being true and the other false, each is true, but relative to the frameworks of different individuals.37 On the other hand, while these doctrines are certainly, on the face of it, compatible with relativism, there is good reason to suppose that they were not, at the time, generally associated with relativism.

Admittedly, Protagoras is spoken of as one of the proponents of these views; and Protagoras, as I have said throughout, is by far the best candi- date for relativism among the Sophists. So it may be that, in Protagoras' case, they were a by-product of relativism-' However, the view that false-

s Plato, Euthydemus 286c2 tells us that, in addition to many others, including those of a prior era, "those around Protagoras" (which presumably is meant to include Protagoras himself) argued that contradiction is impossible; the argument itself is given 285d7- 286b6. An argument against the possibility of falsehood occurs immediately before (283e-285a); it is quite possible that touton ton logon at 286cl, referring to the argument current in Protagoras' circle, is meant to include this as well. Protagoras is also twice mentioned by Aristotle in the course of his argument against those who deny the law of non-contradiction (Metaphysics 1007b22, 1009a6). Aristotle does not spell out the precise nature of the Protagorean position being criticised; but it clearly included a denial of contradiction in the normal sense. (On this, see further note 38.) Prodicus is cited as holding that ouk estin antilegein in a Greek commentary on Ecclesiastes, discovered on a papyrus in 1941; for text and discussion, see Gerhard Binder and Leo Liesenborghs, "Eine Zuweisung der Sentenz ouk estin antilegein an Prodikos von Keos", Museum Helveticum 23 (1966), pp. 37-43, reprinted with additions in Classen (ed.), Sophistik, pp. 452-62. Aristotle ascribes both views together to Antisthenes at Metaphysics 1024b32-35. Finally, Cratylus 429d2-3 says that "plenty of people" have argued against falsehood. 3' By Kerferd, The Sophistic Movement, pp. 72-3, 88-92, and H.D. Rankin, "Ouk Estin Antilgein", in The Sophist and Their Legacy, pp. 25-37. ' This does not, however, eliminate falsehood and contradiction entirely; it does not rule out self-contradiction, and it allows that statements such as "what you just said is false" are true. But we need not pursue these complications. 38 It appears to be on the basis of his "Man the Measure" doctrine that Aristotle (see note 35) construes Protagoras as hostile to the idea of contradiction; it is presumably this which he has in mind when he alludes, in the course of his counter-argument, to Protagoras' assertion that TeA 8oxoOva n&ivta nv &XiOf xa' LA qKuvquw eva

159

hood is impossible is also discussed in the Cratylus (429b-430a); and here, strikingly, it is Cratylus who is represented as its advocate - whose position on language and its relation to the world is, as we noted, strongly anti- relativist. It appears, then, that Plato, at least, did not regard these views as having any special connection with relativism. And indeed, if we examine the surviving arguments offered for them, their main source becomes clear enough, and it has nothing to do with relativism.39 The impossibility of falsehood and of contradiction are just two of the many puzzles generated by that same naive view of language, as consisting wholly of "names", which we mentioned above.' If all language does is name items in the world, then for any particular piece of language, there are just two possibilities; either it correctly names the item or items it is supposed to be referring to, or it does not. If it does, then it hits its target, as it were, and is true.41 If, however, it does not, then it must either have hit a different target by mistake, or else hit no target at all; in other words, either it is actually about something other than it is supposed to be about, or else it is about nothing, and so mere meaningless noise.42 On this picture, there is clearly no room

(1009a8-9). However, as I said at the beginning, it is not certain whether even the "Man the Measure" doctrine is a form of relativism; more on this in section VI. Nor is it clear, in any case, that the "Man the Measure" doctrine was the only basis for Protagoras' denial of contradiction; see further note 39.

On the question of how to reconcile Protagoras' denial of contradiction with his well-known claim that "on every subject there are two logoi opposed to one another" (DL IX.51), see Kerferd, op.cit., pp. 90-2. 3 Arguments either for the impossibility of falsehood, or for the impossibility of contradiction, or both, occur in the passage on Prodicus from the Ecclesiastes commentary, in the Euthydemus (see note 35 for references to these), and in the Cratylus passage just referred to. All three of them show a very strong family resemblance; and in what follows, I am speaking of all three collectively. Recall that it is to Protagoras that the argument in the Euthydemus is ascribed; this raises some doubt as to whether Protagoras' adoption of these views was simply a consequence of the "Man the Measure" doctrine (see note 38). ? Another puzzle from the same source was the notion that only identical predication (or

perhaps even no predication at all) was possible. 41 Note that, on this view, truth is not a property only of sentences; individual words can be "true" as well (this consequence is quite explicit in the Cratylus). The "dream theory" in the Theaetetus (see note 33) attempts to break away from this, when it draws a distinction between individual names and sentences, and between naming and describing; but because Plato persists in regarding sentences as no more than concatenations of names, he is unable to keep the distinction in focus. 42 The first possibility is suggested by the passage from the Ecclesiastes commentary on Prodicus, the second by the arguments in the Euthydemus; both possibilities are raised in Cratvlus 429b-430a.

160

for two statements which are opposed to one another, one being correct and the other not, yet both of them about the same item in the world.

It is in terms of this picture that the impossibility of falsehood and of contradiction are generally treated; such views are a consequence not of relativism, but of a view of language which has not yet made a distinction between subject and predicate.43 In fact, as I suggested earlier, this view of language - as made up of nothing but names - implies an objectivist picture of the world (whether the "names" used to label the independently existing things in the world are regarded as natural or as conventional). One should note, finally, that even Plato, whom no one would accuse of relativism, fails when, in the Theaetetus (187c-200d), he tackles in his own person the subject of false belief; despite repeated attempts, he cannot explain how false belief is possible. Moreover, his problems stem primarily from an over-simplified conception of thinking - a conception having significant parallels, I believe, with the over-simplified "naming" view of language." In this period, then, one did not need to hold avant-garde metaphysical views, in order to experience puzzlement about falsehood; these difficulties were simply a product of meager analytical resources. Once again, it is the Sophist, which does draw (more or less) the distinction between subject and predicate, that marks the major breakthrough in this area.

V

Let us shift now to the Sophists' ethical and political views. In this area, the feature of their thought which is most often taken to be conducive to, if not actually equivalent to, relativism is the sharp distinction many of them drew

4 Another factor contributing to these difficulties (reinforced by, and in turn rein- forcing, the "naming" view of language) are the slippery phrases legein ta onta and legein ta me onta. As is well known, ta onta, "the things which are", may be used to refer both to real entities and to that which is the case; no clear distinction is drawn between these two ideas. As a result (together with some flexibility in the word legein), legein ta onta may be understood both as "speaking of what is real" and as "speaking the truth"; hence legein ta me onta may be understood either as "speaking of what is not real" or as "saying what is not true". In the absence of a clear distinction between these two phrases, it is natural that falsehood would be assimilated with a failure to refer to anything. (The same assimilation is apparent in the tendency to equate to me on, "that which is not", with meden, "nothing"; the locus classicus for this is Parmenides (DK 28) B2.7-8 and B6. 1-2.) 44 I cannot elaborate on these parallels. But if, as I suggested, a version of the "naming" view of language also makes an appearance in the Theaetetus, they would not be surprising.

161

between physis and nomos.45 A great deal has been written about this distinction, and about the meanings of the individual terms. To the extent that translation is necessary, I shall renderphysis by "nature" and nomos by "norm";`6 but we need not concern ourselves further with philological details. For our purposes, the crucial point is this. To distinguish sharply between physis and nomos is, among other things, to recognize that the existing norms of one's society (legal, ethical, religious or whatever47) have no absolute status, but are in some sense a human construct.' (The distinction, therefore, is a natural partner of the numerous accounts of the origin of human society produced in this period.) Now, this recognition does not by itself constitute moral relativism; it is quite compatible with the view that there is an absolute moral order, but one which existing, humanly constructed, norms partially or wholly fail to reflect. However, it is also easy to imagine a form of relativism which does build upon the physis- nomos distinction, along the following lines. "It makes no sense to talk of things being right or wrong physei. Rather, there are merely various sets of nomoi in various different communities; and rightness and wrongness, in any given community, is relative to the nomoi prevalent in that community."

Did any of the Sophists hold views resembling the one just outlined? There is no reason to think so. The ways in which the Sophists build upon the physis-nomos distinction are of broadly two kinds. On the one hand, several of them insisted that there was some kind of "natural" rightness, or justice, and criticized existing norms for diverging from it. Within this first broad category, some took natural justice to consist in the unrestrained pursuit of self-interest, on which nomoi place unnatural curbs; others took natural justice to involve such things as equality between Greeks and barbarians, the free and the enslaved, or people of high and low birth. These views are not mutually exclusive; the large papyrus fragment from Antiphon's work Truth (DK 87B44) seems to contain versions of both.49 " Connections between this distinction and ethical relativism are affirmed by Guthrie, op.cit., p. 59, and by Meiland and Krausz in their Introduction to Relativism: Cognitive and Moral, p. 6. 4 Following Kerferd, op.cit., p. 112, who stresses the prescriptive character of the term nomos.

4' Including, perhaps, linguistic norms; but we have already said enough about the Sophists' ideas on language. 4 The Greeks tended to think of nomoi as having been devised by some single individual, a nomothetes; but this does not affect the present discussion. 4 On the consistency of the two in Antiphon, see David J. Furley, "Antiphon's Case Against Justice", The Sophists and Their Legacy, pp. 81-91. (This had often been seen as

162

Besides him, Callicles in Plato's Gorgias is the clearest specimen of the first idea; and Plato makes Hippias an exponent of the second at Protagoras 337c7-d3. But echoes of one or the other line of thought are apparent in Aristophanes' Clouds, in Thucydides, and elsewhere;51 evidently, both acquired some considerable currency. Equally clearly, though, they are not versions of relativism. To downgrade existing norms by comparison with some supposed natural standard of right - whatever that natural standard might be - is precisely to reject relativism about values; it is to set up an objective, neutral criterion by which existing norms can all be weighed on a common scale.

Other Sophists developed the physis-nomos distinction in a different direction, arguing that, even though nomoi are indeed a human construct, they are vitally important for the survival of human society, and for human well-being. The most famous example of this line of thought occurs in the myth which Plato, presumably with some historical justification, puts in Protagoras' mouth, and which we have already had occasion to mention (Protagoras 322a-323c, 324d-325b, 327c-d). Another, rather less striking instance is the "Anonymus Iamblichi", one part of whose essay deals with the necessity of nomoi for communal stability, which in turn is a necessary condition for the living of a worthwhile human life (DK 89.6-7).51 This second outgrowth of the physis-nomos distinction is not, like the first, in direct opposition with relativism; but neither does it have any essential connection with relativism. Ethical relativism is an answer to the question "What is it that, at bottom, makes some types of behavior right and other types wrong?" In emphasizing the crucial importance of nomoi, these thinkers are not addressing that question. They are not here concerned with whether the nomoi which certify some actions as right and others as wrong have themselves any objective basis, or are instead the end-point of ethical

problematic - but quite needlessly, as Furley demonstrates.) s' In the Clouds, see especially the debate between the Just and Unjust Logoi (889- 1114); in Thucydides, especially the Mitylenean debate (III.38-48) and the Melian Dialogue (V.85-111). Note also Euripides, frr. 433, 920 Nauck. Thrasymachus is often added to the list of the supporters of physis, even though he does not express his position specifically in terms of the physis-nomos distinction; on Thrasymachus' relation with Antiphon and Callicles, see Furley, op.cit. But we have already seen that Thrasymachus is not a relativist, except in the trivial sense. "' Plato's Crito is sometimes taken as a further example. But the Crito does not, in fact, portray nomoi as the product of a human agreement; what is said to derive from an agreement is, rather, the individual's obligation to obey the nomoi of the community in which he or she elects to live. This point is well discussed by Charles H. Kahn, "The Origins of Social Contract Theory", The Sophists and Their Legacy, pp. 92-108.

163

justification, not open to further assessment; they are not interested in the status of nomoi, but in the reasons for caring about them. Compatibly with insisting that nomoi must at all costs be upheld, one might very well (though one need not) also hold that nomoi, though a human construct, are behold- en to some external criterion of correctness,52 and that some nomoi are, in some neutral sense, better than others. As we said, the fact that particular sets of nomoi were created by particular communities does not imply that there can be no objective ranking of different sets of nomoi; and the claim that we cannot do without nomoi does not imply that nomoi are immune from criticism. The position represented in Protagoras' speech, and expressed by the Anonymus, says nothing, one way or the other, about the existence of objective, or natural, standards of rightness; it is a contribution to a quite different debate. (It is misleading, therefore, to divide the Sophists into two opposing groups, "the supporters of nomos" and "the supporters of physis". For the two sets of ideas we have been looking at do not represent conflicting positions on the same subject.)

So neither of the Sophists' major developments of the physis-nomos distinction are evidence for ethical relativism. In the Theaetetus, Protagoras is represented as holding that whatever is considered just in any particular community is just for that community (172a-b); but this is a consequence of his "Man the Measure" doctrine, not of his views about physis and nomos. We have still seen nothing, aside from the "Man the Measure" doctrine, which suggests relativism among the Sophists.

One further related point can be briefly dealt with. Some of the Sophists seem to have offered naturalistic explanations for the origins of religion. Prodicus is said to have held that primitive humans revered as divinities the various parts of the natural world that gave them sustenance - crops, rivers, the sun and so on - and also, according to some reports, the people who made discoveries beneficial to human life.53 And the celebrated fragment of the satyr-play Sisyphus (the authorship of which is debatable) portrays belief in the gods as the product of a clever politician's swindle, designed to induce law-abiding behavior when the authorities were not watching.54 52 The Anonymus Iamblichi actually says at one point (DK 89.6,1) that nomos and justice are "secured by physis". But the train of thought is hardly pellucid. " For reports on Prodicus' views about religion, see the passages collected in DK 84B5, and the larger collection in M. Untersteiner, I Sofisti, Testimonianze e Frammenti, fasc.II, (2nd ed. Florence, 1961), pp. 194-6. 5 The Sisyphus was traditionally thought to be by Critias, on the information of Sextus; but recently opinion has shifted somewhat in favor of Euripides having been the author, as reported by Aetius (both passages in DK 88B25). For the reasons favoring Euripides, see especially A. Dihle, "Das Satyrspiel 'Sisyphus' ", Hermes 105 (1977), pp. 28-42.

164

Both views were routinely labeled "atheism" in the ancient world; and despite the somewhat indiscriminate use of this term by the conventionally pious (both ancient and modern), the label would appear to be fully justified. Both accounts explain away religious belief as the product of an error - a deliberately induced error in the latter case, an error born of confused gratitude in the former. But if so, relativism is once again quite absent. To explain away a belief is to show that it is false (while also showing how it could easily have been thought to be true) - not to show that it is true, but only relative to some framework."

Obviously, I have not discussed every aspect of Sophistic thinking. What I have done, I hope, is examine those aspects of Sophistic thinking which might, with some semblance of plausibility, be thought to support an attribution of some form of relativism,' and shown that they do not support

Either way, the play affords evidence of a view about the origin of religion which was held, or at least contemplated, during the period. " Two texts, both commonly cited in this kind of context, may seem to be troublesome to my position. First, Euripides, Hecuba 800-1 says that "It is by nomos that we believe in the Gods, and distinguish justice and injustice in our lives". Second, Plato, Laws 889e3-890a2 ascribes to certain unnamed thinkers the view that the Gods, and justice, do not existphysei, but only by nomos, and that they vary depending on the different nomoi that hold sway at different times and places. Both seem to express some form of relativism concerning both religion and ethics; and both may well be thought to allude to Sophistic ideas. However, in the Euripides passage, either "nomos" refers (as some have thought) to a Law inherent in the universe - in which case relativism would not be at issue - or the train of thought in the passage is confused. For immediately before (798-800), Hecuba says that we humans are weak, but that the Gods are strong, and that nomos has power over the gods; if "nomos" refers to humanly constructed norms, the passage is saying, very oddly, that these norms are superior, by two removes, to the beings who constructed them. As for the Plato passage, it is again very hard, on closer inspection, to extract any coherent thesis. For the view that Gods exist only by nomos appears to be equated with the view that "there are no Gods such as nomos prescribes that one must believe in" (890a6-7); and the view that justice exists only by nomos appears to be equated with the view that the justest course (89Oa4), also called the kata physin orthon bion (a8), consists in gaining power by force. I doubt, therefore, that either passage can be taken as evidence for any particular philosophical position. Euripides is indulging, as often, in some modish intellectual gymnastics. And Plato is treating us to an indiscriminate tirade against those thinkers (Sophists no doubt included) who, on whatever grounds, might seem to be subverting religious belief; the details of the views he is attacking are not of any great interest to him. ' Two points I have not considered are the obscure fragment DK 87B1, from An- tiphon's Truth, and the portrayal of Prodicus in the pseudo-Platonic Eryxias (397c-399a) . But neither, on examination, affords any evidence of relativism; in saying this, I am following the arguments of J.S. Morrison, "The Truth of Antiphon", Phronesis 8 (1963), pp. 3549, and G.B. Kerferd, "The 'Relativism' of Prodicus", Bulletin of the John

165

anything of the kind. The one Sophist over whom we have left a question- mark is Protagoras. It is time to say a word about Protagoras' "Man the Measure" doctrine, before bringing the paper to a close.

VI

We have already had reason to refer several times to Protagoras' assertion that "Man is the measure of all things, of the things that are, that [or "how"] they are, of the things that are not, that [or "how"] they are not", which is quoted in Plato's Theaetetus (152a2-4), in Sextus (PH 1.216 and M. VII.60) and in Diogenes Laertius (IX.51). Plato and Sextus both offer interpretations of Protagoras' doctrine, and it is these, rather than the exact wording of the fragment itself, which bear on the question whether or not Protagoras was a relativist.57 According to both interpretations, Protagoras held that, whenever someone perceives that something is the case, that thing is the case for, or in relation to, the person whose perception it is (Tht. 152b, M. VII.60). Both agree, also, in extending the scope of the doctrine beyond sense-perception in a narrow sense (even though, in the Theaetetus, it is first introduced as a corollary of the thesis that "knowledge is perception"); according to Sextus' interpretation, it covers all opinions (VII.60), and according to Plato, it at least covers all matters except those having to do with what is healthy or beneficial (171e, 172a5-b2).58 Now, it might seem obvious, at this point, that Protagoras is being interpreted as a relativist; to say that perceptions and opinions are "true for" the person who has them is surely to put forward a radical relativism about truth. But unfortunately, Sextus' interpretation contains a further component which changes the

Rylands Library 37 (1954), pp. 249-56. s7 It has long been a subject of controversy how Protagoras' words are best to be translated; in particular, argument has centered around the words anthropos, chrematon and (especially) hos. Succinct discussions of these issues occur in Laszlo Versenyi, "Protagoras' Man-Measure Fragment", American Journal of Philology 83 (1962), pp. 178-84, and Guthrie, op.cit., pp. 188-92. But the precise manner in which these questions are resolved will not affect whether Protagoras was a relativist; they will only affect what kind of relativist he was, if he was one at all. Whether or not he was any sort of relativist depends on how we are to understand the metaphor of human beings as a "measure" of things; but that takes us far beyond questions of translation. 5 Whether this exception was made by Protagoras himself, or whether Socrates is charitably altering the original position on Protagoras' behalf, so as to render it (as he sees it) more defensible, is a matter of some dispute. McDowell, op.cit., ad loc., holds the latter view; Kerferd, The Sophistic Movement, pp. 104-6, holds the former.

166

picture. According to Sextus (PH 1.217-19), Protagoras held that things are really, in themselves, all the various ways they are perceived as being by different people. Different aspects of reality make themselves manifest to different people, depending on the people's conditions (depending on whether they are young or old, healthy or sick, and so on); and in this sense, any given perception is true only "in relation to" (pros) the person who is in the appropriate condition to apprehend that particular aspect of reality. Nonetheless, all perceptions are true in an objective sense, in that every perception does, indeed, apprehend some aspect of an objective reality - despite the fact that, as we would normally say, some perceptions contra- dict others. And this makes Protagoras not a relativist but, as E.R. Dodds long ago pointed out, "an extreme realist".59 The everyday law of non- contradiction will indeed be suspended; but this is not because of a relativism about truth (which in any case, as we saw at the beginning, could very well uphold the law of non-contradiction), but because of a metaphysical theory which makes reality itself inherently contradictory. Sextus does say that, according to this doctrine, truth is relative (trv 'W s6g tL stvau tiiv &XAXiELav, M. VII.60). But it is relative only in the sense that which of the many (objective) truths about reality it is open to one to apprehend depends on one's physical or mental state. What is true is not itself, on this view, a relative matter.

The really difficult question is whether Plato's interpretation agrees with this non-relativistic interpretation offered by Sextus, or whether Plato actually does portray Protagoras as a relativist. This is a much disputed matter, which I cannot hope to resolve at this point.' (There is also the question whether, if Plato's interpretation does differ, it should necessarily be preferred to Sextus'; but few, I take it, would doubt that the answer to this latter question is "yes".) Part of the trouble is that it is not always clear when Plato is summarizing views that were actually propounded by Protag-

I "The Sophistic Movement and the Failure of Greek Liberalism", in Dodds' collection

The Ancient Concept of Progress (Oxford, 1973); the quoted phrase appears on p. 95. ' Dodds, loc.cit., asserts without argument that Plato, as well as Sextus, makes Protago- ras an "extreme realist". Kerferd, op.cit., ch.9, argues for the same interpretation as Dodds, calling it the "objectivist" interpretation, but calls Protagoras' position "relativism" nonetheless. Myles Burnyeat, "Protagoras and Self-Refutation in Later Greek Philosophy", and "Protagoras and Self-Refutation in Plato's Theaetetus", Philosophical Review 85 (1976), pp. 44-69 and 172-95 respectively, distinguishes relativism from the position Sextus attributes to Protagoras, and argues that Plato, contrary to Sextus, interprets Protagoras (correctly) as a relativist. (Burnyeat (p. 46) sees the mistaken interpretation as beginning with Aristotle, in his discussion of the law of non-contradiction - see notes 35 and 38 above.)

167

oras, and when he is showing us what, in his view, Protagoras is committed to (but did not say), or what, in his view, would be helpful to Protagoras' argument (but was not used by Protagoras himself). Another problem is that the Theaetetus is not, in any case, primarily an analysis of Protagoras' position, but a search for a definition of knowledge. Protagoras' doctrine is said to be equivalent to the claim that "knowledge is perception" (151e2- 152a4, 160dS-e2); but that it is equivalent is presumably Plato's judgement. Suffice it to say that a respectable case can be made for the conclusion that Plato interprets Protagoras as a relativist, in the deep sense we have throughout been interested in6' - but that this conclusion is by no means beyond question.62

If that is indeed Plato's interpretation, Plato's Theaetetus becomes the one piece of evidence for relativism, in the deep sense, among any of the Sophists. As we have seen, none of the other evidence concerning t

Bett the Sophists and Relativism

Documents

Transcript of Bett the Sophists and Relativism