SORITES, ISSN 1135-1349

Issue #13 -- October 2001. Pp. 80-89

Gorgias the Sophist on Not Being: A Wittgensteinian Interpretation

Copyright © by SORITES and Michael Bakaoukas

Gorgias the Sophist on Not Being: A Wittgensteinian Interpretation

Michael Bakaoukas

Provocative aphorisms of the most notable fifth century Sophist, Gorgias, such as «Nothing actually exists» or his realist tenet that «it is not speech (logos) that serves to reveal the external object, but the external object that proves to be explanatory of speech» (DK 82B 85) have been subject to endless cycles of interpretation. I do not propose to offer here a full analysis of Gorgias' thought, but rather some new suggestions as to how to interpret Gorgias on the basis of how scholars have interpreted Gorgias.

Untersteiner (1954: 163-5) and Kerferd (1981a: 93-95) distinguish between three approaches to Gorgianic texts. According to the first approach, Gorgias' treatise On What is Not is just a rhetorical parody of philosophical doctrines (philological or rhetorical approach) [Bux, 1941: 403 ff]. Following the rhetorical approach, researchers in speech communication and rhetoric attempt to attribute to Gorgias an epistemology and a genuine philosophy of rhetoric (Gronbeck, 1972: 36 -- Engnell, 1973 -- Enos, 1976 -- Cascardi, 1983 -- Walters, 1994).

According to the second approach, Gorgias is just a nihilist (or a negative dogmatic or a forerunner of scepticism) attacking the doctrines of the Eleatics and the Presocratics (ontological approach). There are many interpreters who hold that Gorgias is attacking the ontological doctrines of the Presocratics: Grote (1869: VII 331 ff and 1875: 107-109), Gomperz Th. (1901: 480-496), Maier (1913: 223-226), Reinhardt (1916: 39 ff), Joel (1921: 726), Nestle (1922: 554), Lattanzi (1932), Calogero (1932: 157-222), Brocker (1958: 438), Mondolfo (1936: 177-182), Levi (1941: 32-34 and 1966: 204 ff), Zeller (1963: 1305-1310), Sicking (1964: 225 ff), Guthrie (1969: 199 and 1971: ch. 11), etc.

According to Bakaoukas' Ph.D. dissertation, both Gorgias and Aristotle refer to the contradicting views of some presocratic philosophers who argue with each other about one and the same thing, i.e. the «being» (on). For Aristotle, «we cannot be right in holding the contradicting views [sc. of Heracleitus and Anaxagoras]. If we could, it would follow that contraries are predicable of the same subject [sc. which is not the case]» (Metaph. K 1063b24-26). In the same way, Gorgias says in his rhetorical work Palamedes that we should not believe those people who contradict themselves (Pal. 25). Obviously, the «quarrelling» philosophers at issue (in Gorgias' time) are the Atomists and the Eleatics. As far as we can tell from Gorgias' treatise On What is Not, the Gorgianic arguments and counter arguments refer to the Eleatics who had engaged in a controversy with the Atomists about being and non-being (or kenon).

In the third approach (which attributes to Gorgias an interesting philosophic position), Gorgias is seriously interested in the problems of predication and meaning (philosophical approach) [Kerferd, Mourelatos, etc.]. According to Kerferd, «there is nothing in the treatise (sc. of Gorgias) which might not have been expressed by Gorgias in the fifth century and there the matter is perhaps best left [5] (...) there have indeed been those who have treated the work seriously. But its interpretation undoubtedly presents quite extraordinary difficulties, and those who have treated it seriously have arrived at very different views as to what Gorgias is saying [3]» (Kerferd, 1955: 3, 5). So, «what is needed, I believe, at the present stage of Gorgianic scholarship is a programme of discussion and research (...) -- this is to identify certain broad philosophic features in Gorgias' thought in order to provide a kind of philosophic sketch-map» (Kerferd, 1981[b]: 322-3).

For Mourelatos the presocratic fragmentary texts encourage the study of commentaries and interpretations, that is to say the study of «secondary literature». The original works are lost, so one should «seek to come to terms with alternative views already on record» (Mourelatos, 1993: 1). This goes for Gorgias' paraphrased, fragmentary texts as well, i.e. Sextus' sceptical paraphrase of Gorgias' treatise On What is Not (2nd cent. AD) [hereafter DK B30] and the pseudo-Aristotelian paraphrase of Gorgias' On What is not De Melisso Xenophane Gorgia (2nd cent. BC) [hereafter MXG)

Following the third approach, Grote (1869: VII 331 ff) and then Joel (1921: 726) were the first scholars who attempted to interpret Gorgias from a modern philosophical point of view. They put forward a Kantian interpretation according to which Gorgias distinguished between the phenomenal and the noumenal. According to them, the Gorgianic word `being' refers to ultra-phenomenal or noumenal object of which Gorgias denied the existence (not being). In this regard, according to Hamberger (1914: 53, 55), Plato misunderstood the Gorgianic theory on the relationship between noumenal and phenomenal objects. Newiger (1973: 186) emphasises the same interpretative line. But, as Chiapelli (1890) points out, we should translate the Gorgianic «not being» into «unknowable material Being», for the distinction between «phenomenal» and «noumenal» is foreign to all ancient thought before Plato (cf. Untersteiner, 1954: 164, n. 2).

A modern philosophical interpretation is also adopted by Mansfeld, stated as follows: «The point of Gorgias' argument seems to be that the only knowledge (...) is absolute or unqualified knowledge, i.e., knowledge of things as they are in themselves. Personal knowledge, being relative and not of things as they really are in themselves, is not knowledge in the required sense ... is not absolute knowledge of things as they really are but personal knowledge of things as they are experienced. This knowledge cannot be communicated to someone else» (Mansfeld 1985: 252). In this regard, Mansfeld (1985: 258) holds that «some of Gorgias' points (...) are philosophically immensely interesting because they deal with the problem of private vs public knowledge».

Recent interpretations of Gorgias' texts treat Gorgianic arguments as serious and valid. For example Schiappa and Hoffman say that «we ought to treat the On What is Not as a work of careful argumentation and not of inconsiderable philosophical significance» (Schiappa and Hoffman, 1994: 160). According to them, Gorgias refutes successfully the Parmenidean premise «if (A) can mention (O) or can think of (O), then (O) exists». Along this line of reasoning, Gorgias refutes the claim that what is thought of is necessarily existent (DK B3 79); that is he argues «against the existence of thought-about-objects» using a reductio ad absurdum. Namely, the Parmenidean premise «if (A) can mention (O) or can think of (O), then (O) exists» is refuted, for we can think of non-existent things like chimera or chariots running over the sea. Hence, there is no «identity relationship between things-thought-about and things-that-are» (Schiappa and Hoffman, 1994: 157-8). For Barnes as well (1993: 171) Parmenides' premise is fallacious, «for Scylla and Chimera, and many non-entities are, as the Sophist Gorgias says, thought upon». Also it is noteworthy that logicians like Bochenski (1951: 17) and Thom (1986) take Gorgias' arguments into serious consideration.

As regards the relation between Plato and Gorgias it has been argued by Newiger (1973: 177-188) and Hays (1990: 336-7) that there are some important parallels between Gorgias' On What is not and Plato's Parmenides, Meno, Theaetetus and Sophist. These parallels have not yet been investigated in detail. As Calogero and Mansfeld point out, «there is not a systematic comparison of concrete parallels between Gorgias and Plato» (Calogero, 269 ff., 311ff; Mansfeld, 1985: 258, n. 48). In this respect, the philosophical implications of Gorgias' views at issue are very important for future studies, for in order to compare Plato's and Gorgias' arguments we should first examine Gorgias' own views (M. Angelini: 2).

In Hay's words (1990: 336-7), «it would seem prudent for scholars of Plato to re-acquaint themselves with the treatise and to keep in mind that Plato had to respond to these Gorgianic arguments». Crivelli (1996) who holds that the target of Plato's Sophist is Gorgias has done himself this to a certain extent. Many parallels between Plato's Sophist and Gorgias' works corroborate this approach. That is, Gorgias' example of «thinking of a non-existent entity» is «a flying man» (DK B3 79) which is reminiscent of Plato's example of «flying men» (Theaetetus 158b3-4; Sophist 263a8). Also Gorgias' treatment of the contradictory and contrary properties (DK B3 67, 80) is reminiscent of the sophistic argument in the Sophist (240b5, 240d6-8, 257b3-4, 258e6). Furthermore, Gorgias' arguments «had posed formidable challenges to Eleatic philosophy, and (...) [Plato's] quest for forms was particularly vulnerable to the same arguments, because its ontological assumptions were similar to those of Eleatics» (Hays, 1990: 336).

Researchers in Gorgianic scholarship recognise the philosophical significance of Gorgias. In this respect, Mourelatos and Kerferd attribute to Gorgias a sophisticated theory of meaning and interpret Gorgias in terms of a theory of meaning. Mourelatos detects in Gorgias' On What is Not and Helen a behavioural account of the nature of meaning as opposed to a referential and an ideational one. For Mourelatos, «Gorgias attacks two captivating conceptions of the nature of linguistic meaning, viz., that meaning is reference, and that meaning is mental image or idea. The attack is in the form of a series of puzzles. These are by no means sophistic in the pejorative sense. Indeed, the puzzles have recurred in the history of philosophy and have specifically played a role in the development of twentieth-century philosophical analysis» (Mourelatos, 1987: 136). So, according to this linguistic interpretation, Gorgias has an interest in questions such as «do words acquire their meaning by their association with external things, ideas, or behaviour?» or «what do words stand for?»

The third part of the On What is not (DK B3 83-87; MXG 980a18-b19) refers to «incommunicability» (But even if they are known, how could anyone communicate them to another? (MXG 980 a19-20; cf. 979a11-14; DK B3 65). According to Mourelatos, Gorgias here deals with «the inability of logos (speech) to communicate reality to another person, (...) [for] logos cannot furnish, constitute, or represent the external reality» to the effect that communication is undercut (Mourelatos, 1987: 138). Gorgias states «incommunicability» as follows: «Thus (...) since the existent subsists externally, it will not become our speech; and not being speech it will not be made clear to another person» (DK B3 84).

But does Gorgias actually undercut intelligent verbal communication? This is not actually the case, since for Mourelatos «if both speaker and listener have seen (or heard, as the case may be), the thing to which the speaker's words refer communication should be perfectly possible after all» (Mourelatos, 1987: 139-140). For Kerferd as well, «such communication is impossible unless the listener has himself seen the visible object, [so that] one man can learn from another» (Kerferd, 1981b: 324).

However, as Gorgias put it, «and the speaker speaks, but he does not speak a colour or a thing. Anything, then, which a man has not in his own consciousness, how can he acquire it from the word from another, or by any sign which is different from the thing except by seeing it if it is a colour, or hearing it if it is a sound?» (MXG 980 b2-8 tr. Hett). Gorgias herewith attacks a referential theory of meaning according to which «if words are to have meaning, they must refer to things in the real (at least extra-linguistic and perhaps also extramental) world» (Mourelatos, 1987: 151). He says bluntly that «the speaker speaks not a colour nor a sound, nor any other thing; he speaks logos (combining lines 980b2-3 and b6). Blunt, even simplistic though the formulation may be, the argument is by no means trivial. As an elenchus of the referential conception, the argument has fully as much force as the refined modern version of it: we do not eat the meaning of `cake'» (Mourelatos, 1987: 153). Furthermore, in Mourelatos' view, Gorgias herewith objects to an empiricist, ideational conception of meaning according to which words have or acquire their meaning by «some sort of tie or pairing with perceptions (sensory impressions or mental images or thoughts)» (Mourelatos, 1987: 146, 151).

For Mourelatos (1987: 145), Gorgias possesses the concept of mental image or sensory impression (Helen 17: image of the things that are seen); on the basis of this concept, Gorgias uses the argument from perceptual identity or perceptual sameness to show that sensations, sensory impressions or mental images (eikones) are not the same to different observers and in different perceptual conditions. So given the assumption that meaning is mental image, «there would always be doubts as to whether a given word has the same meaning when used by different speakers, or when used by the same speaker at different times» to the effect that intelligent verbal communication would be impossible (Mourelatos, 1987: 154).

That is, as Wittgenstein would put it, if words acquire their meaning by their association to sensations, we are under the spell of a misguided metaphysics, for «when we think about the relation of objects to our experiences of objects (...) we are tempted to conceive of two distinct kinds of worlds the mental and the physical (...) It is against such temptations that the private language argument is directed. But concentrating on S (sensation) while enunciating `S' (a word) does not bring it about that I will remember that `S' means S, unless concentrating on S will transform the [verbal] sound `S' into the expression of a concept. If it does not, then subsequent enunciations of `S' will be empty noises, `whatever is going to seem right to me is right' (Phil. Inv. 258), for no standard has been established by reference to which the subsequent use of `S' can be evaluated as correct or incorrect» (Hacker, 1972: 223-4).

That is to say, sensations as such are meaningless. Consequently, if words acquired their meaning by their association to sensations, we would be in a state of «incommunicability» and «meaninglessness» in which, according to Gorgias: «Even if anything is apprehensible, yet of a surety it is inexpressible and incommunicable to one's neighbour» (DK B3 65), and «no one could say anything false» (MXG 980a11). In the same way, Lazerowitz and Mourelatos use Wittgestein's lectures on private experience and sense data in interpreting Gorgias (Lazerowitz, 1968: 37, Mourelatos, 1987: 154-5, n. 45).

Gorgias also emphasises the common sense fact that it is not necessary for many observers to see exactly the same thing at the same time (MXG 980b 9-19). In this case, he says that there is nothing to prevent a thing from seeming different to many persons MXG 980b12), since these persons are supposed to be neither exactly the same nor probably on the same vantage ground (MXG 980b12-13). He simply points out that two persons can perceive the same object differently, and thus there may be two different appearances of the same object (MXG 980b10-11). This Gorgianic argument is a simple formulation of the argument from illusion according to which: «a physical object may at the same time look A to one observer and B to another observer; but it cannot be both A and B, for that would be a self contradiction» (Hirst, 1959: 46).

In this case, Gorgias says, it is difficult for someone to have exactly the same sense-experience with somebody else's sense-experience of the same thing. As Gorgias put it, «for it is impossible for the same thing to exist in several separate persons; for the one would be two ... there is nothing to prevent it from not being the same in them all, seeing that they are not in every way alike, nor in the same place; for if anything were this, it would be one and not two [...] so that one man can hardly perceive the same thing as another» (MXG 980b10-18).

In modern terms, what Gorgias says is simple. That is, if one physical object has two different appearances, when perceived by two different persons, then what could explain its phenomenal duplication is the possibility of there being two objects with two different appearances -- which, as he says, is absurd, since the one thing in question would be two different things (MXG 980b12-13). Therefore, what changes appearances should not be the object itself, but the sensible things (aistheta), which vary from man to man. Sensible things are as many as the percipients, they are subjective, private to their owners, unobserved by others, and consequently they cannot be identified with the unique thing. In consequence, our subjective sense-experiences of a single thing and the thing itself are regarded as two separate items (Bakaoukas, 2001).

To interpret this passage Mourelatos adopts a phenomenological reading. He uses the argument from perceptual identity. For Mourelatos, «the `one' and the `same' which cannot become `two' and `different' is not an external third thing; it is simply the perception or experience or thought» (Mourelatos, 1987: 143). So Gorgias formulates an epistemological puzzle: «even we should allow that the same external thing should somehow also be `in' two knowing subjects, it need not appear the same to them, because the two subjects are differently constituted and differently placed» (Mourelatos, 1987: 143).

So the question raised by Gorgias is «how can two minds have the same perception?» or «is perceptual identity or sameness possible?» According to Mourelatos, to solve this puzzle Gorgias uses a metaphysical device. He says that two different subjects do not have the same perceptions (tauton) but similar ones (homoion). He substitutes similar (homoion) for numerically the same (tauton). So, as Mourelatos put it, «and since similarity admits of degrees (...) perceptions may not be exactly similar, after all» (Mourelatos, 1987: 144). So, for Mourelatos, if we assume that meaning is mental (or sensory) image, «there would always be doubts as to whether a given word has the same meaning when used by different speakers, or when used by the same speaker at different times» (Mourelatos, 1987:154). This phenomenological reading is justified by Kerferd's view that what concerns Gorgias is «the status of objects of perception ... with primary reference to phenomenal objects» (Kerferd, 1955: 5, 24).

For Mourelatos, «Gorgias has denied the proposition that language has the function of `representing' or `exhibiting', or `setting forth'» (parastatikos) something that is extra-linguistic (first half of the concluding statement in section DK B3 85)» (Mourelatos, 1987: 160). The Sophist does not espouse an ideational theory of meaning either. So what is left is a behavioural theory of meaning. In Mourelatos' words, «it is rather uncanny how closely the vocabulary of section 85 resembles the vocabulary of modern behaviourist theory. External objects (...) `fall upon us' or `make an impact on us' or `impinge upon us' (prospiptonton, hypoptoseos)» (Mourelatos, 1987: 163).

So, according to Mourelatos, Gorgias espouses a behavioural conception of meaning. Gorgias believes that a word has effect on other speakers of the language. For example, he says: «in response to the happy and unhappy occurrences affecting things and bodies, the soul comes itself to experience a certain emotion, through logos» (Helen 9, tr. Mourelatos, 1987: 156-7). For Mourelatos, this Gorgianic position «is an illustration of the conception of words as substitute stimuli (Mourelatos, 1987: 157). Furthermore, Gorgias compares the power of logos with that of drugs («just as different drugs draw different humours from the body (...) so too with logoi» Helen 14 tr. Mourelatos, 1987: 157). As Mourelatos put it, «if only we changed the archaic expression `drawing out humours' to the behaviourist idiom of `eliciting a physiological reaction' this sentence could just as well have been written by such advocates of the stimulus-response conception of meaning as Leonard Bloomfield, or B.F. Skinner, or C.L. Stevenson» (Mourelatos, 1987: 158).

Furthermore, Kerferd interprets DK B3 83-85 as follows: «communication is exclusively by means of speech or words, and the externally existing objects are not words. There is no possibility of converting things into words, and as a result there is no possibility of communicating things through, or by means of, words. This sets up an unabridged and unbridgeable gulf between words and things» (Kerferd, 1984: 218. Cf. Mazzara, 1983: 130 ff.). The text speaks clearly about words being ontologically different from things (Kyrkos, 1993: 299 -- Jaekel, 1988 -- Rodríguez-Adrados, 1981). Such a gulf or difference implies that a referential theory of meaning is ungrounded or at least that words are not «related to things as proper names -- onomata» (Kerferd, 1984, 218). For Kerferd this passage proves (a) that Gorgias rejects «a referential theory of meaning- the view that words possess meaning, because they refer to (externally existing) things», and (b) that words, according to Gorgias, could not be used to communicate information about objects outside us, so that the possibility of communication by means of logos is eliminated (Kerferd, 1984: 218).

In addition, Gorgias says: «if anything exists, it cannot be known, and if it is known, no one could show it to another; because things are not words, and because no one thinks the same thing as another» (MXG 980b 17-19). For Kerferd this Gorgianic view posits a gap between the logos and the sense impressions or thoughts (Kerferd, 1981 [b]: 324). So, in Kerferd's view, «Gorgias has introduced a decisive breach into the relation between words and things, and by so doing also between words and sense-impressions. Yet from Parmenides onwards it was part of the received wisdom that words must refer to something (...) all thinkers in the fifth century BC were still imprisoned in the constraints imposed by the search for a referential theory of meaning; (...) in default of any other possible objects of reference for words [Plato] ended up by proposing fresh entities, the Platonic Forms. No such solution was available to Gorgias. The furthest that he was able to go was to suppose that it is thoughts in our minds which function as objects of reference» (Kerferd, 1981 b: 325-6).

To conclude, as shown, scholars have to deal with multiple frames of interpretation before they can offer any settled account of what Gorgias meant to say to his audience.


Primary Sources

Secondary Sources

Acknowledgements: This paper was originally written at the University of Edinburgh and at the University of Athens under the supervision of Dr Theodore Scaltsas and Prof. Basilios Kyrkos to whom I am indebted.

Michael Bakaoukas MSC, PHD

University of Piraeus
6 Poseidonos, Hagios Nikolaos, 18900, SALAMIS, GREECE
Tel/fax.: 03014651273