"Plato and Delphi," Futures, Vol. 5  (June 1973), pp. 281-286.

by Gordon Welty
Wright State University
Dayton, OH 45435 USA

[/281] Plato had an unparalleled influence on political and utopian thought through his Republic. Recently, Professor Clarke has emphasized the "Primacy of Plato" for the literature of the ideal state.(1) Some qualifications are offered here to Clarke's appraisal of Plato. These qualifications bear upon his characterization of Plato's "extraordinary social insights and his exceptional power of thought". They should be of particular interest to the futures researcher, as they are related to Delphi.

It has been widely recognized that Plato propounded the use of the political lie, whereby he proposed that "With the help of one single lordly lie we may, if we are lucky, persuade even the Rulers themselves -- but at any rate the rest of the city" (Republic III, 414 b-c)(2) of the necessity for maintaining the social status quo. It is important to note that the lie was hoped to result in the Ruler's self-deception, as well as the propogandistic deception of the masses. This lie, then, can be viewed as manifesting Plato's insight into the possibility and the functions of political mass-deception as well as of psychological (or personal) self-deception as `legitimations'. Both of these deceptions accord with Plato's psychological and epistemological rationalism which denies the significance either of one's opinion or the knowledge (or belief) of other's opinions. Of course, if there is significance accorded to opinion, then significance will need be accorded to difference of opinion. The lie is not presented as anyone's opinion, but rather as fact; thus, it can function to deceive either the liar (the Ruler) or his audience (the masses). There is no need for Plato to accept differences of opinion, nor are there grounds for questioning the legitimacy of the political and social system.

In light of the rationalism of Plato, it is perhaps not surprising that he shows no inkling of the possibility of socio-psychological deception, operating through influence processes. A sensitivity to social reality would be a prerequisite to such insight; to question the sensitivity of Plato's insight is to question Clarke's characterization noted above. It might be remarked here that whatever sort of political theorist or utopian Plato might have been, he wasn't much of a social scientist, even when compared to other Hellenes such as the much [281/282] abused Sophists. One citation will illustrate this remark. Plato was convinced that "the introduction of novel fashions in music is a thing to beware of as endangering the whole fabric of society" (Republic IV, 424 c). This lack of insight into the direction of social causation is symptomatic of Plato's elevation of cultural factors to a position of supreme causal agents in society rather than as, at best, barometers of more basic social change.

Given Plato's anti-empiricism and downright contempt for factual evidence, we would expect him to overlook evidence that socio-psychological influence processes in fact distorted the utterances of the Oracles, the founts of wisdom themselves. Such evidence was available to Plato in the tale of Ino.

The Case of Ino
The Greek historians tell the story of King Athamus of Alus in Thessaly who had two wives, first Nephele and then Ino. Ino was jealous of her step-children, and planned their death. A famine occurred after Ino convinced the local women to roast secretly the seed grain, and no crops grew. Athamus sent a messenger to the Oracle of Delphi to find the cause of the famine. Ino bribed the messenger to lie upon his return. The messenger consented and said falsely the Oracle proclaimed the famine would cease only when Nephele's children were sacrificed to Zeus. Thus, we find deception.

The children, however, escaped to Colchis in Asia Minor, on the Golden Ram whose fleece was later retrieved by Jason and the Argonauts. On a second round of oracular utterances, it was proclaimed that Athamus must be sacrificed for the country in lieu of his children. The wicked Ino and her children met unhappy ends, and the king went insane and left the country. Thereafter the eldest male heir in each generation of the family of Athamus was sacrificed, since Athamus' sacrifice never properly occurred.(3) While this tale has an element of myth about it, the point we would like to emphasize is that an Oracle, once deception had occurred, was apparently unable to rectify its utterances.

Implications for the History of Social Thought
For the systematic elaboration of his ethical and political theory, Plato needed a psychology. Windelband points out that the earlier psychology of the pre-Socratics which had arisen out of the presuppositions of natural philosophy, emphasizing individual perceptions and opinions, was not acceptable to Plato.(4) Such a naturalistic (as opposed to rationalistic) psychology might have been the Protagorean phenomenalism.(5) The psychology that Plato found more compatible was an absolute individualism which, as we have noted above, virtually precluded socio-psychological interaction or an appreciation of the social context of knowledge. As Clarke has rightly mentioned, "complete rationality, it seems, does not find all the answers for common humanity."(6)

Plato knew the tale of Ino (cf. Minos, 315 c). How did he deal with the fact of the deception of the Oracle?

In the sociologist Mills' terms, a social system needs normative artifacts or [282/283] `legitimations' to "lead persons into roles and sanction their enactment of them."(7)

Let us first document the nature of legitimations for Platonic society. Plato's observation that the working and development of religious institutions and observances are of such paramount importance that they must be left to the wisdom of the Delphic Oracle is relevant to the problem that faced Alus in its time of famine (cf Plato's Republic IV, 427 b; also Laws V, 738 b). In the case of Alus, it was supposed that some god had been offended and required propitiation. As a measure of his innocence, Plato accepted the fundamental mystery in this sort of affair. The rites and religious institutions cited by Plato included "services to propitiate the powers of the other world. These matters we do not understand ourselves" (Republic IV, 427 c) at which point he implicitly sympathises with King Athamus' uncertainty about the appropriate response to the famine, and concurs with the king's subsequent decision to consult the Oracle. This was apparently standard procedure in Hellenic Greece, and was presumably undisputed by Plato.

By way of further illustration of Plato's acceptance of frequent recourse to the Oracles and susceptibility to legitimations themselves (rather than critically examining the symbols and institutions), he states that in the case of illegal appropriation of buried treasure, the case is resolved when "the state shall send a deputation to Delphi. Accordingly as the god shall pronounce...so the state shall act" (Laws XI, 914 a). Recall that King Athamus sent a deputation (of one) to Delphi to resolve a different problem, that of famine. From that point, of course, the King's troubles began.

Thus, Plato accepted the oracular utterances unquestioningly. The Oracle did have direct contact with the god -- in the case of Delphi with Apollo. This was its legitimation for Plato. In spite of the tale of Ino, he still was of the opinion that the Oracle was "an unimpeachable authority" (Apology, 20 c). This statement was of descriptive and not normative import! The case of Ino illustrates, however, that his faith was poorly founded. Plato's profession of faith tells more of Plato than of men in the world.

Thus a problem of Platonic social thought appears to be the intrusion of a personal belief in the legitimations of certain institutions, particularly those institutions legitimized by oracular utterances.(8) This personal belief of Plato has led to a bias in social analysis to suppose that these legitimations were causally efficacious in society at large. Of course, if these were not causally efficacious, if persons at large did not find their behaviour sanctioned by these artifacts, then the Platonic society (even as plan) will risk disintegration. Apologists for an order may themselves be causally efficacious in preventing disintegration, of course, but the question here is that of the inference warrant, not the personal persuasiveness, in Platonic thought. What does this mean for Plato's social and political theory?

For Plato, the Oracle has knowledge of fact. Hence, influence processes, based as they are on the acceptance of other's opinions in the absence of one's own knowledge of fact, will not be relevant to Plato's sociology. The relational [283/284] nature of the logic of opinion formation is unknown. The Oracle is unimpeachable. But the faith that no one would tamper with the utterance of the Oracle is clearly questionable on socio-psychological grounds.

If Plato had recognized either that the Oracle's utterance itself was a social artifact and subject to molding, or else that a deputation's report might have been opinionated, he could perhaps have dealt with the fact of Ino's deception and the inability of the oracular institution to rectify that deception, in an appropriate Platonic epistemology and social theory. Of course, he did neither.

Elsewhere, in recognition of deception in the political (i.e., non-epistemic) arena, Plato has said: "If anyone, then, is to practice deception, either on the country's enemies or on its citizens, it must be the Rulers" (Republic III, 389 b) and continues that "if anyone else in our commonwealth...is caught not telling the truth, the Rulers will punish him" (ibid). Apparently the punishment will be quite severe, as the offence is seen as potentially fatal to the commonwealth. But Plato has no conception of a mechanism to uncover falsehood and deception, much less one to prevent it. This shortcoming, it appears, was due to his seriously flawed view of man in society. Plato has little more to say than the passionate charge that the deceiver shall be dealt with by "victorious encounter and repulse, and stern correction" (Laws V, 731 b). Retribution could be proposed by Plato as a deterrent to political deception, as he assumes that the Ruler's access to the ideas will provide a criterion for the truth, hence a means of identifying falsehood, after the fact. There is, however, no reason to think it could serve that function in the case of epistemic deception, since the utterance of the Oracle is its own truth warrant. Recall that the Athenian Stranger concludes that the gods "are never to be seduced from the path of right" (Laws X, 907 b).

Thus a potential breakthrough in Greek social thought, to have been effected by the consideration of Ino's deception of the Oracle, was never realized. Clarke's conclusion that "The Republic was...a communications device capable of handling all the diverse and complicated factors at work in a society"(9) can then scarcely be sustained, even when the work, or the Platonic corpus, is viewed in historical context. Important factors such as `credibility gaps' which follow from a careful analysis of the political lie were clumsily handled by Plato, if discussed at all. Yet such concerns were not unknown to Greek thought, as a perusal of the Sophistic writings shows.(10)

Implications for contemporary futurology
The anonymity of the experts who participate in a Delphi exercise may have a unique effect, precisely the same as that related in the Greek tale of Ino, deception, and the Oracle, that warrants consideration. With the contemporary Delphi technique, not only is rectification of erroneous assumptions perhaps not possible but the mechanism may actually facilitate reaching erroneous conclusions. Indeed, F. Cyphert and W. Gant have presented experimental evidence suggesting that this is the case.

Suppose of a Delphi exercise that the median of the first round of forecasts [284/285] indicates a specific central tendency in the expert's judgments. If the manager of the Delphi exercise chooses arbitrarily and deceitfully to change the forecast, as did Ino, and feeds back a median value for the second round substantially different from the `true' value, then there are two interesting alternatives to consider. On the one hand, it might be supposed that the expertise of the respondents would permit their immediate recognition of the deception, in which case they would seek to reestablish the `true' value, refuse to participate in an obviously corrupt exercise, etc. In the Greek tale, this would have been illustrated by the Oracle's refusal on the second round to prescribe the sacrifice of Athamus.

On the other hand, it might also be expected that the experts would not return, in subsequent rounds, to the `true' value. Under this alternative, they would more likely reflect the arbitrarily chosen median in the second, etc, rounds, as the Greeks relate the Oracle in the second round did, and even rationalize their first round 'deviance' from what they suppose to be a group judgment. Hence, on this alternative, the experts in the Delphi exercise are deceived.

The second alternative might be expected because the expert who has been deceived in this fashion will not have prepared a rationale for his deviance and status of `minority of one', as would have an `intellectual maverick'. After all, the deceived was (and still is) part of the deceived majority. Hence, he will bc inclined to change his `deviant' judgment to accord with what he supposes to be the group judgment, rather than generate a rationalization for an unanticipated iconoclasm. Once he has made this accommodation, he can then rationalize his new estimate by denouncing his earlier assumptions.

As we have noted, Cyphert and Gant conducted an experiment which bore on this problem. While undertaking a Delphic exercise on the goals of the School of Education of the University of Virginia at Charlottesville, they introduced a `bogus goal' which was initially rated as having a low priority among all goals considered by the Delphi participants. The consensus was distorted and reported in later rounds as positive, and the final consensus showed the bogus goal rated considerably above the average. They concluded that "the hypothesis that the [Delphi] technique can be used to mold opinion as well as to collect it was supported."(11)

It might be objected here that deception and misrepresentation is possible in any research whatever, hence the Delphi technique is no more susceptible to such fraud than other forecasting and predictive techniques. Of course the issue is a matter of degree. The extent to which a predictive mechanism such as the Delphi technique (or an explicit econometric model, say) permits or facilitates its own improvement by the identification of sources of forecasting error, can be taken as a rough measure of what Professor Morgenstern has called epistemic `openness'. The more `open' a mechanism is in this sense, the less likely deception can occur. In Delphi, recourse is had to the implicit knowledge base of experts, so an initial confounding of personalities, culture, and information renders identification of sources of error unlikely.(12) Furthermore, what replication, traditionally the primary means of identifying sources of error, would mean for a Delphic exercise, is unclear.(13) In conclusion, the Delphi technique does not seem to be very open. Thus the tale of Ino can serve as a caution to contemporary futurologists; err not in your Delphi as Plato did! [285/286]

Acknowledgement
We would like to thank Professor Jiri Nehnevajsa for his acute criticism and comprehensive comments on this essay. Of course he is not responsible for any errors remaining or opinions expressed.

Notes
(1) I. F. Clarke, "Prophets and predictors: The primacy of Plato", Futures, Vol. 4, No. 1, 1972, page 79.

(2) We cite Plato's text without explicit footnotes. The translations are standard, including Cornford, Hackforth, Popper, Taylor and Tredennick.

(3) James Frazer The Golden Bough, Part III: The Dying God (London: Macmillan, 1923), pages 161-163; Herodotus Historia VII, 197; Apollodorus Bibliotheca I, ix, i.

(4) Wilhelm Windelband History of Philosophy (New York, Harpers, 1958), Vol. 1, page 123.

(5) See e.g., Mario Untersteiner The Sophists (Oxford, Blackwell, 1954), Chapter III. This material provides a useful summary of the social thought of Plato's contemporaries.

(6) Clarke, op. cit., page 80. See also G.W.F. Hegel Rechtsphilosophie (Frankfurt am Main, Fischer, 1968), pages 245-246.

(7) C. Wright Mills, The Sociological Imagination (New York, Grove Press, 1961), page 36.

(8) Mills, op. cit., page 37.

(9) Clarke, op cit, pages 79-80.

(10) Contrast Plato's conceptions with Protagoras' recognition of the "dependence of Right on Opinion" and the implication of this for the theory of norm formation. Cf. Untersteiner, op cit, page 31. On the Sophists' recognition of the social context of knowledge, cf. the analysis of Protagoras' thought in A. D. Winspear, Genesis of Plato's Thought (New York, Russell and Russell, 1956), page 141.

(11) Frederick Cyphert and Walter Gant, "The Delphi technique", Journal of Teacher Education, Vol. 21, No. 3, 1970, page 422. An early proposal of such use of Delphi is found in Olaf Helmer, "The Delphi technique and educational innovation", in Werner Hirsch (ed.), Inventing Education for the Future (San Francisco, Chandler, 1967), page 81.

(12) The reader might want to consult V. Lewis Bassie, "Recent developments in short-term forecasting", Short-term Forecasting: Studies in Income and Wealth, Vol. 17 (Princeton, Princeton University Press, 1955) especially page 9 on the contrast between individualistic and socio-psychological orientations towards forecasting studies.

(13) Consider that experts for an exercise, and for a replication, must be selected. See Gordon Welty, "Problems of selecting experts for Delphi exercises", Academy of Management Journal, Vol. 15, No. 1 (1972), pages 121-124.