Hamlyn (1990) Aristotle on Dialectic

Hamlyn, D. W. (1990). Aristotle on Dialectic. Philosophy, 65(254), 465–476. http://www.jstor.org/stable/3751284

 

Owen him self was in part reacting to what I suppose is the traditional view of how Aristotle regarded dialectic, as revealed in Topics I. 1. On that view dialectic is for Aristotle a lesser way of proceeding than is demonstra tion, the method of science. For demonstration proceeds from premises which are accepted as true in themselves (that is to say that they are essentially and thus in some sense necessarily true) and moves from them to conclusions which follow necessarily from those premises; and the middle term of such a demonstrative syllogism then provides the 'reason why' for the truth of the conclusion. Dialectic proceeds from premises which are accepted on a lesser basis 'by everyone or by the majority or by the wise, i.e. by all, or by the majority, or by the most notable and reputable of them' (Topics I. 1, 100b21-3), and proceeds deductively from them to further conclusions.

Let us first turn to Owen. What is it to tithenai ta phainomena (to set out what appears to be so), as Nicomachean Ethics VII. 1, 1 145b3, puts it in a passage which is given central importance by Martha Nussbaum too? It was a cardinal point of Owen's argument that by 'phainomena' Aristotle meant not just the phenomena evident to experience, but, as is clear from Aristotle's actual scientific practice, opinions also, and not always opinions of established worth. They are the endoxa to which the Topics refers in its account of dialectic. Hence, in Owen's view, there is not a sharp distinction between science and dialectic, induction is itself essentially dialectical, whatever happens thereafter in the use of dem onstration based on the results of induction. I have myself presented a thesis about induction, combined with an interpretation of the last chapter of the PosteriorAnalytics, elsewhere.5 While I am fairly unre pentant, despite criticisms, about the central argument of that paper, I shall not repeat it here. There remains a question how Aristotle saw himself as proceeding from the endoxa laid out as preliminaries to the pursuit of a science. What are the principles for the use of such material? It is such questions and the lack of answers to them which motivate what Martha Nussbaum has to say in Chapter8

There she points out that Owen was forced to attribute to Aristotle an equivocation over what phainomena are-sometimes opinions, but at other times the evidence of our senses. The view that knowledge must ultimately be based on the latter she sees as a form of Baconianism, and it is the attribution of such a view to Aristotle that she rightly wishes to reject. Phainomena, she suggests, constitute a loose notion, but not an ambiguous one, simply because Aristotle does not have the notion of a foundation for knowledge as the Baconian view suggests, and does not reject the sceptic by appeal to supposed items of certain knowledge, as later post-Aristotelians were to do. All this seems to me correct. But, so far it does not reveal exactly what Aristotle thought that one ought to do with phainomena, except that by the end of the day it was desirable to save as many of them as possible; and while that outlines a philosophi cal perspective it scarcely describes a philosophical method.

 

If I am right it should now be apparent why I do not think that what Martha Nussbaum has to say about the 'saving of appearances' deals with the problems which she herself rightly identifies in Owen's account of what is in effect the assimilation of science to dialectic. Why should a science begin and proceed from an appeal to endoxa, as Owen points out seems to be the Aristotelian practice? Why sh such a science by setting out thephainomena, and how, if one does, can true knowledge be distinguished from mere supposition, from indeed 'old wives' tales'? For this purpose one does not need a delimitation of the realm or field which constitutes reality, whether this presupposes an internal or an external conception of reality. One needs a method for the determination of the truth, or at least, given what I said earlier, a method for determining what should be accepted as true.

But we are now confronted with the same problem as that which concerned Martha Nussbaum. How should dialectic proceed if it starts from endoxa but is to result in an intuition of the truth? On what valid basis can mere opinions lead to truth about an objective reality? Irwin recognizes the force of the question and in trying to deal with it makes a distinction between weak and strong dialectic. Weak dialectic is just any old argument from endoxa, and is weak just because it has no criteria of validity. Strong dialectic, though Irwin is not entirely clear about this, is typified by the argument for the Principle of Non-Contradiction in Metaphysics IV. Since this argument is one in which the sceptic is shown as presupposing what he denies, it does not simply proceed from sure premises to valid conclusion and does not thus conform to the pattern of foundationalism. So, even if the theory of science presu poses foundationalism, this is not the case of a science of being-qua being, which is arguably what is at stake with the Principle of Non Contradiction. Nevertheless, Irwin thinks that what holds good of that Principle carries over to discussion of other things which are part of what we should call 'metaphysics', the discussion of substance etc., and from that to its applications in psychology and ethics. This is a very large claim and one which is too complex to deal with here. Let us confine our attention to the distinction between strong and weak dialectic itself.

First, it is not at all clear that Aristotle himself makes such a distinc tion, although it is evident that the argument for the Principle of Non Contradiction is quite a different sort of argument from any which moves directly from endoxa to the first principles of the sciences. Hence, if both sorts of argument are dialectical, there is certainly a distinction to be made between different kinds of dialectical argument. The question remains how general in its application the argument which Irwin calls 'strong dialectic' is. Second, and perhaps more perti nent for present purposes, science is left with weak dialectic only to provide its basis, and weak dialectic has little if anything in the way of criteria of validity. However, intuitions are presumably self-verifying. If one has an intuition of something it must be true, and in consequence it does not matter for its validity how it is arrived at. Indeed, that is what makes a system based on intuitions so conceived foundationalist. If it were only a matter of psychology what counts as an intuition, that is to say if it were only a matter of how we felt about the supposed truths on which the science is based, intuition would provide no foundation for knowledge objectively speaking. Intuitions must then, for this purpose, be true and seen to be so. To that end weak dialectic, indeed any kind of dialectic, is unimportant, even irrelevant. It may be part of the psychology of scientific discovery and theory-building; it is not part of its logic, nor of any relevance to its epistemology.

But the understanding involved in knowing the reason why is relative to what else the person concerned understands. That is to say that, in the context of demonstration, it is relative to the place which the propositions giving the reason why have in the chain of syllogisms going back to whatever is the starting-point, the archai or first principles for that particular chain of demonstration. Where the exercise is simply one of dialogue, whereby one person attempts to bring about that understanding in another by discussion and argument, the archai must be whatever is mutually accepted as such a starting-point. But when the exposition of the subject-matter is done in, say, a treatise, assumptions will have to be made by the author as to what the readers are likely to accept as such a starting-point. This is where dialectic comes in again. Aristotle will set out the endoxa and then say in effect 'So don't you see. .., and given that, doesn't it follow that... ?'. The appeal to endoxa is, as it were, a setting of the scene, providing the context for argument out of which, it is hoped, will emerge the insights from which demonstra tion and thus further understanding can follow.

What, to sum up, is dialectic for Aristotle? The answer is that, as with Socrates, it is a form of argument starting from whatever agree ment is available and seeking to produce some insight (nous, intuition) as to the truths from which demonstration can possibly start, so-called first principles, thereby furnishing an understanding of why things are necessarily as they are. But nothing can be laid down as to where that agreement is to be found. What is involved is at best what has been called 'inference to the best explanation'. At any rate, at least one part of what may be involved is that, an appeal to whatever is available that may lead to acceptance as true of that from which can be derived an adequate understanding of why things are so. But dialectic is not part of a search for truth itself. What is produced is indeed 'clearer and more knowable by us'; for that-its being so for us-is all that something which is very close to being persuasion can produce. But if an account of a part of nature can be derived from that in a way which seems to show that things are necessarily so and also why they are so, then indeed one might say that one has arrived at what is 'more knowable and clear by nature'. At least, that is the hope.