To Table of Contents No.2

A Dialogue on the Role of Logic
for the Theory and Art of Argumentation

The dialogue includes Jan Wolenski's review of Witold Marciszewski's Logic from a Rhetorical Point of View (de Gruyter, Berlin, etc. 1994. Series "Foundations of Communication and Cognition". pp. XVI+312).
Wolenski's text, originally published in Grazer Philosophische Studien (1994, vol 48, p.243 ff.) is here segmented; each segment, handling a particular theme, is marked with the initial "W", and followed by the author's ("M") responses. The text has been slightly altered to the extent required by the form of dialogue.

The dialogue contributes to the logic and intelligence issue, though in an indirect way. The question of how to make arguments more argumentative, i.e. more suggestive, persuasive, convincing, can be answered only then, when a thorough theory of efficient problem-solving is provided; and this amounts to providing a theory of intelligence. The relation of argumentation to problem-solving is that of inclusion: every argument is expected to solve a problem (though not every solution consists in giving arguments).

-- W: You attempt to unify two approaches to human mind and its functioning: Cartesian (human mind is working intuitively in principle) and Lebnizian (human mind acts a symbolic calculating machine). As a result, the essay outlines new foundations for rhetoric, based on logic.
-- M: This is a good rendering of what I mean, except for the attempt of unifying. These points are not liable to be unified. They are rather complementary, one giving rise to the phenomenological, the other to the data-processing, or cognitive, approach.

-- W: Let us consider some vital issues in the order of chapters. The first, "On the Rhetorical Point of View", discusses historical fortunes and misfortunes of rhetoric, and why this subject, flourishing in the antiquity and the middle ages, became neglected in the post-Leibnizian philosophy. It also gives reasons why rhetoric should revive in a close link with logic, linguistics and cognitive science.
-- M: Thanks for so excellent summing up. Since the last point may seem controversial, let me explain that what I call rhethoric is a modern theory of argumentative communication, which can profit from the history of relations between logic and the traditional rhetoric.

-- W: In Ch.II, "Mind-Philosophical Logic as a Theory of Intelligence", you consider the mind-philosophical logic as a kind of philosophical logic, so related to the philosophy of mind as deontic logic to the legal philosophy and ethics or temporal logic to the philosophy of time. Note that the analogy of the mind-philosophical logic and deontic or temporal logic is not quite clear, because these kinds of logic are based on explicit calculuses, whereas the mind-philosophical logic is rather a composition of various topics which are not integrated by a formal theory.
-- M: The point requires clarification, indeed. I coined the term "mind-philosophical logic" to render von Neumann's claim that "a new, essentially logical theory is called for in order to understand high-complication automata and, in particular, the central nervous system". That such a logic would rule the mind too (not only the brain), follows from the Leibnizian philosophy of mind. I continue to believe that such a field of research is needed, but I do not insist any longer on those analogies with deontic or temporal logic.

-- W: The same chapter contains the basic distinction of objectual and symbolic reasoning which is very closely related to concepts of formal reasoning and material reasoning. You claim that most ordinary reasonings are objectual for their connections with special subject-matters and situations. On the other hand, objectual reasonings can be represented or at least approximated by symbolic reasonings after suitable additions.
-- M: Now I see the difference between objectual and symbolic reasoning as a deeper one than I did previously. I appreciate the use of logical calculi for studying objectual reasoning, but rather those which succeeded in discovering some non-verbal acts in reasonings, as did, e.g., Beth's semantic tableaux. Thus, one does not put the conjunction sign between sentence; instead, he makes a list of facts. Instead of using the disjunction sign, one branches a reasoning. Instead of using the negation sign, one rejects a proposition, and so on. This is an example of how the progress of symbolic logic makes it closer to actual arguments.

-- W: Ch.III, "Formalized versus Intuitive Arguments, the Historical Background", outlines a historical perspective for the distinction of formalized and intuitive arguments. Let me stress the point that the book discussed is full of very interesting historical remarks and details; this chapter is a very impressive example of a good use of history in studies which are basically systematic.
-- M: Thanks for your understanding of my methodological approach. This is the approach of an evolutionist who believes that the current knowledge can be best understood in the context of historical development.

-- W: Chapters IV-VI constitute a logical interlude. Their titles, namely, "Towards the Logic of General Names", "The Truth-Functional Calculus and the Ordinary Use of Connectives", and "The Predicate Calculus", hint at an outline of logic which, according to you, provides a logical skeleton for rhetoric. Although these chapters are not intended as a contribution to logical theory, one can find in chapter IV an original treatment of the problem of existential import of categorical sentences and their quantified counterparts. A comparison of logical schemata with ordinary meanings of logical particles is a very important feature of this part of the book.
-- M: The phrase "logical skeleton for rhetoric" requires a comment, but I prefer to make it at the end of our discussion.

-- W: In the next two chapters (VII-VIII), entitled "Reasoning, Logic, and Intelligence" and "Defining, Logic, and Intelligence", you offer a general discussion of how logic is related to rhetoric. This discussion is based on some results taken from cognitive psychology; in particular, you accept the connectionist model of mind and its activity.
-- M: I understand your résumé as a request to explain my points more clearly. In fact, I am not so eager to base them on cognitive psychology. When referring to it, I try to explain my ideas in a context expected to be familiar to prospective readers. Though these ideas go back to von Neumann, they are far from being assimilated by logicians, semioticians, etc. Hence my reference to views being more in vogue, and my insistence on case studies to present less familiar notions in familiar examples. After such case studies in the chapter "Symbolic Logic and Objectual Reasoning", there follow those in the last chapter "Implicit Definitions and Conceptual Networks". And this last is, in a sense, most important.

-- W: ?
-- M: I assume that general connectionist point that the concept of a thing is not likely to be represented in the brain by something like a picture located at a definite spot. A concept, an image, etc. is rather a functional totality resulting from a vast set of impulses distributed to various places, and being synthetised by a data-processing unit.

-- W: What this all has to do with logic?
-- M: Look at what we logicians call a system of postulates, or an implicit definitions. In such a system, the components of a concept are distributed to particular postulates. True, these postulates are put down in a spatial neibourghood, but in network a neibourghood means something different, to wit the facility of quick communication. I claim that the whole knowledge of an individual is like an axiomatic system as far as the way of the defining of concepts is concerned, with the difference that for each concept the set of relevant defining postulates has to be immense, and that those quasi-axiomatic systems are not isolated from each other; instead, they get dynamically connected by a gigantic number of impulses. Nevertheless, implicit definitions as studied by logic provide us with an excellent paradigm of knowledge representation and engineering.

-- W: Hence, your project of rhetoric is cognitive in nature. You even use the term "cognitive rhetoric" as a label for your own ideas as well as similar conceptions. In general, cognitive rhetoric is distinguished by an extensive use of logic in the description of arguments. In this respect, you closely follow the Polish analytic pattern: logic controls all human rational activities. Thus, cognitive rhetoric must be sharply distinguished from Perelman's project on which rhetoric and logic are quite different activities. On Perelman's account, rhetoric studies arguments which are effective in persuasive actions, independently of their logical correctness. Obviously, your vision of rhetoric is much more normative than Perelman's is. By the way, it is a pity that you did not compare your ideas with Perelman's views; as a matter of fact, Perelman is even not mentioned in the book.
-- M: I do like the idea that "logic controls all human rational activities", but we have to consider the sense both of "logic" and "rational activities". If logic is regarded as the most general theory of rational activities, the statement is tautological. If logic is construed as predicate logic, even with the host of its extensions, and non-classical alternatives, then the statemnt is not true, since a lot of rational acivities is controlled by intuitive arithmetic, for instance. I fancy , again following von Neumann's ideas, that a more extensive logic, taking into account more subtle mechanisms than logical gates alone, does control a good deal of human activities.

-- W: And what about the relation of cognitive rhetoric to that postulated by Perelman?
-- M: I agree with your suggestion that a mention of Perelman's approach would better emphasize that of mine.

-- W: My main objection concerns the concept of logic. Although you mention twice that truth-preservation is a mark of formal logical arguments, you, in my opinion, overlook that this a fundamental point when relations of logic and rhetoric are discussed. Sooner or later, we must ask whether there is a sort of rhetorical validity or correctness. Unfortunately, I am not sure whether, on the view of yours, rhetorical validity, if any, is reducible to logic or not. There is a real dilemma: if rhetorical validity does not exceed logical correctness, it is too weak for needs of rhetoric in its traditional understanding, but if rhetorical validity is something more than truth-preservation by the form of arguments, it is undefinable by logical means. This dilemma is well-illustrated by the Golden Rule of Argumentation (p. 215): Show the things itself, that is a universal and give it the most illuminating name you can find in the language of your audience. Well, but what if the name is not "the most illuminating"? Is the argument in question simply incorrect from the logical point of view or improper for different extralogical reasons? And what is role of logic in incorectness of arguments which is different from logical invalidity?
-- M: Note that my essay is not entitled "Rhetoric as Entirely Founded in Logic". Instead, I ask how can symbolic logic (i.e., that taken in the strictest sense) contribute to a better understanding of objectual intuitive reasoning, as one most relevant to rhetorical applications (p. X). Hence I do not pretend to introduce any generalized concept of validity, like Descartes did not pretend to do when stating his "rules to guide the mind". The Golden Rule as suggested by myself is just a rule of a proper mental conduct. I agree that the phrase "most illuminating name" as used there is not most illuminating by itself, but I hope the context sheds some light on the question.

-- W: Nevertheless, my suspicion is that you employ the label "logic" in two various senses and that you sometimes shift from one to another. Firstly, logic is a theory of correct (valid, truth-preserving) arguments, and secondly, it is a theory which studies various devices useful in arguments, for example, definitions, classifications, etc. It is obvious that relations between logic and rhetoric look differently, depending of which understanding of logic is employed. When you distinguish objectual and symbolic reasonings, you probably think about logic in its narrow meaning (logic as the theory of valid arguments), but when you formulate the Golden Rule, the second concept of logic is taken into account. This ambiguity is also present, when the concept of the mind-philosophical logic is introduced.
-- M: I endorse this distinction which belongs to the basic vocabulary of people dealing with logic. What may be non-trivial, is how people see the relation between, say, logic-one (which handles truth-preserving data-processing) and logic-two (which studies devices useful in arguments). It was Descartes and his followers who claimed that logic-one is without any use for logic-two. Such an extreme attitude may not appear nowadays, but when Ch. Perelman gives a book of him the title "Logique juridique. Nouvelle rhetorique", and in that book we do not encounter any mention of logical constants, etc., we find this use of the word "logic" rather close to that praised by the old Cartesians.

-- W: Well, how do you see the relation between the "two logics"?
-- M: I see logical calculi, i.e. logic-one, as a necessary tool for logic-two. It has been said how the theory of axiomatic systems becomes a paradigm of conceptual networks. The theory of normal definitions needs the formalism of predicate logic. Furthermore, the basic way to inquire into objectual reasoning consists in a contrastive study starting from symbolic reasoning. And so on. Owing to a progress in preciseness, arguments become, so to speak, more argumentative, and this is a contribution of logic-one, via logic-two, into rhetoric. Thereby logic contributes to the theory and to practical development of human intelligence, for most abilities required for arguments are involved in any efficient problem-solving (being the measure of intelligence).

-- W: Let me add that your case studies, although very interesting in themselves, do not justify the whole project as logical, because they only show that an amount of logic is used in arguments and logic can be used in their evaluation. Also translating some arguments into MIZAR or other computer softwares does not justify that the translated pattern is logical in the strict sense; of course, I do not deny that such a translation is important.
-- M: This is exactly what I wish: to show that a considerable amount of logic is needed for evaluating arguments, and for making them more efficient, i.e. more argumentative. That I did not project any more is due not so much to a virtue of modesty, but rather to a theoretical option regarding the very nature of logic. There are two (at least) possible options. According to one of them, logic is a hardware-independent system, while according to the other, that von Neumann's (and mine), a system of logic can apply to all the cognition and communication activities, only if it proves adjusted to that hardware with which it is realized. Whereas such a device-dependent system is required for rhetorical purposes, the current symbolic logic does not meet the challenge. Nevertheless, it is the only possible starting point in the way towards that future full-fledged logic, expected to render all the kinds and subtleties of arguments which the humans owe to their miraculous hardware.

To the top of this page