Rand's definition of logic

[Roger Bissell]

I am going to frame my remarks around excerpts from Rodney Rawlings' post on "Popper Talk" of 9/20/07:

As promised, I’ll present my thoughts on Ayn Rand’s definition of logic, and why she chose it, as against the conventional definition offered by the other posters....To repeat, the conventional definition is very proper and correct: logic is the study of the principles of valid inference (or, maybe more precisely, deduction), quite apart from the truth or falsity of premises and conclusions.

Comments on this:

1. The conventional definition...Even as conventionally defined, art as a field of study covers more than just forms and principles of valid inference, though these are the point of everything else that is included. Both the mainstream texts (e.g., Copi and Hurley) and the logic lectures and texts by Peikoff and Kelley include considerable material on concepts and terms, as well as on propositions. For that reason, Aristotelian logicians usually use a broader definition than the conventional one for the science of logic. As the Aristotelian logician Peter Coffey notes in The Science of Logic (1912):

f we are to judge well--to form sound and true judgments--we must make a careful study of the materials of thought in order to see how they ought to be employed in those mental processes of ours which lead us to a knowledge of the truth. Those thought-materials, and thought-processes, and thought-products, are therefore the object (or subject-matter) with which logic has to deal. They are briefly: our elementary notions, concepts, ideas, in the first place; then, the judgments [and propositions] we form by comparing our concepts with one another; next, the processes of reasoning or inference by which we compare our judgments together for the purpose of arriving at other and more complex judgments; and, finally, those more elaborate mental or rational constructions built up by our reasoning processes and commonly called sciences, or philosophy itself, as the case may be. The science which studies the proper order and arrangement of all those mental functions which lead us into the possession of truth or knowledge, is the science of logic. [vol. 1, pp. 13-14, bold/underscore emphasis added]

To have truth or knowledge as the result of a process of logic, i.e., of inference or reasoning, we must have not only valid reasoning (which is the core of formal logic), but also valid concepts and true propositions -- including definitions and premises of one's inferences. Absent any of the latter, and all the valid reasoning in the world will only give you gigo (garbage in, garbage out). That is why the Aristotelian logicians conceive of logic more broadly to include not only formal logic, which focuses on valid forms of reasoning, but also material logic,

2. art as against science...The reader might get the impression from Rodney's comment that Rand preferred to define logic this way, as an art, rather than as a field of study, as though there were some error in regarding it as the latter. Leonard Peikoff addresses this issue directly in his 1974 lectures on logic, which were presented (and probably also developed) under the watchful eye of Ayn Rand. He says in his opening lecture that logic is both an art and a science, because there are two perspectives from which one can see it. When viewed as a method, logic is an art. When viewed as a set of principles, logic is a science. Clearly, in all of her references to logic, Rand was considering and defining logic as a method, because that is what she wanted to talk about when she used the term, not the discipline that studies the nature of that method. Equally clearly, judging from Peikoff's remarks just cited, she was well aware that there was a science of logic, and that it was far more than a set of guidelines for arriving at conclusions in accord with reality.

Coffey offers a delicious elaboration on this theme, and I will quote it at length from pages 13-16 of volume 1 of The Science of Logic:

Is logic a science or an art? It is both; or rather there is a Science of logic--a practical science--and an Art of logic.

To have the art of doing a certain thing is simply to know and possess and make use of all the requisite means for doing that thing well and properly. "An art," says St. Thomas, "is nothing more than a right conception of the way to do certain things." Logic, considered as an art is therefore the collection of those practical rules which should regulate our thinking and reasoning processes. And as it is these latter functions which frame the rules for all departments of external activity--the rules which constitute all the arts--the art of right thinking will underlie and direct all the other arts. For this reason logic is somtimes called the Ars Artium, the art of arts. [Compare this with Rand's comment: "...the fundamental concept of method, the one on which all the others depend, is logic" (Introduction to Objectivist Epistemology, p. 36).]

A scientific knowledge of any subject-matter is a knowledge of it through its causes, and reasons, and principles, a knowledge of its laws, a systematized, co-ordinated knowledge of it, got by mental application, analysis, demonstration...Now, manifestly, logic is a science, for it studies and analyses our mental processes and teaches us a systematized body of truths concerning those processes...

An art, according to the ordinary use of the term, is understood to mean a collection of practical rules or canons or precepts for our guidance in the performance of some work...But it also commonly means practical skill derived from experience in the application of those principles or rules...If we extend the term ["art"] to those rules which direct even internal, mental activity, we may legitimately call [logic] an art--the art of correct thinking, of accurate reasoning. That is to say, the discovery and formulation of those rules or canons--which are no less the outcome of experience in thinking than of an analytic study of the processes of thought--would be the practical science of logic; and the application of those rules, the actual reasoning according to those precepts (whether unconsciously or consciously) would be the art of logic. [bold/underscore emphasis added][Compare this with Peikoff's Q-A comments after lecture 3 of "The Philosophy of Objectivism" (1976): Logic as "the art of non-contradictory identification" [is] "a utilitarian, not an esthetic, art, which simply means: a learned technique or skill, a process of performing a certain method according to certain rules in order to achieve a certain product. In this case, you don't produce shoes, but knowledge..."]

Every art has some background of theoretical truths or principles behind it; every department of external experience has some counterpart or complement of internal, rational study. The system of practical rules and laws arrived at by the study of our mental processes was called by the Scholastics Logica Docens--logic in the teaching; the application of those fruits of study for the guidance of those processes, they called Logica Utens--logic in action.

When I first heard Ayn Rand’s definition, I myself was puzzled about why she did not define logic as “The art of noncontradictory statement.”...I concluded—as I said earlier here—that she wished to emphasize the purpose and use of logic. That is, she saw no point in using reason apart from the goal of obtaining knowledge...However, thinking about it some more, I see other factors in Rand’s definition.

Rand would have been aghast at, or at least scornful of, the idea of calling logic "the art of noncontradictory statement," and I think that Rodney's surmise here is quite correct, that Rand was concerned with the cognitive process and not just the linguistic expression out of context of any cognitive process that might have produced it.

In his recent lectures Induction in Physics and Philosophy (1992), Peikoff amplifies this point in regard to propositions (viz., generalizations):

You cannot start with generalizations already in existence, and then, ignoring their origin or genesis, attempt to evaluate them. But this last is the common practice among philosophers today. They will tell you quite openly that they are indifferent to questions of genesis, of where generalizations come from, of how they’re reached, and they will say quite blithely, “Those questions belong to psychology, not to philosophy. We have a division of labor. Philosophy doesn’t care where generalizations come from, or how you get them. It’s concerned only with how do you validate them.” And if you ask them the source or origin of the propositions that they’re so busy struggling to validate, their attitude is, and this is exact, paraphrasing Ayn Rand, “The generalizations are here.” How did they get here? “Somehow.” What caused it? “Nothing has causes.” But you know, she said [in Atlas Shrugged, about the Marxist "mystics of muscle"], “The wealth is here,” but that’s exactly their attitude toward generalizations. Now, my answer to this claim, that generalizations are here, is that they should not be here, unless one knows how they got here, by what steps, in accordance with what method, and whether the method used is valid or not. Nothing else can enable one to know whether the product of that method does or does not correspond to reality. [bold/underscore emphasis added]

In other words, you can't know whether something is a "non-contradictory statement," unless you more fundamentally know that it is a non-contradictory identification of reality.

[T]he science or study of logic as the term is popularly and professionally understood...is a universe unto itself, and it does not have to take cognizance of any outside truth in the premises it juggles. False and arbitrary premises and false conclusions do not matter, because the discipline assumes we are starting from a stance of ignorance and confusion; it cannot be in the position of dictating in advance what we must think about the subject matter to which we want to apply it.

Qua formal logic, logic does not dictate truth or falsity, but only validity or invalidity of inference. But this is why earlier Scholastic philosophers discussed:

...the whole question of the validity of thought in the larger sense of its truth...immediately after the questions of Formal logic and described them as forming Material or Real or Applied logic, or again as Logica Critica, in opposition to Logica Dialectica, which was the title they gave to the logic of formally valid inference. Modern scholastics are inclined to state and expound briefly the practical rules and canons which constitute Method--both inductive and deductive--in Logic, and to leave the fuller discussion of all the great underlying principles of knowledge to a special treatise which they call Criteriology or Epistemology or Theory of Knowledge, and which they claim to be a special branch or department of psychology, or of metaphysics, rather than a section of logic. It is quite true that the treatment of many of these questions must be largely psychological and ontological. Nor is it possible or desirable to draw a sharp line of demarcation between these sciences. [Coffey, vol. 1, pp. 28-29]

Coffey suggests that "larger questions concerning the ultimate criteria of truth and the ultimate motives of certitude" should be left to epistemology, but that logic should include "the questions that have an immediate bearing on the truth, as well as those concerning the consistency, of our thoughts." (ibid, p. 29) For this reason, he includes not only a discussion of concepts and definitions -- as do virtually all logicians of whatever stripe -- but also (as do H. W. B. Joseph and other Aristotelian logicians) a substantial discussion of presuppositions of induction, including the concepts of "reason," "cause," and "uniformity of nature."

If one believed that proper definitions are to be arrived at by consulting dictionaries and encyclopedias (and some here do believe it), that would be all there is to say on the topic. However, as Ayn Rand maintained, definitions must be revised and new concepts must be formed as our knowledge increases. (According to my own thinking, these two processes may actually be combined in what I would call development of a concept. I have more to say on this in the Lulu.com essay “Understanding Imaginaries Through Hidden Numbers.”) Thus, in her definition of logic, Ayn Rand formed a wider, more fundamental concept of it, which related to the conventional one without contradicting it.

Rand's definition was about the methodology or process of reasoning, not the field of study, which is why her definition treats it as a species of art, rather than a species of science. However, qua methodology, this is correct. Rand viewed logic as more than just formal logic, but also (a la the Scholastics) as including a good amount of "material logic," i.e., attention to the truth of one's premises and the other presuppositions of one's reasoning, including the Law of Identity and its corollaries, especially, the Law of Causality. (Or, as Joseph and Coffey and others, characterized it: uniformity of nature and cause-and-effect.)

For these reasons, Ayn Rand did not want to describe logic as “the study of the principles of valid inference,” even though as a detailed, scientific discipline, that is what the term means. That definition puts the focus on the possibility of error in logical chains, which is a special danger with long chains and deserves to be studied independently. (She approved Leonard Peikoff’s taped lecture courses on such logic, which incidentally I have listened to.) Rather, she wanted to characterize logic at its interface with reality and with human life.

Correct. Logic is a tool for producing consistent thoughts, but more fundamentally, it is a tool for producing knowledge of reality. Non-contradictory identification necessarily includes consistent thinking/reasoning. Rand stressed the importance of defining concepts in terms of fundamentals, and she was consistent with this in how she defined logic as an art/method. She insisted that one's thinking method, i.e., one's logic include not only logically coherent patterns of inference, but also a careful checking of the truth of one's premises and the validity of the concepts from which they are formed.

REB

[John Dailey]

REB:

~ This clarification of yours re Rand's definitional (to be sure, re what *she* meant) use of the terms art and identification, as both being necessary and sufficient in any and every person's use of...'logic'...was worth reading.

~ Your distinction re (personally-worked-out?) 'method' and (established?) 'principles', plus the place of development-vs-developed, or, the cognitive process vs the linguistic expression of the end results (as if they're supposedly now really 'ended'!) explicates much re the nowadays perspectives on logic. --- I've nearly forgotten that the term soundness re logic-arguments used to be stressed as distinguished from valid. Don't see much use of the term sound nowadays; only the mental-masturbation of the term valid. Like, internal consistency is all that's worth paying attention to; let's not go 'out-of-the-box' re even thinking what external might...dare I say(?)...'imply.'

LLAP

J:D

[BaalChatzaf]

REB:

~ This clarification of yours re Rand's definitional (to be sure, re what *she* meant) use of the terms art and identification, as both being necessary and sufficient in any and every person's use of...'logic'...was worth reading.

~ Your distinction re (personally-worked-out?) 'method' and (established?) 'principles', plus the place of development-vs-developed, or, the cognitive process vs the linguistic expression of the end results (as if they're supposedly now really 'ended'!) explicates much re the nowadays perspectives on logic. --- I've nearly forgotten that the term soundness re logic-arguments used to be stressed as distinguished from valid. Don't see much use of the term sound nowadays; only the mental-masturbation of the term valid. Like, internal consistency is all that's worth paying attention to; let's not go 'out-of-the-box' re even thinking what external might...dare I say(?)...'imply.'

LLAP

J:D

Soundness and Validity are dealt with in just about every introductory graduate or advanced text on first or logic, particularly where model of formal languages are discussed. But that is all mental masturbation, right?

Ba'al Chatzaf