To OSG, all: The following article was written in the fall of 1989. It was published in the Winter 1989/1990 (Vol. 2, No. 3) issue of Objectively Speaking. I have made some minor revisions here. OSG members my give paper copies of this posting to friends for personal (i.e., non-commercial) use as long as the copy is complete, including this introduction. (Headers above the “To OSG, all:” line can be deleted.) I ask that this posting not be forwarded electronically.
– Bennett Karp
by Bennett C. Karp
At the base of every Objectivist principle is the Objectivist theory of concepts. It is Ayn Rand’s unique and revolutionary theory of concepts that makes the rest of her philosophy possible. Similarly, it is a wrong theory of concepts, whether explicit or implicit, that explains the distorted views put forth by other philosophies. A recent example of the crucial role of concepts and the necessity of epistemological precision is provided by David Kelley’s break with Objectivism.
In “A Question of Sanction,” Kelley says:
When we formulate moral principles, we may abstract from such differences in degree; we omit measurements, as Ayn Rand explained. But when we apply the principles in forming moral judgments about particulars, we must reintroduce the relevant measurements. [Last emphasis added.]
Anyone familiar with Objectivist epistemology knows that measurement omission is the essence of Ayn Rand’s theory of concept formation. But is David Kelley applying that idea properly? What does he mean by, and what is the significance of “reintroducing the measurements,” terminology not mentioned in the Objectivist literature?
Ayn Rand explains the principle of measurement omission in Chapter 2 of Introduction to Objectivist Epistemology (ItOE).
If a child considers a match, a pencil and a stick, he observes that length is the attribute they have in common, but their specific lengths differ. The difference is one of measurement. In order to form the concept “length,” the child’s mind retains the attribute and omits its particular measurements. Or, more precisely, if the process were identified in words, it would consist of the following: “Length must exist in some quantity, but may exist in any quantity … .” [p. 11, all references to ItOE are from the Expanded Second Edition]
This “some, but any” principle is what Ayn Rand means by measurement omission. Integration (in cognition) is based on measurement omission. It is important to be clear here on the precise role played by measurement omission. As Harry Binswanger explained in his lectures at The Jefferson School in August 1989, measurement omission is not an act of deletion or excision. When you omit measurements, you don’t leave out anything; you grasp something more. In essence, measurement omission is measurement inclusion, in that it is seen that a thing can be included in the class regardless of its specific measurements. Ayn Rand explains the point as follows:
Bear firmly in mind that the term “measurements omitted” does not mean, in this context, that measurements are regarded as non-existent; it means that measurements exist, but are not specified. That measurements must exist is an essential part of the process. The principle is: the relevant measurements must exist in some quantity, but may exist in any quantity. [ItOE, p. 12]
Understanding the role of measurement omission in concept formation is crucial to understanding concept application. Only with such an understanding can one grasp the nature of concepts and thus understand how to apply them properly. Ayn Rand explains:
The formation of a concept provides man with the means of identifying, not only the concretes he has observed, but all the concretes of that kind which he may encounter in the future. Thus, when he has formed or grasped the concept “man,” he does not have to regard every man he meets as a new phenomenon to be studied from scratch: he identifies him as “man” and applies to him the knowledge he has acquired about man (which leaves him free to study the particular, individual characteristics of the newcomer, i.e., the individual measurements within the categories established by the concept “man”). [ItOE, p. 27-28]
It is this integration, provided by the measurement-omission aspect of concept formation, that makes concepts useful cognitive devices. When a man sees a table, he recognizes it as an instance of a concept he has formed, even if he never saw a table of that particular shape, size, or color. This new table is not a new phenomenon. It is a new instance of a known phenomenon. The term “reintroducing the measurements” might be used in a legitimate sense to describe the process of recognizing a concrete as an instance of a concept. But, as we shall see, that is not how David Kelley is using the term. (Throughout the rest of this article, I am referring to Kelley’s use of the term.)
Such is the power of objective concepts. By forming concepts, man can store vast knowledge beyond that which he can hold perceptually and beyond that which he can grasp on the perceptual level. Measurement omission allows him to integrate concretes into classes based on their essential similarities. By applying concepts, man can concentrate on gaining new knowledge because his old knowledge is readily available to him.
David Kelley, in contrast, makes it clear that he holds a different view. For him, measurement omission is not a tool of integration, but of disintegration. Measurement omission is not a means of uniting concretes, but of leaving something out, of stepping away from reality. On his view, when we apply concepts, the concretes subsumed under a concept have no essential similarity. Once we have “reintroduce[d] the relevant measurements,” we only have separate concretes, and all concretes are different. Whatever we have learned about a class need not apply to the members of the class. Why bother forming concepts at all?
Observe Kelley’s application of this idea to moral judgment. “Even if we accepted the premise that libertarianism as such is a vice, there would be a vast difference of degree between libertarians and a regime that has the blood of millions on its hands.” (“A Question of Sanction”) Having identified two instances of evil, libertarianism and communism, we are, according to Kelley, unable to apply anything that we know about one to the other. These may be two instances of evil (though Kelley grants that much only for the sake of argument), but once we put the measurements back in, we are left with distinct groups that are much more different than they are similar. Kelley’s call for “a sense of proportion,” is not, in fact, an argument for distinguishing degrees of evil. To the extent and in the respect that someone, a group, or an ideology is evil, he (or they or it) must be recognized and treated as such. Evil is evil. Yet Kelley argues that a group that he considers less evil than dictators can be treated as if it were not evil.
On a proper view, the concept of “evil” encompasses many diverse aspects of reality, e.g., murder, communism, ESP. The particular measurements that differentiate instances of evil are omitted and the essence of evil- —the volitional destruction of human values—is retained. The measurements omitted include the particular values destroyed, the degree of the destruction or evasion, whether it is a concrete instance of evil (like a murder) or an evil ideology. Measurement omission allows us to integrate the various instances of evil into a single unit, to which we can relate all the knowledge we have or later gather about evil. Then, when we encounter evil, even a variant that we had not seen before, we can recognize it as an instance of evil. Having done so, we can apply all that we we know about evil to this new case. We can pronounce it as evil, refuse to compromise with it, and grant it no sanction.
For Kelley though, the concept “evil” does not unite the units of a wide class. Under his version of measurement omission, all that we retain is an empty shell. We have not integrated the units by essentials; if we had, all the units subsumed by “evil” would have an essential similarity. We have merely given the units a common name. According to him, we cannot make any statements about the class; that would, in his words, “exhibit a kind of zealotry.” We must reintroduce the measurements and deal with the concretes on a case-by-case basis. On such a view, how could he see any similarity between an academic Marxist and someone who puts those ideas into practice?
What then, do concepts (and principles) mean to Kelley? They are not tools of cognition in the sense that Ayn Rand has shown them to be. Rather, they are regarded as human conveniences; this is a nominalist view—at best, a way of holding general knowledge between the time when a concept is formed and when it is applied. But when we apply the concept, that general knowledge no longer is relevant because we have reintroduced the measurements and are now dealing with distinct concretes. It is also an intrinsicist or Platonist view in that concepts are held to be part of a separate realm. Under Kelley’s version of measurement omission, the concept is not a classification of particulars, but an essence without the accidental characteristics; the concretes we deal with when we reintroduce the measurements are imperfect reflections of the concept. Hence, any knowledge about the class need not apply to the concretes.
Nominalism and Platonism, of course, are not incompatible; they are merely two sides of the same non-objective coin. In practice, they reduce to the same thing: a severing of concepts from percepts. (A more explicit statement of someone’s approach to concepts would be needed to determine to which of these two views he subscribes; when accepted implicitly, there is no distinction.)
In spite of the window dressing provided by his Objectivist-like terminology, Kelley does not treat concepts as tools of cognition, as a means of unit reduction. Conceptual knowledge is insignificant; in any new case, the relevant measurements will confront us with a phenomenon we have not encountered before.
It is little wonder then that Kelley’s article is replete with pragmatism. (Note especially his endorsement of the tradeoff approach to ethics, i.e., his “weighing” of the costs against the gains on a case- by-case basis in accepting speaking engagements.) Principles are basic generalizations, and, as such, their scope is enormous. Wide ranges of concrete instances are subsumed under a single principle. But, if in applying a principle, one has to reintroduce the relevant measurements before one can deal with the concretes, which are then seen as essentially different from one another, a pragmatic approach is all that is possible. Having rejected his conceptual faculty’s integrating power, all he has left is the range of the moment. No matter how carefully he tries to deal with the concretes, and despite his claimed adherence to principle, he is essentially in the same position as the Congressional committees described in Leonard Peikoff’s article “Why Should One Act on Principle?” (The Intellectual Activist [TIA], February 27, 1989)
Concepts (and principles) are valuable because they integrate concretes on such a wide scale. Newton’s theory of gravity is a great scientific achievement because it explains why an apple falls from a tree and why the planets follow elliptic orbits around the sun (and so much more). It is a great theory because it integrates so much. The principle of justice tells us to be nice to our friends and not to act as if Soviet dictators are civilized people. It, too, is a tremendous integration. To demand that one “reintroduce the measurements” is to wipe out this integration.
It is true that degrees of good or evil can be evaluated, but that does not subvert the principle; such an evaluation depends on the principle. In the case of evil, the scope of one’s evasion and its existential consequences vary from case to case. In the opposite case (that of evaluating the good), Leonard Peikoff points out:
There are degrees in this issue; there are modest attainments and enormous ones; but the differences pertain only to measurement. The principle in all such cases is the same: correspondence to reality (and its causes and effects) deserves and must be given a positive moral evaluation. [“Fact and Value,” TIA, May 18, 1989, p. 3]
In spite of the similarity in the words, there is a complete difference of emphasis and hierarchy between this formulation and Kelley’s. According to Objectivism, we can make distinctions within the principle, but the principle is more fundamental. If we identify something as evil, we refuse to grant it moral sanction. Once it is seen to be irrational and anti-life, it is treated as such. Differences in degree are a secondary matter, but even these can be determined only by reference to the principle. For instance, a murderer is more evil than a thief (in a typical context), because the former destroys a more fundamental value than the latter. A dictator is more evil still because he attacks these same values on a much wider scale. But that does not mean that the thief is to be regarded as anything less than evil.
David Kelley’s approach inverts this hierarchy. Once the measurements have been reintroduced, we are left with distinct concretes, and the principle is forgotten or ignored. Thus, he holds that any comparison between the Soviet regime and Libertarianism is “wild” and “preposterous.” Yet, Libertarianism is an evil doctrine that can lead only to the destruction of liberty. “If the Libertarian movement were ever to come to power, widespread death would be the consequence.” (Peter Schwartz, “On Moral Sanctions,” TIA, May 18, 1989, p. 7) But, these essential similarities are beyond the grasp of anyone who rejects principles, whether implicitly or explicitly.
This approach to concepts helps to explain Kelley’s failure to grasp the nature of Libertarianism. “Libertarianism is a broadly defined movement. The subjectivists represent one definite wing of the movement … ” (“A Question of Sanction”) He does not see that “[s]ubjectivism, amoralism and anarchism are not merely present in certain ’wings’ of the Libertarian movement; they are integral to it.” (Peter Schwartz, “On Moral Sanctions,” p. 6) Once he has reintroduced the measurements, he is unable to see the common denominator uniting these various Libertarian “wings.” The same phenomenon is at work when Kelley says that Laissez Faire Books is not subjectivist, but “ … is a book service, selling works that take many different positions on philosophical issues.” (“A Question of Sanction”) And that, “L-F Books is not an agent of the Libertarian Party. Nor does it represent subjectivism as opposed to any other philosophy.” (private correspondence, May 14, 1989) He cannot recognize individuals or a group as subjectivists because when he reintroduces the measurements, he is left with people who hold some specific views.
Perhaps one of the the most dramatic consequences of Kelley’s approach to concepts is his statement “[t]he concept of evil applies primarily to actions, and to the people who perform them.” (“A Question of Sanction”) It is shocking to hear this from someone who is supposedly a knowledgeable Objectivist. Actions are consequences of ideas. Yet Kelley is forced to divorce cognition from action given his view of concepts. His view of measurement omission cuts off concepts from reality. Ideas are not condensations of reality, but imperfect reflections of it. Concepts are not fully real to him. He must reintroduce the measurements and deal with concretes, thereby limiting himself to the perceptual level. For him, the connection between ideas and actions (i.e., reality) is at best tenuous.
Kelley’s call for “tolerance” and his claim that those who passionately uphold Objectivism “indulge in moral hysteria” can also be understood in view of his approach to concepts. As Leonard Peikoff points out, the concept that Kelley and the other advocates of tolerance fail to grasp is “objectivity”—the central concept in Objectivism. “In the minds of the ’tolerance’-people, there are only two possibilities in regard to moral judgment: moralizing or emotionalism, dogma or whim, i.e., intrinsicism or subjectivism. ” (“Fact and Value,” p. 4) But “objectivity” has as its basis the fact that concepts are objective, i.e., that they represent facts of reality identified by a human consciousness. If one subscribes to a view that concepts are non-objective, that one only has knowledge of reality when he “reintroduces the measurements,” then he will be unable to grasp the concept of “objectivity.”
Ayn Rand’s theory of concepts is a magnificent achievement. But that theory is deep and complex, and subtle errors in applying it, whether made honestly or not, can have disastrous repercussions, as David Kelley’s article clearly demonstrates.
The essence of that theory is measurement omission. But measurement omission has a very specific meaning. Ayn Rand made that meaning clear, and I have tried to indicate it here. Students of Objectivism should be on guard for interpretations or applications of Ayn Rand’s ideas that may sound legitimate at first, but conflict with her actual intent (and with reality).
Ultimately, there are only two wrong theories of epistemology: intrinsicism and subjectivism. In this sense, the fallacy of “reintroducing the measurements” is nothing new. But these wrong theories exist in countless variations. It is crucial for Objectivists to recognize this most recent variation for what it is.
Copyright © 1990 Objectively Speaking, reprinted by permission Copyright © 1991 Bennett C. Karp