Thinking in Definitions

From the time I first read Ayn Rand’s theory of concepts, almost sixty years ago, I have thought of myself as someone who thinks in definitions.
 
Why? All our knowledge, with the exception of proper nouns, consists of concepts. Combinations of concepts that make important connections to each other are called principles. When concepts and principles are well crafted, they become extremely illuminating, to myself, of course, but also to others when I write and teach. In simple terms, thinking in definitions means I know what I am thinking and talking about and, I would hope, others, my students and readers, also understand what I am saying.
 
Unfortunately, definitions today, if any attempt at a definition is made, fall into the “kinda, sorta, maybe, let’s assume” category, or are the positivists’ nonobjective arbitrary constructs, useful only in pushing words around in rationalistic discussions or writing. Textbooks, in particular, provide definitions that are too long and wordy, nonessential, and violate most of the rules of good definition.
 
What is a good definition? One that identifies genus and differentia of the referents being defined, that is, the general category within which the referents are distinguished and their essential distinguishing characteristic, which as Ayn Rand states, is the one that explains and causes most or all of the other distinguishing characteristics.
 
Human beings, for example, are animals that possess the capacity to reason. “Animal” is the genus and “capacity to reason” is the differentia, the characteristic that explains and causes all or most of the others that humans share, such as the ability to speak a language or to build skyscrapers. The definition is concise and, as Rand says, summarizes and condenses all our knowledge of humans in the “file folder,” to use Rand’s metaphor, stored in our minds.
 
Note that we all have our own file folders of humans that can be thin (for a child), thick (for most adults), or fat (for, say, a biologist or psychologist). The definition is written, to continue the metaphor, on the outside of the folder with the word that denotes the concept written on the tab.
 
The rules of a good definition are straightforward—identify genus and differentia but also ensure that the definition is truly essential and is not too broad or narrow, circular, vague, obscure, metaphorical, or, if possible, negative. It does require practice to become competent at the skill.
 
Thinking in definitions can also be thought of as thinking in ranges of measurement. Ayn Rand identifies abstraction as measurement omission, but that does not mean we forget about the measurements. It means that in the process of identifying the differentia of a concept we must be aware of the “stopping points” or boundaries of those measurements that are omitted. In this process, we limit the concept to a certain range. 
 
Tables, for example, are not a mile long, nor are they only one inch long. Bridges are flat with supports and hold objects called automobiles, but bridges are not tables. A toy table may be flat with supports and hold tiny teacups and saucers, but it is not the type of table adults eat dinner on.
 
The concept “human” is based on the range of measurements that describes our distinctive type of mental processing and distinguishes it from that of the higher mammals, the lower end of the continuum being the point at which we observe humans using reason to form and apply concepts, but not the higher mammals.
 
When the boundaries of cholera, to use three examples from McCaskey’s article “Induction in the Socratic Tradition,” were narrowed to an illness involving an “organic poison,” after dismissing season of year and foul water as competing causes, Robert Koch further narrowed the range of measurements, the cause and differentia, to the bacterium he called the comma bacillus. When Charles Wells, after many experiments, narrowed the boundaries of dew to droplets created in accordance with the laws of heat, particularly liquefaction, he recognized that condensation from vapor in the air was the cause and differentia. In both cases, the range of measurements of the respective differentia, comma bacillus and water condensation, were sufficiently limited to establish the concepts as universal.
 
Lord Kelvin concluded, after two hundred years of imprecise and often inaccurate tidal measurements, that Newton had been right, namely that tides are ocean movements that result from gravitational pull of the sun and moon, measurements omitted.
 
There are options in the formation process, as Ayn Rand points out, because different concepts may use different “stopping points” in the continuum. Hence, the English word “man” can be a synonym of human being or, using the range of measurements describing male and female anatomy, can mean a male person. The Yiddish word “mensch,” now an accepted part of the English language, means a rational, volitional human being who acts with integrity and honor. 
 
Options in concept formation are what make translation from one language to another challenging, because the range of measurements used in the concept’s formation differ from language to language. Keeping in mind the continuums that constitute our concepts enables us to maintain contact with reality, the reality of the different-but-similar referents we are discussing.
 
Thinking in well-constructed definitions especially enables us quickly to bring to mind the meaning of broad abstractions—abstractions from abstractions, and concepts of consciousness. Thinking in not-so-well-constructed definitions leaves us with the “kinda, sorta, maybe, let’s assume” category that even academics are guilty of, rationalists and positivists in particular.
 
The favorite vehicle of argumentation of these rationalists and positivists is to make grossly unrealistic assumptions and then announce, “Let’s assume, by definition, that dogs can fly or that humans are “brains in a vat,” and see where that takes us. If, by definition, they mean, and they usually do, arbitrary, disconnected floating abstractions and flimsily constructed tautologies, which violate all the rules of good definition, knowledge can be anything they want it to be, which is where we are today as consequence of Immanuel Kant’s inability to find reality and his positivist followers who restricted awareness of reality to the level of perceived concretes.
 
Flying dogs (and maybe brains in a vat) may make interesting fiction, but they do not make good science. In economics, the doctrine of pure and perfect competition does not even make good fiction—“perfect” information, that is, omniscience, just being one of its absurd assumptions. (See Kirkpatrick, In Defense of Advertising, 118–21.)
 
Concepts, to emphasize, are not “true by definition.” A concept, and its definition, is “true by reality,” that is, true by our having correctly identified the referents in reality and having summarized and condensed the file folder’s content for ease of use. They are open-ended in the sense that they apply to all referents of the same kind, past, present, and future—because they are classifications and cross-classifications (in our minds) of reality.
 
Concepts and their definitions are cognitive necessities, and they are constructed to meet cognitive needs. They are not “out there,” embedded in the concretes of the world. As Rand states, concepts and essences are epistemological, not metaphysical.
 

Positivism

The philosophy of positivism—also called logical positivism and logical empiricism—has for decades been the bane of science, especially the human sciences.
 
The concept “positivism” that I am using is broader than the logical positivism of the Vienna Circle in the1920s–1930s, which attempted, but failed, to find Immanuel Kant’s noumenal world through symbolic logic applied to sense data. Positivism (495–98), as it is still used today and derives from the work of August Comte, means the use of the so-called scientific methods of physics and chemistry to attempt to discover laws of human behavior and events.
 
All science, that is to say, must be quantitative, and if it is not quantitative, it is not science. The post-Kantian premises of positivists, in their attempt to defend science, are the following: the entire branch of philosophy called metaphysics is meaningless—that is, the concept cannot be “empirically verified”—and universals, facts, values, and, especially, truth are similarly unknowable and therefore meaningless. Awareness of reality is restricted to directly perceived concretes. At best, say the positivists, we can try to come up with “successive approximations” of statistical probabilities, but we will never reach “absolute” certainty or “eternal” truth, and, of course, all science must be “value-free.”
 
Thus, an endless parade of statistical studies to incrementally—and slowly—pretend to develop science, never achieving the status of universality because universals, well, as stated above, are metaphysically meaningless. This method, or epistemology, is gospel in the so-called social sciences today, which I prefer to call human sciences, the sciences of homo sapiens that emphasize and study the animal that possesses a volitional consciousness. The human sciences, in other words, study human nature as the being who exercises, or does not exercise, the capacity to reason.
 
The two fundamental special sciences of human nature are psychology and economics, that is, respectively, the science of the motivation and behavior of individual human beings, healthy or unhealthy, and the science of cooperating individuals in a social setting the aim of which is to generate and produce peace and prosperity.
 
Austrian economists do not agree with the positivist mantra, preferring the method of logical reasoning, the process of identifying correct concepts and integrating them into principles, then integrating principles into theories—leaving Newtonian-type algebraic equations to the physicists and chemists. Sigmund Freud essentially did the same thing in the development of psychoanalysis.
 
Austrian economist F. A. Hayek in his 1974 Nobel laureate address did not use the word “positivism,” but did say that scientism and the scientistic attitude of insisting on using the methods of physical science in the social sciences “quite arbitrarily limits the facts which are to be admitted as possible causes of the events which occur in the real world.” Scientism is a close synonym of positivism and is defined by Hayek as a pretense of knowledge.
 
Positivism over the years has been criticized, notably for its contradiction that metaphysics is meaningless—because that statement itself is a metaphysical assertion.
 
What positivism in particular lacks, among its many other faults, is a sound theory of universals and theory of abstraction from abstractions, which I have written about in previous posts (1, 2, 3).
 
According to Ayn Rand, universal concepts are formed by omitting the measurable differences of the concept’s referents—length, width, and height, for example, of several perceived, concrete objects we come to identify and label as “tables.” After identifying other similar but, in some respects, different perceived concretes—such as chairs and beds—we omit the measurements of size and focus on, but also omit, the different purposes of tables, chairs, and beds, retaining their use in a human household to form the concept “furniture,” which is an abstraction from abstractions.
 
When we move to broad abstractions in psychology and economics, such as depression and the business cycle, we require focused attention on the chain of abstractions from perceived concrete to broad abstraction—and back to the perceptual level. This mental effort—and it is mental effort—is precisely what is lacking in positivism because, as they say, the abstractions are “analytic,” arbitrary constructs, subjectively defined. Only “synthetic” perceived concretes can be true or false.
 
The mental effort in forming and correcting our abstractions from abstractions is precisely what led Ayn Rand (67–68) to say, rather pointedly:
 
Like a spoiled, disillusioned child, who had expected predigested capsules of automatic knowledge, a logical positivist stamps his foot at reality and cries that context, integration, mental effort and first-hand inquiry are too much to expect of him, that he rejects so demanding a method of cognition, and that he will manufacture his own “constructs” from now on.
 
Karl Popper (chap. 1) denied that he was a positivist though his premises are positivistic. He attempted to defend science by proposing the still widely accepted criterion of “falsifiability,” that is, the notion that a concept is valid only if it can be falsified.
 
Meaning precisely what? In the human sciences, the answer to this question is that concepts, or rather “constructs,” must be “operationalized” in order to test them using the experimental-statistical methods of the physical sciences. If that cannot be done, the concept cannot be falsified.
 
Operationalization means to make measurable. What is the measuring instrument to be used in the human sciences, where volition is the fundamental premise? Well, first of all, volition is not relevant because it cannot be “falsified” and, besides, real science does not acknowledge such “pseudoscience,” which, as Popper might say, is akin to astrology.
 
Therefore, following the Popperian method, such constructs as “attitude” from psychology and “utility” from economics must be measured by administering a “paper and pencil” instrument called a questionnaire to a sample of five hundred or more people. The collected data is then analyzed, perhaps using sophisticated multivariate statistics. Probability estimates and confidence intervals are calculated. Universals are never discussed or considered relevant or possible.
 
If a concept cannot be measured using this method, it cannot be falsified. It is therefore meaningless and invalid.
 
Popper states that science begins with highly informed guesses” (conjectures) and then attempts to falsify or refute them (chap. 3; my italics). Similarly, economist Milton Friedman (14) says that economic research, in the positivist-Popper tradition, begins with “wildly inaccurate” assumptions that we test and attempt to “empirically verify,” or rather “falsify.”*
 
The problem here is that the terms “guess” and “hypothesis” do not have the same referents. A hypothesis begins with significant evidence about what we are studying, but not enough to draw a definite conclusion. A guess has little or no evidence.
 
 “Empirical verifiability” of the positivists and Popper’s alternative of “falsification” are not essentially different. They both constrict science—all science, not just the human sciences—to perceived concretes and the experimental-positivistic-behavioristic methodology, essentially dismissing broad abstractions and logical reasoning.**
 
As Rand’s theory indicates, the essence of science is measurement omission, not measurement, and the proper method of doing science is conceptualization. The criterion of what is true or false, that is, the correct identification of reality, is Aristotle’s law of non-contradiction. Whatever violates the law in the name of science should be called pseudoscience.
 
Which seems to be an appropriate descriptive of positivism.
 
 
* Friedman, “The Methodology of Positive Economics.” Friedman, however, apparently regretted writing this essay.
 
** The experimental-positivistic-behavioristic terminology is psychologist Abraham Maslow’s disapproving description of positivist epistemology (xlvi).

Mises and Epistemology (part three of a three-part series) (Go to part one)(part two)

The human sciences aim to explain the causes of human motivation and behavior, healthy and effective, as well as harmful, and to guide humans in their choices and action to achieve chosen goals. This includes all applied human sciences whose aims are to get things done and done well. Economics is not just a highly deductive science, but also highly applied in that its aim is to define the principles of cooperation under a division of labor that will secure peace and prosperity in a social setting.*
 
Let us now continue with a few additional epistemological issues in Mises’ writing.
 
 
4. The specific understanding. The two fundamental methods of cognition, according to Mises and the Kantian philosophers, are conception and understanding. Concepts, we have seen from part one of this series, according to Kant, are limited to the phenomenal world, which means we cannot know concretes. This poses a problem in particular for historians, since their work is to explain concrete human events of the past. The solution in German philosophy is the specific understanding (Human Action, 49-50). Mises writes:

It is the method which all historians and all other people always apply in commenting upon human events of the past and in forecasting future events….The scope of understanding is the mental grasp of phenomena which cannot be totally elucidated by logic, mathematics, praxeology, and the natural sciences.

The philosopher Bergson, according to Mises, called this cognitive power “intuition,” which sounds antithetical to reason and mystical. And it is, because for Kant concepts and reason, which he had to limit to make room for faith, cannot know true reality.
 
But a correct theory of concepts proves we can know reality. This cognitive power is not properly described as a “specific understanding,” but as the application of previously formed universal concepts, a deductive process. Since our concepts refer to all concretes of a particular type, we live our daily lives applying our previously learned knowledge to identify correctly the specifics we confront. Thus, Sherlock Holmes deduced that Watson just came back from Afghanistan, based on Watson’s tan and signs of having been wounded. And our medical doctor deduces that our cough and runny nose are instances—symptoms—of a cold.
 
Application (1, 2) is the mental process of identifying a this (concrete) as an instance of a that (concept). It is what historians do and what we all do every day.
 
 
5. Regularity in human action, but no constants. Mises (Human Action, 1) is correct to state that there is “a regularity in the sequence and interdependence of market phenomena,” but no “constant relations” (56), or standard unit that can be used to make quantitatively exact measurements of human behavior. Which means no algebraic equations can be identified or used as in physics and chemistry.
 
Human behavior is not deterministic because of free will, which Mises does not explicitly endorse, though he does say that human action is teleological, which is another instance of his Aristotelianism. But because of the observed regularity that is present in human nature, principles and laws of economics, such as the laws of marginal utility and supply and demand, can be formulated and used to interpret current and past economic events and to make predictions, albeit not mathematically precise predictions. This is Mises’ rejection of the positivist demands for a mathematical economics.** (See In Defense of Advertising, 127-30.)
 
Indeed, the inability to formulate mathematically precise propositions and equations is true of all human sciences, basic and applied, not to mention of all the branches of philosophy. Positivism has corrupted our understanding of science, the actual essence of which is conceptualization, which means measurement omission even in the formulation of the equations of physical science. (See In Defense of Advertising, 153-58; Independent Judgment and Introspection, 72-81.) Most measurements, especially statistics, produce historical data, not universal knowledge of cause-effect relationships.
 
 
6. Subjective value. I touched on this subject in a previous post based on what I call Ayn Rand’s general theory of value, but more needs to be said.
 
Mises (Human Action, 96) is mostly correct when he states, “Value is not intrinsic, it is not in things. It is within us; it is the way in which man reacts to the conditions of his environment.” Values, and all concepts are within us. Essences and values are not “out there” in the thing. They are not metaphysical. We, using the tools, or mental processes of our consciousness, create, construct, or form concepts, including the concept of what is valuable. But values are objective in the epistemological sense—if the identification of what is valuable contributes to the flourishing of human life.
 
Yes, there is a distinction between object and subject, meaning objects are “out there” and whatever goes on in our minds is subjective, but that is a different usage. Customer value judgments are subjective because their judgments come from inside their heads, but they may or may not be epistemologically objective (correctly identified as beneficial for or harmful to them).
 
Mises—and the positivists—talk about the necessity of all sciences being “value-free,” which means “shoulds” and “oughts” do not belong in science. This derives from Kant who faced a dilemma between ethical values that come from the nature of the mind (not the requirements of human life) and free will that comes from the noumenal world, but is unknowable. Positivists “resolved” the dilemma by explicitly making all values subjective.
 
Economics for the most part is a practical science, establishing principles of how to increase wealth for everyone. One can say that price controls are both immoral and impractical—immoral because of the harm to human life caused by injecting force into voluntary trading and the consequent shortages that result, thus introducing values into the discussion, and impractical because wealth for everyone is decreased, not increased. It is only increased, if in fact it is, for certain privileged groups.
 
Yes, there are those today who say that all issues are moral and political, proclaiming “consumers should or ought not to buy that product” or “that bridge should not be built in that location.” In a free society where individual rights, especially property rights, are respected, such shoulds and oughts would be minimal. Applied sciences assume that actions to be taken are moral and legal, then provide guidance to the achievement of particular goals. The actions are assumed to be moral and legal because the aim of applied science, all sciences actually, is to improve life.
 
Two final quotations from Mises about values: “By means of its subjectivism, the modern theory becomes objective science” (Epistemological Problems of Economics, 192). And: “it is in this subjectivism that the objectivity of our science lies” (Human Action, 21).
 
Mises is stating that the theory of marginal utility answered the water-diamond paradox of the classical economists who had assumed values are “out there” intrinsic to reality. If “subjective” is understood as explained above and can be viewed as epistemologically objective, then Mises is correct.
 
I prefer to call our values, market values in particular, “psychological,” which makes them epistemological, not metaphysical. Restating Mises’ second quotation, I would say, “it is in this recognition that values are psychological, not intrinsic or metaphysical, that the validation of economics lies.” (See In Defense of Advertising, 175-78.)***
 
 
In the first paragraph of the Preface to the Ultimate Foundation of Economic Science, Mises writes, “This essay is not a contribution to philosophy. It is merely the exposition of certain ideas that attempts to deal with the theory of knowledge ought to take into full account.” Mises in dealing with these issues is taking steps toward a post-Kantian Aristotelian foundation of economics. Those who would advance economics further can build on his work.
 
The present three-part series has been my attempt to deal with some of the issues, based on Ayn Rand’s epistemology. It is also my encouragement to students of Ayn Rand to take the work of Mises as both profound and serious. There is much to be learned from him.
 
 
* All sciences, basic and applied, use induction and deduction throughout (cf. Aristotle, Posterior Analytics, in McKeon). The applied sciences—medicine, engineering, psychotherapy, and the business disciplines, among others—are predominantly deductive, drawing their basic principles from the more fundamental sciences.
 
** In technical terms, the physical sciences measure the world using interval and ratio scales, the former using equal intervals from one number to the next (1 to 2 is equal to 4 to 5) and the latter requiring equal intervals and a true zero point to produce valid ratios (for example, Kelvin temperature, but not Fahrenheit or Celsius, which are scaled at the interval level). In the human sciences, measurement is teleological using ordinal numbers and ordinal scales (A is better than or preferred to B and B to C). Relationships of means to ends are graded or ranked because we do not have a unit that serves as a standard of measure in the way interval and ratio scales do. The intensity of preference in the intervals of two people preferring A to B may or may not be equal; we do not know. See Stevens and cf. Ayn Rand, Introduction to Objectivist Epistemology, 32-34, 223-25.
 
*** Ayn Rand in Capitalism: The Unknown Ideal (16-17) distinguishes between market value, or price, as socially objective value, “the sum of the individual judgments of all the men involved in trade at a given time” and philosophically objective value, “as value estimated from the standpoint of the best possible to man.” Understanding this last means philosophically objective values are not much involved in economics, as no such value should or ought to be mandated by law, or priced a certain way.

Read part one, part two.

On the Correct Roles of Induction and Deduction in Human Life: Two Sentences from Ayn Rand’s Theory of Concepts

Original thinkers often state their identifications succinctly.

Ayn Rand’s notion of measurement omission (Introduction to Objectivist Epistemology, chap. 2) in the formation of concepts is one such identification. Here is another (p. 28):

The process of observing the facts of reality and of integrating them into concepts is, in essence, a process of induction. The process of subsuming new instances under a known concept is, in essence, a process of deduction.

These brief and to the point sentences state not just the two fundamental methods of cognition, but more importantly, the correct roles of induction and deduction in human life.

And by “human life,” I mean science as well as everyday life.

Induction is the process of generalization, of forming universal concepts based on our observation of particular objects or events. The definition of a single concept states a principle—all humans possess the capacity to reason, for example—and the combination of several or many concepts and principles builds our knowledge of reality and, in some cases, establishes the physical, biological, and human sciences.

Induction is conceptualization. From an early age, probably before we can assign words to them, we all practice the inductive formation of universal concepts.

This was my example in an earlier post of our daughter, before she could walk or talk, laughing heartily at her first sight of a bouncing ball (Applying Principles, pp. 322-24). She identified a universal, because her mind, to quote Aristotle (Posterior Analytics, 100a13), “is so constituted as to be capable of this process” (though the universal is not “in the thing,” as Aristotle assumed).

Rand’s identification describes in general terms the true nature of induction and makes the biological and human sciences as exact and valid as the physical sciences.*

Deduction is the process of identifying particular objects or events as instances of the general knowledge we have already acquired. The process, more accurately, is one of application.

Deduction is what Sherlock Holmes did and what medical doctors do, and what we all do in our everyday lives. We apply general knowledge to specific cases to guide us in making choices and taking actions.

Technology and the applied sciences are sciences of method and therefore are largely deductive, deriving their basic principles from the more fundamental sciences on which they rest, for example, engineering from physics and chemistry, medicine from biology, and economics from psychology with several business disciplines drawing their basic principles from both psychology and economics.

This identification of deduction as application dispenses with the detached-from-reality deduction for the sake of deduction that has dominated the academic world since the Middle Ages. Deduction as application demonstrates how much deduction we practice in our everyday lives.

We all induce and deduce—some of us better (more accurately) and at greater length (in intensive study) than others. What Ayn Rand’s identifications mean is that induction and deduction are not a monopoly of scientists, philosophers, or academics in general.

Where then does measurement fit in the sciences? Conceptualization is universalization, which means its essence is measurement omission, which means the essence of theoretical science is measurement omission. This means that measurement is an aid to theoretical science, not its essence.

Measurement is crucial in the applied physical sciences when, for example, we want to send astronauts to the moon and back. Measurement in the biological and human sciences, however, is not quantitatively exact in the sense of constructing advanced mathematical equations to predict the behavior of animals or humans.

In the human sciences it is that annoying thing called free will—annoying to many human scientists, most of whom are materialists and determinists—that prevents the human scientists’ “elegant” equations from making any practical sense, or from being replicated in subsequent studies.**

The biological and human sciences are exact and valid, if the conceptualizations made by the scientists working in those fields have correctly identified the aspect of reality they are studying. The identifications are not equations, but they are quantitative. For example, psychological depression can be severe or mild.

“Measurement omission” does not mean that conceptualization ignores measurements. One individual case is quantitatively distinct from the next one, as two balls can be two different sizes and can be made of different materials.

Precise measurement is what technology and applied science, especially in the physical sciences, must critically pay attention to. Measurement in the applied biological and human sciences does not have to be so precise—because it cannot be.

“Truth,” to quote another succinct identification of Ayn Rand (Objectivist Epistemology, p. 48), “is the product of the recognition (i.e., the identification) of the facts of reality.” Truth, for Rand, is not a correspondence theory, but one that identifies facts. It is a recognition or identification theory.

And what is our guide to truth? Logic, of course, as “the art of non-contradictory identification” (Objectivist Epistemology, p. 36), not the mathematical or symbolic stuff that is taught in universities today or the medieval rationalism that permeates the older logic textbooks.

Induction and deduction are what we all use every day in our practical lives.

Induction and deduction, respectively, are conceptualization and application. Measurement is an important component of the two, but it is not their essence.


* See John P. McCaskey, “Induction in the Socratic Tradition,” in Shifting the Paradigm: Alternative Perspectives on Induction, ed. Louis F. Groarke & Paolo C. Biondi (Berlin: De Gruyter, 2014), 161–192, on his efforts to revive Socratic induction, a tradition promoted and debated both before and after Francis Bacon, but eventually overtaken by the nineteenth-century positivistic, Millian hypothetico-deductive method, a form of rationalistic, propositional inference. Socratic induction—generalization from particular things or concretes to universal abstract ideas—is consistent with Ayn Rand’s epistemology as inductive concept formation through measurement omission.

** “Man is that which fits economic equations,” as Ayn Rand (Capitalism: The Unknown Ideal, p. 7) so simply and aptly caricatured the very rationalistic, pseudo-deductive doctrine of pure and perfect competition in economics.