NOTES ON PHILOSOPHY

HERE THERE BE MONSTERS:

Expeditions of an Amateur Philosopher

ANTAEUS SHRUGGED

The Dismemberment of Philosophy

People commonly think of philosophy as the study of the most basic or fundamental issues; and, simultaneously, as the study of the most abstract or abstruse issues. This might be restated as the claim that philosophy is concerned with the most general or wide-ranging questions that humans may be concerned with.

Professional philosophers show a tendency to assume a corollary to this characterization of their field: That philosophical questions are the most difficult of all questions -- so difficult, indeed, that we may not reasonably expect them ever to be solved. [Note Kant's claims in the introduction to Prolegomena.]

If, however, we look at the field of philosophy in a historical context, we are more likely to be struck by a very unusual phenomenon. Philosophy is virtually unique among the disciplines in that it is the only field of knowledge, the area of concern of which is shrinking. All other fields characteristically expand their charters as new knowledge is discovered. Physics, for instance, has in the last century added quantum mechanics, particle physics, relativity theory, and several other major new fields of study. Nor is this behavior confined to the sciences; history, for instance, now includes econometric studies of past societies and other subjects that had not been conceived of in the time of Macauley or Gibbon.

Philosophy, by contrast, is a field that history has developed by subtraction. Originally, the term engrossed all fields of human study not immediately of practical application. Thus Aristotle's philosophical writings covered physics and biology. Philosophers argued questions, such as the existence of atoms or the development of embryos, that today are clearly outside the domain of philosophy. As what we now call the sciences began to take shape, the term "natural philosophy" came into currency to refer to them. In the Nineteenth Century, the "hard" sciences such as physics and chemistry broke away from philosophy completely; in the Twentieth, the "soft" sciences such as psychology and sociology began to detach themselves. So modern philosophy consists of what is left.

But even the remnant is increasingly threatened. Consider the five disciplines of philosophy: metaphysics, epistemology, ethics, politics, and esthetics.

Modern physics, working in the fields of cosmology and particle physics, has begun to nibble -- nay, to munch -- at the edges of metaphysics. If modern philosophers complain that metaphysics is "otiose", it is perhaps because all the real questions of the field have been ceded to the scientists.

Meanwhile psychologists, though their own territory is not yet completely secured, have begun to invade the field of epistemology. At the same time an unorganized, almost unintentional guerilla army of scientists is operating behind the lines of professional philosophy. Despite the best efforts of Kuhn and his followers, scientists are starting to develop by their own efforts an increasingly coherent system of scientific epistemology.

In the field of ethics, philosophers have recognized the threat and rallied to the barricades. The first probes of sociobiology were repulsed by a savage counterattack. But how long can philosophers hold the line against the daily increasing strength of the enemy, even if they resort to an alliance with their traditional opponents in the religious community?

The field of political science announces its aggressive intentions in its very name. Here the invading army is still weak, yet philosophers already show signs of readiness to run up the white flag.

All that remains is esthetics. No external challenge threatens here, and yet philosophers seem hardly to want to keep, let alone exploit, this territory. Perhaps the natives -- the artists, and even worse the critics -- have made it so impoverished and unhealthy that no one wishes to conquer it.

So philosophers -- by their own choice, oddly enough -- continue to defend their gradually shrinking territory of confusion and uncertainty.

The Heirarchy of Investigation

When we wish to learn something about the universe, our method of operation will depend on how far we have progressed.

At the lowest level, we have very little knowledge of the subject and probably do not even know how to properly define the boundaries of investigation. At this level, the only available method is trial and error.

Once we have developed our concepts sufficiently that we can communicate thoughts about the subject with some degree of efficiency, we may progress to the next stage: the dialectical method. This allows us to find errors more quickly by using criticism from other people, who are more likely to spot mistakes than the creator.

Further development of our concepts gives us the ability to develop true theories about our subject matter, in which large aggregates of knowledge can be organized into a logical structure. At this point we are able to use the scientific method.

Finally, we begin to approach full understanding of the roots of the subject. At this point we create axiomatic systems and investigate the subject by means of deductive reasoning.

(Note that these levels correspond to conjecture, hypothesis, theory, and law.)

Antaeus Shrugged: or, What's the Question?

Most people who have been exposed to a college philosophy course tend to come away from it feeling a great deal less sure of themselves, a great deal less inclined to feel certain about things. Indeed, teachers of philosophy frequently state, quite explicitly, that making students feel this way is one of their primary goals.

In the academic view, the student enters a philosophy course as a cocksure teenager. He has opinions about everything, and he feels himself standing on firm ground, able to express certainty on all fundamental questions. The function of the teacher is to subtly and gradually cut the ground from beneath the student's feet, demonstrating to him that he does not know what he thinks he knows, until there is nothing left to stand on. Sinking into the quicksand, the student finally concedes, like Socrates, that he knows only that he knows nothing. At this point he has grown up.

Not everyone chooses to accept this epistemological model. Many of us see the proper function of philosophy very differently. We come to the philosopher as a student, having only a small platform of certainty on which to stand. We know only a few things for sure -- that reality really is "out there", that contradictions cannot exist, that, being conscious, we are capable of knowledge. With his aid, we hope to expand our little patch of solid ground, adding to the structure and strengthening it. We recognize that occasionally we will make a mistake -- that some section will not be nailed together properly; but our response will be, when we feel the unsteadiness beneath our feet, to draw back and rebuild, until the defective area becomes firm and reliable.

How, though, are we to defend ourselves from the skeptical philosophers? For we are up against skilled professionals, people who debate ideas for a living, experts who know all the tricks. Such people have dominated the field of philosophy since the very beginning:

Mastery of this sort of stuff would by no means lead to increased knowledge of how things are, but only to ability to play games with people, tripping them up and flooring them with different senses of words, just like those who derive pleasure and amusement from pulling stools from under people when they are about to sit down.

Socrates, Euthydemus

Ultimately, there is no substitute for learning the techniques of intellectual self-defense. One of the most valuable is to take a leaf from Antaeus.

Just as Antaeus could never be defeated as long as he was in contact with the earth, so the student gains an inestimable advantage by maintaining the question in contact with reality.

One of the favorite tricks of philosophy is to challenge the student with a question that sounds like it means something, but does not. For instance: "What is truth?" or, "What is the meaning of life?" or, "What is the origin of time?" The student who allows himself to be drawn into the question will, like Antaeus when he was lifted off the ground by Hercules, soon find himself helpless. Each time he makes a stab at a solution, his teacher will subtly, almost imperceptibly, shift the context so that the answer given appears absurd. Soon he will be suspended in a cloud of confusion, bewildered and helpless.

Let him, however, insist on getting back to earth, and the tables are quickly turned. When real questions are asked about real issues, rationality returns. We may not know the answers -- but we know that there are answers, and we get at least a hint of how to find them.

Let's take some examples. "What is truth?" Well, begin by asking, what does this question have to do with the price of eggs in Peoria? When, in any real-life situation, would it matter to me "what is truth"?

Now, when I say "real-life", I don't necessarily mean some sort of physical, "practical" situation. We are not going to deny that there are important questions that do not have to do with how to make money or how to build a house or how to avoid being eaten by a tiger. Rather, I am saying, to what end in reality is the answer to this question going to be a means?

Very abstruse questions indeed can have immense practical consequences. [examples from physics and technology]

Returning, then, to the question "what is truth?", let us ask: How could the answer matter? If we look at classic efforts on this question -- such as, "truth is congruence to reality" -- we see that they are just as much floating abstractions as is the question itself.

Suppose we demand a re-formulation of the question in practical terms. For instance, if we ask, "By what criterion can I tell whether a statement is true?" we now have a question that is very abstract, very general, very universal -- and very real. Anyone can see that the answer to this question, unlike the answer to "what is truth?", is of crucial importance every day to every human being. What's more, the question is now in a form in which it can be attacked effectively. Instead of floundering around in a fog of hazy ideas, we can ask things like: What kind of statements? For to determine whether "no mammals lay eggs" is true will, it seems likely, require a different procedure than determining whether the Pythagorean Theorem is true, which in turn is a different matter from finding out whether Rachmaninoff was satisfied with the cadenza for his Third Concerto. So we see another advantage to grounding our inquiry in reality: by doing so, we bring out the real complexity of the issues and get to problems that have some meat on them.

Take another question: "What is the meaning of life?" Here it may be argued that "ordinary people" have a definite interest in the answer to this question. It will be said that Joe Sixpack will make very definable changes in his behavior if, for instance, he truly believes that his life has some particular meaning. But -- as soon as we ask, what could be a "meaning" in this context, all this apparent concreteness evaporates. When we are told that service to others or the worship of God or accomplishment of great things "gives meaning" to a life, what does that mean in reality? What it means, we will soon find, is that the person in question has certain feelings or emotions as a consequence of doing or trying to do something with his life. But now we have got the question in contact with reality, and we may phrase it, say: "How are people affected emotionally by the purposes they choose in life?" We now have a question that is important, interesting, answerable (though in many cases with difficulty) -- and quite useless for confusing philosophy students, which is why philosophy instructors never ask it.

Necessary and Contingent

The distinction between "necessary" and "contingent" statements, or "a priori" and "a posteriori", is invalid. The best way to see this is to look at a well-known type of example.

When it is said that "2+2=4" is a priori, but "men have two legs" is a posteriori, what is really meant? We must reject any talk about whether a statement "could be" false in some other universe; this has no meaning and is not testable. That "we can imagine" men with three legs is not a strong argument; some people seem to be able to imagine the inconceivable, and others cannot imagine things that are known to be true. (Note that it is easy to imagine a town in which there is a barber who shaves all those and only those who do not shave themselves.) It may be said that we know that 2+2=4 "without reference to experience"; but I defy anyone to learn this truth, or any other, without experience.

If we are to make any sense of this, or any, distinction between types of statements or truths or knowledge, it must be based on a difference of procedure by which we develop them. But in that case we should have far more than two different kinds of statements. (For instance, what about knowledge developed by means of computer simulation?) This might be a productive area of investigation for epistemology, but the "necessary/contingent" distinction does not by any means fall out of it.

Consider another mathematical statement: "The interior angles of a triangle add up to 180 degrees." Now, is this "necessary" or "contingent"? Is it "a priori" or "a posteriori"? Come to that, is it true or false? Well, if we regard the statement as "a priori", its truth depends on whether we accept Euclidean or non-Euclidean geometry. If we regard it as "a posteriori", we find by experiment that our universe is non-Euclidean, so we can definitely say the statement is false. But we are told that in some other universe it might be true. This is only superficially plausible. Let he who asserts this construct a complete physics with Euclidean space-time and prove that his hypothetical physics is self-consistent! For Einstein's argument for the theory of relativity is to a great extent a reductio ad absurdem of Newtonian physics. Thus the "a priori" is contingent -- and the "a posteriori" is necessary!

The "2+2=4" example is merely a less obvious case of the same situation. Can we be sure that there is no "non-Peano arithmetic" in which 2+2 does not equal 4? Mathematical statements are never "true" except in the sense that they follow from certain axioms. Only when we ask, "Does 2+2=4 in reality? If I have two legs, and you have two legs, do the two of us together have four legs?" -- then we may be absolutely sure that it is true.

We thus find that the supposedly certain and "necessary" mathematical statements are actually impossible to establish in an "a priori" context. Indeed, it is not really meaningful to speak of their "truth" in our everyday usage of the word.

Realism

[Notes on Realism with a Human Face, by Hilary Putnam, edited by James Conant. Cambridge, MA: Harvard Univ. Press, 1990.]

Putnam is a respected modern philosopher, a noted eclectic -- or rather, one who has sampled many philosophical variants and changed his mind frequently. Currently he appears most attracted to a Kantian viewpoint.

The editor's Introduction makes some admissions about the state of modern philosophy -- notably the continued dominance of "analytic" philosophy, at least in the U.S. He quotes with some unease Putnam: "Of course philosophical problems are unsolvable."

One interesting point: Kant was the antithesis of the Objectivist in that he saw philosophy as esoteric -- so difficult that only a few professionals could master it.

Putnam begins by discussing paradox, in quantum mechanics and logic. He concludes that QM shows us the only way to eliminate paradox is to make a sharp dichotomy between observer and reality; no paradigm that sees observers as part of the reality they are observing can be consistent (just as no language that trys to evaluate the truth of its own statements can be consistent).

Putnam cites the "Strong Liar": "This statement is either false or meaningless." Here is a paradox that admits of no resolution on its own terms. Tarski's solution was to demand a "metalanguage" to discuss the truth of statements. But this results in an infinite regress of metalanguages, and, as Putnam points out, one gets into difficulties. In what language does one assert the need for metalanguages?

It is interesting to watch Putnam, as a typical philosopher, grapple with this problem. Any normal person would, at this point, have perceived the fundamental silliness of the problem and dropped it. Putnam can't bear to concede that the evident self-contradiction is inescapable. He continues to struggle. "Maybe we can get around it this way. No, that won't work. What if we try this? No, still doesn't work. How about . . ."

All problems of this sort -- linguistic follies, Russell's "set of all set" paradoxes, the "necessary vs. contingent" distinction -- arise because one fails to recognize that there is a category error involved. When we say, "The ice cube intends to go to school," we construct a grammatically correct statement in which all the terms are well-defined and meaningful -- and yet the statement is not "well-constructed". The predicate "intends" is not one that is applicable to the subject "ice cube".

Philosophers continually get into trouble by trying to "pre-qualify" their questions, rather like a salesman trying to ensure he has only qualified leads. They are attracted to "grammar" in the sense that a "grammatical" sentence can be guaranteed to be meaningful -- true or false -- and thus a proper subject for investigation. This is a natural desire; the problem is, a guy named Goedel showed that it simply cannot be done.

Indeed, there increasingly appears to be a fundamental problem involved when one attempts to separate meaning and truth. One might conclude that the universe is, sort of, Cartesian -- we live not in the best of all possible worlds, but in the only possible world. In this view, any statement that is not true about reality would have to be, ultimately, self-contradictory and thus meaningless.

I think, however, that this is a bit too radical. Instead, perhaps we should re-think what we mean when we say a statement is "meaningful". Classically, "meaningful" means "either true or false". But in the Objectivist view, this could apply only to statements about reality which are (in principle at least) testable. (A position generally associated with Logical Positivism!) As we've seen, this definition eliminates all of mathematics and other abstract fields from the domain of the meaningful. Moreover, Goedel shows us that, under this definition, we cannot determine whether a statement is meaningful except by showing that it is true (or false); so the term loses its usefulness.

Should we not return to the inherent meaning of "meaningful"? Does a statement have "meaning" -- can it be used by a human mind as part of the reasoning process? This is a question that can be answered, and answered in a useful way. Analysis of statements for meaning in this sense is possible independent of determination of their truth; and moreover it opens up new and useful questions about the way people think and can learn.

The Strong Liar and other logical paradoxes make the same type of error. "True" and "false" are predicates that refer only to statements about reality. Statements about statements, self-referential or otherwise, have no truth or falsehood. Statements about sets or other mathematical entities, again, have no truth or falsehood. All we can say is that they "follow" logically, or do not follow, from the postulates of the system to which they belong. Logic as an abstract study is valid only so long as it confines itself to statements of the form, "If p then q."

The "necessary vs. contingent" distinction results from the same fundamental error. It attempts to set up a distinction between two kinds of truth -- and make real truth "contingent" or unreliable. In fact, as noted previously, it is the "necessary" that is unreliable -- or, more precisely, not really true at all.

A Theory of Certainty

Let's begin by noting that "certainty", like "knowledge", is personal rather than impersonal. When we use the passive voice -- "it is certain that . . ." -- what we really mean is that many or most people hold it as certain that . . . One person may be certain of something, another not.

Note that in this context we are using "certain" to mean not just "something I am sure of" but "something I am sure of that really is true". We have no interest in false convictions; that would be more a psychological than a philosophical topic.

Now, a useful procedure in general when examining a concept X is to ask, what are the consequences in reality of X or not X? How does X affect our thinking or behavior? Let's apply this to the concept of certainty.

How do we behave about "certain" knowledge? I am certain that 7 X 12 = 84. I am certain that I cannot make gold by mixing mercury and sulfur in just the right proportions. I am certain that I am not a butterfly dreaming that I am Ron Merrill. How does that certainty affect my behavior? It means that I will not consider any challenge to the truth of these statements. They are not debatable. No argument, proof, or evidence that may be offered to contradict them will be accepted, or even considered. Certain means unchallengeable.

Our next step is to ask, what is the source or nature of certainty?

Is something certain because it is proved? But I could not construct a rigorous mathematical proof that 7 X 12 = 84. And if I were shown a proof that 7 X 12 = 85, I would not accept it, even if on going through it I could find nothing wrong with any step. I would instead insist that there was some error too subtle for me to detect.

Neither can certainty derive from empirical evidence. If you mix mercury and sulfur in a bubbling pot, right before my eyes, and at the end of the process pull out a nugget of pure gold, will I say, "Well! Obviously I was wrong, transmutation is possible after all!" No, I will not. I will rather assume that you pulled some slight-of-hand that I failed to detect.

I want to suggest that certainty is, so to speak, not "local". When a statement is certain for a given person, that certainty lies not in the evidence or logical inference directly associated with the statement, but with its status in that person's overall structure of knowledge.

Suppose, for instance, that I did not regard it as certain that I am not a butterfly dreaming that I am Ron Merrill. I would then be in no position to know anything at all. I would be epistemologically wiped out. This is generally true of certain knowledge. When we say, "I am certain of X," the meaning is, "If X is not true, then I know, and can know, nothing."

Those facts which we take as certain are the corollaries, in the personal context of each individual, of the axiom of consciousness. To deny something of which one is certain would be to deny that one exists as a conscious entity.

We may take a statement to be certain either because: (1) the statement itself must be true in order for knowledge to be possible; or (2) the means by which we justified the statement must be reliable in order for knowledge to be possible.

Note: We should develop for objectivism a biocentric (or rather anthropocentric) epistemology. That is, the key question is not, "How can one know?", but "How can humans know?". Note that we are not designed to be "rational animals", we are designed to survive and reproduce. Our senses and brains are optimized to make sense of the world rapidly from insufficient data -- not at all the same thing as acquiring knowledge in the more general sense of the phrase.

Biocentric (or rather Anthropocentric) Epistemology

The subject matter of epistemology, in its most general sense, is "what can be known, and how". But for all practical purposes, the question is "what can we -- human beings -- know, and how". In other words, epistemology needs to more orientation to a biocentric or specifically anthropocentric approach.

This means the following. (1) We ought to ask, what are the means of knowledge available to human beings? (2) We ought to ask, why do humans need knowledge and what do we use it for? (3) We ought to ask, how does being human interfere with our pursuit of knowledge, and how can we deal with such problems?

For evolution did not design us to be rational animals; it designed us to survive and reproduce.

(An analogy may help. Suppose we were considering the best way to program a computer for a certain kind of task, or a robot to explore a hostile planet. Certainly the specific characteristics, limitations, and special features of the machine in question would be taken into account. We must do the same if we wish to maximize our knowledge as human beings.)

So, what knowledge does a higher primate need to survive and reproduce?

First, we use knowledge for survival. We are designed primarily to develop useful information about the physical world in which we live, and about the other people who are the most important features of that world for our survival. Our senses, our concept-formation process, and our natural methods of reasoning are designed to deal with real-world problems and may easily be fooled by abstract, unrealistic questions. Just as "optical illusions" result from the human brain's priority effort to make sense of visual inputs in terms of the expected environment, so much of the "irrationality" exhibited by humans may be due to a similar process on a higher conceptual level.

Second, we use knowledge to guide action. Life is not a game and those who are good sports about losing die without descendants. Reason cannot be separated from emotion, for emotion drives action and action is essential for survival. We are designed to be easily certain, and to need certainty. One does not commit one's life, and the lives of one's family, on the basis of a reasonable probability -- not as a regular state of affairs. The mind would soon crack under the strain. So we tend to "jump to conclusions", to snap to the best available solution to a problem, and commit to it -- and we are right to do so.

We should discuss the epistemolical habits of scientists. The popular idea of the scientist as an impersonal, robotlike investigator, who never commits himself to any idea until it is proved, is quite false. Emotional commitment is necessary to creativity, in science as in art. The idea that emotions play no role in reasoning is a reflection of the mind-body dichotomy. We need to commit ourselves in daily life, as well as science, in order to function. A willingness to make mistakes, to take chances, is an epistemological necessity.

On defying Kuhn

Scientists should follow the lead of Hank Rearden. When he was accused of selfishness, he did not deny the charge; he attacked the premise that selfishness is bad.

We should respond in an analogous way to the accusations of Kuhn and his fellow relativists. We are indeed guilty as charged -- except that our crimes are actually virtues.

Kuhn points out that scientists do not, in reality, accept the facts as final arbiters of truth for their theories. Instead, we give top priority to self-consistency of the theoretical apparatus, to maintaining a consistent worldview. He is absolutely right. This is what we do -- and it is what we should do! (Non-scientists, incidentally, are prone to underestimate the difficulty of getting reliable experimental results.)

Again, Kuhn says that scientists do not rely on inductive reasoning. Their observations are "theory-laden". Right -- and that's as it should be. Induction, and maybe deduction, are old-fashioned models that don't properly describe how people really think and investigate and learn.

Or, to sum it up more briefly: Although Kuhn's model is the currently accepted paradigm in scientific epistemology, that does not mean it is leading us any closer to the truth . . .

THE WHICHNESS OF THE WHY

The Whichness of the Why; or, The Problem of Primaries

There is a game that is played by young children and scientists. When a child learns the use of the word "why", he soon learns the game of asking endless "why"s. To each answer or explanation offered by his exasperated parent, the child asks again, "Why?" Eventually, of course, the answer must be either, "I don't know," or "It just is, that's all."

Now, scientists play this same game, admittedly in a much more serious and sophisticated way. Scientists, whose business it is to deal with these chains of "why", have an informal principle they call "elegance". It states that we must seek to minimize the number, and complexity, of primaries (a restatement, of course, of Occam's Razor). It might be paraphrased, "The explainability of the universe is a maximum." And science has repeatedly been able to reduce highly complex phenomena to outcomes of very simple laws. We therefore have a need to know where the game will end.

Science is the systematic study of causality. As Aristotle points out, causality is the source of answers to the question "why?"--and he presumed that there were four types of causes, corresponding to the different ways in which the question "why?" can be answered.

Causality is axiomatic. We are constantly faced with the need to identify connections in the behavior of entities that are separated either by space or by time. Thus you and I are different entities, and what I do may affect you--for instance, based on what I say, you may decide to study philosophy. But also, if I say, "I am a different person from what I was 30 years ago when I was in college," you will find this quite coherent. And you can easily believe that something I did then--such as choosing to major in chemistry--affects the way I am now.

The first of the two final answers given above is, so to speak, the epistemological terminus of the game. For the scientist, "I don't know" doesn't actually end the game; to find out what we don't know is our job. But we admit that there are certain things we can't know, for practical reasons.

More fundamental, and thus more interesting, is the metaphysical terminus of the game: "There is no 'why'; it just is, that's all." Does the chain of explanations have an end -- a statement that explains other truths, but has no explanation itself? We may call such a statement a "primary". Are there primaries? And if so, what are they?

It is important to understand that the problem of primaries is not equivalent to the issue of Cartesian rationalism. Some philosophers, such as Descartes, have attempted to derive all knowledge by deduction from a few basic principles or axioms. It is by no means clear that any such program can be carried out. But the question of induction vs. deduction is not the issue we are addressing in the problem of primaries. Here we are asking: Are there any statements about reality about which it is not meaningful to ask "why?" The issue is not how we know that the proposition is true (whether by deduction, induction, or any other means); we may take it as given that it is true. But we wish to know why it is true--which is not the same thing as how it is known (though the issues may be related).

Before tackling this issue, it will be instructive to ask a couple of "meta-questions" on the subject. What answers to the question of primaries could there be? What answers would we consider satisfactory?

To begin with, either primaries do not exist, or they do.

If primaries do not exist, there are only two possibilities. There could be an infinite chain of explanations. Every "why", we might claim, has another "why" on its back, like the famous fleas, ad infinitum. This possibility was considered, and rejected, by Aristotle. His refusal to accept an infinite chain of explanation does not carry so much conviction to the modern mind, however. We are more used to the concept of infinity, and might feel no more uncomfortable with saying there is no First Cause than with saying there is no largest number. A stronger objection to an infinite chain of explanation is empirical, not a priori. Our experience shows that each level of explanation is enormously more powerful than the lower level. One cannot imagine going very many steps before coming to a single, all-encompassing explanation.

A second possibility is that the chain of explanation is circular. The "anthropic principle" is an important recent example of this solution to the problem of primaries. The "anthropic principle" states that there is only one possible set of laws of physics (and boundary conditions) that allows for the existence of intelligent life. Thus our world, in which life is possible and exists, is explained ultimately by the laws of physics, and the laws of physics are explained by the existence of our world and the intelligence it makes possible. We may of course indignantly retort that a circular explanation is no explanation at all. But is a circular explanation really any worse than an infinite chain of explanations? Or explanations ultimately dependent on arbitrary postulates?

I begin to think that the solution lies in the notion of transcendent causality. There comes a point in the causal chain in which a new kind of causality comes into play. Call it the "Immovable Mover" as Aristotle did, or the "Star Maker" as Stapleton did, or something else if you don't like the religious connotations. But when we talk about the "cause" of the laws of physics being the way they are, are we not using "cause" in a new sense? (Rand, of course, would say there is no such cause.) I would characterize this transcendent causality as being related to common or garden causality in a way analogous to the way infinite numbers are related to finite numbers.

After all, the discoveries of modern physics, particularly in quantum mechanics, show that we must either abandon the idea of universal causality, or develop an expanded notion of causality. It is noteworthy that, though QM allows causality violations, these are always organized such that they can never be used to violate free will. As in the Baseball Diamond Experiment, a QM entity can predict my actions--but that prediction can never be used, by any other person or by myself, to create a contradiction or compel me to act in a certain way.

Well, then, what if primaries do exist? Again, there are two possibilities. One is that the chain of explanation might terminate with a highest level of arbitrary principles. That is, we might have to accept some small number of postulates as true but unexplainable -- they "just are", for no reason. This has some problems also. For one thing, it is not clear that under such a system anything would be truly explained; every explanation would ultimately reduce to "it just is, that's all". Primaries are a special class -- the First Cause, or God, or perhaps Plato's forms. Essentially, this adopts the counsel of despair. We concede that a "why" is needed, but that it is not possible to produce any "why" for the primaries. For if we adopt the Aristotelian or Randian principle that "the what of an existent is its why"--how is this any less mystical than the postulation of a First Cause?

But if primaries are not arbitrary, what are they? To be non-arbitrary, a primary would have to have an explanation -- but by definition, a primary does not have an explanation. Is there any way out? Just one: We might assert that a primary explains itself -- that is, that it is an axiom in the Aristotelean or Objectivist sense. This is more or less the approach of Descartes. Some modern physicists are also attracted to it. The idea is that given "A is A" and other similar axioms, all else about the universe follows. There is only one non-self-contradictory set of laws of physics (and boundary conditions) and hence only one non-self-contradictory universe.

The axiomatic principle depends on what you will consider axioms. Look at Rand's metaphysical axioms. "Existence exists." As we now know, the requirement of a stable, objective, reality puts severe constraints on possible laws of physics. So do "A is A" and "Non-contradiction." But not enough, it would seem, to really pin things down. However, Rand also, as I recall, took consciousness as axiomatic. If -- and it's a big if -- we regard the existence (or at least possibility) of consciousness as an axiom, then the Anthropic Principle cuts in and the laws of physics are very tightly circumscribed. But can we demonstrate that the existence of consciousness is, so to speak, objectively axiomatic rather than subjectively axiomatic -- or, it might be better to say, metaphysically rather than epistemologically axiomatic? Was the "Axiom of Consciousness" an axiom before there were any humans or other conscious beings?

There is indeed something attractive about this solution -- that the chain of explanation might terminate with a set of axioms, that is, statements which are "self-explanatory". But there are three problems. To begin with, if we characterize axioms as "self-explanatory" we need to tighten up the terminology a bit. More important, it is by no means obvious that we live in the only conceivable universe. It is true that modern discoveries show that reality is far more tightly constrained by logic and mathematics than was previously appreciated. Still, it is hard to see how logic could force, say, the speed of light to have only one value out of an infinite number of possibilities. It seems we are likely to need at least one arbitrary postulate, such as Merrill's Conjecture ("We live in the most interesting of all possible universes."). Finally, to reduce the problem of primaries to "self-explanatory" axioms is, after all, a resort to circular explanation -- it's just a very small circle!

Let's get back to real-life questions. What does "explanation" mean? In science, as in practical life, we connect the ability to explain things to the ability to make predictions about them. Pure induction doesn't work; to really predict the sunrise, we must understand why the sun appears to rise.

But when scientists talk about "prediction" we do not mean the term in the sense of fortune-telling. We may make "predictions" about the past (what would happen to a species of dinosaur under certain environmental conditions, eg), and may even make probabilistic "predictions" (as we do all the time in quantum mechanics).

By "prediction", the scientist means the ability to say, "A system S will exhibit phenomenon P." For instance: "If a 22.4-liter container holds one mole of hydrogen at 25 degrees the pressure inside will be one atmosphere." These statements, of the form "S implies P", are a consequence of the rationality of the universe. Existence does not just exist; it exists in accord with reason, and every phenomenon of existence has a "reason", ie, a cause.

So, in a scientific context, we might state the Axiom of Causality as follows: "There exist a set of laws, L, such that for every possible system S in the universe, L will specify what phenomena P can occur in the system." (Note that we say "can", not "will", on the assumption that the universe is non-deterministic.)

A digression on quantum mechanics: QM phenomena have bothered many scientist precisely because they are, or at least appear to be, stochastic, and thus violating causality. I wish to suggest that this is merely an artifact which appears because we impose macroscopic measurements on entities which do not possess the characteristics we are trying to measure.

Imagine a world in which physicists study billiard balls, and in which only black balls and white balls exist. It is taken for granted that all balls are either black or white. Now, to investigate phenomena which to them are microscopic -- let us call it the "Q-world" -- they adapt their standard measuring apparatus, a photocomputer which determines whether a ball is black or white. Unfortunately, in the Q-world, all balls are black on one hemisphere and white on the other. The apparatus cannot deal with this; when observing a "Q-ball", it locks onto either the black or the white setting, depending on which face of the Q-ball happens to be most facing it. The physicists thus find that Q-balls exhibit a stochastic behavior! (Note that this is not a "hidden variable" theory.)

In the same way, we find elementary particles exhibiting stochastic behavior because we insist on assuming that they have properties such as position.

Note also that we assert that causality is metaphysical, not epistemological; the rationality of the universe is a reality, not an illusion imposed on it by the structure of our minds.

The value of this approach is that it suggests that we can dodge the "chain of explanations" argument entirely. The laws L are in a different category from phenomena P; the Axiom of Causality (but can we demonstrate that, as stated, this is an axiom?!) requires that every P be explainable in terms of L, but imposes no requirement that L be explainable; indeed, in these terms, "explanation" does not even apply to L. What's more, S need not be explainable; thus parameters such as the mass of the universe, and possibly even the speed of light, are excluded from causal requirements.

We therefore might adopt the position that the Laws of Physics, L -- God's Equation List, so to speak -- and the associated boundary conditions S, are primaries of the arbitrary sort.

The axiom of causality says that one draws the line at entities in general. Events have causes; it is meaningful to ask, "why did it happen?", and all such questions have an answer. But entities do not have causes (or are their own causes). There is no "why" of an entity; it just is, that's all. It is irreducible, unexplainable, a primary. (But is this really true? Why does this computer exist? Somebody made it. Why does, let us say, a particular photon coming from the computer screen exist? Because an electron hit an atom under certain conditions.)

There are various approaches here. For instance, religious people say, in essence, that we need a special type of entity, God, to serve as the "First Cause". In general, we need to ask, what is special about the class of irreducibles or "primaries" -- what makes them, unlike other phenomena of the universe, immune to "why?"

The important question arises when we are dealing with things which are neither entities nor events. The laws of physics, or physical constants such as the speed of light -- do these require explanation? Here science departs from Aristotle. Science provides explanation for the existence of entities (eg, the earth or the sun) as well as events. But can science supply an explanation -- a cause -- for its own laws? Can anything?

What do we mean when we say something can be explained? What sorts of things can be explained? We accept that events can be explained. But note the nature of the explanation. If we ask why this billiard ball is moving at 1 meter/second, we explain it by saying that it has just been hit by another billiard ball, and it is the nature of billiard balls to . . . and so on. So each event is explained by reference to:

1. a prior event

2. the entities involved

3. the rules which define how entities behave under

certain circumstances.

Now, contrary to Aristotle, entities can be explained in the same way, for an entity (as Aristotle conceived it) is simply a configuration of matter resultant from past events. Thus our billiard ball is explained by the events which manufactured it and the raw materials from which it was made, in combination with the rules defining how those raw materials behave under the relevant processing conditions, and so on.

The problem is: Is there a First Rule? The enterprise of science has been to reduce explanatory rules to a minimum. But what is this minimum? And if we can find a First Rule, is it arbitrary, or axiomatic?

Taking another look: We may regard science, or common sense, as involving a hierarchy of explanation. The phenomena we directly observe are explained by "first-level explanations" (eg, the fall of objects is explained by the existence of gravity). The first-order explanations in turn have second-order explanations (eg, gravity must obey an inverse-square law because there are three spatial dimensions and energy is conserved). And so on. Now: where will it end? There are just four possibilities.

First, explanations might be circular. This is clearly unsatisfactory. [Or is it? An axiom is "self-explaining" -- and this is a circular argument, it just involves a very small circle!]

Second, there might be an infinite number of levels of explanation. I don't necessarily subscribe to Aristotle's immediate rejection of this, but certainly it raises problems. And, empirically, we find that each level of explanation is much simpler than the level below it, so that an infinite chain of explanation looks unlikely.

A better way to say this is that an explanation (ie, a cause) is always more general than that which it explains. The chain of causality does not lead deeper and deeper into specifics of the entity. On the contrary, it always branches and expands. The cause always explains more than the effect which we originally wanted to explain. This is true of the formal cause, the effective cause, the final cause, and the material cause.

Third, the chain of explanation might terminate with a highest level of arbitrary principles. That is, we might have to accept some small number of postulates as true but unexplainable -- they "just are", for no reason. This has some problems also. For one thing, it is not clear that under such a system anything would be truly explained; every explanation would ultimately reduce to an "it just is, that's all".

Now, could it be that all this is a problem only because I am carelessly accepting an Aristotelean rather than Objectivist epistemological principle? Rand says "essence" is epistemological, not metaphysical as in Aristotle's analysis. This could be taken to mean that the "why" of an existent is not in the existent itself, but arises only when we try to explain it. That is, the "why" is an aspect of our conceptualization of the existent, not an intrinsic property of the existent. But at this point we would seem to be getting perilously close to some sort of Kantian "category".

Notes on Adler's Intellect

(Mortimer J. Adler, Intellect: Mind over Matter. New York: Macmillan, 1990.)

Definitely not up to Adler's usual standard. From stupid errors (claiming modern computers contain only about 10 E 3 transistors) to general ignorance (insisting on the old-style Heisenberg interpretation), the book is full of very weak material.

Adler's big problem is that, like Aristotle, he wants to insist on the immortality of the human soul. The only way to do it is to accept, at least within the limited domain of psychology, Plato's forms. The result is a lot of fuzzy text about the "problem of universals".

Adler wants to show that intellect is real -- an existent -- but not material. His argument is that our concepts are "universal" -- Platonic forms more or less -- but our brains are not, they are particulars. This is somehow supposed to prove that our intellects therefore cannot be solely functions of our brains.

What is of interest here is not "the problem of universals" but what might be called "the problem of connectedness". Let's begin by asking how we wish to use the term "exist". We say that an apple exists -- it is made of matter, it is part of reality. The apple is an existent; it has attributes, such as being red and occupying a certain position in space. The attributes are part of reality, but are not matter and not "existents" in the same sense the apple is. Now, another entity, say a watch, may exhibit more complex behavior than an apple; behavior is just a different way of looking at attributes. For instance, when we say an apple is red we mean that when white light shines upon it the apple behaves in a certain manner -- it reflects all except the red frequencies, which it absorbs. A computer exhibits still more complex behavior than a watch.

A human's behavior is still more complex, and indeed has crossed the threshhold into volitional behavior -- ie, the output cannot be predicted from the input, even in principle. This behavior is an attribute of the human brain, and another way of referring to the attribute is to call it intellect.

So much for Adler's problem. But note that science relies on this kind of distinction, though it is phrased in different ways:

entity has attributes which reflect its interaction with its environment

system has states which it moves between in the process of interacting with the surroundings

device contains algorithms which control the relationship between input and output

The question is, to what extent can we ensure that the boundary (to use a term from thermodynamics) is sufficiently controlled and well-defined that we can reason about the system's behavior? There is no such thing as an isolated system; everything in the universe is connected to everything else. Only thing is, it is connected so that interactions may be made too slow to interfere with the phenomena we wish to observe (by using the speed of light limitation).

Adler, like Rand, has a problem with the "indeterminacy" of quantum-mechanical phenomena. To my mind, one cannot escape paradoxes (such as the violations of causality and volition implied by the baseball-diamond thought experiment) unless one is willing to jettison relativity (and the results of the actual EPR experiment suggest we should do so) and say that wave functions really do extend to infinity.

In this view, the indeterminacy is a mathematical consequence. Note that in, say, calculus, every problem has a determinate answer; or (as in the case of roots of a quadratic) a finite number of discrete, possible answers; except when we encounter a singularity. Once an infinity enters into the calculation, we end up with a stochastic situation; an infinite number of possible (though not necessarily equally likely) solutions. The same thing is true in physics! As long as there are no singularities, we have simple deterministic systems. But if the system has one or more singularities, there will always be the potential for stochastic behavior. This is true even in "Newtonian" mechanics; if we have perfectly hard billiard balls, we can only statistically predict the results of their collisions.

Incidentally, it is interesting to consider whether it is possible to construct a physics (at least, one which even vaguely resembles the real world and is also non-trivial) in which singularities play no part.

Another interesting side-light: Aristotle may have been right about the impossibility of a vacuum. Consider Dirac's idea of a universe of negative energy levels completely filled with particles. Modern physics has a very peculiar conception of the term "vacuum".

VOLITION AND PHYSICAL REALITY

THE TABOO SUBJECT: REPRODUCTION AND THE OBJECTIVIST ETHICS

THE IRRATIONAL ANIMAL

On Evaluating Morality

Kelley brings up the issue of evaluating the morality of an action. As he points out, the historical controversy has been over whether an action should be evaluated on the actor's intentions, or on the action's results. Let's look into this a little more deeply.

First of all, note that the problem exists only if we are doing this evaluation retrospectively. If the action has not yet been taken, then the problem reduces to the normal query of ethics, "what should I do?" For this query, intention must be the decisive factor (though the intention to take a risk may have to be taken into account). Prior to action, one cannot be certain what the result of the action will be. One cannot ask the question of the result vs. the intention, because the result does not yet exist. So moral evaluation comes down to: (1) what is intended?, and (2) how skillfully is it intended?

Note that objectivism emphasizes skillfulness as a moral issue -- you are reponsible not only for wanting and trying to do good, but for learning how to do good. If your action produces unintended bad consequences because you are naive or ignorant, and you had the choice not to be naive or ignorant, you are morally responsible. The classical dilemma presents a naive actor; the dilemma exists only because of the failure to ask, "why was he naive and what are the moral implications?" If you accept that skillfulness is part of morality -- indeed, the essence of morality -- the dilemma goes away.

Let's turn to the retrospective evaluation of the morality of an action. At this point we must note that such evaluation may be either internal or external -- that is, by the actor or by some other person.

The internal case presents nothing new; it is simply the previous situation viewed from a later time frame. If one's past action, while intended to do good, actually produced evil, one must ask whether one committed a moral fault in acting on insufficient knowledge.

The case that Kelley is really concerned with is the evaluation of the action(s) of another person from a retrospective position. Following Rand's principle that "why" suggests "how", let's ask: Why does one need to evaluate the morality of other people's actions? What does it accomplish?

The first thing to realize is that it matters only from a predictive point of view. The only practical, that is, moral, value of such an evaluation is that it may allow us to more reliably predict how a person, whose actions may affect us, will act in the future. A mental condemnation of past actions accomplishes nothing, in itself, toward maximizing my life. The only thing of use is to get some idea of how a person will treat me. (Exception: Evaluating irrelevant past actions -- eg, the atrocities of Hitler, which ended before I was born -- is worthwhile as a means of training or educating the moral sense, of improving my skill at evaluating people. Thus moral evaluation may be of direct or indirect value.)

But if our purpose is to predict the actions of the people who affect us, why do we do so on the basis of moral evaluation, rather than simple statistical analysis or psychological examination? Here we have two possible answers.

First, we may claim that moral evaluation gives us a better insight into the person than, say, psychoanalysis, and thus allows us to make more reliable predictions. Note, for instance, that Freud found it necessary to "import" moral evaluation into his psychological theory, in the form of the "superego", in order to make it more predictive.

Second, we may say that, because men are social animals, moral evaluation has not only a predictive but a proscriptive value. If we can convince others to agree with our moral evaluation, we can enlist their aid in dealing with the malefactor. Of course, the same is true to some extent with psychological or statistical evaluation. But morality may claim to fill a more natural or fundamental role in social interactions, and certainly in political interactions.

However we regard it, we must realize that what is really important is the evaluation of persons rather than the evaluation of actions. Evaluating an action as immoral is strictly a means to the end of evaluating the moral stature of the actor.

Now, what about intent vs. result in this context? We immediately see that the question is not properly posed. As in the internal case, it is both intent and skill -- that is, efficacy and efficiency -- that must be taken into account. Lack of either is a moral failing.

But there is something more here. Judging the moral stature of actions and people is a universally valuable skill! Each of us has a moral obligation to develop this skill, in our own self-interest. But it is very difficult to do. You must know a lot about human nature to properly evaluate such complex questions. So this reduces to a special case of the more general issue: How ought one to proceed when one must make important, even crucial decisions, on the basis of limited knowledge or skill? Once again, ethics reduces to epistemology.

Sexual Ethics

Suppose we take up the question of sexual ethics -- that is, ethical questions involving sexual actions or relationships. (Another context -- the issue of how ethical prescriptions differ for men and women -- is also important.)

First, let us invoke the biocentric principle. The most obvious starting point is that sex plays an extraordinarily large role in life for members of our species. Indeed, there is probably no other animal species that copulates as often, or even comes close to it. What's more, sexual considerations permeate much of our behavior, including areas that are by no means directly relevant to sex. For most animals, sex is a rare occurrence and something outside of, and disconnected from, their daily lives. Finally, for most animals -- outside the higher primates, at least -- sex serves a single function, reproduction. In our species, sex contributes to our life in many ways.

Let us next apply the principle of rationality. This might be phrased as, make your action as intentional as possible. That is, be aware of what you are doing, try to understand why you do it, and make conscious choices rather than just drifting.

The principle of productiveness suggests that we ought to expect -- and to welcome -- that the benefits of sex have to be produced through effort. Thus, in a sexual relationship as in a business relationship, it is wrong to be a parasite who consumes the goods produced by others.

The principle of pride brings out a very important point -- that in sexual activities of various sorts, competence cannot be taken for granted.

Objectivism and Feminism

The issue of "feminism" (a term which its proponents carefully leave undefined) requires us to consider this question: To what extent (if any) and how, should political and social arrangements treat men and women differently? There are three principles of Objectivism which are relevant here.

First, Objectivist ethics is biocentric and contextual. How one ought to act as a man or as a woman, and how one ought to behave toward members of the male and female sexes, depends on the facts of the nature of male and female human beings.

Second, Objectivism rejects the mind-body dichotomy. The plasticity of the human mind is quite high. This does not mean that human software is hardware-independent. Part of becoming the most that one can is adapting and shaping one's mind to fit the characteristics -- including the sex -- of one's body. Thus data indicating that males and females can be very similar is irrelevant.

Third, Objectivism takes an Aristotelean, not a Hobbesian, view of humankind. The reflection of ethics in a social context is custom. Thus, again, the fact that individual males and females have the option of being effectively neuter is irrelevant. The question is, how ought rational people to organize society so as to create an environment suitable for the maximization of human life? [This is worth an essay in itself, and the principle is relevant to environmentalism and what might be called "racialism", as well as feminism. "The New Totalitarians" would be a good title.]

ON MORAL DILEMMAS

On Moral Dilemmas

John F. Kennedy said that his big problem in the presidency was not choosing between right and wrong but choosing between right and right. One need not be especially cynical to dismiss this; obviously his real dilemma was between doing right and gaining votes. However, it does stimulate an interesting question: What is the nature of moral dilemmas? We can classify them into several types.

Before starting, we should exclude from consideration -- though with great care -- the "roc's egg" dilemmas. We cannot require of morality that it settle an issue based on a contradiction of the basis of the morality. (Eg, "emergency ethics" for Objectivism)

First, there is the dilemma of temptation. We know what is right, but do not want to do it.

Second comes the dilemma of insufficient data. We would know what is right if we had all the important information about the situation -- but we don't. Note that some moral codes are codes of "intention" and inherently immune to this dilemma. Others are codes of "outcome" and face it in heightened form. Still others are codes of "action". Objectivism is primarily a code of "being", that is, of becoming the maximum of one's potential; this is the only approach which truly addresses this type of moral dilemma.

Third, we may face the dilemma of contradiction. If our moral code is inconsistent, it may require conflicting courses of action in some situations. Here we may wish to distinguish cases where the moral code is not self-consistent, and cases where it is inconsistent with reality, ie, it requires one to do the impossible or upholds the desirability of certain effects while prescribing actions which will prevent them.

Fourth, there is the dilemma of incompleteness. We may encounter situations for which our moral code simply has no prescription. We may break this down further into situations where (1) the code is complete, but the actor does not fully know it; (2) the code is incomplete, but a prescription for the situation in question can be derived (at least in principle) from the basic principles of the code; (3) the principles of the code are incomplete. [Note that, if we are really rigorous, no non-trivial moral code can be complete, according to Goedel's Theorem.]

When we hear that "hard cases make bad law", we know that the moral basis of the speaker is flawed. Hard cases -- regardless of the type of dilemma, even temptation -- indicate a deficiency in the moral code.

THE PROBLEM OF FORCE

On Welfare "Rights"

It is argued even by many who would characterize themselves as basically libertarian that the truly disadvantaged have a right to be supported by others. Often this is formulated as an appeal to "decency" or "compassion".

Let us consider what is really involved here. The situation is described as one in which we have a small number of blamelessly, helplessly, hopelessly destitute people in the midst of a large, rich society. Surely it is permissible that the State extract by force a tiny portion of the "surplus" of the productive to support the existence of those who cannot support themselves?

But who are the ones who need this support? They are the ones who are not supported by private, voluntary charitable activities. With all the many well-off people in society, not one can be found to contribute a little to preserve these human lives. What kind of "needy" would evoke no pity? Obvious answer: those who are evil, by the standards of the productive.

The situation as presented in the hypothesis contains an inherent contradiction. A productive, rich society will not be made up of people who cheerfully watch innocent people starve. That does not make sense theoretically, and it does not happen empirically. The ones who are allowed to starve are the criminals.

The important point here is that welfare rights are supposed to be grounded in the moral views of the population. The argument is that rights are implied by people's morality, and therefore they cannot, so to speak, pick and choose; having willed liberty on the basis of morality, they must also will welfare. But the hypothesis of the argument can only be consistent if it is assumed that the "needy" who receive no voluntary help are evaluated as evil by the population in question.

THE NEW TOTALITARIANS

A good example of the way communism has become more efficient by becoming less virulent is its handling of border control. For years, communist countries confined their citizens with Iron Curtains. In the Seventies, they became more sophisticated. They allowed a certain amount of emigration as a safety valve. Moreover, when reproached by Western countries, they released (as in the case of Cuba) or threatened to release (as in the case of China) large, indigestible hordes of refugees.

While we congratulate ourselves on this reduction of virulence in old ideologies, we must be aware that the new totalitarian ideologies, because they are new, will be highly virulent if they ever come to power.

PLANNING THE SECOND CRUSADE

The Radicalism of Ayn Rand

Why has there always been such a viciously antagonistic reaction to Ayn Rand? Her advocacy of laissez faire? But von Mises or Milton Friedman or even Murray Rothbard induces no such response. Her egoism? But other egoistic philosophers do not provoke such a violent reaction. And when we look at where the objections focus, we find it at her metaethics: She says that she has a logically defensible ethics.

In Galt's speech, we find: "There are two sides to every question: the right side, and the wrong side." That is the radicalism of Ayn Rand. That is what academics, in particular, find intolerable in her philosophy.

Consider a well-known fact about lawyers. Two attorneys on opposing sides of a case vilify each other and their clients without mercy. Then, when court has adjourned for the day, the two "opponents" leave arm-in-arm for their golf date. Whatever their opposition during litigation -- and it can on occasion be real and personal -- lawyers are united in their exploitation of their clients and the control of the legal system which makes it possible.

In the same way, academics may cut one another up in print or at a panel discussion. One advocates reason, freedom, and capitalism; another mysticism and totalitarianism. But at the faculty tea, one finds them on perfectly good terms. Whatever divisions they may have, they are united against their students.

Nathaniel Branden's instinct was right when he set up NBI. The only effective way to spread objectivist ideas is to avoid the academic mainstream, to go over the heads of the professors and appeal directly to those who seek the guidance of philosophy in their lives. Partly this is a matter of sanction of the victim. It is also part of the more general principle of alternative education. If we ask what we can do for objectivism, a big part of the answer is alternative education. Not "educational reform"; it is much too late for that. Neither the public schools nor the universities are salvageable at this late date. We must build a new educational system from scratch. The Montessori system is an excellent model for pre-school and the early grades. But we need to invent models for high school and university-level education.

Another item: We should be making a special effort to write objectivist fiction for children.

***

The radicalism of objectivism has implications for our struggle. For one thing, it means we cannot "make converts" efficiently among the adult population. The example of Kingsley in The Black Cloud is relevant here. Our only real recruiting ground is the colleges, and -- for the brightest students -- the high schools. Beyond that, we may wish to look for people who are in philosophical crisis.

Our radicalism also means we should forget about reform of education. We must build our own educational system from scratch. And this is true of other institutions too. We must build our own social, business, and political organizations along new lines. The rule is: Do not reform -- bypass. Do not argue -- explain.

Another consequence of radicalism is the need for social support. No radically new philosophy in history has ever sent its adherents out into the world to live in isolation in a hostile society. Institutions are desperately needed.

Another consequence, as mentioned in IAR -- we must renounce our desire for respectability. Radicals are not respectable.

The Concept of Sense of Life (Comments on Will Wilkinson's Essay)

"Metaphysics" and the Definition of Sense of Life

Rand's phrase "metaphysical value judgment" is rather infelicitous.

(Let's note in passing that Objectivist metaphysics is not trivial, merely undeveloped. It need not be "dreary" and there is plenty of "fun conjecture" available--for instance, Rand asserts that quantum-mechanical tunneling cannot occur, and that a vacuum cannot exist, on metaphysical grounds.)

Let me suggest that what Rand is trying to get at here is this: Each person forms [ethical] value judgments starting in early childhood. These are of the form: "X is good (or bad) for me." At the same time, each person forms metaphysical value judgments, which have the form: "X is possible/impossible (or likely/unlikely, or easy/difficult, etc.)" These latter judgments are "metaphysical" in the sense that they refer to what is, or is possible, in reality. They are value judgments because they deal with facts that are of relevance to the good for the individual.

Illustration: When I think or feel: "Success in business is good (for me)," that is a value judgment. When I think or feel: "Success in business is possible (for me)," that is a metaphysical value judgment. When I think or feel: "The existence of black holes in distant galaxies is possible," that is a metaphysical judgment, but not a metaphysical *value* judgment, because it has no relevance to my well-being.

Now, I think Rand used the phrase "metaphysical value judgment" to bring out the point that these judgments, like [ethical] value judgments, have emotional impact. When we make metaphysical value judgments, we feel an emotional response, because we are thereby estimating our chances of achieving our values.

Sense of Life as an Emotion

Rand was very much aware that our emotions reflect a view of facts of reality, but that they may not be congruent to the facts as determined by our reason.

(Note: She recognized that in some cases the

emotion could be right, and the conscious analysis wrong. For instance, see the discussion in *Atlas Shrugged* between Dagny and Reardon regarding her visit to Dr.

Robert Stadler.)

Thus our [ethical] value judgments on the emotional level may not agree with our conscious ideas. For instance, a smoker may consciously know that cigarettes are bad for him, yet still, emotionally, value smoking. The same is true of the metaphysical value judgments that underlie one's sense of life. Thus a person may be consciously convinced that he is competent to succeed in business, but emotionally he may have the feeling that success is not possible.

The key personal objective for the Objectivist is to bring emotions and reason into congruence. This must be done with regard to both ethical and metaphysical value judgments. But a prerequisite for this is having a clear and accurate understanding, on the conscious level, of what one's emotional responses *are*. Rand repeatedly emphasized the importance of knowing what one feels, and why one feels it. I would say that for her a great part of the value of art is that it can facilitate this understanding.

The Formation of Sense of Life

Rand describes formation of a sense of life as a "a subconscious counterpart of a process of abstraction," in which units are classified by emotional (rather than conceptual) common denominators. Thus we unify "things that make me feel ecstatic" or "things that make me feel optimistic" or "things that make me feel sad." This classification scheme results in a sense of life because it assigns "importance" (ie, emotional consequence) to facts of reality.