THE ART OF MAKING SENSE
A Primer on Reason and Critical Thinking

Table of Contents

Part One: The Word
Chapter 1 - Logic Isn't Everything-But It Helps
Chapter 2- Words and Ambiguity
Chapter 3- Define Your Terms
Chapter 4- What Kind of Language Are You Using?

Part Two: The Argument
Chapter 5- How Not to Argue
Chapter 6 - Putting Up a Logical argument
Chapter 7 - Some Patterns of Reasoning

Part Three: Truth and Falsity
Chapter 8- Truth and Evidence
Chapter 9- Knowing the Causes of Things
Chapter 10- Are All Generalizations False?
Chapter 11- On Matters of Taste and Opinion

Glossary

---

PART ONE
The Word

CHAPTER 1
Logic Isn't Everything-But It Helps

Let us begin with a few disclaimers. Study of logic is not going to be a panacea. It won't make you a better lover, it won't improve your digestion, nor will it be much comfort when you wreck your car or lose your job. But as an effective weapon in what James Thurber called "the end-less war between meaning and gobbledegook," or as a guide when you lose your way in the morass of doubletalk, conflicting claims, and verbal diarrhea that meets us almost daily in our society, logic is unbeatable.

What is logic and what isn't it?
Logic is correct reasoning. More precisely, logic is the process by which statements are supported with adequate proof by being tested against the right amount and kind of evidence, the process by which knowledge is rendered reliable-in short, the "science of proof." It follows then, that to be logical is to argue reasonably. What the logical man insists on is simply this: if you claim that you have proved a point-about anything at all-then your conclusion must be examined in terms of the adequacy and reliability of your evidence. Logic shows us what kinds of tests to apply in each case and how to apply them. In other words, if you claim that Jones is a thief, or that blondes are inferior in intelligence to redheads, or that the testing of nuclear weapons leads to destruction of the environment, logic provides ways for judging the "truth" of your assertion and tile reliability of your evidence.

Logic encourages taking a somewhat skeptical attitude toward what-ever you cannot verify directly by your own experience and in this regard it is akin to the skepticism implied in phrases ranging from the man from Missouri who challenges "Show me," to the simple "Oh yeah?" Obviously this approach is not useful in all situations and the logician acknowledges that many statements cannot be or do not need to be approached logically. In fact, all human discourse can be divided into two categories the logical and the nonlogical When someone says "I'm getting fat. I must eat less," he is using a well known form of argument (a unit of discourse that purports to prove something) called a syllogism lie is claiming a relationship between his weight and the 'amount of food he eats Such a claim can be subjected to certain tests, proof can be called for and examined, and an objective judgment made about its truth. Note that we are not using argument in its popular sense, that is, a dispute or disagreement. Rather, "argument" in the language of logic is discourse that contains at least two asserted statements and the claim that one statement ought to be believed because another is true. Logic is helpful then in handling certain types of information and resolving certain kinds of questions. It is appropriate when we deal with patterns of proof or disproof, claims and counterclaims.

On the other hand, statements like "I like to travel" or "I love you" are ordinarily regarded as nonargumentative. There can be no useful debate, in other words, about whether you like to travel the proof such as it is, is solely in your own head. Such statements are regarded by the logician as "nonlogical" and do not require supporting evidence. (Note that "nonlogical" does not mean illogical; an illogical statement is one which violates the rules of sound reasoning-it's like adding one and one incorrectly.)

Do we really want or need logic?
Man used logic long before he was defined by Aristotle as "the reasoning animal." He has continued to develop and refine the rules of logic not as an intellectual exercise, but as a means of getting at the truth. It must be admitted at this point, however, that logic is often disregarded when it could be most helpful in conducting man's business. There are a number of possible causes for this disinclination to make logic part of our lives. In the first ace, many persons think of logic as a branch of philosophy, which everyone knows is an academic pursuit far removed from life and living. According to this point of view, there is no more reason for the common man to study logic than to study abstract art or nuclear physics. In an age when swamis, parapsychologists, back-to-nature enthusiasts, sensitivity group followers, and mystics of various stripes advocate their respective paths to "knowledge," it is not surprising that logic, which requires discipline and eschews emotion, should be brushed aside or ignored. Nor does it help the case for logic to note that logical thinking can lead to unexpected and uncomfortable ends. By following the rules of logic we are often led to unpleasant conclusions, and are sometimes confronted with our own prejudices and opinions masking as "facts."

What are the advantages of using logic?
The point to remember is that logic is not a straitjacket or a religion. It is a tool. Without logic you would have available to you no knowledge other than your own experiences. Most of what you have learned or been taught was compiled through the process of logic. No one person can personally verify everything: the number of employed women college graduates in the United States, the correlation between crime and poverty, the major parties' popular and electoral votes in the last presidential election. Thus, when a sociology textbook says, "The lessons of psychology induce a change in people's behavior when they grasp them," the author is basing his conclusion on data, fact, and conclusions that other people supplied him. And you, in turn, use his conclusions to arrive at your own.

Since logic gives you a method for assessing the truth of many of the statements that other people make to you (whether in print, in person, or oil television or radio), it is perhaps the most relevant study-and the most basic one-that you will ever engage in. It is eminently practical in that its whole focus is on solving problems and arriving at conclusions. In helping you distinguish between facts and inferences, it will enable you to arrive at a sound basis for judgment and (if necessary) for action.

We are inundated daily by claims of all kinds, many of them contradictory. Opposing politicians state conflicting "facts"; advertisers make soaring claims for their products. Some claims are insidious in their subtle disregard of logic. Consider the following letter, recently published in a California newspaper:

Editor:
I wonder how many people, especially women, realize that if the Equal Rights Amendment is put into law the American woman may look forward to absolute horror While Russia and Red China favor treating their women as animals I shudder to think of what will happen to our women Frankly I prefer our girls to stay as sweet as they are.

A careful reader (and thinker) will ask himself how such words as "horror," "animals," and "sweet" are used he will determine what claim or argument is being asserted by the writer of the letter and he will determine how this claim or argument can be proven or verified. To be able to distinguish fact from fiction is more than an abstract exercise. It can determine the products you purchase, the candidates you elect, even the safeguarding of your rights as a citizen. Learning more about logic will give you the tool you need.

How can we become more logical?
We are all capable of being reasonable and logical. But this does not mean that we are always reasonable, or that our reasoning is always correct. All of us are guilty of some common bad habits of thinking; by identifying these habits and examining their causes we can tackle our problems more methodically and learn to test our conclusions by critical standards and rules.

A starting point would be to eliminate the personal equation from our thinking when our thinking should be concerned with facts. The anthropologist Franz Boas, in his Mind of Primitive Man, tells us that primitive man has the same kind of mind that we have, except that he is more likely to be influenced by emotion. Civilization, Boas says, does not improve the mind, but it decreases emotional association with ideas and thus helps us to think more clearly. If we wish to make sense we must try to eliminate emotion when emotion is irrelevant. Emotion is out of place when we are considering information contained in the multiplication table, and it is equally irrelevant in assessing a problem of legal evidence. Was Ezra Pound a traitor? If we say "No, because I like his writing," or "Yes, because I can't stand his poetry," we are evading the fundamental duty of a rational person, which is to study the evidence.

Another aid to clear thinking is the avoidance of too much rationalization. Our reasons should lead to our conclusions rather than serve later as an excuse for our conclusions. And we should be very clear, at least in our ova minds, as to which came first. We should be honest with ourselves in identifying the reasons for voting for a particular candidate, purchasing a certain model car, or selecting a mate.

It is hard to exercise our critical powers in matters that involve our emotions and self-interest. When asked to do so, we have a tendency to become dogmatic and to make positive or arrogant assertions that cannot be proved. We may become blind fanatics and stop thinking altogether, automatically regarding others' ideas as outlandish or perverse or dangerous. But instead of rejecting evidence that threatens to disprove our ideas, we should welcome it as an opportunity to increase our knowledge.

All too often we rely on the word of "authorities" for guidance without ever inquiring into the validity of their evidence. We should heed Bertrand Russell's observation concerning the role of expert opinion: "When the experts are agreed, the opposite opinion cannot be held to be certain; when the experts are not agreed, then no opinion can be held to be certain." In other words, let us respect the experts, but let us not follow them blindly.

While it is hard for human nature to sustain the tension of never having anything settled with finality or certainty, it is the safest and most realistic course to reserve final judgment on almost everything. The most reasonable attitude is that of the scientist who says, "Let us test our beliefs by the evidence, showing a willingness to revise these beliefs as the evidence changes, never claiming finality for our beliefs, but recognizing that the probabilities are sometimes so overwhelming that we can occasionally identify some truths. Thus we should be ready to believe when there is sufficient evidence, but we should suspend judgment when evidence is lacking."

We might start our study of logic by practicing the rules that Rene Descartes, the French mathematician and philosopher, used to guide his thinking:

1. Never accept anything as true which you do not clearly know to such; that is, avoid hasty judgments and prejudice.
2. Divide each difficulty under examination into as many parts as possible, or into as many as necessary for the solution of the problem.
3. Begin with the things that are simplest and easiest to understand, and then ascend to knowledge of the more complex.
4. Make enumerations so complete, and reviews so comprehensive, that you may be reasonably assured that nothing is omitted.

Now, whatever we may think of the usefulness of these simple precepts for attacking a problem, the assumption that underlies them is encouraging. We are all capable of understanding, and we can improve our understanding by using the right methods of thinking. These methods will help us solve our problems more efficiently. Our problems may be "scientific" ones, in the narrow sense of that term, but they may also arise in business and social relations in politics and in love

How this book will help
A few words now concerning what you are going to find in this book. The art of making sense involves an understanding of semantics and scientific methods of thinking in addition to the analysis of reasoning or logic. The good thinker will make a threefold analysis of a discussion. He will interest himself in the meanings of the words used, he will look for the "argument" in what he reads or hears; and he will ask himself whether what he hears and reads is true or false, and how it can be verified. Each of these steps represents a question to be asked, a problem to be solved.

To illustrate the manner in which these questions arise, let us examine an "argument," that is, a unit of discourse that purports to prove something: "Polygraph (lie-detector) tests are not twentieth-century witch-craft. After MI, they are used by many good police departments, and the results are accepted as evidence in some courts Let us put this argument to the test of our three questions

1. The semantical question: What do the words mean? For example, what do you understand by twentieth century witchcraft? What is a "good police department"? How many are "many"? What is meant by some courts? The principles which will help us clarify questions of this kind are discussed in chapters 1 through 4.

2. Let us consider the argument as a whole. What is the author trying to prove? What evidence does he give? Do you think his argument is logical, that is, does his conclusion follow from his facts and assumptions? Does your agreement or disagreement with his assumptions have anything to do with the "logic" of the argument? A discussion of questions of this kind will be found in chapters 5 through 7.

3. Finally, another type of question: Is it true that many good police departments use polygraph tests? Is it true that the results of such tests are used in some courts? what makes a police department "good"? If you have beliefs on these matters, and your beliefs are challenged, do you know how to support your position? The concluding chapters of this book deal with such questions.

One thing at a time, however. Let us begin our examination with a basic factor in the human communication process. Charles Lamb said that logic is nothing more than a knowledge of words, and it is with words that we shall begin.

FOR DISCUSSION AND WRITING
1. This chapter points out that logicians divide all discourse into the categories of "logical" and "nonlogical." Using these terms, identify the following statements:
a. The senior senator from California is bald.
b. I was born in East St. Louis, Illinois.
c. I have a stomachache.
d. She said she has a stomachache.
e. A rose is a rose is a rose.
f. According to the Bureau of tile Census, 16 percent of all black males over twenty-five years of age with a college education earned over $15,000 in 1971.
g. Help!
h. A bird in the hand is worth two in the bush.
i. Please open the door.
j. The newspaper story described the fuel shortage facing this country.

2. Using a dictionary when necessary, define each of the following terms: hypothesis, inference, argument, guess, intuition, hunch, theory.

3. As you consider the history of man and his actions, does it appear that he is a logical creature? What evidence can you suggest to support your view?

4. This chapter suggests three questions the careful thinker asks when analyzing an argument. Apply those questions to the following passages:
a. We must not lose faith in man's future. It was faith that spurred on the pioneers to settle this country It was faith that made this nation great. That same faith led man to scientific discoveries and inventions And that same faith will help him overcome the problems of pollution, crime, the depletion of natural resources and the threat of nuclear war.

b. When men are spoken of as kings and subjects or when government is mentioned under the distinct and combined heads of monarchy aristocracy, and democracy what is it that reasoning man is to under stand by the terms? If there really existed in the world two or more distinct and separate elements of human power, we should then see the several origins to which those terms would descriptively apply, but as there is but one species of man there can be but one element of human power and that element is man himself. Monarchy, aristocracy, and democracy are but creatures of imagination, and a thousand such may be contrived as well as three. (From The' Rights of Man, by Thomas Paine)

5. What are some everyday problems that could profit from the application of Descartes' four rules? Do they apply chiefly to scientific problems, or to problems in sociology, politics, and psychology as well?

6. Within one week two newspapers published conflicting accounts of workers' attitudes toward their jobs. One article reported that "a representative sampling of adult Americans from coast to coast" indicated that most people like their jobs and do not find them boring. The second article gave summaries of interviews with a sampling of production-plant workers and concluded that most of these workers find that their jobs are "like being in prison." How would you go about finding the truth concerning the accuracy of these stories? How would you determine which article more nearly correctly reports the attitude of most workers in this country?

7. Can you think of any decisions you have made recently that were made primarily on the basis of reason and logic? In general, how often is logic the basis of your most important decisions?

---

CHAPTER 2
Words and Ambiguity

Man has always regarded words with reverence and mystery. In primitive societies a belief persisted that a person or thing had one right or true name, and knowledge of that name gave power over the person or thing. In the Gospel of John we are told that, "In the beginning was the Word, and the Word was with God, and the Word was God." Because of this tendency to invest words with a magical quality, man often forgets an important truth: words have little importance on their own account, or for their own sake. Their importance derives from their meanings, and their function is to act as signs or symbols of something outside themselves. Perhaps this is what the English philosopher John Locke meant when he said, "We should have a great many fewer disputes in the world if words were taken for what they are, the signs of our ideas only, and not for things themselves."

In this chapter we shall examine some fallacies and myths about the relationship words have to the things they stand for. We shall also consider some of the difficulties and ambiguities that arise-whether in writing a theme for class or speaking informally to friends-when we are care-less in choosing our words. Such an examination will begin by reviewing the principles of semantics: the study of words (and symbols generally) in relation to their meanings.

Semantics: The study of "the meaning of meaning"
In 1933 Alfred Korzybski, a Polish-American scholar and mathematician, published Science and Sanity, a difficult and provocative study which presented a method by which people might think and speak more clearly through an analysis of language habits. By analyzing how we use words and how we react to words, semantics (as Korzybski's system is called) has come to designate the study of words as signs or symbols, that is, as things that usually stand for something other than themselves. An-other way of describing semantics is to say that it is the study of "the meaning of meaning."

Although the study of semantics has received a great deal of attention in recent years from philosophers, logicians, scientists, and teachers of English, its subject is not a new one. A recognition of the importance of a proper understanding of language for clear thinking is as old as philosophy itself, going back to Socrates in ancient Greece. Philosophy, for Socrates, was the "pursuit of meanings," and he sought adequate definitions of words like "justice," "good," "right," and "wrong." He realized that unless there is agreement on the meaning of the words we use, communication is impossible.

Since the time of Socrates a concern with meaning and language has been a constant preoccupation of philosophers and logicians, from Aristotle to Bertrand Russell and Noam Chomsky. Today semantics covers a vast and complex field of investigation, with widely diverse branches of study, and it often employs the methods of anthropology, linguistics, logic, psychology, and many other disciplines. But we shall not be concerned with a systematic investigation of semantics. Nor do we share the hopes of some semanticists that this subject offers a panacea for all the world's ills. We are interested in some practical applications of the subject: to show how an understanding of semantical principles concerning the uses and functions of words may aid us in thinking and writing more clearly) and thus in making sense. A sound understanding of language, the instrument of communication, is a means to this end.

No one really knows how much thinking human beings could do without language, but it is undeniable that our thinking would be very limited. Human intelligence is based on our ability to think and talk about things that are not in our actual surroundings. We use words to "point to" these Things. Though animals can think, they have no words, and so their intelligence is limited. We can think and talk about Mary even though we don't see her. But suppose we mention the name "Mary" to a dog, and Mary is his mistress. The dog will react to the familiar sound, and will wag his tail in eager anticipation of seeing Mary. The dog reacts to the name much as he does to Mary. But human beings can think and talk about Mary, knowing that she is absent. In fact, knowing that Mary is absent makes it a good deal easier to talk about her. A girl can pick at a daisy and say, "He loves me, he loves me not," but no animal can do that.

Language is thus indispensable to human thinking, but this does not mean that mental ability is the same thing as having a large vocabulary. A limited vocabulary, it is true, restricts the range of our thinking. And this may lead us, mistakenly, to think of people as being unintelligent merely because certain words are unfamiliar to them. Because vocabulary is determined largely by cultural and socioeconomic factors, there is a tendency today to regard intelligence tests that rely heavily on verbal skills as unreliable means of determining native intelligence. The children in an underprivileged neighborhood, for example, did very poorly in an intelligence test which contained questions such as this. "A hand is to an arm as a foot is to a ___." Only a few children filled the blank with "leg," and most of them were scored low in intelligence. But later it was learned that the expression "is to" was unfamiliar to these children. They would have said "goes with," that is, a hand "goes with" an arm, and when "goes with" was substituted for "is to" in the same test, they did very well, and scored high in "intelligence."

Language also influences the content of our thoughts. The phrasing of a question will influence our thinking about the subject matter. It is well to remember this point when we evaluate the results of public opinion polls. It makes a difference whether one is asked: "Do you favor clemency for those American men who obeyed their consciences and refused to participate in what they believed to be an illegal, immoral war?" or "Are you opposed to letting draft dodgers and deserters go unpunished?" The phrasing of a question may convey an emotional tone, and most of us are suggestible, so that we have a tendency to agree with what the questioner seems to expect of us. Consider how we are apt to respond to the question: "How do you account for the fact that the great majority of American people disagree with the motives and actions of the men who refused to serve in the Vietnam war?" We may fail to stop short at this point to ask the question: "Is it really true that the great majority of Americans do disagree with the motives and actions of the men who refused to serve in the Vietnam war?"

Semantics, then, is concerned with language insofar as language is relevant to problems of thinking and communication. We shall now examine certain misunderstandings concerning words, and some of the fallacies and myths that cluster around the relationship that words have to the things they stand for.

Three semantical errors (or, a warning to the unwary student)
Words are not mysterious things. They are events in space and time; that is, they have a physical dimension, and they have meanings. Those meanings, however, are arbitrary. Further, words are merely symbolic sounds, and they do not possess inherent qualities or defects. Let us now examine the consequences of these facts.

1. The word and the thing
When we say that words have meanings, we say that human beings agree that a certain word, like "nylon," for example, shall refer to a certain kind of material. This material could have been called by any other name, but the Du Pont Company christened it "nylon." So now, when I think of this material, and say "nylon," the sound comes to your brain, and your mind is referred to the material I am thinking of, namely, the fabric used in women's hosiery, etc.

Now, this relationship of the word to the thing is an "arbitrary" relationship, in the sense that the word could have been any other word. There may be aesthetic or other reasons for choosing an "appropriate" name, but these considerations are never compelling. This principle, that the relationship of words to things is an arbitrary one, may seem absurdly simple to you, but it is a basic principle of semantics, and, though simple, it is unknown to some, and forgotten by others. Children, for example, are apt to be unfamiliar with the principle. When a French semanticist asked a child whether the moon could have been called "the sun" and the sun "moon," the child said, "No, because the sun makes it warm and the moon gives light." Another little boy once asked his mother: "Mother, when I was born, how did you know that I was really Charlie, and not some other little boy?" Children often believe that a word is necessarily connected with a thing, so that it would be impossible to call the thing by any name other than the one by which it is known. Now, of course, after words come to be associated with specific things, a connection is established, and we would create confusion if we did not use these words in their customary meanings. But the point is that words, when they first come into being, can be anything at all.

And here is another amusing example of the same type of error. When the planet Pluto was discovered in 1930, the story goes, a young lady was reported to have asked an astronomer: "Professor, when you astronomers discovered the new planet, how did you know that this planet was really Pluto, and not some other planet?" Now this new planet, of course, could have been called by an' other name, even Mickey Mouse, though in some ways that name would not have been so appropriate, for it is customary to name the planets after ancient Roman gods.

The kind of error we have just noted was the subject of an amusing bit of spoofing by Mark Twain. Somewhere in his writings he discussed chapter 2, verses 19 and 20, of the Book of Genesis, which deal with a semantical matter: "And out of the ground the Lord God formed every beast of the field, and every fowl of the air. And brought them unto Adam to see what he would call them, and whatsoever Adam called every living creature, that was the name thereof. And Adam gave names to all the cattle and to the fowl of the air, and to every beast of the field." The semantics of this account is unexceptionable, but Mark Twain, in his little jest, imagines that these animals pass in review before Adam. He gets along fine, until he is finally baffled by one animal. He can't think of a name for it. In desperation (like a man) he turns to his wife for help. "Eve," he asks, "what name shall I give this animal?" Without a moment 5 hesitation, Eve answers, "Call it a horse." "But why a horse?" Adam asks her. "Well," says Eve, "it looks like a horse, doesn't it?"

We have been dealing with rather obvious examples. But the same kind of error also occurs on subtler levels than those just considered. For example, the system of government in the United States is commonly referred to as a democracy. There are some people, however, who argue that it is wholly improper to call our system of government a democracy. Sonic have gone so far as to demand that a law be passed forbidding the use of the word in this sense. These people argue that we are a republic, that we have a republican rather than a democratic form of government. The dictionary defines a republic as a system in which sovereignty resides in the people and in which legislative and administrative functions are carried on by elected representatives of the people. "Democracy," the argument goes, "means the rule of the people directly, and not through representatives," as in a small town where ever' voter has his say in a town meeting.

Now, there is no question that we are a republic, but we may also be a democracy as that word is not used. If the people of the United States, England, and France wish to refer to their systems as democracies, and define a democracy as "a system of representative government based on the principles of freedom, and legal and political equality," there is no one who can or should prevent the people from doing so. In other words, the word "democracy" has been broadened in its meaning by usage, and today it usually refers to a system of government in which the people elect representatives in regularly scheduled free elections, and in which there is a basic devotion to freedom and political equality. This at least is one meaning that democracy has acquired, and to argue that it is wrong to use the word in thismanner represents a failure to acknowledge the arbitrary relationship of words and things. Language is like a living thing in its growth and development, but its life depends on human decisions.

The uses and alleged misuses of the word "democracy" may appear to involve only a theoretical problem of semantics. The heat engendered by this apparently trivial matter, however, indicates that other issues may be involved. These other issues are not far to seek. The issue is the age-old one of conservatism versus liberalism. To call the United States a democracy is to emphasize our ideals of freedom and equality, in addition to our representative system. Conservatives, who believe that an emphasis on equality will work against the public good, prefer the word "republic," which does not suggest equality. Many liberals, who want more equality, prefer "democracy."

Words, then, can have as many uses as people give them. As language grows and develops, it becomes permissible to use new names for things, or to use old names in new ways. Thus, the word "surgeon," from Greek roots meaning "one who works with his hands," once meant a laborer; today it means one who operates on a living organism. The word "doctor" today means a practitioner of medicine and surgery, among other things; but originally it meant a teacher, especially one of great learning. These are examples of old words that have acquired new senses. It would sound rather odd if we said that medical men were not really doctors, on the ground that "doctor" really means a teacher of great learning.

New names are invented not only for new things, like new drugs and synthetics (Streptomycin, Orlon) and space-travelers (astronauts), but we also give old things new names. Military invasions are called "interdictions," and bombing raids "protective reactions." Inflammable materials are now called "flammable," and a question concerning the meaning of a word is now called a "semantical question."

The principle that words are arbitrarily associated with things should of course not be abused. If we desire successful communication we should not capriciously assign new meanings to old words. The use of words in their customary senses also enables readers or hearers to devote more attention to the thought and less to the vocabulary. That famous character known as Humpty Dumpty, however, was unconcerned with whether anyone understood him or not, and accordingly he was free to abuse the principle, as evidenced by the following colloquy:

Humpty Dumpty said ...:"There's glory for you."
"I don't know what you mean by glory," Alice said.
Humpty Dumpty smiled contemptuously. "Of course you don't-till I tell you. I meant, 'There's a nice knock-down argument for you.
"But 'glory' doesn't mean a 'nice knock-down argument,"' Alice objected.
"When I use a word," Humpty Dumpty said in rather a scornful tone, "it means just what I choose it to mean-neither more nor less." (Lewis Carroll, Through the Looking Glass.)

2. The word and magic
We have been discussing the failure to take note of the arbitrary relationship between words and things. We shall now examine a second type of error which arises from the failure to recognize the fact that words are merely symbolic sounds. This error is the belief in the magical power of words. It is the practice of certain primitive tribesmen to change their names after being cursed, so that they may escape the evil which has become attached to their names, and thus to themselves. Another example is the case of the benighted primitive who, when cursed by a fellow tribesman, dropped flat on the ground so that the words would fly harmlessly over his head. And do you remember the story in the Arabian Nights, about Ah Baba and the Forty Thieves? Ah said, "Open, Sesame," and lo! the cave door opened. These examples indicate the nature of the belief in the magical power of words: words have potencies to do things all by themselves; a name can become infected with evil; it can harm a man if it actually strikes him, and it can open cave doors.

Are these superstitious beliefs in the magic of words confined to primitive man? I am sure that you are not guilty of similar superstitions, but how about your friends? And, candidly, aren't we all, just a little bit? Don't we all know people who say, "Speak of the devil and he's sure to appear, and who really believe that there way be something in that expression? Don't we go to the racetrack and find a horse whose name has struck our fancy, and bet on that horse regardless of the form charts? And if this has happened to you, and your horse won, weren't you just a little bit persuaded that there really is some kind of magic in a name? And why do we say, "Knock on wood," when we express a thought concerning our good fortune? Well, obviously, because we assume that there are forces in the universe which don't like to hear people talk about their good luck. Again, the belief in the magic of words.

Have you ever watched a group of men engaged in a form of wagering known as "shooting craps"? It is a highly instructive area of investigation for the student of language. One of the players is hoping that he will throw a seven or eleven. He pleads with the dice, and informs them that his very young infant is in need of protective covering for its little feet. But alas! The dice roll regardless of his words. For words have no magical powers.

Or consider the radio broadcasting or the telecasting of baseball games. Many of the listeners and television viewers believe that words have magical powers. If the home team's pitcher is pitching a no-hit game, these believers in magic regard it as a terrible crime for the announcer to mention this fact, for if he does, then the spell will be broken, and the next batter will be sure to make a hit. Because of this superstition, which many announcers are afraid to challenge, millions of listeners are cheated, for they are denied knowledge of the dramatic intensity on the baseball field.

3. Words, truth, and beauty
A third error is the assumption that words give us guarantees concerning things. This error usually occurs in this way: we assume that a fine-sounding name proves the fine quality of the thing referred to. This is a frequent source of deception. An organization may call itself "The People's Committee for Peace," or some such name, because the words "people's" and "peace" sound trustworthy and good. In the same way, a group seeking to foster race prejudice may use words like "Christian" or "fellowship in its name in order to convey the impression that it upholds the principles of religion and love for one's fellow beings. We all know of the many fancy names which used to be common in the fur business, such as Hudson Seal for muskrat fur. The Federal Trade Commission may be going too far, however, in forbidding the expression "imitation fur" for materials that are not made of fur, on the ground that the word "fur" cannot be used for materials not made of fur! This reminds one of a famous cartoon showing a policeman beating a hapless-looking individual at a Communist demonstration. The man: "But, officer, I'm an "anti-Communist." The policeman: "I don't care what kind of Communist you are!" and continued to whack.

One must guard against these sources of deception, but there are also many harmless forms of this sort of thing which involve willing self-deception. Thus on transatlantic liners, many years ago, third class" was changed to "tourist class" because third class sounded too inferior. In the Soviet Union, we may note, the name "third class" was changed to "third category," for there can be no class distinctions in a communist society! The accommodations are the same, objectively, but the change of name may have a pleasing psychological effect, for it seems :0 eliminate a stigma.

We also often assume that evil-sounding names imply that the thing is evil. The implication does not follow. But again, we must emphasize the fact that "brutal" and ugly names may have undesirable psychological consequences, and this is particularly unfortunate when such names are not strictly accurate in their descriptive aspects. For ample, the expression "home for incurables." This name was based n the assumption that certain diseases are actually incurable, but this is a big assumption, and the present tendency is to change the name to institution for chronic diseases." Not only is such a name truer to the facts, but it also gives the patients more hope, and hope may have beneficial psychosomatic consequences. Furthermore, who knows but that science may some day find cures for such diseases? In other words, let us not call a spade a spade even when it isn't one.
Names, then, guarantee nothing in themselves. As Shakespeare's Juliet remarked:

What's in a name? That which we call a rose
By any other name would smell as sweet.

Words and etymologies
So much for three semantical errors. One of the basic points of this chapter is that the meanings of words are based on arbitrary human choices. An understanding of this point will also help to clarify the semantic relevance of the etymologies of words. The word "etymology" refers to the study of the history of words, to their derivations from their roots, with all their changes of form, spelling, and meaning. Etymology describes the manner in which words came to acquire their present meanings. For example, the word "philosophy" is based on two Greek roots, philein and sophia. Philein means "the love of," and sophia means "knowledge" or wisdom," so that the etymological derivation of the word "philosophy" indicates that it means the "love of wisdom." "Philanthropist" is based on philein and anthropos, meaning "man," so that a philanthropist means a "lover of mankind." "Sophomore" comes from sophos and moros, the latter root meaning "fool." A sophomore is a wise fool.

Other examples could be multiplied endlessly. The word "assassin" is based on the Arabic word hashahin, which is derived from "hashish," an intoxicating drug found in the Middle East. The first assassins were hashish-addictive eleventh century Syrians and Persians who were members of secret societies which murdered their political enemies. "Planet" comes from a Greek word meaning "wanderer," for planets change their positions among the fixed stars.

But the important point for semantics is this: etymologies enlighten us concerning meanings, and sometimes give precision to words whose meanings are somewhat vague to us. The etymology of "definition," for example. This word comes from the Latin roots de (off) and finis (end, limit, boundary). A definition, then delimits the meaning of a word. The study of etymologies will help us in using the right word to express a precise shade of meaning. But etymologies do not control the use of language. No matter how a word may have originated, it means today what people use it for. Custom is king in matters of language, and if human beings customarily use words in new senses, it cannot be said that they are wrong in doing so, for words are noises arbitrarily associated with things. An assassin today means one who commits murder because of fanaticism or for a reward, not a person who smokes hashish. And etymologies may also be misleading. The word "etymology" itself, for example. The word is based on etymon, meaning "the true sense", plus logos, or "word," but there are no "true" or "false" senses of words. There are only customary or uncustomary senses. And so, though etymologies illuminate the meaning of many words-like "philanthropist," and "planet"-they do not establish the "real" meaning of the word. Words mean what people intend that they shall mean. We are the masters of the words, not the words of us.

Words are wonderful engines of communication, but we must know what they mean, and how to handle them. And we must guard against being "taken in" by them. As that wise old English philosopher Thomas Hobbes wrote: 'Words are wise men's counters-they do but reckon with them, but they are the money of fools."

Ambiguity: Or, I wonder what he meant by that?
Words, we have seen, can have any meaning we assign to them. This characteristic, called "ambiguity," is responsible for many unnecessary disagreements. We shall now examine the ways in which ambiguity is a major cause of failures of communication, whether in writing or in conversation.

Let us imagine that we are listening to a conversation between two men, Bill and Jim, at one of those informal debating societies known S "cocktail parties." They are arguing a frequently debated topic, the principle expressed in our Declaration of Independence, that all men are created equal. Bill has the floor:

"Jim, I tell you that men aren't equal. Don't let anyone tell you hey are. They don't know what they are talking about. Use your own yes! Do you see the equality of mankind? Do you find in your own experience that people have equal abilities, or equal characters? Do you find that they are equal in any respect whatsoever? Everyone is different from everyone else. In my opinion Thomas Jefferson uttered preposterous nonsense when he said, and I quote, 'We hold these truths to be self-evident, that all men are created equal.' This so-called truth is not self-evident to me, so it can't be self-evident. In my opinion this so-called truth is actually a falsehood."

Let us now hear from Jim: "Just a minute, Bill. You are the one who doesn't know what he is talking about. Men are equal, and I agree completely with Jefferson. The equality of mankind is the foundation our democracy. No man has the right to think of himself as better

than any other, or as entitled to special privileges which others aren't entitled to. Every person is entitled to equal opportunities, and no one should suffer discrimination because of his race, color, or creed. This is he basis of our legal system, which tells us that all men are equal before the law. A legal decision should not depend on the color of a man's ,in. Do you deny that? Are you in favor of racial and religious discrimination?"

And so on. Let us be merciful, and tune out Bill and Jim at this point, though they are probably still arguing, unless they have already ached the point of mutual exhaustion. Now, the argument we have just overhead can never have an ending, because it was really not an argument at all, but just two fellows talking at cross-purposes. In order to have a genuine argument there must first of all be an agreement or meeting of minds about the issue in dispute; that is, there must be a common understanding of the question to be answered. But there was no such agreement between Bill and Jim. They were talking about different things, and so their minds never really met. They were engaged in what we shall call a "verbal dispute."

A verbal dispute is one in which the two speakers engage in what merely looks like an argument, but really is not, because the speakers o not understand each other. The reason why they do not understand each other is that they are using a key word in two different senses. The key word in this ease was "equal." In other words, there can be no argument concerning whether or not men are equal, if Bill means one thing by "equal" and Jim means something else. Let us recall what they were saying.

Bill said that men were not equal. By "equal" he meant having the same size, shape, mental and physical powers, talents, and so on. Jim said that men were equal. By "equal" he meant that all men should be given the same opportunities, and that they should have the same chance of getting justice in a court of law. Bill's mind and Jim's mind did not meet, for they were thinking about different things. Though each used the same word, "equal," they meant quite different things by the word, and so were engaged in a verbal dispute, rather than in a genuine discussion or argument.

The point is at we should not disagree with anyone until we first find out what he means by his words. To understand before we disagree is not only a rule of courtesy hut also good sense.

The basis for the troubles we have just described is the fact that words are ambiguous. The key words in many disputes have more than one meaning, or more than one sense, and this leads to misunderstandings.

Words stand for things, but we don't have just one word for each thing, like a buttonhole for each button. More than one word may stand for the same thing: such words are synonyms. One word, on the other hand, may stand for several different things. When there is uncertainty as to the meaning which e speaker or writer intends, there is ambiguity. For example, the word "secretary" usually means "a person who attends to correspondence." But a big-game hunter may tell you that on his last trip to Africa he captured two secretaries. If you look the word up in the dictionary, you will find that "secretary" also means "a South African bird with very long legs." Or, a business acquaintance may tell you that lie recently moved his blonde secretary into his home. He is referring to a writing desk made of light-colored wood. The ambiguity of words may create embarrassing misunderstandings! The word "equal," similarly, may be understood in different senses, and this opens the door to misunderstandings of the kind we have just described. Words are not ambiguous by themselves but only in a context which makes their meaning uncertain. There is no ambiguity, of course, in "I wish to dictate a letter to my secretary."

When words are spoken rather than read, their phonetic sounds may be ambiguous. The sound "tears," for example, in "The audience sat in tears." Tears or tiers? The next example also involves phonetic ambiguity: "Some people pray on their knees on Sundays and on their neighbors the rest of the week."

Let us now look at another example of a discussion in which a key word is used in different senses. Let us suppose that there is a disagreement over the number of unemployed in the United States at the present. time, and that two collectors of statistics have reached different results in counting the unemployed. The difference between the statisticians may be due to biased figures, or unrepresentative samples. But the be a semantical one-the statisticians may have defined the word "unemployed" in different ways. There are some industries which employ seasonal workers, such as the canning industry. Is a seasonal worker unemployed during the winter months, when he is regularly laid off? One statistician may consider him unemployed; the other may say he is employed, for he expects to return to his job in the spring. Or the statisticians may differ with respect to the classification of workers who are ill, or on strike. These matters should be settled by definitions, other-wise a verbal dispute may occur because of the different meaning given to the word "unemployed." If we are not agreed on what we mean by a word, we shall talk at cross-purposes.

A verbal dispute frequently engaged in by college students arises over the old chestnut: en a tree falls in an uninhabited forest, does the crash make a sound? The argument goes on and on. One side agrees that there is no sound because there is no one present to hear it; the other that longitudinal air waves, known as "sound waves," will occur in the air whether or not anyone is present to hear them, so sound is present. Now, the sciences of physics and psychology tell us that "sound" occurs when waves in the air hit our eardrums and cause motions in our nervous systems, finally reaching the brain. When motion finally reaches the auditory nerve, we experience what is called "sound." This analysis reveals the presence of two elements: (a) a certain kind of mental experience and (b) a physical cause of that experience. In the dispute we just noted, "sound" was used in both of these senses: (a) for the experience itself and (b) for the physical cause, the sound waves that cause the mental experience. In sense (a) the crash does not make a sound in the uninhabited forest; in sense (b) it does.

Verbal disputes indicate the manner in which the ambiguity of a word may result in our talking at cross-purposes. Precisely the same sort of thing happens in verbal agreements, as distinguished from disputes. Verbal agreements are "merely apparent agreements. We may find ourselves in apparent agreement with another person only because of ambiguity and speaking at cross-purposes. Just as a verbal dispute conceals a possible real agreement, so a verbal agreement may mask a real dispute. Consider the agreements reached between this country and Russia after World War II, or with No Vietnam. In both instances ostensible agreements were found to be based on words, rather than on a meeting of minds. Once again, talking at cross-purposes.

So much for verbal disputes and verbal agreements. The problem of ambiguity has much wider ramifications, some of which we shall now explore. Ambiguity is an ever-present aspect of language, for most words have many meanings. This leads to difficulties in communications, as noted, but it also vastly enriches language. A word like "fast," for example, which refers to abstention from food, to a quality of colors, to certain kinds of characters, and so on, is the equivalent of many words. But we are primarily interested in ambiguity insofar as it is an obstacle to communication. There are four types of ambiguity that cause trouble of this kind, and we shall briefly survey each type. The four types: the ambiguity of single words, of sentences, of emphasis, and of significance.

1. The ambiguity of single words
Verbal disputes usually involve the ambiguity of a single word or expression. One of the ways in which we can detect the presence of this kind of ambiguity is to ask a question containing the suspected word, and phrase the question so that it can be answered by Yes or No. If a Yes or No answer requires a specification of the sense, then the question is ambiguous. "Are all men equal?" Whether we answer by Yes or No, we must specify the sense of "equal." The best answer: "Yes and No, depending on the sense of 'equal.'"

Similarly, the question, "Do you believe in God?" requires clarification of the sense of the ambiguous term "God." Now, there are many people who object to this kind of analysis, and who say that they want no quibbling; they want a simple Yes or No answer. But philosophers have defined the word "Cod" in different ways. When the American philosopher Arthur 0. Lovejoy applied for his first teaching position, the application asked the question "Do you believe in God?" Lovejoy appended a list of more than thirty philosophical definitions of God and asked, "In which of these senses is the question to be answered?" (He got the job.) Or consider the conception of God held by the philosopher Spinoza, who was a pantheist. The pantheistic conception of God holds that God is the system of Nature as a whole, in all its existential and dynamic aspects. For Spinoza, everything in the world is part of God, and God is everything. Spinoza regarded himself as a profoundly religious man; for him, God is the only Being who can be loved by man without fear of man's ever being disappointed, for God is eternal, infinite, and perfect. But most people think of God as a personal Being, as the Creator of the Universe, and pantheism denies that God is a Person. It is thus apparent that one may believe in God, in one sense of that term, and yet be considered not to believe in God, in a different sense. To the question, "Do you believe in God?" Spinoza would have answered, "Yes, in one sense; no, in another."

Unless we are alert to the possibility of ambiguity we may find no sense where sense is present, as in this sentence from Paul's First Epistle to the Corinthians: "And though I bestow all my goods to feed the poor and have not charity, it profiteth me nothing." This is an apparent contradiction, for "charity" means almsgiving. But in an older sense, charity means love (from the Latin caritas), and certainly charity in the modern sense is possible without love.

Ambiguity may also result in fallacious reasoning, as in this example of a bad argument: Science has discovered many laws of nature. This is proof that there is a God, for a law implies the existence of a lawgiver, and God is the great Lawgiver of the universe." This argument is vitiated by the ambiguity of the word "law." In "laws of nature," law is used in its scientific sense. It means "a description of the uniform behavior of natural events." In another sense, that is, in the sense of "legal law," law does imply the existence of a lawgiver, for law in this sense means "regulations emanating from legislatures and courts, rules enforceable by the state, etc.,' A simple way of differentiating the two is to say that a natural law is a description; a civil law is a prescription. We need a lawgiver to prescribe, or issue a command, but scientists use the term "law of nature" to mean that they have found some kind of invariant behavior in nature. Their description of this behavior has nothing to do with commands. To use an ambiguous term in two different senses, as this argument does, and to draw a conclusion based on this confusion of two senses, is to be guilty of the fallacy of "equivocation."

The first type of ambiguity, the ambiguity of single words, is also a prolific source of humor: If I had a mind to," Wordsworth said, "I could write like Shakespeare." the following example, the humor was unintentional: It was reported that at Oxford and Cambridge, "half of the school population are poor students on scholarships." "Poor": in money or studies? Good, bad, and indifferent puns are based on the same kind of ambiguity. Sydney Smith, the wittiest of English divines, once described how two women used to lean out of their windows, on opposite sides of the street, and argue with each other. "They will never agree," he said, "for they are arguing from different premises." Benjamin Franklin coined a famous pun when he warned his fellow colonists that they had better work in cooperation against England. "We must indeed all hang together," -he said, "or, most assuredly, we shall all hang separately."

2. The ambiguity of sentences
Let us now look at the second type of ambiguity, the ambiguity of sentences, as distinguished from the ambiguity of single words or expressions. Ambiguous sentences are statements whose grammatical construction may lead to possible misinterpretation. Such sentences are called "amphibolous." This is the kind of ambiguity involved in the "help wanted" sign at the factory entrance: "Wanted: Young Girls to sew Lace Trimmings on the 4th Floor." This "grammatical ambiguity" is the kind employed in the messages which diplomats like to send to each other. The ambiguous language permits the writer to claim that he did not mean what the reader thought he meant. Literally and strictly interpreted, his language may commit him to nothing. For example, a diplomatic editor wrote a short note to a would-be author: "I shall lose no time in reading your manuscript."

This type of ambiguity was used by the famous oracle at Delphi in ancient Greece. The oracle made predictions of things to come, and the predictions always came true, because of the form in which they were cast. For example: If the oracle were questioned concerning the outcome of a battle between the Greeks and the Persians, the oracle would deliberately cast its prediction in amphibolous language, something like this: "The God Apollo says that the Greeks the Persians shall subdue."

3. The ambiguity of emphasis
Our third type of ambiguity we call the "ambiguity of emphasis." It occurs when we are uncertain as to the emphasis which words require. This is one of the reasons why it is so much more interesting to hear a play performed by competent actors than to read the play, for the actors give the words their proper emphasis. An example of this type of ambiguity is found in "Nothing is too good for her." Consider how the meaning will vary with the emphasis!

Errors in emphasis occur when we stress the wrong words in a sentence and thus distort the meaning of the writer. Consider the Ninth Commandment: "Thou shall not bear false witness against thy neighbor." Now, if one reads this commandment with the accent on neighbor, this suggests that it is permissible to bear false witness against those who are not our neighbors, a meaning obviously not intended.

Errors in emphasis also occur in our writing when we misrepresent another author's meaning by making biased selections or quotations from his work. Such improper emphases, however, are usually due either to carelessness or to deliberate design, rather than to actual ambiguity. For example, a professor made the following comments on a student's thesis:

"Your thesis is both good and original. Unfortunately, the good things in it are not original and the original things are not good." The student, an expert excerpt-lifter, quoted his professor's remark: "Your thesis is both good and original."

4. The ambiguity of significance
Our fourth and final type of ambiguity is called the "ambiguity of significance"; ambiguity, that is, concerning the significance of what is being said. This type of ambiguity occurs when a perfectly true statement has misleading connotations, as if one were to say, "John didn't beat his wife last week." (Was this his usual practice?) Or this one: "British statesmen always put the interests of Britain first." True, but what is its significance? It insinuates that British statesmen are more selfish than others. But every statesman puts the interests of his country first. It is not the statesman's job to sacrifice his country's interest to the interests of other countries. And when a characteristic is true of everyone, as this one is, specific individuals or nations deserve neither credit nor discredit for sharing it with everyone else.

An old story about a sea captain and his first mate will serve as a final illustration. The captain and his mate alternated in writing the happenings of each day in the ship's log. One day the mate drank too much, and the next day he found the entry, "The mate was drunk today." Very much annoyed, the mate asked the captain why he had made that entry. t's true, isn't it?" the captain asked. The mate admitted it was. "Well, then," said the captain, "if it's true then it is properly entered in the log." The next day the captain (who was a sober man) opened the log and found the mate's revenge in the notation, "The captain was not drunk today.,"

So much, then, for the subject of ambiguity and its four forms, ambiguity of single words, of sentences, of emphasis, and of significance. Is there anything we can do to avoid these fertile sources of misunderstanding and thus improve the process of communication and our writing? There is. Whenever we find ourselves in a dispute, we can ask whether a key word is being used in different senses. Whenever we read editorials or other discourse containing opinions and arguments, we can check to determine whether any of the four forms of ambiguity are present, and if so, whether our first interpretation of the meaning is the only possible one that makes sense in the given context.

The cure for the troublesome aspects of ambiguity lies in making our ideas clear. And this means that we must define our terms. The general problem of definition will be discussed in the next chapter.

FOR DISCUSSION AND WRITING
1. Meanings of words are based on arbitrary human choices. As a result, few words have fixed meanings but are in a constant state of flux. For example, the word "silly" has taken on a somewhat derogatory meaning in recent times, but it originally meant 'blessed" or 'holy"; later, it came to mean "defenseless," and now signifies "foolish." Using a good dictionary (the Oxford English Dictionary if available), trace the histories of the following words:

propaganda
liberal
brave
counterfeit
nice
democracy
crafty
fret
villain
buxom

What conclusion can you draw concerning the reliability of etymology as a clue to present meaning?

2. As we have seen, the meaning of a word is based upon convention; that is, the speakers of a given language accept by general agreement that a certain symbol signifies a particular thing. Speakers of English will accept 'how-wow" as the word which describes the sound of a barking dog. Indeed, it is generally agreed among speakers of English that 'how-wow" is what they hear when a dog barks. But even in such a case where the word is apparently based on onomatopoeia, convention enters in. For example, a Chinese dog goes "wang-wang, and a Netherlandish dog "waf-waf." In the following exercise can you match the words which signify the sound with the animal that makes that sound?

1. Polish dog	a. miih
2. Latvian dog	b. liau-liau
3. Japanese dog	c. kikeriki
4. American cow	d. mu-bu
5. German cow	e. snof-snof
6. French cow	f. meck-meck
7. Hungarian cow	g. vau-vau
8. Portuguese pig	h. meu
9. Polish pig	i. kukeleku
10. Hungarian pig	j. cue-cue
11. Finnish pig	k. be
12. French sheep	l. ming-ming
13. Russian sheep	m. bo-bo
14. German goat	n. roff-roff
15. Turkish goat	o. moo
16. Dutch rooster	p. kwick-kwick
17. German rooster	q. beh

Answers: 1-b; 2-g; 3-1; 4-o; 5-a; 6-h; 7-d; 8-j;
9-p; 10-n; 11-e; 12-k; 13-q; 14-f; 15-m; 16-i; 17-c

3. In the following statements a word becomes more than an arbitrary symbol for a thing. In some instances the word takes on magical power; in others, the word attempts to give us a guarantee as to the worth (or lack of worth) of a thing. In still others, the word is considered as necessarily connected with a thing. Analyze the statements, attempting to describe the author's assumptions in using his words. Does he recognize that words are merely symbols? Is he consciously trying to establish a connection between the word and the thing? For what purpose?

a. Imperial margarine is fit for a king.

b. Professor: If you were to create a new language, which of the following words would you use to designate "iron": "sig" or "sug"? Student: "Sug"! The word sounds like "iron."

c. I baptize you in the name of the Father, the Son, and the Holy Ghost. Amen.

d. The allied forces made a protective reaction strike yesterday as they bombed enemy territory.

e. Dr. Henry Gibbons described a kiss as "the anatomical juxtaposition of two orbicularis oris muscles in a state of contraction." Of this definition a newspaper editor remarked, "A kiss may be one of those things, but it doesn't taste like it."

f. Capri. The first sexy European under $2600.

g. Notice to all department heads: In all reports, the work designation "garbage collector" will be changed to "sanitation engineer."

h. I refuse to salute the flag. It represents all those things I consider anti-American.

i. "Who could find anything better than hum, or buzz, or whir? Who could think of anything more sloppy than slop? Is not the word sweet a kiss in itself, and what could suggest a more peremptory obstacle than stop?" (From Englishmen, Frenchmen, and Spaniards, by Salvador de Madariaga. New York: Oxford University Press, 1937)

4. Consider the following statements. What does the author of each suggest about the nature of language? Is he correct in his assessment?

a. "Dilapidated: said of a building or other structure. But the word is from the Latin lapis, a stone, and cannot properly be used of any but a stone structure." (Ambrose Bierce)

b. "As the language is, so also is the nation." (Otto Jespersen)

c. What's in a name? That which we call a rose by any other name would smell as sweet." (Shakespeare, Romeo and Juliet)

d. "Names belong to things by nature and the user of words must keep in mind that the name of a thing belongs necessarily to the thing itself." (Plato, Gratylus)

e. "So God formed out of the ground all the wild animals and all the birds of heaven. He brought them to the man to see what he would call them, and whatever the man called each living creature, that was its name. Thus the man gave names to all cattle, to the birds of heaven, and to every wild animal." (Genesis 2, 19-20)

5. Verbal disputes usually involve the ambiguity of a single word or expression. One of the ways in which we can detect the presence of this kind of ambiguity is to ask a question containing the suspected word and phrase the question so that it can be answered by Yes or No. If a Yes or No answer requires a specification of the sense, then the question is ambiguous. In the following sentences, apply the above test. Which words in the statement require clarification before a Yes or No answer can be given?

a. God is dead.
b. Welfare programs destroy the recipients' desire to succeed.
c. Anti-pornography laws should be strengthened.
d. Love makes the world go round.
e. There is no such thing as a moral war.

6. Select one of the above statements and analyze the crucial words, their various possible meanings, their purpose in the context of the statement, and the author's intention in using those words. In a paragraph, attempt to describe the conditions under which one could answer Yes or No to the statement.

7. The following newspaper headlines are ambiguous because of their grammar or diction. Rephrase each headline, indicating the two meanings which it may suggest.

a. Student rates high
b. Man eating shark found
c. Nude swims tonight
d. Player shoots ace in tournament
e. Students stoned in Hollywood
f. Men's trousers half off Tuesday only
g. Race held inferior
h. Giraffe tastes sweet

8. The ambiguity of significance occurs in a statement in which the semantic value of the words is clear, but the full significance of the factual statement is not. For example, in the sentence, 'e number of heroin addicts in the United States rose 20 percent during the present administration," the fact is verifiable, but its significance is ambiguous. Does the figure 20 percent represent a drastic increase when compared with other comparable periods? Or does it represent a drastic decrease? How was the figure arrived at? Does the speaker have a special interest in making the statement, that is, is he pro-administration or anti-administration? Analyze the possible ambiguities in the following statements. what questions need to be asked about the fact contained in each?

a. The enemy suffered 735 fatalities last week; the friendly forces suffered seven.
b. Professionally performed abortions are about eight times safer than full-term pregnancies.
c. Unemployment is up to 2.5 million in the first quarter of this year.
d. They were soundly beaten in the roll call vote by 246 votes.
e. The 14 percent increase in hospital room rates this year indicates that hospitals are doing everything in their power to hold prices down.

9. In a short paragraph, write an analysis of one of the preceding statements, indicating the possible ambiguities which may be present.

---

CHAPTER 3
Define Your Terms!

The great practical problem of semantics, as we saw in chapter 2, is communication. We have seen how ambiguity is responsible for failures in communication. But communication fails for other reasons, too: we don't know what the other fellow's words mean; he doesn't understand what we mean; and we may not even understand what we ourselves are saying.

The point is that we ought to know what we are talking about. When we use high-level abstractions, words like "democracy," "freedom," "capitalism," or "communism," we should make our meanings clear. We should never forget that a word is like a check drawn against the world of experience, and that it has no meaning for us unless we can "cash it in" by pointing eventually to that to which it refers. When we speak or write, let us beware of "glittering generalities" which we do not understand, and let us not speak unless we know what we are Saying. It is this necessity for making our meanings clear that was in the mind of the French philosopher Voltaire when he said, "Before I will discuss anything with you, you must define your terms."

Definition is one of the convenient and natural techniques for developing your ideas in an essay. In your college writing you will frequently need to use words and phrases that have various meanings, or shades of meanings. To avoid vagueness or ambiguity, you will need to define your terms. Some words can be defined by a synonym or a phrase. Others, however, are too complex or abstract, and you will want to extend their definitions for a paragraph or longer. In any event, once you have faced the need to understand or explain a term-once you have defined your subject you will see the direction your discussion or theme should take. Equally important, your reader or listener will understand what you mean.

Vagueness
Communication often fails because words are ambiguous, or vague, or because they are used loosely, or carelessly, or without meaning. Ambiguity is not quite the same thing as vagueness. An ambiguous word is one capable of being understood in more than one sense in a given context. We are not sure which sense is intended. By a vague word, we mean one whose meaning may be fairly well understood, but whose limitations are unclear. Thus, the term "obscene" is vague, for we don't know where to "draw the line." A statute which forbids vehicles in a public park obviously applies to motorcycles. Bicycles also? Tricycles, Roller skates?

The use of most of the words in ordinary language becomes vague when we are confronted with "fringe" applications, but each word also refers to clear-cut examples, called "paradigm" cases, to which the word typically refers. There is no question that automobiles, airplanes, and railroad trains are vehicles, as are boats.

In mathematics and the sciences, and also in law, terms are precisely defined, but such words are taken out of everyday circulation. Words in ordinary speech cannot be defined with such exactness.

Vagueness can sometimes be eliminated by "drawing a' line." The expression "the West," as used in American history, is a vague term. Bernard DeVoto gave this term a specific meaning when he said that "the West begins where the average rainfall drops below twenty inches. when you reach the line which marks that drop-for convenience, the one hundredth meridian-you have reached the West."

There are some important words which are both ambiguous and vague, that is, they have several unclear meanings. This is a kind of "double-barreled" ambiguity. The words "freedom" and "liberty" fall into this category. Consider the famous "Declaration of the Four Freedoms" which mentions Freedom of Speech, Freedom of Religion, Freedom from Hunger, and Freedom from Fear. Note that the first two of these Freedoms have a "negative" connotation, referring to the limitation of the power of the state with respect to the rights of the individual. The third and fourth freedoms, on the other hand, imply that the state must take positive action to provide the people with food and security. Or consider words such as "liberal" or conservative." There are no universally accepted definitions of these terms; dictionaries can merely list their various uses.

Defining your terms
In serious discussions, when key words are subject to varying interpretations, i.e., when words like "liberal" or "conservative" are used, the speaker or writer who wishes to make his meaning clear should stipulate definitions of his terms. By "stipulation" we mean "specifying or particularizing": The speaker announces the precise sense in which he will use the word. lie should begin by saying, "In this discussion the word X will mean . . ." This is flow the meaning the word will have whenever he uses it. Three possibilities are open to the stipulator: (1) He may find one of the several customary meanings of a word adequate, (2) he may choose to stipulate a modified version of a customary meaning, or (3) he may stipulate a brand-new meaning.

Stipulations would be unnecessary if words had one and only one meaning. But since this is not the case, stipulations of definitions are often an indispensable element in making our meanings clear. The stipulator has a great deal of freedom, but great caution should be used in stipulating new meanings. Readers and audiences find it difficult to follow a speaker who uses words in unfamiliar ways, and the speaker's freedom is limited by his desire to hold his audience. Confusion is likely to result when new meanings are used, for old habits are hard to get rid of and we usually persist in giving familiar words their customary meanings. And worse: The stipulation of new meanings may involve dangerous traps for the innocent. There ought to be a code of linguistic ethics for all speakers and writers; when they stipulate new meanings, they should warn the reader that they are doing just that. The danger is that the stipulator may give us a new meaning as if he were merely giving us one of the customary meanings that all of us accept, and thereby get us to believe things we otherwise would not.

As an example of the stipulation of a new meaning without prior notification that this is being done, let us examine the following paragraph:

Many readers are quick to dismiss "confession" magazines as containing little of permanent literary worth. They forget that the confession story has provided us with some of the greatest realistic, revealing, personal literature of the ages-Cellini's Autobiography, The Confessions of Jean Jacques Rousseau, De Quincey's Opium Eater. Boil down to its essence any of the great enduring classics and you will find-a confession. Anna Karenina, Madame Bovary, Tess of the D'Urbervilles, Sister Carrie, Of Human Bondage-these are all confession stories-all based on human emotional problems, personal conflict, human desire, human greed, human passion.

The net effect of this passage is to make the reader think that the stories he finds in confession magazines on any newsstand-with their predictable plots, one-dimensional characters, banal dialogue, and superficial treatment of the human condition-are in the same genre and of the same quality as the world's greatest literature. How does the writer of this passage do this? By implying, in his last sentence, that any literary work based on "human emotional problems, personal conflict, human desire, human greed, human passion" fulfills the requirements of the confession story. In other words, because confession stories deal with the same topics as the classics he cites, they belong to the same literary classification. But such a definition ignores the differences in technique, depth of feeling and emotion, use of language, originality of plot, portrayal of characters, and all of the other characteristics we associate with great literature.

There is another type of trick that may be used in the stipulation of definition. The stipulator may take familiar words having favorable connotations-like "democracy," or "freedom"-and redefine them with a quite different content than is customary. The unwary may fall into the trap of carrying the favorable connotations of the old word to the new content, thinking that the new must be as good as the old because it is called by the same name. This, of course, is the fallacy of thinking that words are guarantees of things.

For example, we sometimes hear people say, "The Russians (or Chinese) have democracy, just as we have democracy, except in a different way. They have democracy as they define the word, namely, as a system in which 'the people,' rather than private capitalists, own the industrial plant." Now, of course, the Russians (or Chinese) may call their system a "people's democracy," and they have "democracy" in accordance with their stipulated meaning. There is no international law which forbids this kind of linguistic behavior. But they mean something quite different from what we mean, and they adopted this term "democracy" for very special purposes. Having their own aims in view, and knowing that the slogan of "democracy has great appeal to people because of the libertarian and egalitarian ideals it usually stands for, they adopted this word, but gave it a new meaning. The fact that the same word is used makes people feel that the thing is the same. And since we regard democracy as good, and they use the same word, the innocent may conclude that their system must be as good as ours, though in a different way.

When confusions of this kind occur in a discussion it is better to discard the word "democracy" altogether, and thus avoid the possibility of these confusions. The important thing is to compare the actual practices of the rival systems, regardless of what they are called.

We have been discussing some problems with respect to the stipulation of meanings. It is desirable for speakers and writers to tell us how they will use important words. They may stipulate a customary meaning or a new one. But when they stipulate brand new meanings without informing us that these are new meanings, we may be misled. Danger lurks in the failure to give us such warning.

Definitions, then, contain more than meets the eye. The definer's purpose, in other words, may be other than informative. lie may have propagandistic aims, and his definitions may be slanted according to his purposes.

There is also a type of expression which looks like a definition but which really aims at imaginative insight, to capture the "essence" of things. "Architecture is frozen music" is an example. "Poetry is music plus imagination" would not pass muster as an exact definition, although it may be more enlightening for some purposes than a more literal statement. And inventive imaginations have coined many amusing "definitions" which contain sharp social satire: "A politician is a man who sits on a fence with his ear on the ground"; "A wedding ring is a matrimonial tourniquet designed to stop circulation"; "An explorer is a bum with an excuse.

A definition is a statement that says, "For this word A, substitute these words, X, Y, Z." For the word 'perjury," substitute its legal definition "the wilful and voluntary giving of false testimony under oath or affirmation, with regard to a material matter, in a judicial proceeding." The single word is substitutable for the long expression, and vice versa. But there are different ways in which a definition may clarify meanings. When we turn to a dictionary for the meaning of an unfamiliar word, we find three kinds of definitions. We find synonyms, examples, and analyses of meanings. For example, if we look up the word "troglodyte," we may get a synonym: "hermit." Or suppose we look up "oxymoron." The American Heritage Dictionary tells us that an oxymoron is "a rhetorical figure in which an epigrammatic effect is created by the conjunction of incongruous or contradictory terms." This is an analysis of the meaning. But since this analysis is somewhat obscure, our dictionary helps us out by furnishing us with an example of an oxymoron: "for example, 'a mournful optimist."' (Other examples are, "To make haste slowly," "His kindness was cruel.")

Definition by analysis
Definitions, then, may consist of synonyms, examples, or analyses. Any one of these will do, provided that it makes the meaning clear and helps us to understand what the other fellow is trying to say. But the most enlightening type of definition is the analytical. This type is sometimes called definition per genus et differentiam, that is, it states the general class of things to which a term belongs, and then shows its differentiating characteristics within that class. When we define a triangle as a "plane figure having three sides," we have noted the class of things to which triangles belong: plane figures; and we have noted how they differ from other plane figures, that is, in being three-sided.

The most important logical problem in connection with analytical definitions concerns the nature of an adequate definition of this type. "A triangle is a plane figure having three sides" is a perfect definition of this type because it has the attribute of "convertibility." A convertible definition is one that can be "turned around" and still yield a true statement. Thus, we can say, "Any plane figure having three sides is a triangle." Convertible definitions show an "equivalence" between the definition and the word being defined. Thus, in Aristotle's definition of man as a "rational ,animal," if we agree that all men are rational animals and that all rational animals are men, this definition will exhibit equivalence and convertibility. Another way of checking for convertibility is to use the "All and Only" test. Can we say all men are rational animals and that only men are? If we can, then the definition is convertible.

A definition is inadequate if it lacks convertibility. It is then either too broad or too narrow. "Too broad" means that it covers too much ground, as in a definition of propaganda as "any talk or action which influences anyone toward some predetermined end." This definition covers things it does not mean to cover, as when I say to my neighbor at dinner:

"Please pass the salt." I have influenced him; have I thereby become a propagandist? (We cannot say that only propaganda involves influencing people or that all influence is propaganda; we may influence people with-out being propagandists.) On the other hand, a definition may be too restrictive, and not cover enough ground. But most faulty definitions will be found to be too broad. We find some element common to the items we are interested in and hastily define the thing in terms of those common elements without pausing to notice that our net has swept all sorts of other things into our definition.

Definition by synonym and example
We noted earlier that definitions by synonym or by example are often quite adequate. This is the case when all we desire is a reference to the sort of thing for which a word stands. But synonyms and examples will be wholly inadequate when an analytic definition is called for. For example: In a discussion of religion, someone may raise the question, "Just what do you mean by God?" An answer like "By the term, God, I mean the Deity" would be inadequate in most cases, for the questioner wanted an explication or analysis of the term. Definition by synonym is often as insufficient as was Polonius' response to the King and Queen when they questioned him concerning Hamlet's strange condition:

Your noble son is mad:
Mad call I it; for, to define true madness,
What is it but to be nothing else but mad?

Definition by synonyms is sometimes called "circular" definition, and results in a "begging of the question," a matter to which we will return in Chapter 7. An example: "A morally good man is one who acts virtuously." "Morally good" and "virtuous" are synonymous terms in this context, so that the definition merely repeats the word that is being defined. Circular definitions are of course "convertible equivalents," but they are faulty in that they offer no clarification of the meaning of the word being defined. If we are looking for clarification of the meaning of "morally good" so that we may know to what kinds of conduct it refers, it is not helpful to be told that "morally good" refers to "virtuous actions." This is like saying that 'virtuous actions are virtuous actions." When Hamlet tells his friends that he brings "wonderful news," namely, that "there's ne'er a villain dwelling in all Denmark, but he's an arrant knave," Horatio answers: "There needs no ghost, my lord, come from the grave to tell us this."

Closely resembling these faults in definition are such things as pleonasms and rhetorical tautologies, as in saying, "lie is writing his own autobiography," or "I have one small son, a boy." True, but foolishly superfluous. Even more amusing, usually, is the "Irish bull," a good example which occurred in the movie The Quiet Man: "He'll regret this to his dying day, if he lives that long." The absurdity of this lies in the fact that it denies a tautology, and a tautology, though vacuous, is necessarily true. And here is one more example, from Pliny, the ancient Roman: "It is better to be idle than to do nothing."

Synonyms, then, do not always satisfy our demand for a definition. "Definition by example" may also be inadequate: "What is poetry'?" "Milton's Paradise Lost is an example of a poem." This does not tell us much about the nature of poetry. Or suppose that we are asked to define "free enterprise," and we point to the United States as an example of a nation having free enterprise. We shall give more than we intend by this example of "pointing," for a stranger might conclude that legislation in aid of farmers was an essential part of a free-enterprise system. The vagueness of the reference that may accompany the gesture of pointing is well illustrated by a story, that has become a classic among students of language. It is narrated by J. H. Weeks, in his Among Congo Cannibals:

I remember on one occasion wanting the word for Table. There were five or six boys standing round, and, tapping the table with my forefinger, I asked, "What is this?" One boy said it was a dodela, another that it was an etanda, a third stated that it was bokali, a fourth that it was elamba, and the fifth said it was meza. These various words we wrote in our notebook, and congratulated ourselves that we were working among a people who possessed so rich a language that they had five words for one article.

But later Weeks discovered that

one lad thought we wanted the word for tapping; another understood that we were seeking the word for the material of which the table was made; another had an idea that we required the word for hardness; another thought we wished a name for that which covered the table; and the last, not being able, perhaps, to think of anything else, gave us the word meza, table--the very word we were seeking.

There are of course some situations in which an analytic definition cannot even be attempted, and where only a definition by example (or pointing) is possible. If we are asked what "chrome yellow" means, no words can designate its sense qualities. If this is not obvious, ask your-self how you would explain what any color is, to a person who had been blind from birth. We can explain what sound and light waves are to a blind person, for he can understand these things in terms of his sense of touch, but he cannot understand what we mean by color, for this depends on a sense he lacks.

In concluding our discussion of definition we must issue one final warning. The history of philosophy may be regarded as the record of man's search for adequate analytical definitions of the key terms in human discourse, words such as "truth," "beauty," and "goodness." Let us not hastily assume that "the last word" has been spoken in defining these terms. The "last word" has not been spoken on these matters, and probably never will be. And let us not dogmatically assume that we have adequate definitions of any term whose meaning is a matter of controversy.

Nor should we demand precise definitions of that which lacks precisely determined characteristics. "Art" is in this category. After attending a few exhibitions of "modern art" many spectators experience a sense of bewilderment, and they are apt to raise the question: "Just what is art?" The dictionary will not be very helpful here. According to the Random House Dictionary, art is "the class of objects subject to aesthetic criteria; works or objects belonging to this realm, as paintings, drawings, etc.: a museum of art." Not a very useful instrument for distinguishing art from not-art! There is no definition which can precisely delimit art from that which is not art. And the same problem will be found to arise in connection with many familiar words. To insist on formulating and applying precise definitions in situations in which they are inappropriate is to fail to make sense.


FOR DISCUSSION AND WRITING
1. The meaning of a word is determined largely by its context. As the context changes, so too a word's signification may change, however slightly. Interpretation of a writer's or speaker's definition of a word often depends upon its use in the sentence, paragraph, or larger unit of communication. If you were to define the word grable as it is used in the following five statements, what definition could you provide? Write a twenty-word definition, paying careful attention to the possible shades of meaning the word may suggest.

a. To grable before the enemy is hardly heroic.
b. She grabled for several years, but her psychiatrist finally was able to help her.
c. I am not sick; I grable because I am naturally high-strung.
d. The tender leaf grabled before the wind's force.
e. Transmission grabling? See Honest Doug!

In a similar manner, define krafic on the basis of its use in the following sentences:
a. It was krafic enough to see the mountains.
b. His argument was krafic but his language was atrocious.
c. I have never seen a more krafic young lady. I would like my son to marry her.
d. Did you krafic the inside as well?
e. The vernal glade kraficked my soul.

2. How is democracy used in each of the following sentences?
a. The United States is a democracy.
b. In a democracy any man or woman can speak freely.
c. In spite of the fact that he is a member of the nobility, he takes a very democratic view of the man in the street.
d. The Democratic party uses the donkey as a symbol.
e. The Students for a Democratic Society did not allow the speaker to continue.

3. Which of the following offer a definition of some term? What method is used to explain the meaning of the word (definition by synonym, analysis, example, or other method)?
a. Happiness is an activity of the soul in accordance with virtue in a complete life. (Aristotle, The Nicomachean Ethics)
b. "babble": chatter
c. Science is what we know and philosophy is what we don't know. (Bertrand Russell, Bertrand Russell Speaks His Mind)
d. "Network" means anything reticulated or decussated, at equal distances, with interstices between the intersections. (Samuel Johnson's Dictionary)
e. "cosmetic": rouge, lipstick, face powder
f. A liberal is a man who cultivates the skills that make freedom operational. He is always a man on special assignment. (Max Ascoli, editor and publisher of The Reporter magazine)
g. The root difference between the Conservatives and the Liberals of today is that Conservatives take account of the whole man, while the Liberals tend to look only at the material side of man's nature. (Barry Goldwater, The Conscience of a Conservative)
h. "awkward": clumsy
i. Philosophy is a battle against the bewitchment of our intelligence by means of language. (Ludwig Wittgenstein, Philosophical Investigations)
j. "philosopher": such men as Aristotle, Plato, Kant, Dewey, and Jaspers

4. Select one of the following words or phrases and, in a paragraph of no more than 200 words, define it so that your reader knows exactly the limit of your use of the word.
a. adolescence
b. university
c. the common man
d. conservative
e. religion
f. law
g. liberty
h. idealism
i. the American way of life
j. love

5. For each of the following terms supply a synonym, an example, and an analysis.
a. pop art
b. automobile
c. sin
d. honor
e. a delicious meal

6. Give examples of definitions that are circular, tautological, and lacking convertibility. You may cite those you encounter in your reading or supply your own.

---

CHAPTER 4
What Kind of Language Are You Using?

A misunderstanding may occur because we fail to grasp the meaning of a word, or the thoughts embodied in a sentence. But there is another obstacle to communication: we misunderstand the purpose of speech. The most typical form of this mistake is to treat every use of language as if it were intended to convey information. For example, many people read a poem as if it were a scientific treatise. But the criteria of truth and falsity may be irrelevant to the poet's purpose, so far as his actual statements are concerned. He may be trying to evoke a mood, or a state of feeling, or attitude, rather than to give us literal truths. When Shelley writes of the skylark,

Hail to thee, blithe spirit!
Bird thou never wert,

he did not really mean to deny that the skylark is a bird. His language was "emotive" rather than scientific in its purpose. What we are emphasizing here is that we shall do well to look for the purposes and intentions of speakers and writers. In this chapter we shall consider the different functions, purposes, and uses of language.

The functions of language
Language, in other words, has more than one purpose. We might say that language operates on different levels, except that the word "levels" suggests higher and lower planes in a scale of value, and this is not intended here. We shall deal with three functions: the informative the expressive, and the directive. To say that language has three functions is to say that there arc three different reasons for speaking. One reason, or purpose, is to communicate factual information. This is the informative function, probably the most important for our purposes. We speak also in order to express our feelings, to "blow off steam," or to stir the feelings and attitudes of the person we are talking to. We shall call this the expressive or "emotive" function. And, finally, we speak in order to get people to act. This is the directive function.

Some illustrations are in order. A book on astronomy describes the solar system and the stars. We learn that the diameter of the earth is about 8,000 miles, that of the sun, about 800,000 miles, a ratio of 100 to 1. We learn that the star Betelgeuse has a diameter three hundred times that of the sun. This means that if the earth is represented by a baseball, about three inches in diameter, then Betelgeuse would have a diameter of almost a mile and a half. We may learn that there are as many stars in the heavens as there are grains of sand on all the seashores of the world . . . I have just been using language to communicate information. This use of language tries to avoid words with private connotations or words that unnecessarily stir up feelings. If you want your writing to be clear, do not use words that distract your readers' attention from what you are saying to how you are saying it.

Expressive language is a second type. When I talk about the United States senator I like least, I may let off some steam and relieve my pent-up feelings. I may even infect you with my feelings, making you feel as I feel. The poet, of course, is a specialist in expressive language, as in the lines:

Comes the blind Fury with th' abhorred shears
And slits the thin-spun life.

These lines give expression to John Milton's feelings and perhaps make us feel as he felt. When we tell our friends a funny story to get a laugh, we express our feelings too and affect theirs. Expressive language often occurs in political speeches, in plays, and in advertising. When you are writing an argument, you should avoid whenever possible the use of expressive language.

The third type, directive or action-provoking speech, is illustrated by examples like: "Do unto others as you would have others do unto you, or "Drink to me only with thine eyes." We say these things to get action. Ceremonial language, such as "I am happy to meet you, "What a beautiful baby!" and conversation about the weather, also have a directive purpose: to establish social rapport and to get a friendly response. We use directive language when we are explicitly and overtly trying to get some-one to act a certain way or to refrain from acting in that way.

There are, then, at least three different purposes of discourse. We may also make a somewhat similar classification for words, that is, for words taken by themselves. A basic distinction here is between what we shall call neutral words and emotive words. Neutral words merely convey ideas to us, as when I say, "The sun rose at six this morning." The words in this sentence do not arouse our emotions. But words like "God," "love," "freedom," and "family," are so closely connected with our total attitudes to life that they are likely to arouse emotional reactions. This division of words into neutral and emotive, however, is relative to our personal experiences, for there is nothing in the world itself which makes it neutral or emotive. If a word conveys nothing but an idea to you, then it is neutral to you; if it arouses your emotions, then it is emotive to you. The word "bread" is a neutral word to me, but to a "fat boy" or a starving man, it may be fraught with emotion. Nevertheless there are some words which can be counted on to make almost everyone "see red," so to speak, like the word "traitor.

This classification of words is independent of our classification of the functions of language, for those who wish to inform may use either type, as may those who want to express their feelings, or to get action. In general, however, neutral words will be used when we wish merely to inform; emotive words when we wish to be expressive.

Let us return to our classification of the purposes of language. And let us avoid the vice of oversimplification. In life, or living speech, the functions of language are seldom found in a pure or unmixed form. In life things are rarely simple and never pure. Speech and writing usually present mixtures of the informative, expressive, and directive functions of language. Consider the informative item concerning the diameters of the sun and the earth. Though this language informs, and though it is not its primary purpose to stir our emotions, our feelings may nevertheless be stirred when we learn from Sir James Jeans that there are as many stars as there are grains of sand on all the seashores of the world, for this knowledge may make us realize how infinitely vast is space and how infinitesimally small and feeble is man, crawling for his brief day upon an insignificant planet.

Nevertheless, informative language is the type most likely to be found in a relatively pure form. The writing of scientists is apt to be purely informative, especially in the physical sciences, like physics and chemistry, though less so in political science. But expressive language is rarely, if ever, used exclusively. Expressive language is usually mixed with something else. The lines from Milton's "Lycidas" did not aim primarily at giving us information, but they do say something that has the ring of truth in it. They tell us that men are mortal and that there are forces beyond our control at work upon us, which give us the helpless feeling that we are the pawns of fate. Alexander Pope's "Essay on Man" is philosophical discourse in rhyming couplets and contains a developed system of thought, as in the closing lines of the First Epistle:

All Nature is but art unknown to thee,
All chance, direction which thou canst not see;
All discord, harmony not understood,
All partial evil, universal good.
And, spite of pride, in erring Reason's spite,
One thing is clear, whatever is, is right.
The poet may also mix a directive purpose with the expressive one. He may want us to do something, as in so-called "inspirational poetry." Clough's lines are an example:

Say not, the struggle naught availeth,
The labour and the wounds are vain
The enemy faints not, nor faileth,
And as things have been, they remain.
If hopes were dupes, fears may be liars;
It may be, in yon smoke concealed,
Your comrades chase e'en now the fliers,
And, but for you, possess the field.
The last line of this poem, "But westward, look, the land is bright!" was quoted very effectively by Winston Churchill during the dark days of the Battle of Britain in 1940. This poem mixes the directive and expressive types of discourse. But one should not expect all poems to give us a moral message, or practical advice, or scientific information. Poets are not necessarily preachers, though some are, and they are not necessarily scientists or philosophers, though some may be. Wordsworth's "Daffodils" should not be read as if it were a botanical treatise on the "Narcissus pseudo-narcissus of the amaryllis family," to use the technical name for the daffodil. One of the aims of poetry is to communicate feelings and attitudes toward life, to convey to us the poet's feelings concerning his experiences, and to make us aware of life's mystery and wonder.

More about directive language
We shall now examine the directive type of language in some detail. When a speaker wants action from his audience, he may tell them to do what he wants them to do. But, as every parent knows, it is often better to use an indirect approach to get action. Instead of saying, Johnny, eat your spinach" (or whatever other torture-food happens to be the vogue at the moment), we say, "My, what a gorgeous-looking dish of spinach that is." (The method suggested here cannot, unfortunately, carry a guarantee with it.) Similarly a politician may ask us to work for him, to help him win election; but he may also use an indirect approach. He may stir our emotions by painting a vivid picture of the horrible crimes committed by the opposition party. He arouses our emotions, and these emotions, the psychologists tell us, will demand an outlet. The person thus aroused wants to go out and do something.

Now, this technique of getting action by working on men's emotions has been known to mankind since time immemorial and has been practiced by politicians ever since they made themselves indispensable. An excellent illustration of this technique is found in Mark Antony's funeral oration in Shakespeare's Julius Caesar. Brutus, you will recall, assassinated Caesar because he feared that Caesar had ambitions to make himself dictator of Rome. Brutus is now in power, but graciously permits Mark Antony, Caesar's friend, to make the funeral oration over Caesar's dead body. Antony, however, is not primarily concerned with eulogizing Caesar. He uses the occasion as a step toward seizing power for himself and seeks to turn the Roman populace from hatred of Caesar to hatred of Brutus and his fellow conspirators. The oration begins with the famous lines:

Friends, Romans, countrymen, lend me your ears:
I come to bury Caesar, not to praise him...
He was my friend, faithful and just to me.
But Brutus says he was ambitious;
And Brutus is an honorable man....
You all did love him once, not without cause.
What cause withholds you then to mourn for him? ...
My heart is in the coffin there with Caesar,
And I must pause till it come back to me.

Antony is beginning to awaken the emotions of his hearers. He quotes Brutus, sarcastically. He mentions the sacred bonds of friendship which bound him to Caesar and which once bound the crowd to Caesar. Antony then goes on to mention Caesar's will and says that it leaves generous bequests to the people, but, he adds, he cannot possibly read this will aloud. If the people but knew what Caesar had done for them, he says, they would not be able to control themselves, for they are not made of "wood" or "stones" but of flesh and blood. The mob is now inflamed with expectancy and demands that the will be read, but Antony puts them off, first showing them the cloak which Caesar wore when he was killed. He points to the holes made by the stabbing daggers:

Look! in this place ran Cassius' dagger through...
Through this, the well-beloved Brutus stabb'd...
And, as he pluck'd his cursed steel away,
Mark how the blood of Caesar follow'd it.
For Brutus, as you know, was Caesar's angel:
Judge, O you gods, how dearly Caesar lov'd him.
This was the most unkindest cut of all.

Tears begin to flow, and Antony now reads the will, with its generous bequests to the people. The crowd leaves Antony in a fury, resolved to destroy Brutus and the assassins who killed their beloved Caesar. As they leave, Antony, knowing that he has accomplished his purpose, mutters to himself:

Now let it work: mischief, thou are afoot,
Take thou what course thou wilt!

Antony, like other demagogues, is a master of practical human psychology who knows how to move the masses to his own ends. This kind of psychological insight is often turned to bad uses. But it should also be part of the equipment of anyone who wishes to be effective in moving people to action. Even in the best of causes human beings need some stimulus to action.

In the next chapter we shall distinguish between legitimate and illegitimate appeals to emotion. It should be obvious, however, that emotional appeals are sometimes quite proper, as when the Community Fund and organized charities get us to contribute by appealing to our hearts.

It is well to know when speakers are trying to get us to do something, rather to get us to inform us. And also-to know when they are trying to get us to do nothing. This is a reverse kind of directive language. And just as ordinary directive language seeks to arouse emotions, so the reverse type seeks to neutralize emotions, or to de-emotionalize a situation. For example, a shocking crime occurs in our city: an important public official is arrested for embezzlement. There will be a great public clamor for drastic action. Those who fear for the status quo will seek to d am pen public indignation. We will be told that "the authorities are investigating," etc.

The negative use of directive language often characterizes official government papers" concerning actions which seem immorally brutal. The actions will be described in a calm, unemotional manner. Neutral words will prevail, and emotive terms will be avoided. This use of language was often resorted to by the Johnson and Nixon administration spokesmen when they sought to gloss over the harsh realities of the Vietnam war. "In our time," George Orwell says, "political speech and writing are largely the defense of the indefensible." Neutral language helps to immobilize our emotions.

Propaganda techniques
Our distinctions between types of language may help to clarify some aspects of "propaganda analysis." The meaning of propaganda is extraordinarily vague, at least so far as common usage is concerned. Most people use the word "propaganda" without having any precisely defined meaning in mind. In everyday speech this word may mean anything from "a pack of lies" to "the attempt to influence anyone about anything." But these are inadequate definitions. Not every liar is a propagandist, and we do not usually think of a hostess urging us to have another helping of her culinary specialty as a propagandist. She is trying to influence us about something; but unless the word "propaganda" is more limited in its meaning, it will be quite useless, and ought to be dropped from the language. "Trying to influence others" is almost coterminous with speaking.

The usual dictionary definition is better: "Propaganda is a systematic effort for the gaining of public support for a course of action." But even this definition is not accepted by everyone. This lack of agreement concerning the meaning of propaganda may be demonstrated by the following experiment: Co to any library and find twenty books in the fields of government, political science, sociology, and social psychology which contain the word "propaganda" in the indices. You will find twenty different definitions of the word. Some years ago a public-spirited citizen of New York offered a thousand-dollar prize to anyone who would define the word in such a way as to win general acceptance. There were no applicants.

We shall not offer a twenty-first definition here. But we can point out some of the things we ought to have in mind when we suspect that we are on the receiving end of propaganda. The word, after all, is of little consequence; it is the thing that counts. Just what is it that we are suspicious of, when we think of propaganda as being in some fashion dishonest, as most people do? This is what concerns us here.

As thoughtful citizens, we want to know the facts, so that we can come to intelligent decisions. We are afraid that there are organized efforts to deceive us so that we will act blindly and unintelligently. What can we do about this? The answer is quite a simple one: become as well-informed as possible. Knowledge is needed to see through a lie, whether a big or a small one, and nothing else will do the job.

In recent years, with the public's increasing awareness that it is being victimized by "propaganda"-understood as a form of organized deception-many writers have sought to help the public in its search for protection against deception. These analysts have sought to give us a method for detecting propaganda and thereby to achieve security against its harmful effects. We are told to watch for such techniques or "tags" as "name-calling," "glittering generalities," "testimonials," "band-wagon appeals," etc. Let us look at the meaning of these categories.

Name-calling, or "labeling," refers to the practice or attaching "bad" names to individuals or groups or ideas: names such as "Fascist, reactionary," "totalitarian, racist. These words stir feelings of fear and hatred for the persons and ideas to whom they are attached. The abusive terminology makes people forget about the necessity for knowing what the facts are. But name-calling, though a popular instrument of propaganda, is not confined to propagandists. We all indulge to some extent) even when we call a man a "name-caller."

We ought not to call names, of course, as a substitute for giving evidence. And the rules of courtesy (not to speak of the laws against slander and libel) tell us that calling names is a boorish as well as an illogical way of making a point. But it is also very easy to exaggerate these warnings. Some people use the category "name-calling" for the use of any uncomplimentary term applied to anyone. But it is not improper to characterize a man as a Fascist when this word is adequately defined and there is proof that the statement is true. When a man is called a Fascist merely because he espouses views that are less liberal than our own, however, we exhibit intellectual irresponsibility.

Similar considerations apply to the other techniques. A "glittering generality" is a smug generalization, illustrated by "Woman's place is in the home," or "What is good for business is good for the country." There is a wisecrack which says that generalizations are always false, and statements like the ones just quoted undoubtedly ought to be qualified. But certainly propagandists have no monopoly on the use of such generalizations. Finally, consider the "testimonial." This may be worthless as evidence, especially when it comes from a non-expert or unqualified source, but the opinions of qualified persons, on the other hand, are worthy of respect. We should always consider the source, that is, the competence of the testifier, the probability of his being prejudiced, etc. But testimonials are not necessarily illegitimate.

The analysis of these "propaganda techniques" calls for our attention to the possibility that we may be permitting emotion to sway us, and we are reminded that we ought to look into the evidence. But language alone does not distinguish propaganda from other forms of discourse. Propagandists may avoid these techniques altogether, and non-propagandists may use them.

What is meant by propaganda in this discussion? Its meaning may be made clearer by contrasting the propagandist and the educator. There are some, of course, who deny the distinction and who tell us that "the advocacy of what we believe in is education; the advocacy of what we don't believe in is propaganda." But this is not what most of us mean by these words. An educator is one who, in the phrase of Robert Hutchins, seeks to teach people how to think for themselves." An educator wants people to seek the truth. To this end he will present them with facts; he will appeal to their reason; he will follow an argument to whatever conclusions may be warranted by the evidence. An educator will, or ought to, have his own point of view, his own preferences, and he will recommend his personal beliefs to his audience, but he will state the grounds on which he holds these beliefs, and he will state the major objections to them. Thus his students will be able to judge for themselves concerning the validity of his arguments and the truth or falsity of his beliefs.

We have been speaking of an ideal educator, a truth-seeker. But a propagandist, in the strict sense, is not interested in the truth for its own sake, or in spreading it. his purpose is different. He wants a certain kind of action from us. He doesn't want people to think for themselves. He seeks to mold their minds so that they will think as he wants them to think and act as he wants them to act. lie prefers that they should not think for themselves. If the knowledge of certain facts will cast doubts in the minds of his hearers, he will conceal or ignore these facts.

It may be said that there are no educators in this ideal sense and that, really, everyone is a propagandist. "The propaganda with which we agree is called education; the propaganda with which we disagree is called propaganda." If we accept this notion, we are forced to deny the distinction between tricksters and truth-seekers. This confusion of categories may be fostered by those who are afraid of the truth and who therefore want us to disbelieve whatever we read about them. If they can get us to believe that "everything is propaganda," we will believe nothing, including the truth about them. But to accept the wholesale skepticism suggested by the phrase "everything is propaganda" is just as foolish an attitude as to be completely unskeptical. There are two errors we ought to avoid: to be too trusting and to be too skeptical. Some people believe everything they read in the papers, and others believe nothing. We must learn to be discriminating-to distinguish between what it is reasonable to believe, what it is reasonable to doubt, and what we ought to dismiss as probably false.

To sum up. We have noted that there are three kinds of language, or purposes in speaking, and we examined the distinction between neutral and emotive words. We saw how emotion is employed in order to get action. Propaganda, we saw, has a directive purpose, for the propagandist wants us to act. The question, "What shall we do in order to protect ourselves against propaganda?" is misleading. The question assumes that propaganda is bad, and we cannot say that this is so without making certain distinctions. Insofar as propaganda seeks to get us to act by emotional appeals coupled with a concealment of facts-facts that might make us think about the merits of the proposal-it is "bad" as a method. It is possible of course that the propagandist may have our best interests at heart, so that his goal may be a good one.

Another reason why it is misleading to speak of "protection against propaganda" is that this implies that there is a special kind of defense against propaganda There is no magic amulet whereby one may exorcise its evils. The only defense against harmful propaganda is to add to our knowledge and to sharpen our critical abilities. We shall then know how to protect ourselves against the various forms of hokum.

The Chinese have a proverb which says that there are three sides to every question: my side, your side, and the right side. Though the right side is hard to find, to seek for it is legitimate, and we should have the confidence that we may find it if we try hard enough.


FOR DISCUSSION AND WRITING

1. This chapter describes the three functions of language: informative, expressive, and directive. Using these terms, clarify the following passages according to their function.

a. He ran the last mile in a blistering four minutes eight seconds, finished in 13:22.8-and, in the process, clipped seven seconds off his old record.

b. Go and catch a falling star,
Get with child a mandrake root,
Tell me where all past years are,
Or who cleft the Devil's foot,
Teach me to hear mermaids singing,
Or to keep off envy's stinging,
And find what wind
Serves to advance an honest maid. (John Donne, Song)

c. "I was obligated to hire a team and a man for the plowing though I held the plow myself. My farm outgoes for the first season were, for implements, seed, work, etc., $14.7232'. The seed corn was given me. This never costs anything to speak of, unless you plant more than enough. I got twelve bushels of beans, and eighteen bushels of potatoes, beside some peas and sweet corn. The yellow corn and turnips were too late to come to anything." (Henry David Thoreau, Walden)
d. Light up with a Lucky.
e. For America to survive at all in the future it must develop representative proportional government and evolve to equitable distribution of wealth. That is, simply, democracy and cooperative economics.

2. You have been asked to evaluate a product for a consumer protection organization. Write an informative report describing the article, its performance, durability, value, etc.

3. You have been asked to write a description for the product above (#2) which will be used in a sales brochure to be used by salesmen. Write a persuasive (directive) report.

4. In the following advertisements, classify the copy writer's use of language according to the divisions established in this chapter. Are the distinctions clearly delineated, or blurred? To what purpose?
a. Nearly half of the new cars sold in America last year were hard-tops. The public, it seems, is in love with hardtops. At Volvo, we're not. As far as we're concerned, the best way to build a safe car is to build a strong car. So Volvos have six steel pillars holding up the roof. Each one is strong enough to support the weight of the entire car. These pillars are part of a box construction that surrounds and protects the passenger compartment. A Volvo's body is fused together by 10,000 spot welds. And when you build this kind of strength into a car's body, it holds up. Are you in the market for a hardtop? Or is what you really want a hard top?
b. Salem uses only natural menthol, not the kind made in a laboratory. Like our rich, full-flavored tobaccos, our menthol is naturally grown. Then we blend natural menthol with our superb golden tobaccos. It is a unique blend found in no other cigarette. A blend that gives Salem a taste that's never harsh or hot . . . . A taste as naturally cool and fresh as Springtime.
c. Ever since the U.S. Government began testing the tar and nicotine levels of cigarettes, True has been lower in both tar and nicotine than 99% of all other cigarettes sold. It was True in 1969. True in 1970. True in 1971. And it's still True in 1972.
d. This isn't just a new model of an old favorite. This Jeep Commando is a whole new vehicle. Take its strength. Jeep guts make this Commando the strongest one ever built. With a hefty 232 CID 6-cylinder engine as standard equipment. And a 304 CID V-8 as a mighty option. That's power. They'll take you places you never dared go before. Take its looks That new front end makes the Commando more stylish than ever-along with the nine bright, up-to-date colors you have to choose from. This 4-wheel drive vehicle looks at home-at home! And take the interior. The Commando adds more of everything you want . . . . And more luxury, too, with extra trims and options. Altogether, it's the new Jeep Commando-the most exciting new 4-wheel drive vehicle in America.

5. In the following pairs of words, each word signifies approximately the same meaning as its mate. However, one word has a pejorative connotation, and the other an ameliorative or more pleasant meaning. Using a dictionary, describe the differences in meaning.
a. satisfaction-surfeit
b. statesman-politician
c. indoctrinate-educate
d. learned-sophistic
e. foreign-alien
f. beliefs-dogma
g. innocent-naive
h. provincial-local
i. hint-insinuate

6. The following sentences are emotionally neutral. Rewrite each sentence keeping its original meaning, but giving the sentence (a) an unfavorable connotation, and (b) a favorable connotation.
a. Frank enjoys words, and his vocabulary reflects his enjoyment.
b. The mayor asked the protesters to allow him to speak.
c. The cost of living rose three percent last month.
d. Professor Hogins always spends three weeks discussing human reproduction with his biology class.
e. The businessmen of Collinsville petitioned the state legislature to lower corporate taxes.

7. Write a description of an automobile accident from these different perspectives: from the point of view of the newspaper reporter who covered the story, from the view of a legislator who cites the accident to support his thesis that fifteen-year-olds are too young to drive, and from the point of view of the fifteen-year-old driver who believes that he is innocent.

---

PART TWO
The Argument

CHAPTER 5
How Not to Argue

Much of your college writing will be in the form of argument--that is, a series of statements, some of which serve as reasons for others, leading to a conclusion based on those reasons. Good argumentative writing depends, to a large extent, on logic. But this raises an important question: must all writing and speech rely exclusively on logic and reason? Is there no place for emotion or feeling? The following historical sketch presents this dilemma rather succinctly.

In the spring of the year 399 B.C., a famous Greek philosopher was put on trial for having committed two crimes. One was impiety to the gods of the state; the other was the corruption of youth, by teaching them impiety. The penalty for a conviction on these charges was a severe one, possibly death. The prisoner's name was Socrates, and he was seventy years old at the time.

There were other reasons, political reasons, for trying Socrates. He had been associated with the old aristocratic regime, now overthrown by the democracy, and he was held in suspicion as a critic of the democracy. Among other things, he became unpopular for his strange doctrine that even politicians ought to know what they are doing.

Socrates was reputed to be the wisest man of his time. This reputation surprised him, he said, for he considered himself to be an ignorant man, ignorant of the answers to the supreme questions concerning human happiness and human destiny. But he was also sure that no one else knew the answers to these questions, and this furnished him with an explanation of his reputation as a wise man. Though he was ignorant, he alone knew that he was ignorant, whereas other ignorant men did not know that they were, thinking they had all the answers.

The court which tried Socrates was made up of 501 of his fellow Athenians, the 501st man being added in order to avoid a tie vote. As Plato reports the trial in his short masterpiece "The Apology," Socrates vigorously denied that he had done anything wrong. He denied that he was guilty of impiety to the gods, and he denied that it was wrong to carry on free and open discussions with young men. He was convinced that he was right in what he was doing, and he was sure that he could convince his judges, by logic and reason, of the justice of his cause. Throughout his life he acted on the principle that clear thinking is in-dispensable for right living, and that human life without the joy of thinking is a life not worth living. Let us look at one passage from his speech to the judges at his trial:

"Perhaps some of you," he says to his judges, "when you appeared before the judges in a similar situation, begged and besought the judges with many tears, and perhaps you brought your children into court to arouse the compassion of the judges. But I will do none of these things, though I am in peril of losing my life. I too have children, but nevertheless I shall not bring any of them into court to beg you to acquit me. This is not because I am stubborn, my fellow Athenians, or because I lack respect for you, but because I think it disgraceful for respected people to act in that manner. But, apart from the question of reputation, gentlemen, I think it is not right to implore the judge or to get acquitted by begging; we ought rather to inform the judge and convince him. For the judge is not here to grant favors in matters of justice, but to give judgment; and his oath binds him not to grant favors according to his pleasure, but to judge according to the laws."

Socrates, in other words, used a rational and logical approach in presenting his case to the jury. This was the speech of a rational man. Socrates presented the evidence, and refused to indulge in an emotional harangue in his own defense. In other parts of his speech,. however, he did present an eloquent defense of the free pursuit of truth, and he also goaded his accusers with withering sarcasm. Socrates' spirit of reasonableness and, incidentally, his sense of humor, did not desert him even after he heard the sentence of the court that he be condemned to die by drinking a poison made of hemlock. An anecdote bears witness to this. After the verdict of guilty was returned, one of his disciples, Apollodorus, who was at the trial with him, exclaimed, "But what I find it hardest to bear, Socrates, is that I see you being put to death unjustly." Socrates replied, "Was it your preference, Apollodorus, to see me put to death justly?'.

Logic or emotion?

The question may be raised: Was Socrates' rational approach the proper one under the circumstances? Should one use a "logical" approach, or should one use an emotional appeal in presenting a case to a jury? We are all of us familiar with that favorite of the cartoonists, the glamorous blonde witness who sits in the witness chair with her beautiful legs crossed provocatively. This, one might say, is the proper technique for presenting a case to a jury, especially when the jury is an all-male one.

Before we try to answer our question, it will be instructive to compare Socrates' approach with that of Mark Antony (see the example). The funeral oration was an emotional appeal. Antony achieved the goal he sought. Now, can one say that Socrates used the wrong technique and Antony the right one? If Socrates had had you as his lawyer, would you have advised him to appeal to the emotions of the court, consisting, as it did, of a large crowd? It all depends, of course, on what one is after. The question, "Did Socrates use the proper approach?" is somewhat ambiguous. "Proper approach" has at least two meanings. "Proper" may refer to an efficient technique for achieving a goal, whatever it may be, or it may mean what we ought (in a moral sense) to do.

Taking people as they are, and desiring to mold them to ones purpose, an emotional appeal may be more effective than a rational one. Unscrupulous demagoguery may get results. One might wish it were other-wise, but in the real world, as distinguished from an ideal society, such is often the case. But talk also has a moral aspect. There is a moral obligation to tell the truth. And there is also the clement we call honorable conduct." Though few would have condemned Socrates if he had obtained an acquittal by an appeal to the court's emotions, he preferred death to what he considered dishonorable conduct. His sense of moral integrity did not permit him to compromise with his principles, and he became a martyr to the principle of unswerving devotion to the truth. To the truth, that is, as he saw it.

Thus the question, "Who used the proper approach?" depends upon the meaning we give to "proper." But there is also a larger issue. Logic has its proper place, and so has emotion. A purely intellectual approach to life is as insufficient as a purely emotional one. The activities of life may be divided into two broad categories, the logical and the nonlogical. By "nonlogical" we do not mean 'illogical," but rather activities that have nothing to do with logic. There are times when we reason and argue and draw inferences. But, for the most part, we are engaged in nonlogical activities, like eating and sleeping or narrating the events of the day, and so on. Logic enters only when we give reasons for our beliefs.

When we give reasons for our beliefs, we are reasoning. Reasoning is either logical or illogical. Illogical reasoning is bad reasoning, but the nonlogical has nothing to do with reasoning. When we seek to prove that something is or is not the case, then we engage in argument, in which we say: This is true because that is true, or this is so because that is so. When the reasoning is adequate, we say it is logical, when not, illogical.

The law of rationality
When we assert beliefs which may be questioned, then we have an obligation to be rational. This common human obligation may be stated in the form of a "law of rationality" or "law of argument," that we ought to support our beliefs by adequate evidence. When we say that we know that something is true, we ought to be able to justify our belief by adequate evidence. What is adequate evidence? This term is best defined by example, and we shall give examples as we go along, but we shall assume here that we agree pretty well as to the distinction between evidence that is good and sufficient and that which is not. In the end there is only one court of final appeal in settling a problem concerning what is rational and what is not-the community of reasonable men. Fortunately, the human race has always agreed pretty well on which of its members are reason-able and which not.

Logic is not all, then, but we have a common obligation to be logical when logic is relevant. In the last chapter we saw how emotional language is used to get action, and we raised the question as to when emotional appeals are appropriate and when not. An emotional appeal under circumstances like those portrayed by Shakespeare is highly improper, for the mob did not know what the facts were. They came believing that Caesar was an evildoer, and they left convinced that Brutus was. But they did not revise their judgment on the basis of evidence. When the facts' are in dispute, we ought to demand information and evidence rather than emotion. The Roman mob thus fell short of the human obligation to be rational. Before acting they should have asked themselves some questions concerning the facts: Did Caesar really aspire to be a dictator? Did Brutus seek to save their freedom? What evidence is there for or against the is-sues involved in these questions? But Antony foreclosed the inquiry by substituting emotion and passion for reason. And when we act on emotion without concerning ourselves with the facts, we are likely to rush into disaster.

Usually, when a politician does what Antony did, that is, when he substitutes emotional appeals for proof, propaganda for rational persuasion, when he inflames rather than informs, we shall find that he does so for one of two reasons. Either he has a contempt for the people, treating them as if they were children, incapable of understanding the issues, or he doesn't want them to know the truth.

We are not saying that emotional appeals are never appropriate. On the contrary. When the facts are not in question, and action is desired, then an emotional appeal is appropriate, even indispensable. In the critical days during the "Battle of Britain," Prime Minister Winston Churchill made his great 'blood, toil, tears, and sweat" speech to the British people. He inspired his people and spurred them to heroic efforts. Emotion is the best fuel for this kind of energy, and this kind of stimulus is needed even the best of causes.

How not to argue
Let us pause for a moment to get our bearings. The law of ration-is the central core of the rational approach: When we assert that a belief is true, we should be prepared to support our belief by adequate evidence. But the law of rationality is frequently violated, and it may be evaded. All of us hold many beliefs that are unsupported by evidence, and we sometimes argue as if evidence were unnecessary. One of the most common violations of the law is one that we have been discussing in this chapter: to make an emotional appeal at a time when evidence is required. We have not condemned emotional appeals under all circumstances, but only when we substitute emotion for proof when proof is required. The latter form of behavior is the essence of what is meant by "How Not to Argue."

Students of logic were provided many examples of the use of emotion substituted for evidence during the 1950s (the "McCarthy era"), when public officials were often attacked for grotesquely irrelevant reasons. In some instances, they, their parents, or wives had been born in a

Communist-occupied country; in others, they were attacked because friends or associates had once belonged to organizations which numbered among their members individuals who had belonged to the Communist party. Such tactics were attempts to divert attention from the is-sues at hand.

Appeal to emotion
The appeal to emotion sometimes takes the special form called "the appeal to laughter." If one is unable to refute an opponent's arguments by evidence, it is always possible to make him the butt of a joke and thereby evade the necessity of presenting evidence. A notorious example of sort of thing, which apparently misfired, occurred in a celebrated debate over the theory of evolution in 1860. Bishop Wilberforce scored when he asked Thomas Huxley, who was defending the Darwinian theory, whether it was through his grandfather or his grandmother that he claimed descent from a monkey? Huxley, who was in no mood to appreciate the Bishop's humor, retorted that he preferred descent from monkey to descent from a man who used his great gifts and versatile intellect to distract the attention of his hearers from the real point at issue by eloquent digressions and skilled appeals to prejudice.

We have been discussing bad logical behavior on the part of speakers and writers who try to divert our attention from the need for evidence by working on our emotions. They fool us in this way. But we also fool ourselves. We rationalize; we engage in a wishful thinking"; we may accept unfounded beliefs because they satisfy us emotionally. For example, do we find ourselves saying, "I believe thus and so because it makes me feel good so to believe"? Or do we say, "I must believe as I do because I just couldn't bear to think my belief false"? We deceive ourselves if we believe that our emotions guarantee truth. A beautiful passion, which makes its object appear not only handsome or beautiful, but also good, reliable, noble, and intelligent, really guarantees nothing of the kind. Perhaps it is not wholly undesirable to use a little logic even in love.

When one says, 'This must be true because I feel so strongly about it, and if it were not true I could not feel as I do," he may be misleading himself. For alas, wishes are fathers to thoughts that just aren't so. The fact that we want something very strongly apparently does not guarantee that it will come our way. This would be a much nicer world, of course, if our wishes could make things come true. There would then be no broken hearts, unsatisfied ambitions, or even lack of the wherewithal to own yachts, including the cost of the upkeep.

Our emotions, in other words, may interfere with our logic and prevent us from seeing the truth. This is why we fail to see ourselves as others see us. "I am firm; you are stubborn; he is pigheaded." And do you know women whose attitude might be expressed in the following way: I am beautiful; you have quite good-looking features; while she isn't bad looking, if you like that type"?

It is our emotions that make us adopt a double standard of intellectual morality--one for ourselves, another for the other fellow. The Democrats are naturally enraged when they are unjustly attacked by Republican speakers; the Republicans may candidly acknowledge that the spokesmen of their party have been guilty of some exaggerations, but they will say that such attacks are justified in political debate. The Republicans, on 'the other hand, find unjust Democratic charges unbelievably vicious, while the Democrats will say of their own speakers that, though they may have made it a little strong, nevertheless all's fair in love and politics.

And then there are those who think that they have transcended their emotions, who "see both sides," but all too often what they tell us is, in the words of an unknown poet,

In matters controversial,
My perception's rather fine.
I always see both points of view,
The one that's wrong, and mine.

So much for one of the major ways in which we evade what we have called the law of rationality" or the law of argument." We should aim to support our beliefs by adequate evidence. The form of the evasion we have been discussing is called the "appeal to emotion." There are of course many other ways in which the law of rationality is evaded. To draw an analogy from ethics: Aristotle once said that good men are good m one way, but the evil are evil in many ways; that is, good men resemble each other in their actions, but there is great variety in wickedness. Perhaps that is why we read so much more crime fiction than stories about virtuous men. Variety is more interesting. Aristotle also uses the image of the archer shooting his arrow at the target: there is just one way to get a bull's-eye, but many different ways in which to miss.

So with arguments. A good argument must hit the point exactly, but there are many ways in which we can miss. Logicians have catalogued many types of errors of reasoning, but it would be impossible to list every possible kind of error, for there are an infinite number of ways in which we can miss the target. In arguments, too, it sometimes seems as if the archer has turned his back on the target and shot in the opposite direction!

Argumentum ad hominem
Let us now consider another major evasion of the law of rationality, the "argumentum ad hominem." This term, from the Latin, means "an argument directed to the man." To the man, that is, as distinguished from the point at issue.3 For example, let us suppose that we disagree with what a speaker says. Now, we may try to disprove what he says by presenting the evidence. Instead, we simply verbally attack the speaker to cast doubt on his statement. Such attacks are often on the behavior, motives, family, dress, or other characteristics of the person who disagrees with us.

If we believe that a statement is false, we ought to attack the statement, not the man who utters it. A speaker, let us say, attacks the so-called "right-to-work" laws, which forbid compulsory union membership. He argues that the law unfairly discriminates against unions) on the ground that workers who benefit from union activities ought to pay for these benefits. Now, if you disagree with the speaker, you should support the position that the law does not unfairly discriminate against labor. But suppose, instead, you say to the speaker, "By the way, you re a union man, aren't you?" The question implies that the speaker's views must be false, on the ground that his union membership makes him so biased and prejudiced that it would be a waste of time to take his remarks seriously-they simply must be false.

As another example of this sort of thing, let us examine some remarks made by the German philosopher Arthur Schopenhauer in his "Essay on Women." In reading what follows, it may be helpful to remember that Schopenhauer was a pessimist who believed that life is a painful and very sad affair. here are a few lines from the essay:

It is only the man whose intellect is clouded by his sexual impulses that could give the name of the fair sex to that undersized, narrow-shouldered, broad-hipped and short-legged race: for the whole beauty of the sex is bound up with this impulse. Instead of calling them beautiful, there would be more warrant for describing women as the unaesthetic sex. Neither for music, nor for poetry, nor for fine art, have they really and truly any sense or susceptibility; it is a mere mockery if they make a pretense of it in order to please. Hence, as a result of this, they are incapable of taking a purely objective interest in anything.

Schopenhauer continues on and on in the same vein. He tells us that women are interested only in acquiring husbands and that to this end they develop their real interests-in cosmetics, in clothing, and in jewelry, to the exclusion of all higher interests.

Perhaps at this point I should emphasize, as strongly as I can, that I happen to disagree with Schopenhauer. I am using this example for illustrative purposes only. The point is this: How do many women react to these remarks? Do they present evidence to disprove what he says, as required by the fundamental law of rationality? No. Rather, they attack Schopenhauer himself, with remarks like these: "That fellow must have ad very little success in his love life"; "He must have been refused by very woman he proposed to"; "He must have been psychologically frustrated, and suffered from an anxiety neurosis"; "He should have been psychoanalyzed."

Now, these remarks give us an example of the argumentum ad hominem, for the argument attacks the man instead of disproving what he says. But even an unpleasant fellow like Schopenhauer may be stating the truth, so if you disagree with what he says, present the evidence. It is not true, for instance, that all women have short legs.

Perhaps we ought not to take Schopenhauer's remarks so seriously, for they are of the nature of an emotional diatribe. His purpose may have been merely to express his splenetic feelings. The point is that a man's statements are logically independent of who the man is, or what he is, and that we do not disprove what he says by raising doubts concerning his parentage. Logically, a statement stands or falls on its own merits, regardless of who makes it. Truth and falsity are determined only by evidence. Personalities do not determine logical issues, and discussions should not degenerate into name-calling.

The reader may have a question at this point: Is it always wrong to attack the speaker personally? Is it wrong to cite the speaker's history, background, and associations in order to discredit what he says? This question requires a distinction between the argumentum ad hominem and a different kind of attack against a speaker. A witness testifies in a courtroom and the other side believes that he is lying. If they cannot directly disprove his alleged eyewitness testimony, then they will seek to attack his character. They show that he was once convicted of perjury, after testimony in another trial. Now, what effect should this have on a jury? The jury learns that the witness once lied under oath, but this does not prove that he is lying now. He may be telling the truth. To say that his conviction for perjury proves that he is lying now would be to commit the argumentum ad hominem. But, though there is no proof that he is lying now, he has been shown unworthy of trust, and the jury should therefore refuse to give much weight to his testimony.

In other words, it is quite legitimate to show that a speaker is unworthy of trust, or that he is prejudiced, or biased, or that special interests have paid him to say what he is now' saying, that he is insincere, and so on. "what you are," we say, "speaks so loudly that it is difficult to hear what you are saying, even though what you are saying may be true." The important thing, however, is that we should clearly distinguish between convicting a speaker of prejudice, on the one hand, and disproving what he has specifically said, on the other. We customarily give a speaker our trust and faith; we assume that he is telling the truth as he sees it. But if. the speaker has open or concealed affiliations, such as paid or unpaid connections with propaganda organizations, or other special interests, which make it impossible for him to tell the whole truth, then we should not give him our trust. He may be telling the truth, but we should not rely on anything he says merely because he says it, since we do not believe in his sincerity.

The history of warfare shows that every new offensive weapon encourages the development of new defensive weapons. The same is true of arguments. The ad hominem is an attack, and this attack often calls forth a counterattack. Logicians call this counterattack, or defensive weapon, the "tu quoque." Translated into less dignified language, this means "You're another." This counterattack is appropriate only when one has been unjustly and irrelevantly attacked with an ad hominem. Here is a simple example: A man in his forties argued that the drafting of men into the army was desirable, since it would make the United States ready for any emergency m the dangerous world situation. A young man, instead of trying to prove that the draft law was unnecessary, used the ad hominem attack. He said to the speaker, "You favor the draft because you are past the draft age and won't have to serve." This ad hominem approach calls for an obvious tu quoque. The older man replied, "By the same token, the only reason you are against the draft is because you are afraid you will have to serve." But the real question should have been, "Is the draft in the best interests of our country? The answer to this question does not depend on who says what. An attack against the speaker proves nothing concerning the merits of what is being discussed.

There is a variant of the ad hominem which furnishes a useful clarification of the tu quoque. Speakers sometimes try to discredit theories by calling them old-fashioned. Now, "old-fashioned" may be a devastating criticism in the field of the exact sciences when experiments have disproved an old theory. But in the field of social and political ideas, most theories have some merit, regardless of their age. Let us assume that a speaker wishes to refute the theory of "free-enterprise." "'That idea is old fashioned," he says, "It goes back to the eighteenth century." Two answers are possible. One may point out that this "criticism" is like the ad hominem approach, for the date of a theory is irrelevant to its validity. Only evidence can disprove a theory. Or one may use the tu quoque here. "If the theory of free-enterprise is discredited because it goes back to the eighteenth century," we may say, "then the theory of government regulation is even more conclusively discredited, for it goes back to the seventeenth century."

One of the rules of a good discussion is that the participants should stick to the point Their remarks should be relevant to the matter at hand. In the following pages we shall be concerned with relevance in the sense of logical rather than causal connection. For example, suppose that I should say that the great majority of Americans enjoy, on the average, the highest living standards of any people in the whole world. An objector says that my statement is false and that he can disprove it. He points to the fact that there are many people in the United States who are quite poor, having yearly incomes of less than $4,000 per year. Now, the facts cited by the objector are logically irrelevant. They are not to the point. Why not? Let us grant the truth of the facts he cites. There are many poor people in the United States. But I did not say that every American was well off." Nor did I say that there were no poor in the United States. I said that the majority of our people have the highest living standards, on average, in the world. To disprove my statement, the objector would have to cite a country in which the masses of the people enjoy a higher living standard. There may be such countries, but it is irrelevant to point to poor people in our country as a disproof of a statement which refers comparative average living standards.

To argue in the manner of the objector involves a slipping away from the point. The objector has committed a "diversion". This practice is also refereed to as "drawing a red herring across the trail of an argument."

A speaker or writer may also slip away from his own point and thus create a diversion. This usually happens when he has undertaken to support a difficult case. He may seek to create the impression that it is sufficient if he proves a point that can more easily be established. But the latter may represent a quite different issue. For example, a senator may speak in behalf of a bill requiring "100 percent parity prices" for farmers He may dwell at length (or even exclusively) on the economic suffering of the farmers before the days of government-supported farm prices. But instead of answering the question, "Is 100 percent parity desirable? he answers one that may be stated as "Is some form of government aid to farmers desirable?" It is much easier to support the latter question than the former. When speakers set out to prove the obvious, look for a diversion

Diversions are traps for the unwary, and one should be constantly on the alert for them. Let us imagine a conversation between a pacifist and a non-pacifist. The pacifist argues that "all wars are morally evil, no matter for what purpose they may be fought." "No war," he goes on, "is ever justified, so we should sternly refuse to distinguish between just wars and unjust wars. It is always wrong to kill a fellow human being." His opponent then asks him whether it would be immoral for a man to fight in defense of his country in the event his country is invaded. To this the pacifist responds. "I can assure you that no one is going to invade us." His opponent then proceeds to argue that there is a real danger that their country may be invaded.

Let us analyze what happened here. The pacifist was asked whether his principles require that a man should refuse to fight to repel an invader. The pacifist slipped away from the point when he said there will be no invasion. This was a diversion. But his opponent did not notice it and fell into the trap, arguing the irrelevant issue as to whether there might be an invasion, and the original question was forgotten.

The diversion sometimes takes the special form called "extension. A speaker says that some corporation executives sympathize with the aims of organized labor. A critic argues that it is false to say that all corporation executives are friendly to labor. But the critic is not attacking the speaker's statement, which said "some." The critic has extended "some" into "all" and attacked his own extended version of the original statement. Similarly, if the issue in debate were, "Are all corporation executives opposed to labor unions?" it would be irrelevant to prove the falsity of the statement that all executives are friendly. For even if it is false that all executives are friendly, this would not necessarily mean that all were unfriendly. Some are and some are not. And the fact that some are unfriendly does not prove that none are friendly. Debaters use this trick of extension because it is so much easier to disprove an extreme statement than a moderate one.

Begging the question
So much for one type of irrelevancy in argument. But before we discuss another type of failure to "stick to the point," we shall discuss an error which may be called an "overdoing" of this business of sticking to the point. This new error, in fact, is opposite in form to the error of slipping away from the point. In this new error we never get away from the point. We merely repeat it, over and over again. Examples: "why do I say that every human being believes in God? How do I know that? The proof is that the belief is universal in the human race." Now, this is sticking to the point with grim tenacity. But it proves exactly as much-no more, no less-as we prove when we argue that a rose is a rose. The speaker says that everyone believes in God, and when asked for proof tells us that everyone does. "How do I know that it is so? Because it is so." In an argument a reason should be given for a belief, and the reason should be a fact from which the belief can be inferred. But in this "argument" the reason (or proof) is exactly the same as the original belief. The reason ("'The belief is universal in the human race") merely repeats the original statement in different words. This is not proof. This kind of "reasoning" is on the level of the child's response to the question "Why?" The child responds with a "because" and nothing more. The demand for a reason is unsatisfied.

This fault in reasoning is called 'begging the question." Its older Latin name is "petitio principii." The error consists in our pretending to prove something when actually we assume, in the "proof," that which we are supposed to prove. "Why do I believe that the Chinese can't be trusted? Because they can't be." (Am I doing all right or am I doing all right?) Now, no formal logical error is committed in these examples, for we do not infer conclusions unwarranted by the facts. We infer no conclusion at all-we merely make an assertion. If the belief in God is universal, then it is surely the case that everyone believes in God. But the repetition of a belief is not the same as proof that it is true.

Here is another simple example. A guardian (self-appointed) of the public morals tells us that it is morally wrong for topless dancers to perform in public bars. We ask, "why is it immoral?" and receive the answer, "Because it isn't right." The answer begs the question. It repeats, in different words; what was supposed to be proved. And if we now asked, "But why isn't it right?" the answer we should expect to receive would be, "Because it's wrong."

When we beg the question, we make a pretense of proving a point but actually merely repeat it. Nietzsche once said that all mankind was corrupted, and when challenged for a proof answered, "The mere fact that you disagree with me is in itself proof that you are corrupted." And here are two passably humorous illustrations. The first is a little story about two men who approached a teller in a bank One of them wished to cash a check. The man was unknown to the banker, who asked if some-one could identify him, "Yes," the man said, "my friend will identify me. "But I don't know your friend, either," the banker objected. "Oh, that S all right," the man replied, "I'll be glad to introduce you to him." The second illustration, a very old tale, is about two medieval Jews who were engaged in a dispute concerning the respective spiritual gifts of their rabbis. To clinch his case one of them said, "And now I'll give you proof positive that my rabbi is the most wonderful rabbi in the whole world Is there another rabbi who dances with angels every night after he falls asleep?" The other was somewhat skeptical of this. "But how do you know," he asked, "that your rabbi really dances with angels?" "Why," replied the other, "because he told us so himself." "But can you believe him?" "What!" the other retorted in indignation, "would a rabbi who dances with angels every night tell a lie?"

"Reasoning in a circle" is a "drawn-out" form of begging the question. It contains intermediate steps. A man says that classical music is better than modern music. Challenged for proof, he answers, "The best critics agree that this is so." Who are the best critics? "Those who prefer classical to modern music." And here is a more complex example: The founder of a new religion tells his followers that he is inspired, so that they may believe whatever he tells them. Now, in the unlikely case that he should be challenged for proof of his inspiration, he might answer, "Because I am inspired." That would be the simplest form of begging the question. But if he "reasoned in a circle," the argument might go like this: "Why do I say that I am inspired? Because here. is a book which says that I speak in God's name. Why should we believe this book?" he is asked. "Because it comes from God," he answers. 'How can we know this?" "Because you can take my word for it." "And why should we take your word?" "Because I am inspired." If we now ask, "How can we know you are?" the circle will start all over again.

The so-called "argument by definition" is a special form of begging the question. Jones asserts that all Christians are good men. Brown disagrees and points to Thwackum, who is a Christian but whose conduct falls very much short of that of a model of virtue. "Ah," answers Jones, "Thwackum may attend church regularly, but he is no real Christian, for if he were, then he would be a completely virtuous man. I reiterate, all Christians are virtuous men." This argument begs the question. What Jones meant was this: "I define a Christian as a good man. Thus I can assert without fear of contradiction that every Christian [defined as a good man) is a good man." Jones' original statement cannot be proved false, for it is not a statement about facts but a "stipulative definition." A stipulation, insofar as it is nothing but a declaration of intention as to how a word is going to be used, is neither true nor false. The question, "Is this statement true?" can be raised only with respect to assertions that purport to describe facts. Observation or experiment may confirm or disprove factual statements-they are true or false-but we cannot raise the question of truth or falsity concerning a man's declaration of his intention to define a word in a certain way. If Jones had said that every churchgoer is faithful to his spouse, that statement would be true or false. But Jones merely tells us that he is going to use the words "Christian" and "virtuous man" interchangeably. By definition, then, no bad man can possibly be a Christian. (If we stipulate a definition of a square as a four-sided figure, it is senseless to ask, "Can a square have more than four sides?") Jones' statement, that "all Christians are good men," is thus tautological. He is saying that all good men are good men. But this is not what Brown took him to mean. Brown understood him to say that a Christian, defined as a member of a church that worships Christ, can always be depended on for his trustworthiness. This might be false. But a definition taken as a stipulation concerning word usage cannot be false.

And here is one more variety of question-begging: the "question begging epithet." The previous varieties were cast in the form of arguments, with a pretense at giving proof. Repetition was substituted for evidence. 'Our new variety merely uses epithets. For example, expressions like "the stupid conservative point of view," or "wild-eyed radicalism," contain question-begging epithets which assume something that may require proof, without even a pretense at proof. Instead of proving, first of all, that Moriarity is a crook, we ask, "What do you think of that crook?" A ready-made conclusion is put into the hearer's mind. This variety of question-begging is perhaps more dangerous than the other forms, for it operates on our reluctance to question a positive assertion that is "unquestionable," especially when asserted in a strong manner. The intimidated listener may also be at a loss to know just what it is that he should question.

Argumentum ad ignorantiam
Another mistake commonly made with respect to the burden of proof in an argument is called the "argumentum ad ignorantiam." As the name suggests, this means an argument based on ignorance, or on an appeal to our lack of knowledge. For example, a man states his belief that God selected Mohammed as the final prophet of His Word. When challenged for a justification of this belief, he asks: "Can you disprove it? If you deny what I say, it's up to you to disprove my statement." Or let us say that an atheist denies the existence of God. When asked for evidence; he answers: "Can you prove that God exists?" But just as failure to disprove is not proof of the opposite (the Mohammed example) so failure to prove is not equivalent to disproof. It all depends on who has the burden of proof. If a man makes the claim that God does not exist, then the burden is on him to prove his point. Atheism is quite different from agnosticism, which merely says, "I don't know." Similarly, the per son who argues that God does exist has the burden of proof for that thesis.

This, then, is the appeal to ignorance. Instead of proving a statement by positive evidence in its favor, we appeal to the fact that our opponent hasn't disproved it. But the law of rationality tells us that we should furnish positive evidence for our beliefs.

An amusing variant on the ad ignorantiam argument is the old story about the justice of the peace who heard a case concerning a man who was accused of stealing a horse. Two witnesses testified that they saw the accused unhitch the horse from a post and lead it away, but five witnesses testified that they had not seen the defendant steal the horse. The J. P. said that since there were more people who had not seen the alleged theft than there were persons who claimed to have seen it, and since he believed that the majority were always right, he was forced to dismiss the case.

"Either-or"
Let us consider a final example of "how not to argue": the "either or" fallacy. These two little words "either" and "or" make trouble for us unless we watch them carefully. They are useful tools in thinking, but a great deal of bad and confused thinking falls into an either-or pattern. A The worst of these vices-the error of insufficient options-is the assumption that there are only two possibilities in a situation, or only two choices, when there are in fact more than two. This can be a very serious matter, indeed. Consider, for example, an application of this assumption to our international problems. There are people who say, "Either other nations are for us or they are against us; either they will take sides with us against our enemies or they will take sides with our enemies against us," But this ignores the possibility of neutrality as in the case of Switzerland, or of other spheres of influence, as in the case of the 'Third World nations that have coalesced around economic and cultural interests in recent years.

Similarly, it is not the case that I am either for you 100 percent or I am against you." I may be for you 99 percent, or I may be neither for you nor against you. There is a middle ground between love and hate called indifference, just as between indifference and hate there is a region called dislike. We cannot say of a man that he is either an angel or a devil, either a god or a beast. As Aristotle said long ago, we are in between, superior to the beasts and inferior to the gods. Let us stop using just two categories for people: the saints and the devils. Many of us are middle-of-the-roaders.

The kind of thinking we have just described is sometimes confused with a law or principle of logic called the "law of the excluded middle." This law tells us that a thing either has a particular characteristic or it does not have that particular characteristic: A man either has a million dollars or he does not; he either owns a home of his own, or he doesn't. Anything, the law tells us, is either A or it is not A; either it has characteristic A or it does not. These alternatives do exhaust the possibilities; the options are sufficient. Note that the examples of "insufficient alternatives" we considered above do not illustrate the law of the excluded middle. It is false that a car's color must be either red or yellow; it may be black or blue or green. But the law of the excluded middle says only that a car is either red or it is not red. A chemist analyzes a solution to determine the presence or absence of arsenic. Either at least one molecule of arsenic is present or no arsenic at all is present. This is in accordance with the law.

The most interesting and troublesome problems involving the use of "either-or," however, concern a type of application in which the law of the excluded middle is not violated-the two alternatives actually do exhaust the possibilities-but in which the "either-or" may be misleading or unrealistic. This type of thinking requires a more extended discussion. One of the most, striking illustrations of this sort of thing is found in the bad habit of thinking which 'we shall call "moral perfectionism." The perfectionist sets up an ideal or standard of moral perfection, and then thinks in terms of only two alternatives, "Either you are a good man, or you are not." Though the perfectionist does not always say so, he usually implies that he himself has achieved perfection and that the rest of the world has not, and he judges everyone else as falling short of the standard. The perfectionist thinks of every man as falling into one or the other of the two classes, the good and the not-good.

There are many illustrations of this kind of perfectionism. To the ancient Stoics there were no degrees in vice; you were either perfectly virtuous or you were not virtuous at all. The theft of a piece of firewood, they said, is just as much a violation of the moral law as ruining a man in a swindle. As one of them put it, "The man who is a hundred miles from Canopus, and the man who is only one mile from Canopus, are both equally not in Canopus." A similar thought is expressed in the New Testament, in the words of James: "For whosoever shall keep the whole law, and yet offend in one point, he is guilty of all." In other words, for the perfectionist you are either good or you are not good-and there is no middle ground.

This kind of thinking permits of no compromise. Either you fulfill the moral law completely or you do not fulfill it completely. Either you are perfectly good or you are classed undiscriminatingly with all the backsliders. And herein lies the most serious fault of this attitude. Though the perfectionist says, "all-or not-all," in practice he means, "all-or nothing," for the man who falls just short of perfection is regarded as being in the same class as the most vicious and hardened criminal. The perfectionist is not interested in the degree in which one falls short.

The perfectionist attitude is found in all sorts of places. Consider the perfectionist's attitude toward philanthropic benefactions. Either .a person is a "perfect giver," they say, or he is not. A perfect giver always gives anonymously, for if it is known that he has made a contribution, then we must assume that he expects to receive applause for what he has done. And if he desires public acclaim, then he deserves no credit for giving, for he didn't give solely for the joy of giving. Those who don't give at all are apparently no more ungenerous than the donors whose names appear in the lists. But surely we should distinguish between the man who gives, hoping for some expression of gratitude, and the man who refuses to give anything at all.

So much for perfectionism, which uses the "either-or" to formulate unrealistic dichotomies. Let us now restate the three types of usage of the "either-or" formulation that we have considered. There is first the "either A or B" statement, in which one says, "He is either a member of the bourgeoisie or he is a member of the proletariat. He must be one or the other." This is the error of "insufficient options." He need not be either; there are other possibilities. This type of usage is characteristically at variance with the facts. The second usage is in the form of the law of the excluded middle: "Either A or not-A." Here we can properly say, "It must be one or the other, A or not-A," for these exhaust the possibilities; either he has a particular characteristic or he does not have it. We then drew a distinction between realistic and unrealistic applications of the law of the excluded middle. "Either A or not-A," we noted, is quite adequate where degrees are not involved, as when the chemist says, "Either arsenic is present or it is not present." But when degrees are involved,' then the expression "either A or not-A" may be unrealistic and misleading. This is the third type of usage we discussed: "He is perfect or he is 4 not perfect" Now, this statement is in accordance with the principle of the excluded middle, and so not incorrect when we understand it is a precise formulation concerning those who have a characteristic and those who do not. Nevertheless, the statement is misleading because the important thing about human conduct is the degree with which conduct approaches a standard of perfection. "He is either perfect or not-perfect" makes us lose sight of the degrees of imperfection.

Let us look at some further illustrations of the application of "either-or" to cases involving degrees. Where there is a continuum of degrees-as in rating the intelligence of human beings-the subjects do not divide into two sharply contrasted opposites: the intelligent and the unintelligent. It would be misleading to use the principle of the excluded middle here, to say, "Every human being is either intelligent or unintelligent," as if there were just two classes, into one of which every human being falls. Or at least this would be a very arbitrary thing to do. But-and this is the other side of the coin-there are occasions when it is necessary to make such arbitrary distinctions between the two classes, the intelligent and the unintelligent, as in the Armed Forces Qualification Tests. It may even be necessary to draw a sharp and arbitrary distinction between sane and insane, as in a trial for murder. Our criminal law draws a distinction between offenders who are sane and those who are not, on the principle that we ought not to treat the two classes in the same manner. And judges need clear-cut definitions, or at least some kind of arbitrary dividing line. In the state of Illinois, for example, the statutes define sanity as the ability to distinguish between right and wrong. Though most psychiatrists consider this definition inadequate, it is at least workable, in a rough sort of way, and gives us an arbitrary dividing line.

Or consider the matter of academic grades. In our schools we use an "either-or" for passing or not-passing. Passing means that a student has mastered the subject matter to the required degree. Thus we can say of any student, "Either he has mastered the course or he hasn't." But, you may protest, this is unrealistic, for mastery is a matter of degree, and it is wrong to divide all students into only two classes. The differences between students lie in a continuum of almost imperceptible differences in degrees.

Let us say that a grade of 65 percent is set as the minimum passing grade. But how much difference is there, after all, between the student who makes sixty-five and the one who makes sixty-four? The latter fails the course, and it seems grossly unfair that a 1 percent difference should have such enormous consequences. Shall we pass the sixty-four percenter? But now, the sixty-three comes to claim equal justice. We pass him too. The logic of this procedure is that we must finally pass even the person who makes zero, for zero is only 1 percent below one. The grades. from zero to 100 are connected by small graduations in degrees. But school administrations need a dividing line, and they arbitrary set it at 65 percent or some other definite figure. If there were no such dividing lines, there could be no definite standards of competence. There could be no bar examinations or C.P.A. examinations. Though we recognize the dividing line as arbitrary, still we must have one. And even the kindest-hearted teacher, who passes the sixty-fours and the sixty-threes and the sixty-twos, must draw a line somewhere. Certainly the twelves and thirteens should not pass.

There are two extremes to avoid with respect to the continuum. One is the danger of making too sharp divisions in reality, of thinking that all people can be divided into just two classes-into capitalists and Communists, or into the good and the not-good. The other extreme is to deny the existence of all distinctions simply because one class passes into another by imperceptible degrees. This leads to a fuzzy-mindedness, which says that the good are really not-good, for they are connected by imperceptible degrees to the not-good, or, conversely, that the not-good are really good, for exactly the same reason. In making the first error we think in terms of two sharply divided natural classes; in the other we see no classes at all. Justice Holmes once referred to the question, "where are you going to draw the line?" as "the tyro's question." We must draw lines, he added, for all life involves "the marking of grades between white and black." There are classes of things even though they merge into each other by imperceptible degrees. The mistake we too often make is in thinking that there can be only two.

There are times when we must make decisions, and the etymology of the word is instructive. It comes from the Latin, meaning "to cut off." We make a sharp break when we decide, for we must decide in one way or another, no matter with how many qualifications. In an election we must decide whether an officeholder has or has not satisfied our standards. Though our minimum standards of decency are vague and ill-defined, they are there. There are times also when we cannot evade a yes or no answer, when we must choose between alternatives. People who dislike yes or no answers dislike saying "Everything is either A or not-A." Some people would even like to "abolish" the law of the excluded middle.

This chapter has emphasized a fundamental paradox, which may be called the "it is and it isn't" situation. There are no sharp divisions between white and black, and yet there are. There are continuities, and there is also the necessity for arbitrary standards. The lesson of this chapter, if there is one, is this: Let us not deny or forget the facts of continuity, and let us not make sharp divisions where these are inappropriate. And let us not deny the distinctions between classes of things.


FOR DISCUSSION AND WRITING
1. The following statements contain errors in logic. Find each error, showing how it violates the law of rationality.
a. As a lawyer, Frank, you are obviously against no-fault automobile insurance because such legislation would cut into your income severely.
b. America: love it or leave it!
c. Your attitude toward morality is founded on an outmoded nineteenth-century notion that there is a double standard for men.
d. You might know Sally would give you advice about your marriage. After all, she's had three husbands herself.
e. In your heart you know he's right.
f. Of course, as a Democrat I'm pro-labor, con, but better that than being a follower of big business like all you Republicans.
g. But Mom, it's a very "now" thing to do.
h. His views on social reform are very radical, but what can you expect from a professor of sociology?
i. You've learned only theory in college; you can't be expected to understand the problems of the man in the street.
j. I suspect his fitness as a candidate for that office. He underwent psychiatric treatment nine years ago.
k. Her proposals for welfare programs are indeed radical; I remember when she entertained the Socialist Club in her own home.
l. Don't explain his voting record in the Senate to me. The fact that he is a Mormon is all I need to know.
m. It's just one of those things you can't explain, but I feel so strong about his proposal I know it must be a good one.
n. His conception of foreign aid is influenced considerably by the hopelessly antiquated theory which was popular at the end of World War II.

2. Can you think of circumstances in which the use of emotion in a speech or argument would be acceptable? Should advertising, for example, be based totally on reason? Can it be?

3. Using the following examples as models, construct your own list of adjectives: "I am firm; you are stubborn; he is pigheaded"; "I am slim; you are thin; she is skinny."

4. Analyze the following statements. In each case, determine whether the argument violates logical thinking.
a. Either you are for us or against us.
b. Either an animal is a canine or it is not.
c. Whosoever shall walk in the paths of righteousness will be saved; he who swerves from that path even for a moment is damned.
d. Either Hamlet is a classic or it is not.
e. Either Jones receives the nomination or the country is in serious trouble.
f. Either you vote for what is good for this country or you will vote for the Democrats.

5. Analyze the following arguments, showing whether they are valid or invalid.
a. While the strike was an effective means of helping the working man in the nineteenth century, it no longer is a necessary tool for labor. In fact, it is a dangerous tool. If we allow dock workers to strike or teachers or the police, nothing will prevent others from doing the same. Imagine what would happen to the country if state employees went on strike or the army or the doctors or even politicians.
b. The argument that our welfare programs will lead to a socialistic state is groundless. We already give aid to the poor, the' aged, the sick. why not then develop the program merely a step further. I would pro-. pose that we increase our aid to the sick by providing them with free medical aid. The state could do this easily if it followed my plan. Because there is little difference between free medical aid financed by the state and state control of medical aid, let the federal government take over alt hospitals. Because there is little or no difference between health insurance for the ill and directly paying the salary of the individual doctors, let the government control all hiring, training, and locating of doctors; in that way we could all be assured of prompt and high quality medical attention.

6. Given the following statements as your sole source of information, how would you define the "law of the excluded middle"?
a. All girls are fickle. No girls are not fickle.
b. All lakes are polluted. No lakes are unpolluted.

7. Write sentences that contain the following violations of the law of rationality:
a. argumentum ad hominem
b. begging the question
c. argumentum ad ignorantiam
d. "either-or" fallacy

---

CHAPTER 6

Putting Up a Logical Argument

Whether we are reading or writing argumentative prose, we should know how to take it apart, that is, we should be able to identify its conclusions and premises. We should always ask-even of our own writing-the following questions: What is the author's point? What reasons does he give to support his point? Is the argument valid? Is it true?

The argument and its parts
But first, a warning. The word "argument" has several senses. To most who hear the term, it is a contest in reasoning in which one person wins and another loses. In contentious arguments one person tries to prove another wrong. An argument in this sense is often contrasted with "discussion," an interchange of ideas in which there is no attempt to defeat an opponent.

Arguments, in the popular sense, often become verbal slugfests, m which one person tries to beat another down. All too often contentious arguments are wrangles in which tempers rise and in which the arguers often put up a mule-like resistance against anything their opponents may say. But we shall not be concerned here with such arguments.

By "argument" we shall mean the basic unit of reasoning. The proof any statement or belief is always presented in the form of argument, defined as "a unit of discourse in which beliefs are supported by reasons." Our interest henceforth is in argument, not for purposes of contention, but insofar as arguments are an indispensable element in the quest of truth. Argument in this sense is the heart and soul of the rational enterprise.

Premises and conclusions
Arguments, then, are not things that are either lost or won, but units proof (or attempted proof) that something is or is not the case. Here is an example: "Only citizens who have registered can vote, and you haven't . registered, so you can't vote." In this argument the speaker seeks to prove that "you can't vote." This statement is supported by reasons. A statement supported by reasons is known technically as the "conclusion" of the argument. This is its "point," what it is "driving at." The statements which sup-port the conclusions (the reasons) are called the "premises" of the argument. The premises are the evidence, or facts, or assumptions, or reasons, on which the conclusion is based. Note that a statement is a premise only in the context of an argument. The mere assertion, "Dust thou art," becomes a premise when it is used in an argument: 'Dust thou art; therefore the body is inferior to the spirit." A statement becomes a premise by virtue of the role it plays in an argument.

In an argument, then, we say, "This, because of that," or "'This is so, therefore that is so." This process is called "inference." But before we engage in further analysis of argument, let us contrast an argument with what we shall call a "mere assertion." Here is an example of the latter:

'There are thousands of young people whose lives are being ruined by marijuana. Marijuana should not be legalized."

By a "mere assertion" we mean any statement for which no "justifying reasons are given. In an argument there are several statements, one of which is supported by others. The supported statement is the conclusion; the supporting statements are the reasons or premises. In the "marijuana" quotation there are no supported and supporting statements. And when. you read, "Dust thou art, and unto dust shalt thou return"-period, this too is an unsupported statement, thus a mere assertion.

The argument, in the sense of "discourse containing inference," is the central core of logic. Only arguments can be called logical or illogical. Not all discourse is argument; perhaps most is not. We ask of a friend, "What's new?" and he tells us. Narration is not argument, and so there is little argument in newspaper reporting and historical writing. But when we read a newspaper editorial, we are likely to find argument. The writer will be trying to prove something, such as the error in present public policy or the desirability of a new course of action. We support our beliefs by argument when we expect to be met with a challenging "Why?"

"Why do you believe that?" "Why do you think so?" "Why ought we to?" We said earlier that it was the mark of a rational man to support his beliefs by adequate evidence. This Is especially the case when his beliefs are of a controversial nature.

Now, when we read argumentative discourse, it is well to know how to "take it apart" with the critical eye of logic. There is perhaps no more important lesson for skilled reading than this: When you read argumentative discourse, find and identify its conclusion, and then note its supporting reasons, or premises. There are two questions which should always be in the forefront of the reader's mind: (1) What is the writer's point, exactly what is he trying to prove, or disprove; what is he trying to "put across"? (2) What reasons does he present to persuade me that he is right, on what basis am I expected to agree with or accept the conclusion? These two questions, of course, constitute only the first lesson in logical analysis. The next lesson will take up the question whether the argument presented by the writer is sound or unsound. But one thing at a time. What we are now concerned with is the analysis of an argument into its parts, and the reader who makes the two questions part of his normal response to argumentative discourse is already on the road to becoming a more critical and intelligent reader.

An argument, then, has two parts: the premises (or premise) and the conclusion. The premises may be stated before the conclusion, or they may be stated after the conclusion. There are certain words, called "logical indicators," which connect the premises and the conclusion. When the premises are stated first, the word "therefore" (or a synonym) will be used: "You haven't registered; therefore, you can't vote." The word "therefore" always precedes the conclusion of an argument, and it always follows the premises. On the other hand, when the conclusion comes first, we use "because" (or a synonym) to connect the parts of the argument:

"You can't vote, because you haven't registered." The word "because" always precedes a premise. The conclusion may also be sandwiched in between two premises: "Only those who have registered can vote, so you can't vote, for you haven't registered." If you are henceforth in doubt as to what the conclusion of an argument is, look for the logical indicators. They may, of course, not be present in expressed form (they may be understood), and one should then look to see where they can be inserted.

There are many synonyms for "therefore" and "because." For example, if you substitute words like "so," "hence," and "consequently" for "therefore," you Will see that they have the same meaning. Sometimes the logical indicator is spelled out more fully, as in "which shows that," "which indicates that," "and so we may conclude that," etc. Synonyms for "because," are "for," "since," or phrases like "in view of the fact that," or "for the reason that," and so on.

In an argument, then, we find the two elements, premises and conclusion. There is no rule as to whether we should state the premise be-fore the conclusion, or vice versa. In an extended argument which aims at persuasion, however, certain strategical considerations may influence the positions of the premises and the conclusion. For example, let w suppose that you are going to argue in behalf of a conclusion which will sound reasonable to your audience. It will then be well to state your conclusion at the outset, before giving your reasons for accepting it. You will have won the confidence of your hearers by the 'soundness" of the conclusion which you are going to prove. On the other hand, let us assume that you wish to attack the popular view. Now, if you were to state your conclusion at the outset, this would arouse strong opposition and perhaps resentment. Many hearers would regard you as so unreasonable that they would not listen carefully to the rest of what you had to say. In this case, may be advisable to build up the evidence with facts that your hearers accept, and then show how these facts logically require a conclusion different from the one they accepted previously. But these strategical considerations have nothing to do with the logic; they relate solely to the strategy of the argument.

Arguments may have more than one premise, and they may also have more than one conclusion. "I believe that Police Captain Blue takes bribes. He associates with gangsters and has become very wealthy." Two premises for' one conclusion here. Note that there are no explicit logical indicators in this argument, but the sense indicates the structure. "The farmers voted for quotas. This will increase government expenditures, and it also signals the end of free enterprise." Two conclusions drawn I here from a single fact.

In a "serial argument," we find a conclusion drawn from a reason, and this conclusion then serves as a reason for drawing a second conclusion. The final conclusion is the main point of the argument. Here is an example: "I am not as wealthy as I used to be, because of the decline in the stock market. And, since I am not as wealthy as I used to be, I shall be unable to buy that new house."

Proof and disproof

An argument has been defined as "discourse containing inference." Inference is used in proof, which we shall now consider. Proof may be used in a broad or in a strict sense. In the broad sense proof means "enough evidence" to justify a conclusion, as in the legal expressions, "proof by preponderance of the evidence" or "proved guilty beyond a reasonable doubt." In the strict sense, as used in the proof of a theorem in geometry, proof means "demonstration," that is, the logically necessary relation between axioms and theorem. We shall use the broader sense unless otherwise noted.

"Proof" also covers "disproof," for a disproof is simply proof that a statement is false. This must be distinguished from "failure to prove, just as a verdict of "not proven guilty" differs from proof of innocence. (We suspect that many guilty defendants have been acquitted in criminal trials.) Disproofs (or refutations) often take a characteristic pattern which resembles the "reductio ad absurdum" (reduction to an absurdity). The principle of the reductio is that, if a statement implies absurd (or false) consequences, then the statement must be false. Here is a sample of the reductio ad absurdum:

It is a common notion that morality simply means conformity to the customs of one's group. But this cannot be the case. If it were, we could never criticize and improve the morals of our group, at least we would have no moral basis for doing so. However superstitious, or stupid, or cruel the customs of our community are, they would be, by definition, morally right-for us. The unthinking conformist would be the moral man, the moral reformer the immoral man. There would be no moral progress. But no one really believes this. We all constantly criticize the morals of our group. (Adapted from an argument by Durant Drake, in Invitation to Philosophy.)

This argument seeks to disprove the theory that morality means conformity to the customs of one's group. The theory is disproved by showing that it entails absurd or false consequences. If this theory were true, the argument runs, then we could never be justified in criticizing the morals of our group, and it would be senseless to speak of "improving" them. But, as the last sentence in the argument notes, we do criticize the morals of our group and assume that we are justified in doing so. The consequences being false, the theory must be false.

The last argument may raise some unanswered questions: Is the argument sound? Do the conclusions really follow from the facts cited? But such questions must be postponed until later in this chapter. What we are here concerned with is simply to illustrate the structure of an argument, not to assess its validity. The criticism of an argument for faulty reasoning must come after we understand exactly what the argument says. To understand what it says we must know what to look for. Thus the understanding of an argument often requires a great deal of cooperation from the reader. It is the duty of a writer to make his meanings clear, and also to make the argumentative structure clear, but there are bad readers as well as bad writers.

The syllogism
We shall now analyze the formal structure of the syllogism, a common form of argument. A syllogism is an argument consisting of two premises and a conclusion: "Only those who have registered can vote, and you haven't registered, so you can't vote." The most famous syllogism of all time is one used in logic texts for the past two thousand years:

All men are mortal
And Socrates is a man;
Therefore, Socrates is mortal.

The syllogism has two premises, one major and the other minor. major premise is simply the premise which contains the "major term," which is defined as the "predicate of the conclusion." "Mortal" is the predicate of the conclusion, so "All men are mortal" is the major premise. The subject of the conclusion is called the "minor" term: Socrates. There is one more term, which appears in both premises but not in the conclusion. This is called the "middle" term: men. The "middle term" is so-called because it is the connecting link between the other two terms. "Man" connects Socrates and mortality.

The syllogism we have just analyzed is only one type of syllogism. It is sometimes called the "Aristotelian" type, because Aristotle was the first logician who analyzed syllogisms having this type of structure. Later logicians have examined other kinds of syllogisms. Here is a different type: "If prices continue to rise, then the unions will ask for further wage increases." Prices are continuing to go up, so we may be confident that the unions will demand further wage increases." And another: "Either a world government with an international police force will be established, or the world will continue in a state of tension. But there are no signs that such a government will be established, hence we can expect tension to continue." Now these last two arguments are syllogisms, for each has two premises leading to a conclusion. But they do not contain major, minor, and middle terms. Different kinds of analyses are required for these types, as we shall see in Chapter 7.

There are many confusions concerning the role of the syllogism in thinking. We are sometimes told that the syllogism is artificial and outmoded," or that no one ever reasons in accordance with the form of the Socrates example above. But the logician does not present the syllogism as a model to be imitated. His point is, rather, that it is a form we actually do use in our reasonings. This may seem surprising to the reader, for no one would normally think in the pattern of the Socrates example. This example is truly in an artificial form, but it is deliberately put into that form for purposes of analysis, so that we may get a clear picture of what we are talking about. In "real life" this is the way the reasoning might look: "Socrates must die: we must lose him some day. For he. is but a man, and mortality is a doom which none of us can escape." If we eliminate the rhetoric here, we shall find the familiar syllogism, which gives us only the bare bones of the argument.

A syllogistic pattern may sometimes add more to our knowledge than the Socrates example does. The American philosopher W. P. Montague once described how, many years ago, he had puzzled over whether women should have the right to vote. He was a firm believer in that famous historical slogan: "No taxation without representation." He suddenly saw the answer to his problem: Taxpayers should have the right to vote, and women pay taxes, so women obviously should have the right to vote. He had found his middle term! Needless to say, his initial puzzlement would have made him a target for today's women's rights groups.

Enthymemes
There is another reason for our surprise when we learn that we reason in syllogisms. This is because few syllogisms are stated completely in everyday talk. The obvious should not be belabored, and so, to avoid boredom and tedium, we leave something to the imagination of the listener. When what is clearly indicated is too obvious to mention, we may omit a premise from a syllogism, or we may even omit the conclusion. The following example is typical of ordinary reasoning: "Joe must be doing well this year, for he took his family to Florida this winter." This is a syllogism, but it is incompletely stated. There was an additional premise in the speaker's mind: "People who take their families for winter vacations in Florida are 'doing well."'

Incompletely stated syllogisms are called "enthymemes" (rhymes with "Bentham-eems"), from the Greek en (in) and thymos (mind). Here are some more examples: "Naturally I consider him an intelligent man; he's an independent voter, isn't he?" This assumes that all independent voters are intelligent. "Our police should not carry guns. This type of police practice has worked well in England." This argument may sound more plausible in its incomplete form than it does when spelled out For the missing premise would be something like "what works well in England will work well in the United States." This last argument closely resembles the next one: "This cough medicine ought to be good for my cough, for, according to the advertisement, it helped a man in Minneapolis." Will whatever helps one man m Minneapolis help you?

And here is an enthymeme with its conclusion missing: "An old Abbot, talking among a party of intimate friends, happened to say, 'A priest has strange experiences; why, ladies, my first penitent was a murderer.' Upon this, the principal nobleman of the neighborhood enters the room. 'Ah, Abbe', here you are; do you know, ladies, I was the Abbot's first penitent, and I may promise you my confession astonished him!"' (From a story by Thackeray)

"Chain arguments"
We shall note one further type of structure here. This is the "chain argument." A chain argument is a serial argument (mentioned earlier), in which a conclusion becomes a premise for a further conclusion. Let us look at a fairly complex example taken from Leibniz:

The human soul is a thing whose activity is thinking. A thing whose activity is thinking is one whose activity is immediately apprehended, and without any representation of parts therein. A thing whose activity is immediately apprehended without any representation of parts therein is a thing whose activity does not contain parts. A thing whose activity does not contain parts is one whose activity is not motion. A thing whose activity is not motion is not a body. What is not a body is not in space. What is not in space is insusceptible of motion. What is insusceptible of motion is indissoluble (for dissolution is a movement of parts). What is indissoluble is incorruptible. What is incorruptible is immortal. Therefore, the human soul is immortal.

There are several variations in chain arguments, but this sample will indicate the general idea. The validity of arguments of this type depends on a property of what logicians call "transitive relations." A transitive relation is one such that, if A has it to B, and B has it to C, then A must have it to C. "Ancestor of " is a transitive relation. If A is the ancestor of B, and B is the ancestor of C, then A is the ancestor of C. WI' en we speak of one class of things as being "included in" another class of things, we are also dealing with a transitive relation, for if class A is included in class B, and B is included in C, then A must be included in the class C. If the "class" of whales is included in that of mammals, and the class of mammals is included in that of warm-blooded creatures, then whales must be warm-blooded. In the Leibniz argument we find this relation of class inclusion. Each sentence can be interpreted in terms of the relations of two classes to each other: The class of "souls" is included in the class of "things whose activity is thinking." The latter class is in-eluded in the class of "things whose activity is immediately apprehended," and so on. The argument is valid because "class inclusion" is a transitive relation.

Not all relations are transitive, of course. Some are "intransitive," such as "being the father of." An intransitive relation means that if A has a certain relation to B and B has it to C, then A cannot have it to C. "Ten percent larger than" is another such relation. Then there are "non-transitive" relations, such as "being a friend of": If A has a relation of this kind to B, and B has it to C, then A may or may not have it to C.

Logicians also classify relations as "symmetrical," "asymmetrical,,' and "non-symmetrical." A "symmetrical" relation is one such that if A has it to B, then B must have it to A: "equal to." (If A equals B, B equals A.) An "asymmetrical" relation is one such that if A has it to B, B cannot have it to A: "mother of." A "non-symmetrical" relation, obviously, is one where B may or may not have the relation to A, when A has it to B: lover of." The reader may find it amusing to discover additional examples for the nine possible combinations of the relations of transitivity and symmetry:

1. Transitive-symmetrical: equal to
2. Transitive-asymmetrical: greater than
3. Transitive-non-symmetrical: included in the class of
4. Intransitive-symmetrical: polygynous spouse of
5. Intransitive-asymmetrical: father of
6. Intransitive-non-symmetrical: nearest blood relative of
7. Nontransitive-symmetrical: cousin of
8. Nontransitive-asymmetrical: unrequited lover of
9. Nontransitive-non-symmetrical: lover of

In this chapter we have been concerned with understanding what an argument is, how to identify one, and how to break it up into its parts Two questions should be asked whenever we find an argument: What is its conclusion? What reasons are presented in support of the conclusion?

But there are other questions we must ask, These new questions center around the goodness or badness of the argument What kinds of arguments do you consider good ones? When you say that an argument is "good," do you mean that you agree with the conclusion? Does it make a difference, in your estimate of an argument, whether or not you agree with the premises? Can you refuse to grant the truth of the premises, and yet accept the truth of the conclusion? Can an argument be a good one though every statement in it is false? Can it be a bad argument though every statement in it is true?

Before we try to answer these questions, let us note an ambiguity in the word "good" when applied to arguments. A "good" argument may mean one which is valid in form, that is, an argument whose structure is such that, if the premises are true, the conclusion must necessarily be true. Such an argument is valid even if the premises are not true. On the other hand, "good argument" may mean one that is completely satisfactory: valid in form and containing true statements. A valid argument, then, may not be completely satisfactory.

Let us now examine the principles which will help us develop and recognize logical arguments.

PROOF

Necessary and probable proofs
Are there two persons in the city of Chicago who have exactly the same number of hairs on their heads? Perhaps you think this highly unlikely, or perhaps you think it likely, but can we prove it one way or the other? It would be highly desirable if we could decide the issue without having to count the hairs on thousands of heads. Logic comes to our aid here. There are two well-known facts that make our answer certain. First, we must acknowledge the fact that a human head can have, as a maximum, about a quarter of a million hairs. A second fact is that there are close to four million persons in Chicago. Now, let us put these facts together, and we see that there must be two persons with the same number of hairs. For suppose that we actually did start counting the number of hairs on people's heads. And suppose that in the first 250,000 heads we counted, each head had a different number of hairs, so that no duplication occurs in the first 250,000 heads. In other words, there will be one head with one hair, a second with two, and so on up to 250,000, the maximum possible. We then come to the 250,001st head, that is, the first head beyond a quarter of a million. We must now duplicate one of the numbers of the earlier subjects, since no one can have a number greater than 250,000. How useful logic is in sparing us tedious investigations!

This was an example of a logical proof. Let us now look at a different kind of "proof," the kind we find in a law court. A man is tried for the murder of his business partner. The accused was the beneficiary of a large insurance policy made out by the victim, and an additional motive is established in the fact that the victim was in love with the accused's wife. Ballistics experts establish that the bullet which killed the deceased was fired from the defendant's gun. The accused man claims that he is innocent, but cannot establish an alibi. His only defense is his claim that he is innocent. The jury must weight his denial against the evidence presented by the state. The jury finds that the state has proved his guilt beyond a reasonable doubt.

We have examined two arguments, each of which uses the term proof." But there is an important difference between them. If it is true that there are x numbers of hairs on a human head and more than that number of persons in Chicago, then it must necessarily be true that there are two persons with the same number of hairs. This is a formally valid, or necessary, argument. But in the second case, if we grant the truth of the premises (the evidence given), the conclusion-that the accused is guilty-may or may not be true. There is a probability that he is, but he is not necessarily guilty. There is a possibility that he is innocent.

In the strictest sense, "proof" means an argument in which the conclusion necessarily follows from the premises. This is "formal" proof, or "demonstration." If we accept the premises in the hair-counting argument, we must accept the conclusion of that argument; if we grant the truth of the premises of that argument, then we must accept the truth of its conclusion. On the other hand, when we say "proved beyond a reason able doubt," we do not mean that the conclusion must be accepted or that it necessarily follows from the premises, but only that it would be unreasonable" to accept the premises and not accept the conclusion. This is "proof' in a less precise sense of the term. We shall use the word "proof" in both senses, but we should recognize the difference between the strict and the looser sense of the word. We shall also use the term valid" for an argument that involves logical necessity, and we shall call the other type a "probable argument." The distinction depends on the relationship of the premises to the conclusion. Granted the truth of the premises, must the conclusion be true? If it is impossible that the conclusion should be false when the premises are true, then the argument is a valid one. If it is true that all reformers are idealists, and that all idealists are nonconformists-if these premises are true-then it is impossible for "all reformers are nonconformists" not to be true, and so this is a valid argument. But if the premises can be true while the conclusion can be false-as in the murder example above-then the argument is not valid, though the premises may make the conclusion highly probable. In a "probable argument," of course, the degree of the probability, whether high) low, or moderate, will depend on the quantity and quality of the evidence.5

Truth and validity
Thus far we have been discussing the distinction between a necessary and a probable argument. We shall now discuss the relationship between truth and validity. The main point is this: The truth or falsity of the premises (or the conclusion) has nothing whatsoever to do with the validity of the argument. This point is perhaps the most important lesson that one can learn about logical thinking: the distinction between the logical structure of an argument, on the one hand, and the truth or falsity of its evidence, on the other.

In connection with this distinction it is well to bear in mind the precise definitions logicians give the terms "truth" and "validity." Statements are true or false: A true statement is one which describes the facts correctly. Arguments are valid or invalid: A valid argument is one in which I the conclusion is necessitated by the premises. Note in particular that only statements are true or false; only arguments are valid or invalid. Thus logicians never (or almost never) say "a true argument" or "a valid statement."

Let us illustrate the distinction between truth and validity. "Human beings can't live on the moon, for there is no oxygen on the moon, and human beings can live only in places that contain oxygen." This is a valid argument, for if the premises are true then it would be impossible for the conclusion to be false. But we may question the truth of the second premise: Astronauts can bring their own oxygen with them when they land on the moon. So, though formally valid, this is not a satisfactory argument.

Further, an argument may be valid though all its assertions are preposterous: "If Eskimos were Cubans, and if every Cuban were an atomic scientist, then every Eskimo would be an atomic scientist." Validity is concerned with form or structure alone. If we symbolize Eskimos by E, Cubans by C, and atomic scientists by A, the form of the argument is: If E's are C's, and C's are A's, then E's are A's. The content of the substitutions for E, C, and A is irrelevant to validity. The rules of logic in relation to the substance of an argument are like the rules of arithmetic in relation to examples of application: "If you had twelve purple cows and seven of them were kidnapped, you would then have five in your possession." It would be an irrelevance, insofar as we were interested in the correctness of the arithmetic here, to tell us that you, speaking for your-self alone, never saw a purple cow.

Let us now apply the same considerations to a probable, or inductive, argument. Another murder trial. The prosecution presents three witnesses who testify that they saw the accused murder the victim. A motive is established, opportunity is proved, and other incriminating circumstances weigh heavily against the accused. The jury brings in a verdict of guilty beyond a reasonable doubt. We say that the jury acted reasonably, that is, they reasoned quite soundly in returning the verdict of guilty. But later we learn that the three witnesses were conspiring against the accused; they had framed him with his own gun, etc. The logical question:

Does this new information affect our previous decision that the jury acted reasonably? It does not. It was reasonable to believe that the accused was guilty on the basis of all the evidence known at the time of the verdict. The quality of the reasoning, then, is independent of the truth of the assertions on which the conclusion is based.

Now we can see that the truth or falsity of the premises is irrelevant to the "logic" of an argument. Just as an argument may be valid though its premises are false, so a conclusion may be probable on the basis of accepted premises, even though these premises turn out to be false or questionable.

The principles we just discussed will perhaps explain some familiar experiences, in which we listen to arguments in which each step follows from the preceding one without a logical flaw-but in the end we are of the same opinion still," unconvinced. This sort of thing often happens when someone is trying to convince us of the errors of our political or religious ways. These unconvincing arguments may actually have been quite sound from a purely logical point of view. But this means only that the conclusion did actually follow from the premises assumed by the speaker. Your refusal to accept his conclusion, then, indicates that you refuse to accept the truth of his premises, and if you examine the premises. carefully, you will be able to spot the place where your disbelief or doubt arises. (We are assuming that the argument is understandable.) Arguments of this kind are obviously unsatisfactory, for a completely satisfactory argument is not only correctly reasoned in the purely logical sense, but it is also one in which the premises are acceptable to us.

If the reader will turn back to the chain argument quoted from Leibniz (click here), he will find that the logical form of this argument is impeccable. You may or may not agree with Leibniz' conclusion that the soul is immortal. But if you do accept the truth of his premises, you must accept the truth of his conclusion, for the argument is valid. It is not the mark of a rational mind to say, "Your argument is valid, and your premises are true, but I refuse to grant the truth of your conclusion." This statement would be self-contradictory, for it first grants the validity of the argument and then denies it. It is like saying, "This is a square, but it has five sides." if an argument is valid, and the premises are true, the conclusion must be true, for this is what validity means: a structure such that if the premises are true, the conclusion must also be true.

But let us suppose that a reader disagrees with Leibniz' conclusion that the soul is immortal, or that he remains uncertain of its truth, although he does see that the argument is a valid one. This must mean, then, that he disagrees with, or is at least uncertain concerning the truth of, at least one of the premises. He may be able to point to the premise he disagrees with. But a reader untrained in metaphysics may find the meaning of the premises in Leibniz' argument obscure, or even unintelligible. A properly worded objection to the argument, in this case, might go like this:

"I am unconvinced of the truth of the conclusion, even though it follows logically from the premises. I must therefore be unconvinced of the truth of a premise or premises. I can't tell you which I consider wrong, for I don't understand them well enough to criticize, but I feel that there must be one premise that I would consider uncertain even if I understood it, for I don't accept the conclusion."

Now, this is not an unreasonable position to take, for, just as it is the mark of a boor to criticize an argument he doesn't understand, so it is the mark of an intellectually irresponsible person to accept an argument 6-he doesn't understand. If we wish to make friends and influence people, we may find intellectual irresponsibility less irritating to others than candor, but an intellectually responsible person will refuse to agree or disagree until he understands.

We have devoted a good deal of attention to the distinction between validity and truth, for this is a much misunderstood point. And there is a special kind of confusion concerning the matter of accepting an unproved premise "for the sake of argument." Suppose someone says, "If the need for large military expenditures should diminish within the next year . . ." and an objector breaks in at this point. "Stop right there," he says. "There's no point in your going on with your argument, for I don't accept your basic premise, and therefore can't accept whatever conclusion you will draw." But this is a refusal to be rational, for it may be very en-lightening to deduce the logical consequences of uncertain, or even false, premises. What the speaker was going to say was: "If the need for military expenditures should diminish within the next year, and we are not prepared with plans for immediate tax reductions and other stimulants to the domestic economy, then we shall have a severe depression." It is worthwhile considering the logic of this argument even if we do not admit the truth of its major assumption. For if the premises should turn out to be correct, it is certainly useful to know the consequences they entail, and thus prepare ourselves accordingly.

The good thinker, then, must often entertain unproved or even false assumption "for the sake of the argument." Scientists do this as a matter of course. Sir Isaac Newton's "first law of motion" tell us that if a moving body is not influenced by outside forces, it will continue in motion forever. The first part of this law contains an assumption which is contrary to fact, for there are no bodies which are "not influenced by outside forces." But physicists find this law useful, for it implies that a body will continue in motion for a longer and longer period as friction is reduced.

Our basic distinction is between is between the truth of the premises or conclusion, on the one hand, and the logical validity of the form, on the other. The distinction may be put in another way. It is like the difference, in an audit, between the soundness of the evaluations of items and the arithmetic used in adding up the totals. I may evaluate a pretzel factory building at $50,000, the machinery at $50,000, and the pretzels on hand at $900,000. Total assets: $1,000,000. The arithmetic is faultless, but the evaluations may be unrealistic.

Invalid argument
The significance of what the logician means by validity may become clearer when we understand the meaning of an "invalid argument." A formally invalid argument is one which is deductive in form, but in which the conclusion is not necessitated by the premises. The conclusion does not "follow" from the premises. This is the meaning of "non sequitur." Two examples:

(1) Cats climb trees
And squirrels climb trees;
Therefore, cats are squirrels.

(2) Manitobans live in the northern part of North America
And Canadians live in the northern part of North America
Therefore, Manitobans are Canadians.

Note that these arguments are similar in form. In each, the premises compare two things, cats with squirrels, and Manitobans with Canadians. In both, the compared entities have common characteristics. In (1) the common characteristic is tree-climbing; in (2) it is living in the northern part of North America. The conclusions are similar in that each tells us that one of the compared entities is identical with, or at least included within, the other.

Both of these arguments are invalid. In neither case does the conclusion follow from the premises. The relevant principle of logic is this:

The mere fact that two things have one or more characteristics in common does not justify us in concluding that the two things are identical, or even that one is included within the other. The fact that cats and fox terriers suckle their young does not justify the inference that cats are fox terriers; the fact that the Chinese and Japanese eat rice does not justify the inference that Chinese, after all, are nothing but Japanese.

Note that the conclusion in (1) is false; that in (2) true. Many people will think that (2) is a better argument than (1), but the arguments are equally bad from a strictly formal point of view. The second appears more plausible because the conclusion is true, but in order to see how bad the logic is, let the reader substitute the word "Alaskans" for "Manitobans" in (2). The new premise: "Alaskans live in the northern part of North America." The new conclusion: Therefore, Alaskans are Canadians."

The error we have just described may appear to be a very simple one, too obvious to mention, but we commit it often. The error occurs most often in complex contexts, and in subject matter that involves our emotions. We are particularly apt to overlook the badness of an argument when we believe the conclusion to be true, and particularly so when we derive emotional gratification from it. Consider the following: Joe Doakes must be a Communist, for he believes that Communists should be permitted to speak on university campuses, and we all know that Communists are in favor of permitting them to speak. This argument will sound plausible to many.

But, once more, the same error as above. Joe agrees with the Communists in one respect, we are told: both believe that Communists should have permission to speak. A common characteristic! But this does not prove that Joe is a Communist, for "the fact that two things have a characteristic in common does not prove that they are identical." Joe may be violently opposed to communism, and yet believe that students should be permitted to hear all points of view.

But a warning signal should be posted at this point. Though an argument is invalid, it may have considerable merit. An invalid argument is one in which the conclusion does not necessarily follow from the premises. We remarked earlier that the phrase "not necessarily" covers a lot of ground. The conclusion that cats are squirrels does not follow necessarily from the premises we noted above. Nor does the conclusion follow necessarily in this argument: "True theories are confirmed by careful experiments, and Einstein's theory has been confirmed by careful experiments, so his theory must be true." This argument, like the cats and squirrels argument, is invalid, and for the same reason. The premises tell us that Einstein's theory shares a characteristic with true theories, namely, that both are confirmed by careful experiments. But this does not guarantee that Einstein's theories must be trite, for the sharing of a characteristic does not prove identity. Scientists will agree that this is a correct analysis of the argument. For it is a well-known fact that many theories have been confirmed by experiments, only to be disproved by later experiments. This is the basic reason why scientists disclaim absolute certainty for their findings.

We are not, of course, singling out Einstein's theory as a special case, but are using it only as an illustration. Every experimental proof in science takes the same form. Nor are we casting the doubt of skepticism over scientific findings, for though the argument we just considered is in-valid from a technical, formal point of view, it differs from the foolish arguments considered in this chapter as sense differs from nonsense. This point requires careful consideration.

"Invalid" means that the conclusion does not follow necessarily from the premises. But in some cases the premises of an invalid argument seem to make the conclusion highly probable, as in the Einstein example; in others they do nothing of the sort. Why this difference? It all depends on the nature of the characteristic in which the two things agree. We say that the sharing of a common characteristic does not prove that two things are identical, or that one of these things is included within the other, but we come nearer to proving this inclusion in some cases than in others. For example, anarchists eat food, and so does Joe. Agreement in "eating food" proves absolutely nothing concerning the political similarities of food-eaters. But if we say, "The anarchists believe in abolishing all government controls, including the police force, and so does Joe," here the shared characteristic is highly significant. The argument is invalid, if we conclude that Joe is necessarily an anarchist, for the premises may be true and the conclusion false. Joe may merely be a "rebel without a cause." But the probability is high that he is an anarchist.

In other words, if the shared characteristic is one that is possessed only by anarchists, or if there is a high probability that anyone having the shared characteristic is an anarchist, then we can translate our invalid argument into a valid one. If we can say, "Anyone who believes in the abolition of all government controls, including the police force, is probably an anarchist," and we find that Joe so believes, then obviously Joe is V, probably an anarchist. This argument is valid for the conclusion necessarily follows from its premises.

If we return now to the argument that led to this discussion, the premise "True theories are confirmed by careful experiments" can be translated into "If a theory is confirmed by careful experiments, then it is probably true." (Note that we did not say "necessarily.") And thus, since Einstein's theory was so confirmed, we can derive the valid conclusion that it is probably true. But this kind of translation is possible only when the shared characteristics are of such a nature as to make it probable that one thing is included within another. We cannot translate our "cats and squirrels" argument in this way, for "Cats climb trees" can't be translated into "If it climbs trees it probably is a cat" (or a squirrel). Thus the conclusion of this argument is not a probable one. The point is this: when we note an argument based on shared characteristics hereafter, let us V-also note the significance of the shared characteristics. Some shared characteristics may yield a probable conclusion. To repeat, "not necessarily' covers a lot of ground, from a highly probable conclusion to a worthless one.

The fallacy of the "shared characteristic," known technically as the fallacy of the "undistributed middle term," resembles the idea of "guilt by association." Just as the sharing of a characteristic does not prove identity, so the fact that a man knows a Communist does not prove that he is one. But just as the significance of the characteristic, or characteristics, is important in establishing probability, so with one's associates. If a state's attorney has gangsters as his constant companions, there may be a justifiable suspicion as to his honesty, and we may want state's attorneys who are not only unconvicted of crimes, but also above suspicion.

In concluding our discussion of validity and its relation to the truth of assertions-the subject matter of "deductive" logic-we shall sum up the matter schematically. There are four possible combinations of premises and conclusion with respect to their truth and falsity:

(1) the premises may be true, and the conclusion true
(2) the premises may be false and the conclusion true
(3) the premises false and the conclusion false
(4) the premises true and the conclusion false.

We shall now illustrate these combinations with examples of invalid and valid arguments, respectively. In arguments which are invalid, we can find each of the four combinations listed above:

(1) The invalid Manitobans argument. (click here)

(2) "Socialists are capitalists, and those who wish to abolish private property are capitalists, so Socialists wish to abolish private property." (The sharing of characteristics does not prove identity.)

(3) Too obvious to illustrate.

(4) The "cats are squirrels" argument. (click here)

In valid arguments, on the other hand, only the first three of these combinations of truth and falsity can be illustrated:

(1) The familiar "All men are mortal" syllogism.

(2) "Socialists are capitalists, and capitalists favor the abolition of private property; therefore, Socialists favor the abolition of private property." This is a valid argument, for if the premises are true, then the conclusion would have to be true. It may be helpful to compare this illustration of a valid argument for combination (2) with the illustration of the invalid argument for combination (2) given above. Each has false premises and a true conclusion. But in the invalid argument the conclusion is not necessitated by the premises; that is, acceptance of the premises as true would not require us to accept the conclusion as true. The conclusion of that argument happens to be true, but we cannot say: If these premises are true, then the conclusion must be true. But this is precisely what we must say of the valid argument illustrating combination (2) in this paragraph.

(3) The valid Eskimos argument (click here)

(4) This combination is impossible when an argument is valid. For consider the meaning of the term "valid argument": one in which the truth of the premises requires us to accept the truth of the conclusion. If we say that an argument has true premises and a false conclusion, we thereby declare that it is not a valid argument

The main points of this chapter may be summed up.

(1) By formal logic alone we cannot prove the truth of any assertion. What formal logic tells us is that if we start with true premises and reason logically from these premises, then our conclusion must be true.

(2) If our premises are false, or even uncertain, then even when we reason logically our conclusion has not been proved to be true.

(3) When the logical form is invalid, that is, when the reasoning is illogical, then even true premises cannot guarantee a true conclusion.

(4) In arguments like the Einstein example, we found that some invalid arguments can be translated into valid arguments yielding probable conclusions.

We also recall our earlier discussion of the ambiguity of "good" as applied to arguments. A completely satisfactory argument, we said, was not only valid in form, but also contained true premises. A bad argument, then, is one which is either invalid or which lacks truth. We have also seen that true conclusions may be supported by either true or false premises in invalid arguments.

A final point. A bad argument, that is, one invalid in form or containing false premises, cannot prove a true thesis. But it is also important to remember that a bad argument does not discredit a true thesis, though it may sometimes appear to, as when a weak premise is attacked in a debate. The refutation of the premise may then seem to be a refutation of the thesis. In general, it is better to present a few good arguments for a thesis, rather than a great many, one or more of which may be weak, for the opponents are apt to seize on the weak premise and, by discrediting it, appear to discredit the conclusion. But the logical person will consider the merits of ideas regardless of the bad arguments used to support them.

FOR DISCUSSION AND WRITING

1. Analyze the following arguments. In each, identify the premises, conclusion, and (if present) logical indicators.
a. I think that Professor Dwyer is a particularly good teacher. His classes are always full, and lie is known to have excellent rapport with his students.
b. I am not as happy as I was before Roberta left me. And because I am unhappy, I will never be able to love again.
c. The candidate has claimed that the United States is a nation of neglected poor. If this were the case, we would suppose that our welfare programs were seriously lacking. But this is not the case. The federal and state governments spend 20 percent of their funds on welfare. Therefore, this cannot be a nation of neglected poor.
d. Con McAuliffe is a successful businessman. He always buys luxury automobiles, and his house is a mansion.
e. The Jets have not played as well since Joe Namath's knees gave out. Because their play is so poor, they will never win the Super Bowl.

2. Analyze the formal structure of the following syllogisms. What are the major and minor premises, the major and minor terms, and the middle term? If any syllogisms are incomplete, furnish the missing parts.
a. All A's are B; C is an A; therefore C is a B.
b. All girls are made of sugar and spice. Alice is a girl. Therefore, Alice is made of sugar and spice.
c. All athletes are well conditioned. Frank is an athlete. Therefore, Frank is well conditioned.
d. Fred must have accepted a promotion, because his salary has in-creased twofold.
e. All newspapers are slanted in their coverage. The Times is a newspaper. Hence, the Times must be slanted.
f. The editorials of that newspaper have a Republican bias; the owner is a Republican, isn't he?
g. All artists are nonconformists. Larry is an artist; thus Larry is a nonconformist.
h. Mary has always been precocious; she started reading at the age of three.
i. That automobile ought to be the best one on the market. All the race drivers recommend it.

3. Analyze the following syllogisms. Are their premises true? Is the conclusion true? Valid?

a. All monkeys have tails; Mr. Smith is a monkey; therefore, Mr. Smith has a tail
b. All petunias have petals; all roses have petals; therefore, all roses are petunias.
c. If all men were viola players, and if every viola player was bow-legged, every man would be bowlegged.
d. Francis lives in Belgium, and Belgium is a European country; therefore, Francis is a European.
e. Gloria is a Belgian, and Belgium is a European country; therefore, Gloria is a European.
f. Birds live in trees, and squirrels live in trees; therefore birds are squirrels.
g. Woodfinches live in trees, and robins live in trees; therefore woodfinches are robins.
h. Woodfinches live in trees, and birds live in trees; therefore woodfinches are birds.
i. A classic is a piece of literature that has undergone the test of time; Shakespeare's King Lear has undergone the test of time; therefore, King Lear is a classic.
j. Communists believe all property should be owned by the state; Burl believes that all property should be owned by the state; therefore Burl is a communist.
k. All Scotsmen are stingy; Bill is Scotch; therefore, Bill is stingy.
l. Most people are basically selfish; selfishness is an evil; therefore, most people are evil.

4. Write valid syllogisms which have the following statements as conclusions.
a. Narcotics are poison.
b. Education can take place outside of school.
c. Some ice cream is delicious.
d. Some professors are interesting.
e. Manitobans are Canadians.
f. A thunderstorm is dangerous.
g. Only education can alleviate prejudice.
h. intelligent women never leap to unwarranted conclusions.

5. Write a syllogism for each of the following combinations of premises and conclusions:
a. The premises are true, and the conclusion is true.
b. The premises (or one premise) may be false and the conclusion true.
c. The premises are false, and the conclusion is false.
d. The premises are true, and the conclusion is false.

6. Test the validity of the following arguments by rewriting the argument as a syllogism.
a. The study of religion always includes the study of how men should act toward each other. Our schools also have the responsibility of teaching the young how to live in society; it is logical to conclude, there-fore, that religion should be taught in our schools.
b. It is a truism that a political candidate should be very wary of declaring his stand on controversial issues. In fact, the successful candidate is often the one who appeals to the largest number of voters by promising everyone what he or she wants. Our candidate, therefore, should avoid controversial issues.
c. The president has a moral obligation to do for the country what is best for the majority. But special interest groups are constantly attempting to persuade him to act in their behalf. To be morally just, however, he should not listen to their appeals.

---

CHAPTER 7
Some Patterns of Reasoning

The ability to construct a valid argument is extremely important when writing convincing argumentative prose. It is time now to examine he notion of validity somewhat more closely and to explain why some structures of reasoning permit us to draw valid deductions whereas others do not. In carrying out this task we shall have to indulge in a technicality or two.

Four patterns of reasoning
We begin by considering a lion, his cage, and a zoo. We shall construct four very simple arguments which reveal some of the basic patterns of reasoning:

1. The lion is in his cage, and the cage is in the zoo. We draw the 'conclusion: The lion is in the zoo. Obviously this conclusion follows necessarily from the premises. The argument is valid.

2. The lion is in his cage, but the cage is not in the zoo. The conclusion: "The lion is not in the zoo,' follows necessarily from these premises. Valid.

3. The cage is in the zoo, but the lion is not in his cage. Now, can we conclude with certainty concerning the whereabouts of the lion? Obviously not, for the lion may or may not be in the zoo. If we draw either one of these conclusions: "He is in the zoo" or "He is not in the zoo," we will draw a conclusion not warranted by the premises, and the argument will be invalid. Neither conclusion follows necessarily from these premises.

4. The cage is in the zoo, and the lion is in the zoo. We cannot draw a necessary conclusion concerning the lion's relationship to his cage. He may be inside it, and he may not be. To draw either conclusion as following from the information given would be illogical.

In these simple examples we find four basic patterns of reasoning, two of which are sound, and two unsound. The information given to us by the premises permits us to draw valid conclusions in patterns 1 and 2; not in 3 and 4. We shall now draw diagrams to illustrate these patterns in order to get a visual picture of the difference between valid and invalid structures of reasoning. We shall combine the premises of each argument in diagrams, in order to see why the conclusions are necessitated by patterns 1 and 2 and not by 3 and 4.

1. Premises: The lion is in his cage, and the cage is in the zoo.

This diagram exhibits the premises: the lion in his cage and the cage in the zoo. The conclusion. "The lion must be in the zoo" is unavoidable.

2. Premises: The lion is in his cage, but the cage is not in the zoo.

This is the only way in which it is possible to diagram the information given by the two premises. The conclusion: The lion cannot be in the zoo.

3. Premises: The cage is in the zoo, but the lion is not in his cage.

Difficulties arise when we try to diagram these premises. Let us start with the first premise:

We must now show "the lion is not in his cage." There are two ways of showing this, in conjunction with the first premise:'

Diagrams "a" and "b" are both faithful to the premises, but neither one is necessarily required by the premises. Diagram "a" shows the lion in the zoo; "b" shows him outside. An argument which drew either conclusion would be invalid.

4. Premises: The cage is in the zoo and the lion is in the zoo. Similar difficulties arise. We begin by diagramming the first premise:

The second premise tells us that the lion is in the zoo, Two ways to show this:

Each diagram shows us both the cage and the lion in the zoo. But the premises give us no information concerning the relation of the lion to his cage. To conclude that either "a" or "b" necessarily follows from these premises is to draw an unjustified inference.

Our first set of illustrations is oversimplified, for we were talking about a particular lion, a particular cage, and a particular zoo. Most reasoning concerns classes of things (cabbage, kings, etc.) rather than exclusively individual objects. Our next four illustrations will exhibit the same general patterns for classes of things. Note the responsive similarities in formal structures for the arguments bearing the same numbers:

1. Whales are mammals, and mammals are animals, so whales are animals. Diagram to illustrate the premises:

If the premises are accepted, then the conclusion must be accepted. Valid.

2. All monarchists are conservatives, and no conservatives are Utopians. We may properly conclude that no monarchists are Utopians. The diagram shows that the argument is valid:

3. All Hindus are vegetarians, and no Sikhs are Hindus, so no Sikhs are vegetarians. The conclusion "No Sikhs are vegetarians" is not justified by the evidence presented. The argument is invalid. As before, we draw a diagram for the first premise:

The second premise tells us to draw Sikhs outside of Hindus, but it tells us nothing about the position of Sikhs with respect to vegetarians. We can draw at least two diagrams to show the possibilities:

Diagram "a" shows Sikhs outside the class of vegetarians. This was conclusion of the argument But "b" is also a possibility, and "b" shows Sikhs inside the vegetarian box. To say that "a" (or "b") follows necessarily from the premises is to draw an unjustified inference.

4. Iranians live in the Near East, and Kurds live in the Near East, so Kurds must be Iranians. Invalid. From the information given in the premises, Kurds may or may not be Iranians. We cannot conclude that they must be.

The first premise gives us:

We must now draw Kurds inside the box "People who live in the Near East" But they may be inside the Iranian box or outside. Neither conclusion is necessitated, so the argument is invalid.

We may sum up our discussion of the four patterns of reasoning in symbolic form. We shall use the symbol "A," "B," "C" for any three classes of things:

1. A is in B and B is in C. That A is in C follows necessarily.

2. A is in B and B is outside of C. A must be outside of C.

3. A is in B and C is outside of A. We cannot conclude that C must be outside of B, for C may be outside of A and inside of B. Nor does any other conclusion follow necessarily.

4. A is in B and C is in B. We cannot conclude that C is in A (or A in C), for, though both must be in B, they may be outside of each other.

Before we continue, a further word or two should be said concerning what tile diagrams show. When we draw diagrams for the premises in a valid argument, the conclusion is seen to be inescapable. But when we draw diagrams for the premises in an invalid argument, we see that no definite conclusion is necessitated. We can draw the premises of an invalid argument without exhibiting the particular conclusion which the argument drew. In other words in an invalid argument we need not accept the conclusion even though we accept the premises. The conclusion may be true, but it is not proven true by the premises.

Let us sum up for a moment. Arguments 1 and 2 were valid in each set; 3 and 4 were invalid. Let us now examine the technical rules of logic violated by patterns 3 and 4. Argument 3 is an example of the fallacy called illicit distribution," and argument 4 illustrates the fallacy of the "undistributed middle term." These are two of the most frequently encountered errors in reasoning. In order to understand the meaning of these fallacies, let us examine the technical logical term known as "distribution."

"Distribution"
The logician speaks of the "distribution" of words that designate classes of things (apples, mortals, emotions). To say that' a term is distributed means that we have said something about all members of the class to which it refers, that we have asserted something about each and every member of that class. Thus, in "All dogs are animals," "dogs" is a distributed term, for we said all dogs. But we did not say anything about all animals in this particular sentence. Dogs constitute only part of the class of animals, so our sentence refers only to some animals. "Animals" is an undistributed term in this sentence. Similarly, if we had said "Some dogs are hunters," both dogs and hunters would be undistributed.

In the sentence "No men are angels" both terms are distributed. The sentence asserts that each and every man is outside the class of an-gels, and it also says that each and every angel is outside the class of men.

To sum up. The distribution of the subject term in a sentence depends on whether its quantifier is "all," "no," or "some." The distribution of the predicate term in a sentence is dependent on whether the sentence is affirmative or negative, and this distinction in turn depends on the copula of the sentence. The copula, a form of the verb "to be," connects the subject and predicate. When the copula is "is" or "are," the sentence is affirmative (S is P); when it is "is not," the sentence is negative (S is not P). "No S are P" is classified as negative, since it states that S are not P, that is, all of S are excluded from all of P. The sentence "All non-S are non-P" is affirmative, since the copula is "are."

The distribution of the predicate term may now be summed up in two rules: (1) Affirmative sentences never distribute the predicate term (P): All S are P, Some S are P; (2) Negative sentences always distribute the predicate term: No S are P, Some S are not P.

The distribution of S and P in the four possible types of subject-predicate sentences are shown below:

All S (distributed) are P (undistributed)

No S (distributed) are P (distributed)

Some S (undistributed) are P (undistributed)

Some S (undistributed) are not P (distributed).

This discussion of "distribution" by no means exhausts the subject, but it is sufficient for our purposes. We are now ready to explain the fallacies of "illicit distribution" (illustrated by argument 3) and "undistributed middle term" (illustrated by argument 4). Let us take number 4 first, since this fallacy is already familiar to us. The reader will recall the many examples of this error-the error involving "shared characteristics"-in the last chapter: the arguments concerning cats and squirrels both being tree-climbers; the argument that told us that Chinese must be Japanese since both eat rice; that Manitobans must be Canadians since both live in North America; that Einstein's theory must be true since it has been confirmed by careful experiments, and all true theories are so confirmed. We shall see, in a moment, how we may apply this particular technique of logic-the "distribution" idea-to these arguments.

Just one further preliminary comment: the "middle term" of a syllogism is the term that appears in the two premises, serving as a connecting link between the other two terms,

Validity of the syllogism
The fallacy of the "undistributed middle term" refers to a rule of logic which tells us that the middle term of a syllogism must be "distributed at least once" in order to permit the drawing of a valid conclusion. This means that, if the middle term is not distributed at all, no valid conclusion is possible. Argument 4 illustrates this fallacy. Let us examine it:

(All) Iranians live in the Near East (middle term)
And (All) Kurds live In the Near East (middle term)
Therefore, Kurds are Iranians.

"Live in the Near East" is the "middle term" (defined as the term which appears in both premises). This term was distributed in neither premise. "Kurds" and "Iranians" were distributed in their respective premises, for we assume that "all" was understood for each, but the premises refer only to "some" people who live in the Near East. Thus this syllogism violates a rule of logic. The reader will find the same technical error in the other examples of this fallacy in the previous chapter. The diagrams above (click here) show us the sense of the rule.

In the patterns of reasoning illustrated in those diagrams, only example number 4 violates this rule. In argument number 1 the middle term is "mammals." This term is not distributed in the premise, "All whales are mammals," but it is distributed in "Mammals are animals," for "MI mammals" is intended here. So the rule is satisfied. In argument number 2 the middle term is "conservatives." It is distributed in "No conservatives are Utopians." In argument number 3 the middle term is "Hindus." This term is distributed twice. This argument, then, does not violate the rule concerning the distribution of the middle term, but it violates a different rule involving distribution. Let us examine it.

A second rule of logic tells us that "if a term is undistributed in the premises, then that term must not be distributed in the conclusion." This is the sense of the rule: when a term is distributed, information is given concerning each and every member of the class referred to. When a term is undistributed, information is given only about some of its members. From information concerning "some" we can draw no conclusion concerning "all." For example, if experience teaches me that some women are fickle, I cannot logically conclude that all are. We commit the fallacy of going from "some" to "all" when the conclusion distributes a term that was not distributed in the premises.

Let us restate number 3, which illustrates the fallacy of "illicit distribution":

All Hindus are vegetarians
And No Sikhs are Hindus
Therefore, No Sikhs are vegetarians.

We examine the distributed terms in the conclusion to detect a possible violation of the rule against illicit distribution. If a term is distributed in the conclusion, then it should have been distributed in the premises. We find that both of the terms in the conclusion are distributed. Now examine these terms in the premises. "Sikhs" was distributed in the second premise. But "vegetarians" was not distributed in its premise, and so the rule is violated. When we conclude that "no Sikhs are vegetarians," we assert something about all vegetarians; "All of them," we say, are outside the class of Sikhs." But the premise gave us information only about some vegetarians.

(All) Senators have traveling privileges
And (All) Senators are politicians
Therefore, (All) politicians have traveling privileges.

"Politicians" was undistributed in the second premise and distributed in the conclusion.

Let us examine the conclusions of arguments 1, 2, and 4, above, in which this second rule is not violated.
Number 1: Whales are animals. Whales is distributed in the premise as well as in the conclusion.
Number 2: No monarchists are Utopians. Both terms distributed in the premises as well as in the conclusion.
Number 4: All Kurds are Iranians. Kurds is distributed in the premises. This last argument, how-ever, violates the rule concerning the distribution of the middle term.

Negative premises and conclusions
So much for two of the basic rules of validity for syllogisms. There are also three additional rules, each of which concerns negative premises or a negative conclusion:

(1) A valid conclusion cannot be drawn from two negative premises;
(2) A negative premise requires a negative conclusion;
(3) A negative conclusion requires a negative premise.

These five rules are like the axioms in Euclidean geometry. They are necessary and sufficient to test the validity of any syllogism involving subject-predicate sentences in ordinary language.
With practice you should now be able to test the validity of almost any deductive argument, to determine whether the conclusion follows from the premises. This skill will be helpful in both writing and reading argumentative prose. By using the syllogism as a model or form, your paper will be compelling in its logic. And in reading someone else's deductive argument, you will be able to reduce it to a syllogism and test it.


FOR DISCUSSION AND WRITING

1. Analyze the following syllogisms by diagraming their premises and conclusions. Which arguments are valid? How many possible diagrams can be drawn to illustrate the invalid arguments?

a. All men are animals.
Emil is a man.
Emil is an animal.

b. All men are animals.
Emil is an animal.
Emil is a man.

c. All men are animals.
All men are reasonable.
Animals are reasonable.

d. All animals are sentient.
Emil is an animal.
Emil is sentient.

e. All students are activists.
Al is an activist.
Al is a student.

f. All students of government are activists.
All politicians are students of government.
All activists are politicians.

g. All students are activists.
All activists are radicals.
All radicals are students.

h. All students are activists.
Al is a student.
Al is an activist.

i. Some students are activists.
Al is a student.
Al is an activist.

j. No student is an activist.
Al is a student.
Al is not an activist.

k. No professor is careless.
Muriel is not a professor.
Muriel is careless.

l. No animal is rational
Larry is an animal.
Larry is not rational.

m. Some animals are dogs.
All dogs are four legged.
All four-legged creatures are animals.

n. Some animals are four-legged
Dogs are four legged.
Dogs are animals.

o. Some animals are not carnivorous.
Don is an animal.
Don is not carnivorous.

p. Some animals are carnivorous.
Don is carnivorous.
Don is an animal.

q. Some humans are not evil
Politicians are human.
Politicians are not evil.

2. In the preceding syllogisms, which terms were distributed? Which were undistributed? Indicate the middle term.

3. Below are five rules of logic governing the syllogism. Apply them to the syllogisms in question 1. If the argument in any syllogism is not valid, indicate the particular rule it violates.
a. The middle term of the syllogism must be distributed at least once.
b. If a term is undistributed in the premises, then that term must not be distributed in the conclusion.
c. A valid conclusion cannot be drawn from two negative premises.
d. A negative premise requires a negative conclusion.
e. A negative conclusion requires a negative premise.

---

PART THREE
Truth and Falsity

CHAPTER 8
Truth and Evidence

The truth is something that all of us would like to know, except, of course, when it displeases us. But what is it?

We have been told that the question, "What is truth?" was asked by jesting Pilate, but that he "stayed not for an answer." His hasty departure implied that the question was an unanswerable one. We have been told that the truth is something which, when crushed to earth, is sure to rise again; that it is something which can never lose out when engaged in a grapple with falsehood; that it is something which, when told, shames the devil; that it is folly to tell women the truth; that there is nothing so powerful as truth, and nothing so strange; and that the truth will make us free. But what is the truth?

Mankind has been struggling to know the truth for some years now, and at last accounts the final results are not yet in. There are some who insist that the truth is absolute; others say it is relative. Some believe that mankind can achieve the truth; others, the skeptics, deny this possibility. Empiricists believe that science can give us reliable knowledge, but they are skeptical with respect to metaphysical truths, that is, the "ultimate" truths concerning man's origin and destiny. Others are skeptical even concerning the evidence of their senses.

Our primary concern is with a more prosaic kind of truth, the kind we shall call "factual." By "factual truth" we mean only this: a truth relating to the facts of human experience. And, prosaically, let us define truth. What do we mean when we say that a statement is true? We mean that a statement agrees with, or corresponds with, the facts. This is the definition Cardinal Newman had in mind when he said, "Truth means facts and their relations, which stand towards each other pretty much as. subjects and predicates in logic." For example: The statement, "Lyndon Johnson was President of the United States from 1963 to 1968" is true, for it agrees with the facts. A true statement describes the facts correctly, something like the way in which a map pictures a territory. When a map shows the relations of towns, rivers, mountains, and valleys just as they exist in reality, then the map is a true map. A true statement is like a true map.

"Logical relativism"
Let us now consider some of the implications of this definition of truth. These implications may be best brought out by considering the theory of "logical relativism." A relativist, in the logical sense, denies the possibility of universal truths. He holds that what is true for one man may be false for another, what is true at one time may be false at another time, and what is true in one place may be false in another. Truth, he holds, is relative to the circumstances of the viewer, to his "frame of reference.

The theory of logical relativism is sometimes associated with the theory of relativity in physics--a familiar theory since Einstein. But they have little in common, though both emphasize the "frame of reference." In physics relativity refers to the importance of establishing a "frame of reference" whenever we describe motion. For example, if we ask, "Is the furniture in your home moving at the present moment?" the answer should be "Yes and no, depending on your frame of reference." Since the earth is moving around the sun at a speed of eighteen miles a second, your furniture is moving at the same speed relative to the sun. But, relative to the earth, your furniture is at rest. In other words, the furniture is moving in one frame of reference, but it is at rest in another. A seated passenger in a moving car is at rest but also in motion.

But the physicist does not say that truth is relative. He believes that it is really or "absolutely" the case that motion is relative to a frame of reference. He does not say that the relativity theory is true for some scientists and false for others. And this is precisely what the logical relativist declares. He says that what is true for anyone will depend on his past experience, his training, his education, and the ideas accepted by his time and environment. Another man, with different experiences, will find a different truth. One man's food, in other words, is another man's poison in matters of truth as well as diet

But when the physicist says that motion is relative, he does not think of the principle or law of relativity as relative. He believes that it must be true for everyone that motion is relative to the frame of reference.

In support of his position the logical relativist points to the differences of opinion that divide the human race in time and place. People once believed that the sun moved around the earth; today we believe other-wise. But was not the former belief true for the people of the middle ages? The relativist notes the different points of view in the Orient and Occident, and in the countries on different sides of the Iron Curtain. He quotes Pascal's aphorism: "Truth on one side of the Pyrenees, error on the other side."

Now, there is a good deal to say in behalf of this relativistic position, for human beliefs do differ. The candid observer will in fact be deeply impressed by the actual variety of opinions that men hold on all matters of real importance, on politics, religion, and so on. Consider the different versions of events that lead to the breaking up of friendships! Each person seems to be right from his point of view. But none of these considerations requires the conclusion that truth is relative, if we accept our definition of truth: a true statement is one that agrees with the facts.

The issue between the relativist and the nonrelativist may be a terminological one. The dispute will then be verbal, depending on how we define the word "truth." By "true" the relativist may mean "believed to be true." Let us try to clear up this semantical confusion. In this discussion we shall assume that it is possible to know whether some factual statements do or do not correspond to the facts. We assume that the statement "Water boils at 212? F. at sea level" does correspond with the facts, and that this is not "just an opinion," or mere "belief." It is statements of this kind that we shall refer to as factual truth, or truth in the strict sense of the term.

There are also looser usages. Thus we sometimes speak of religious or moral truth. "What is right for one person must be right for all similar persons in similar circumstances" is a principle that may appeal to us as a "true" moral principle. What we usually mean is that this moral belief is one that every rational person ought to subscribe to, though we admit that it cannot be verified by the eye, or other senses. It is also difficult to speak of literal truth in assessing the "causes" of great social events, such as the origin of "capitalism" or the causes of a great war. Different theories will appeal to us partly because of our prior sympathies and interests and national origins. Historical facts, as Dwight MacDonald has put it, are not "solid, concrete (and discrete) objects like marbles. Rather are they subtle essences, full of mystery and metaphysics, that change their color and shape, their meanings, according to the context in which they are presented."

When people argue the respective merits of unrestricted free enterprise versus government controls in producing higher living standards, value judgments will influence opinion. Some people prefer freedom, others prefer security, just as some enjoy a touch of danger, while others abhor danger. Our reception of theories may also depend on whether we stand to gain or lose by change. It is difficult to attain literal truth in such discussion, even when "factual" predictions are made. Prophecies of ultimate disaster cannot be verified. It is always safe to make a prediction if one does not set a time limit.

These considerations would seem to support the relativistic position. But even in the complex issues and predictions just discussed some views will seem to be more reasonable than others, some will seem plausible or probable and others preposterous. Our primary interest, however, is in clarifying the meaning of truth in the literal sense. Let us return to our example of a map which purports to delineate an area. A true map will correctly describe; a false map will describe incorrectly. In a largely uncharted area, on the other hand, it may be impossible to determine whether a tentative map is correct or not. Its truth or falsity will then be 'uncertain.

Now consider a map of the world printed in the year 1492 which did not show the continents of North and South America. The people of that time believed that their map was a true map of the world. Shall we say, rather, that in 1492 men were ignorant of the true nature of the globe, for some of its continents were as yet undiscovered, and so, since this map corresponded with their ignorance, this map was false? Though believed true, the map was never true, though it corresponded with the known facts. But it did not correspond with the actual facts. Instead of saying that what was true at one time became false at another time, let us say rather that what was believed to be true was actually false, or at least that what was believed to be true at one time is believed to be false at a different time.

It is quite inappropriate, then, to say that anything can be true here and false elsewhere, if by true we mean that it correctly describes the facts. We would never think of saying that a map of the United States is true for Americans but false for Russians.

But the relativist has another type of argument Take the statement: "It is warm today." Is this statement true at all times and all places? Obviously, the answer is No. The statement, "It is warm today, may be true in July but not in January; it may be true in San Diego but not in the northern woods of Wisconsin. But this does not prove that truth is relative. The plausibility of the relativist's point is based on the fact that "It is warm today" is phrased in the vague language of every-day speech. In order to test the truth of a statement, however, we must first give it the precision of a scientific sentence. A scientific statement is dated and located. A properly formulated scientific statement concerning the weather would read something like this: "The temperature was 89 Fahrenheit on September 28, 1973 at 3:00 P.M., Pacific Daylight Saving Time, at the meteorological station in San Diego, California Now, stated in this form, we have a statement that may or may not correspond to the facts. That it was 89? F. in San Diego at 3:00 P.M. on September 28, however-q true-must be true for the Chinese as well as for the Australians, and it must be true forever, for our statement was dated and located for a specific time and place.

Similarly, when we look at a map in a contemporary history book showing "Europe in 1800," it would be incorrect to say that the map was true in 1800 but false today. It is the case, of course, that the map corresponds to the political boundaries of Europe in 1800, but not to those of today. But when a map is drawn, it carries with it the qualification of a specific time-this is the way the boundaries looked in 1800. And if the map was true in 1800, then it will be true forever, for the year 1800.

Let us examine another type of case that seems to support the relativist's argument-the relativity of perceptions. I look at a book and say that it has a blue cover. A color-blind man says that it has a gray cover. Are not both of us correct, though we state contrary ideas? Is not one man's truth error for another, then? But if we analyze this situation properly, we shall find that it does not support the relativist's position. We should distinguish between two types of statements. When I said, "The cover is blue" I may have meant that the cover will appear blue to a person with normal vision; that is, its pigment reflects light rays measuring about 485 millionths of a millimeter. If the color-blind man denies that it is "blue" in this sense, then at least one of us must be wrong. Whether or not such rays are being reflected cannot be true at a certain time and place, and also be false. It cannot both be true for one physicist and false for another, regardless of whether one is color-blind. But I may have meant something else by "It is blue." Perhaps I meant, "I have the experience of seeing a blue cover." And if the color-blind man meant that he has the experience of seeing a gray cover, then both of us would be right. There is no inconsistency in saying that two persons have different experiences. It is common, even proverbial, knowledge that people feel differently when an ox is gored, depending on the ownership of the ox.

There is a relativity of feeling and of experience, then, but this is not a relativity of truth. People have different experiences, depending upon the physical conditions of their bodies, their past experiences, their conditioning and reconditioning, so that one's perceptions and responses will be relative to one's "frame of reference." A man cannot be wrong if he correctly reports what he has himself experienced. But when we talk about the speed of light, or the light waves reflected from the cover of a book, we are talking about something other than our own experiences, and if our physical frames of reference are the same, then conflicting descriptions cannot both be true.

We do not wish to minimize the great importance of the subjective element in perception. This element is often overlooked, and we frequently objectify subjective experiences. There are many parables and stories that illustrate this point. There is the parable about the blind men and the elephant. The blind men encountered an elephant along the highway and compared notes after examining him. "'The elephant," said one, is something like a stone, cool and smooth, and shaped like a curved cylinder." "No," said another, "the elephant is like a hairy rope." A third said, "No, the elephant is something very massive and solid, full of little hills and valleys." Now, obviously, each of the blind men was right in saying that the elephant had the characteristics he mentioned, though no one had the whole picture. Each had a different conception of the elephant, and each idea was consistent with the other ideas when put together, for one man had come into contact with the tusks, another with the tail, and the third with the side wall of the body. But each was wrong in thinking that the elephant was nothing but what be had experienced it to be.

So much for the doctrine of logical relativity." We have argued that { if a statement is true, then it is always true, and true for everybody. But a new kind of problem arises: How can we be sure that a statement does correspond to the facts? For example, the statement, "'There is buried treasure beneath the house in which I am writing," is either true or false, but no one knows which, for certain. It is one thing to define truth as correspondence with the facts, but another to determine whether or not a statement does so correspond. It is true or false that some world leaders intend to start a third world war, but we do not know whether this statement actually corresponds to the facts. Let us examine the new problem: Can we be certain about the truth of any statement?

Truth and certainty

Are we ever justified in saying, "This belief of mine is absolutely true?" Are there any beliefs of which we can be absolutely certain? Can we ever assert with complete confidence that we are completely right and the other fellow completely wrong? We may deceive ourselves. It is a common experience, in a law court, to find several honest and sincere witnesses giving different reports of how an accident occurred. A person sometimes quite positive that he saw something he did not see. And then there are the distortions, conscious or unconscious, based on personal interest. In the Japanese film Rashomon a murder occurs, and three different versions of the incident are presented by those involved in it as participants. One naturally expects that each of these will give a biased version in order to justify his own behavior. But there is also a fourth witness, a supposedly impartial observer, and he too presents a version that is distorted and subject to doubt.

The question, then, is whether we can ever be certain that any statement is true. Even if a statement is verifiable by observation, and we make the claim of truth in the strict sense, can we be positive that we are not mistaken?

The French philosopher Rene' Descartes raised the question: Can we ever be certain that we really know? Descartes had been brought up with a traditionalist education, and he later came to believe that many of the ancient beliefs he had been taught were really false. If some of these beliefs are proved false, he asked himself, "How can I be sure that any of them is true?" On what basis can I say, "'This or that belief is certainly true"? He then initiated a new method of philosophical study, the method of doubt, and thereby gave a new direction to the course of modern philosophy. He said that he would doubt everything without exception, and then see whether there was anything that could successfully withstand critical scrutiny.

Descartes even doubted the evidence of his senses. Perhaps, he said, they deceive me. The traveler on the desert sees a mirage in the distance, but the wooded oasis he thinks he sees is only an optical illusion. Descartes said that he could even doubt whether he was actually awake at a given moment, for he might merely be dreaming that he was awake. Now do you think that you can be absolutely certain that you are not dreaming at this moment? Is it conceivable to you-I do not mean, is it likely, but is it even conceivable--that your alarm clock will ring shortly, that you will awake, and say: "What a vivid dream that was-I dreamt that I was reading a book-though I forget what it was about"? And perhaps you have heard about the college professor who dreamed that he was lecturing to his college class-and when he awoke he found that that was exactly what he was doing!

In the end, of course, Descartes found that there was one belief he could not doubt, and this was the belief that he himself existed. For, he said, if I doubt my own experience, my doubt implies that there must be a doubter. If I did not exist, then I could not even doubt that I existed.

Skepticism

In its philosophical form, skepticism is a doctrine which denies that the human mind is capable of attaining genuine knowledge about any-thing. In the history of mankind there have been many skeptics who carried doubt very far. In ancient Greece, in the fourth century B.C., there was a sect of skeptics who maintained that they could not be certain of anything whatsoever. The senses are deceivers, they said; they affect us according to the way we feel, and their reports are always uncertain. We see differently in sickness or health, and in joy or sorrow. Nothing seems to be so true, their leader Pyrrho said, but that it has not some-where been thought false, and nothing seems so false but that it has not somewhere been thought true. The Greeks liked to tell amusing anecdotes about Pyrrho, but they undoubtedly embroidered on his behavior and beliefs, for, if the following story were true, it is hardly likely that he could have lived until the ripe old age of ninety. Pyrrho, one story goes, was crossing a road, when he saw a chariot approaching. "That looks like a chariot," he said, 'but how can I be sure that my senses are not deceiving me?" As he considered the matter, the chariot came by, knocking him down. His loyal disciples picked him up and dusted him off. They were always around for just such emergencies. It was also said that when Pyrrho died, his disciples did not mourn him, for they could not be positive that he was dead.

Perhaps we should note here that Pyrrho was really certain about at least one thing, namely, that his senses sometimes deceived him. And this inconsistency confronts all skeptics, for when they say that they know nothing, surely they must know that they know nothing, and this is a self-contradiction. A person who says that he knows nothing, in other words, is thereby saying that he does know something. A consistent skeptic will keep his mouth shut and say nothing at all.

Bertrand Russell in A History of Western Philosophy, neatly disposes of this thoroughgoing type of skepticism. "It should be observed," he writes, "that Skepticism as a philosophy is not merely doubt, but what may be called dogmatic doubt. The man of science says 'I think it is so-and-so, but I am not sure.' The man of intellectual curiosity says 'I don't know how it is, but I hope to find out.' The philosophical Skeptic says 'nobody knows, and nobody ever can know.' It is this element of dogmatism that makes the system vulnerable."

Let us sum up for a moment. We began by defining a true statement as one that corresponds with the facts, and this means that if a statement is true, then, when stated with proper precision, it must be true for all. When we say, "The truth changes," or "The truth of one age is the falsehood of another" what we mean is that beliefs concerning the truth change. We then discussed the difficulties in determining whether a statement really does or does not correspond with the facts. Relativism and skepticism properly emphasize the formidable obstacles to the doctrine that it is possible for us to know the truth. For what one age considers absolutely true, another age often rejects. And are not the ideas of today just as much subject to error as those of the past? How, then, can we ever claim to know the truth about anything?


Probability

The best answer to this question is found in the theory of probability, which harmonizes the doctrines of relativism and skepticism with the search for truth. The scientist, for example, seeks the truth, but he is also very much aware of the difficulties attendant on this search. He thinks of the actual attainment of truth-the actual correspondence of a statement with the facts, where this correspondence is established once and for all-as an ideal which we can never attain, for it is always possible to question any belief whatsoever. We can never be certain that no error has occurred or that some factor has not been overlooked. And to say that a belief may be questioned means that it is not absolutely certain. No scientific statement, then, is exempt from the requirement of proof, and no proof can be final.

But though the scientist believes that the truth can never be completely attained-he believes that we can never be sure that we actually have arrived at a final answer in the sense of a real correspondence of statement and reality-nevertheless he believes that we can come closer and closer to the true answers. By "closer and closer" he means that our answers may acquire higher and higher degrees of probability.

Before we continue, it is important to note that we are concerned here with empirical probability, and not with the a priori probabilities of mathematical calculation. Past experience is of the essence here. We. can have certainty in mathematical probability provided we make certain assumptions. We can be certain that an ideal coin will fall heads or tails with equal probabilities.

There are also other instances of logical certainty, where we deal not with the world of experience, but with the analysis of concepts. Thus we can be certain that no one will ever see a square with five sides, for a five-sided figure is not a square. I also know for certain that 2+2=4, by the meaning of the concepts. But this is pure mathematics, and we cannot have the same certainty when we apply mathematics to real things in the world of experience. For two gallons of alcohol mixed with two gallons of water will add up to less than four gallons of liquid, because of the chemical changes resulting from the mixing.

The scientist, then, thinks in terms of empirical probabilities. He says that we can know the probabilities, and that probability is the guide of life. Some beliefs are warranted, and some are not, depending on the evidence, which establishes probabilities. And he thinks of probability in terms of degrees. The notion of degrees of probability may be clarified by a diagram showing the 'line" of probabilities:

0 .01 .25 .50 .75 .99 1

On this line "1" stands for unity, or certainty that something is the case, as when we can say flatly: "'This is so." "0" means certainty that something is not the case: "'This is not so," or "'The belief is false." ".50" is the middle state, when we say: "It may or may not be the case, and I don't know which." ".75" means something like "It is very likely," and as we move toward unity the probability increases. ".99" stands for an overwhelmingly high probability. As we go from ".50" to "0" we use statements like "It is improbable," or "It is highly unlikely," and when we reach ".01"- we mean "'There is just a theoretical possibility." These are not to be taken as technically exact descriptions of these numbers, but the general idea should be clear.

If we could be sure that we have reached unity or zero, we could be sure that we have discovered what is really true or false) without any question. Now, in common-sense terms, there are many statements about which such certainty seems justified. I know that I am not sitting in a jet bomber as I write these lines, and I know that at least some motion pictures are not exhibited at the bottom of the ocean. I know that there is at least one professor of English who cannot run around the circumference of the earth in less than ten seconds. But in what follows we shall ignore such simple certainties. For when the scientist says that "probability is all: we never can be certain," he usually refers to statements that go beyond our immediate experience; he refers to judgments in which error really is possible. I can be certain that the water I am now drinking is cold--this is an immediate experience--but did I have water with my lunch on Monday of last week? A judgment that I did goes beyond immediate experience; it relies on memory, and memory is notoriously I tricky. Not many of us, of course, are like the fabled university professor who was walking near the campus one day around the noon hour. He stopped a student: "Would you please tell me whether I am walking north or south?" "You are walking north, sir," the student replied, "Ah," said the professor, "then I've had my lunch." And waiters, as Jacques Barzun has observed, are often even more absentminded than professors.

Memory goes beyond immediate experience, but so does every inference from things observed to things unobserved. And scientific generalizations and predictions not only go beyond immediate experience, but beyond all past experience of any kind whatsoever. And so, whenever we deal with a scientific law or prediction, we can have no more than probabilities. It is not certain, for example, that the sun will rise tomorrow; some sort of cosmic cataclysm may occur five minutes from now. That "all men are mortal'- is also only highly probable. When we carefully state what we mean by this generalization, which says that all human beings must someday die, we find that we are making a prediction that all human beings now alive will die before reaching a certain age, say 200. To be testable, a determinate time limit must be placed on a prediction. And this prediction, of course, is not absolutely certain.

Similarly we cannot be absolutely certain that some things are impossible. There is a possibility that life exists elsewhere in the universe. But it is unlikely that men as we know them exist there, and still less likely that they are listening to "acid rock" bands. But even the last fantasy cannot be ruled out as completely impossible. Strange things do occur, and also strange coincidences. There is just enough possibility in the following story to make it amusing: The story is told of two friends who were patients of the same psychoanalyst One day they decided to play a practical joke on him. Each would tell the analyst the details of a fantastic dream they invented, every phrase of which would be narrated in exactly the same way. One patient told his story, and then, two hours later, his friend repeated exactly the same story. Toward the end of the second recital the analyst could no longer contain his amazement, and blurted out, "What a coincidence! In the last twenty-four hours I have heard exactly the same dream told to me three times!"

Truth and probability
There is an important difference between the meaning of truth and the meaning of probability. This difference can be brought out very simply by showing the possible relations of truth and probability: A statement can be true and probable or false and improbable, but it can also be true and improbable or false and probable. By truth, in other words, we mean the actual correspondence of a statement with the facts, but probability is relative to the evidence available to us. The statement, "'The earth is motionless," once appeared highly probable, in the light of the evidence then available, but today we say it is false; "The earth moves, once seemed improbable in the light of the evidence of the senses, but today we think it is true.

The statement, "'There is buried treasure beneath the house in which I write" is either true or false at this moment. Either the facts correspond, in which case the statement is true, or they don't, in which case the statement is false. The truth of a statement is not relative to the evidence-the statement either corresponds with the facts or it doesn't, even though we do not know which. This follows from our definition of truth. But when I say, "There probably is treasure," or "'There probably is no treasure," I mean relative to a certain body of evidence.

A statement cannot be true for one man and false for another, but it can be probable for one and improbable for another, depending upon the available evidence. On the basis of the evidence available to me, it may be reasonable for me to suppose that Jones is guilty of cheating at cards; on the basis of the evidence available to you, it may appear highly unlikely. He is or he is not, but neither of us may really know for certain which. The truth of a statement cannot change, but probability judgments vary with every change in the evidence. In May it may appear likely that there will soon be an end to tension in the Middle East; in July this may appear improbable. At any given moment we make our estimate of probabilities on the basis of the evidence available to us at that time. And, the scientist adds, we can never reach more than a highly probable conclusion, for all the facts can never be known.

What practical applications can we make of these matters? We can try to assess the probabilities of the facts upon which we rely whenever we make decisions. We should be exceedingly careful before we claim we know something for certain. We should be certain when we have the right to be, of course. A reliable observer can give an accurate report of what he actually witnessed. In general, however, we ought to abandon the use of the expression "absolutely true," except for matters within our immediate experience. And where there is controversy, let us use more modest expressions, such as, "The evidence indicates that this or that is probably the case." And let us also remember Jefferson's distinctions between truths, probabilities, possibilities, and lies. There is no golden touchstone to guide us in each case. We can simply try to avoid two extremes. One is a too uncritical attitude, whereby we jump to conclusions whenever we hear an idle rumor; the other is a too skeptical attitude which refuses to believe despite good evidence.

Though it happens that different opinions are sometimes reasonably held by different observers, this does not mean that "everything is just a matter of opinion." This is a form of skepticism which denies both the possibility of knowledge, on the one hand, and of ignorance; on the other. Nor is it the case that everyone is "entitled" to his own opinions, except in a legal sense. When we say "not entitled," we mean from a rational or logical point of view. For consider: A man says that every professor at a certain university is a Communist and an atheist, and then adds the words, "in my opinion." If you disagree with him, he says that simply means that his opinion is not your opinion, but that everyone is "entitled" to his own opinion on the matter. But the facts may make it overwhelmingly probable that his opinion is a false opinion. Unless one is intellectually irresponsible, he will inquire into the truth or probability of his opinions. One is not forgiven for making a dangerously irresponsible statement merely because he tells us that such is his opinion. We ought to examine the evidence before we talk. The question the careful thinker will always ask himself is: "What is the evidence?"


FOR DISCUSSION AND WRITING

1. Read the following statements carefully. In each, how would you define the author's attitudes concerning the possibility of attaining truth? How would you describe his method?
a. The improver of natural knowledge (science) absolutely refuses to acknowledge authority, as such. For him, skepticism is the highest of duties; blind faith the one unpardonable sin. . . . The man of science has learned to believe in justification, not by faith, but by verification. (Thomas Huxley)
b. Strong Son of God, immortal Love,
Whom we, that have not seen thy face, By faith, and faith alone, embrace, Believing where we cannot prove. (Alfred, Lord Tennyson)
c. A struggle for existence inevitably follows from the high rate at which all organic beings tend to increase. Every being, which during its natural life-time produces several eggs or seeds, must suffer destruction during some period of its life, and during some season or occasional year; otherwise, on the principle of geometrical increase, its numbers would quickly become so inordinately great that no country could support the product . . . . Linnaetis has calculated that if an annual plant produced only two seeds-and there is no plant so unproductive as this-and their seedlings next year produced two, and so on, then in twenty years there would be a million plants. (Charles Darwin)
d. We say that the general phenomena of the universe are explained-as far as they can be--by the Newtonian Law of Gravitation. On the one hand, this admirable theory shows us all the immense variety of astronomical facts as only a single fact looked at from different points of view; that fact being the constant tendency of all molecules toward each other, in direct proportion to their masses and inversely as the squares of their distances .... As to determining what attraction and weight are in themselves or what their causes are, these are questions which we regard as insoluble and outside the domain of the Positive Philosophy; we, therefore, rightly abandon them to the imaginations of the theologians or the subtleties of the metaphysicians. (Auguste Comte)

2. This chapter distinguishes between "empirical probability" and a priori probability." How would you classify each of the following examples?
a. If I throw this die, chances are one in six that I will throw a two.
b. The chances of the Republican party nominating a woman for president are mighty slim.
c. It is highly improbable that an ordinary woman could run the mile in three minutes.
d. Thirty-five percent of today's elementary school students will go to college.
e. It is probably true to say that what is highly probable is going to happen sometimes.
f. His batting average is .500; therefore, the chances are one in two that he will get a hit today.
g. If one were to fly continuously for ten years, the chances are very high that he would be involved in an airplane crash.

3. The theory of logical relativism is useful in impressing the observer with the variety of opinions that are held about a controversial topic. It is also useful in that it reminds us of how easily we might close our minds to another point of view. Unless we can appreciate the other side of an argument, we cannot approach any kind of objective evaluation. Construct a statement which expresses you own point of view. Then, changing only the particulars of the statement (persons, places, etc.) reconstruct exactly the same statement. Now you are able to judge the validity of your opinion by seeing it in a new context. For example, we might read of a Russian intellectual who, in a series of novels, attacks the communist system. We might approve of such an action because it fits our own opinions. But, because the statement is recast to describe an American intellectual who, in a series of novels attacks the capitalistic system, our attitude might shift. What does such an exercise illustrate about the relativistic position?

4. The possible relations of truth and probability are four:
a. A statement can be true and probable.
b. A statement can be false and improbable.
c. A statement can be true and improbable.
d. A statement can be false and probable.
Construct a statement for each of these situations, explaining how the meaning of truth and the meaning of probability are related in each instance.

---

CHAPTER 9
Knowing the Causes of Things

In 1926 The Nobel Prize for physiology and medicine was awarded to Johannes Fibiger, a Danish pathologist, for being the first to produce cancer experimentally in a laboratory animal. In a series of postmortem examinations of tubercular rats, he noticed that several had suffered from stomach cancers, a highly unusual occurrence. He learned that the dealer who had supplied his laboratory with the rats had secured them from a sugar refinery. Investigation of the place revealed a high infestation with cockroaches, which formed a large part of the rats' diet. Fibiger suspected there was a connection between roaches, rats, and cancer. He collected thousands of roaches from the refinery and fed them to rats from another breeding establishment. When the rats died, three years later, Fibiger opened them up. To his astonishment, he found many stomach cancers. A microscopic study of the growths revealed in every case that they had formed around a parasitic worm, the same worm to which the roach had been host before is was fed to the rat The larva of the worm, coiled up in the muscles of the cockroach, later developed into an adult worm in the rat's stomach. Around this the tumorous growth had appeared.

Fibiger's study illustrates the "cause and effect" relation. Something unpredicted or unexpected happens; we ask "Why?" We then search for the cause. This is what happens when a housewife finds that a recipe did not turn out as expected. Her problem is also one of finding the cause, and her thinking resembles that of Fibiger, or the thinking of a biologist when he searches for the cause of a disease. The chief difference between the scientists and the housewife is that the former are specialists who have vast stores of knowledge to aid them in solving such problems.

When Francis Bacon said, "Knowledge is power," he meant that science could give man mastery over the forces of nature by discoveries and inventions. Bacon's aphorism is a modern echo of the ancient poet Virgil's saying, "Happy is he who knows the causes of things." For, in large measure, man's power and control over nature have their origins in his understanding of the causal connections among events. Our under-standing of causal connections enables us to improve the fertility of the soil; it is because we know the causes of diseases and other unpleasant occurrences that we can eliminate them. Perhaps some day we will know, more precisely, the causes of business depressions, and then the fear of widespread unemployment will disappear forever.

Before we proceed further, let us note the assumption which is taken for granted by Fibiger, the housewife, and the biologist, in our examples. This assumption is that everything has a cause. We believe that "things don't just happen by themselves," but that something is responsible for every single thing that happens. This assumption is sometimes called the "principle of determinism," and it is a postulate of rational thinking about the events of human experience. It is called an "assumption" or "postulate" because it is obviously impossible to prove that everything has a cause, including all future events. We don't even know the causes of all sorts of contemporary events-cancer, for example-but we are sure that there is a cause. Why are we sure? Because, we say, there must be a cause! The point of the principle is this: We have made up our minds not to regard any events as being beyond explanation. No events in human experience are inherently unexplainable.

Much bad writing stems from careless cause and effect reasoning. Merely making bold assertions about alleged causes and effects will persuade no one; indeed, such assertions usually hide uncertainty and confusion about one's experience and ideas. Carefully marshalling evidence and limiting your topic will force you to scale down your generalizations, and therefore lessen the likelihood of this common fallacy.

Let us look at the semantical aspects of our problem: Exactly what we mean by "cause"? In popular speech a cause means "that which is responsible for a thing's happening," or "the power that produces an event." These are question-begging definitions, however, for we immediately ask: What is meant by "responsible" and "produces"? The notion of cause is involved in the meaning of these words. Nevertheless we shall begin with the common-sense meanings of the term, and try to refine them as best we can.

What is a cause?
Let us also examine the way in which the scientist uses the words cause and effect." When a biologist tells us that yellow fever is caused by a filterable virus transmitted by the bite of a certain kind of mosquito, he means that when this virus enters a host body, the host will suffer from yellow fever, and if an individual suffers from yellow fever, then we know that this virus has entered his body. By "cause" the scientist means the necessary and the sufficient conditions for the occurrence of an event. These terms require definition.

A catarrhal affection of the respiratory tract, commonly called "a cold," is probably due to a virus. But this virus is presumably present in many persons all of the time, without their having "colds." The cold occurs when body resistance is lowered, owing to fatigue or exposure to low temperatures, dampness, or drafts. If we can assume that this is the correct theory concerning some types of colds-there is no universal agreement on the matter-then the virus is not strictly the cause of the cold, but only an indispensable prerequisite. A necessary condition is defined as a condition without which an effect cannot occur. We have assumed that colds cannot occur unless the virus is present, so the virus Is a necessary condition of the cold's occurrence.

A sufficient condition, on the other hand, is illustrated by the following: A murderer administers cyanide of potassium to his victim, and the victim dies. We say that the poison was the cause of death. But the relation of poison to death is quite different from the relation of the virus to the cold. Death may occur in other ways than by poisoning: old age, to give just one example. Poison (of a certain amount under certain conditions) is sufficient in itself to cause death, but it is not a necessary condition of death. In the former example, the virus is not sufficient to cause a cold, but there can be no cold without the virus.

In other words, some conditions are necessary but not sufficient for a given effect: the virus and colds. Some conditions are sufficient but not necessary for a given effect: poison and death. If a person has the virus in his body, he may or may not have a cold, but if he has a cold, then we know he has the virus. If a certain amount of poison has been administered to man, then we know that he will die, but if all we know is that a person is dead, we do not know whether or not he was poisoned.

Though scientists often speak of a sufficient condition as "the cause"-poison is a cause of death-and though they may even speak of a necessary condition as the cause, the scientist seeks for something more than either a necessary or a sufficient condition. He has an ideal conception of a cause as that set of conditions which are both necessary and sufficient to bring about a certain effect. The scientist seeks to know the entire constellation of conditions which will always result in the effect and without which the effect will never occur. C is the cause of E (in the ideal sense) when E always occurs following on C's occurrence, and E never occurs unless C has occurred.

This ideal notion of "cause" may be illustrated by considering a forest fire. A lighted cigarette is thrown into the brush by a camper. A forest fire results. Can we say that the cigarette was the cause of the fire? Common sense tells us that it is quite proper to speak in this way, but the scientist seeks a more accurate type of statement. He notes that no fire would have resulted if the leaves had been damp from a recent rain. Dry leaves, then, are a necessary condition for a forest fire. Dry leaves, plus an igniting element, plus sufficient wind make up "the cause," for these are necessary and sufficient conditions of a forest fire. Without any one of these conditions the forest fire would not occur; when all occur together, a forest fire will always occur.

In law, and in the ordinary affairs of life, of course, the smoker would be held responsible, for in law and ordinary affairs we seek the causes of individual events in order to ascribe responsibility. The lawyer thinks of the cause as some identifiable act or event without which the result would not have occurred. The scientist, on the other hand, does not seek to ascribe "responsibility." He seeks for general causal connections, that is, repeatable patterns in nature, and so he looks for the conditions that are both necessary and sufficient.

In practice, however, even scientists are often satisfied with less than the "ideal" statement of the cause. Nor is it always necessary to know the cause in the strict sense. It may be enough to learn either the necessary or the sufficient condition in order to achieve the purpose m mind. Practical considerations are involved here. It all depends on what one is after: to produce or to prevent. If scientists wish to produce some-thing, such as a specific to cure a disease, or synthetic leather, or a stimulus to business activity, they need only know the sufficient conditions of these effects. If they wish to prevent or eliminate an effect, such as a disease, it is very helpful if they know the necessary conditions without which the disease cannot occur.

Discovering causes

We shall now examine a famous example of the search for the cause of a disease. In the year 1910 the disease known as pellagra was widespread in many of our southern states. Pellagra is characterized by skin eruptions, gastric disturbances, and nervous derangement. The problem became so serious that the United States Public Health Service sent its Dr. Goldberger to the state of Mississippi to search for the cause of the disease.

In the early years of the twentieth century it was the prevailing doctrine among biologists that there was only one possible cause of disease, namely, germs or microbes. This is the "germ theory" of disease. It was therefore assumed, at that time, that pellagra must also be caused by a microbe. The Webster International Unabridged Dictionary, published in 1900, defined pellagra as "a disease caused by a microbic parasite," and added that it was probably carried by a fly. Now, one of the implications of the germ theory is that the disease will be "catching." that is, infectious, and that it will be transmitted from one person to another through contact. Goldberger began with the hypothesis that the disease of pellagra was caused by microbes, and he first investigated to see whether there was personal contact among the victims. There was.

Goldberger began with a preliminary hypothesis which told him what to look for. His first observations seemed to confirm his hypothesis. But he was a man who refused to take anything for granted. and he decided that it would be wise to test the microbe theory further before undertaking a full-scale search for the particular microbe which might be the cause of the disease. And so he reasoned as follows: If pellagra is catching, then whenever we find healthy people coming into contact with victims, under normal circumstances, the hitherto healthy persons will acquire the disease. He decided to test this implication by investigation. He visited a hospital, and he observed that the nurses, orderlies, and doctors were in close contact with the patients, and that they made no efforts to avoid such contacts. But none of the nurses, orderlies, or doctors had ever caught the disease from the patients. Goldberger thereupon decided that he was on the wrong track, and he rejected the hypothesis that microbes were the cause of pellagra.

Goldberger had predicted, on the basis of the then prevailing theory, that contact would cause the acquisition of the disease. The facts were against this theory, and so he abandoned it. For the scientist there is no such thing as "a good theory which does not work." If "the stubborn and irreducible facts" are against a theory, so much the worse for the theory. If a theory does not work, then it is a bad theory and must be discarded. The apparent exception to this rule is the case where the facts are misinterpreted.

Goldberger now had to make a completely fresh start. He was sure now that pellagra was not caused by microbes. He had to look elsewhere for the cause. He investigated further, and noted that only poor people seemed to suffer from the disease. He carefully observed their habits of living, especially their diets. He noted that the pellagra victims lived on a rather uniform diet, consisting of cornmeal mush, hominy grits, and similar foods. On one occasion, while visiting an orphanage, he found that some children suffered from pellagra, and others not. The pellagra victims, further, were all in the six to twelve age group. This was a surprising situation, and he made further inquiries. He was told that the children who were over twelve were required to work on the farm, and they were fed meat to provide them with the necessary energy. The children under six were given milk, since they were regarded as being in the "baby" stage. The orphanage's funds were very limited, and so the six to twelve group received no milk or fresh meat. They were too young to get meat and too old to get milk.

Goldberger now developed a new hypothesis-that a dietary deficiency, specifically, the absence of milk and fresh meat, is the cause of pellagra. He predicted that if the children in the six to twelve age group were supplied with milk or meat, their pellagra would disappear. The Public Health Service supplemented the children's diet with the missing ingredients, and all cases of pellagra in the orphanage were cured. His hypothesis was confirmed.

But science is a never-ending search for truth, and no proof is ever final. Even confirmed hypotheses will be subjected to retesting, again and again, for confirmation in previous experiments may have been due to the presence of special factors. Unknown factors may produce exceptional results in a specific case. So Goldberger did not stop at this point' And there was another reason why further tests were called for: Many scientists were still unconvinced that his investigations had actually disproved the germ theory. After all, they said, the sufferers from pellagra were in contact with each other, even though the attendants did not catch the disease. They may have been blessed with immunity. The critics wanted further proof. And we have often seen that stubborn refusal to give up a theory in the face of contrary evidence may in the end be justified.

Goldberger proceeded with new tests. He was given permission to by out a dietary experiment at one of the state's prison farms. He proposed to feed a group of convicts a special diet for six months, to test his hypothesis that a deficiency in diet is the cause of pellagra. Volunteers were asked for; the reward for participation was to be freedom after the experiment was finished. One or two lifers accepted, then several others, and he finally had twelve subjects, all of them in good health. The twelve convicts were isolated from the others and fed almost nothing but white bread, corn pone, grits, sweet potatoes, salt pork, cane syrup, and cooked cabbage for six months. They received no milk or fresh meat' After several months of this diet, beginning in April, 1915, the twelve convicts became listless; they began to develop severe abdominal pains and finally developed skin eruptions of the pellagra type. But the rest of the convicts at the prison farm suffered no such disorders.

Goldberger had now established a solid confirmation of his hypothesis that nutritional deficiencies alone are sufficient to cause pellagra. The convicts were now fed a proper diet and, as expected, all of them recovered and were given their freedom. No further proof of his hypothesis seemed to be necessary, but there were still some skeptics among the microbe theorists, and so Goldberger and his assistants decided to do one further experiment to convince the doubters. The experimenters performed this test on themselves. They injected blood from pellagra victims into their own bloods streams, and no ill effects followed. This was a crucial experiment, for if pellagra was caused by microbes, then they should have become infected. After this test it was an accepted fact among all biologists that pellagra was caused by nutritional deficiencies. Later research

has confirmed Goldberger's findings; our knowledge today is simply more precise than his. It is not the absence of milk and fresh meat C's such that causes pellagra, but rather the absence from the diet of certain factors in the vitamin B complex These factors are found in muscle meats, milk, liver, kidney, fish, and green vegetables.

Goldberger's experiment with the convicts illustrates one of the most reliable scientific techniques for discovering causes. A careful analysis of the method employed in this case will enlighten us concerning the nature of a scientific proof that one kind of thing is the cause of another. This method requires the setting up of two situations which are identical except for the presence or absence of the factor which is being investigated. There were two sets of convicts at the prison farm-those who received a special diet and those who did not' The living conditions of both groups were identical except for the diet, and only those who were on the restricted diet got pellagra.

Goldberger's method may be made clearer by a somewhat more detailed example. Let us assume that a young woman, call her Susan, suffers from skin irritation and inflammation of the face. She seeks for the cause. The hypothesis occurs to her that it may be due to her use of face powder. She stops using the powder for a period of time, and the irritation disappears. She then uses the powder again, and the irritation reappears. She has solved her problem: She knows the cause of the irritation and how to avoid it' Face powder seems to be a necessary and sufficient condition for the occurrence of the irritation.

But this newfound knowledge does not make Susan completely happy. She hates to give up the use of face powder, for she isn't quite so attractive when she doesn't use it' The thought occurs to her that it may not be the powder as such that is the culprit but rather one of the ingredients contained in face powder. So she consults a chemist friend and asks him for a solution to her problem. The chemist makes an analysis of the powder and finds that it contains six ingredients: talc, kaolin, magnesium carbonate, zinc oxide, ochre (for coloring), and perfume. The chemist now formulates the hypothesis that the perfume may be the cause of the irritation. To test this hypothesis he prepares a batch of face powder containing all the ingredients except the perfume, and he then divides this batch into two parts, to one of which he adds perfume. Susan now uses the powder without perfume and suffers no ill effects. Then she tries the part with the perfume added, and irritation appears. This is sufficient proof that she is allergic to the perfume in the face powder, and not to the powder itself. (Or to some ingredient in the perfume.)

"The method of difference"
The method we have been illustrating is sometimes called the laboratory or "controlled experiment" method for determining cause and effect' The English philosopher John Stuart Mill, who pioneered in the field of scientific logic, called it the "method of difference." The basic idea is to use two cases, identical in all respects except one. In the face powder example this method was used twice, first to determine that the original powder was the cause of the irritation and then to determine that it was a specific element in the powder and not the powder as such that was the cause. Goldberger also used this method, for he kept all living conditions for his two groups of convicts identical except for one factor-the diet' "Identical," of course, means "the same in all respects that are considered relevant to the experiment"' This kind of identity is achieved most perfectly in the sciences of physics and chemistry, where we can be fairly certain that there are no relevant factors outside the conditions of the experiment. This method can also be applied with a large degree of precision in the biological sciences.

The most important point to remember in applying this method is to keep the two groups of things identical (or substantially the same) except for one element. If the "addition" of this element results in the effect, and if the effect never occurs when it is absent, then we have found the probable cause. But it is not always possible to apply this method, for it may be impossible to isolate one factor and keep all others the same. This is especially true in the social sciences. The variables involved in human actions are exceedingly large in number, and human behavior is vastly more complex than is the career of a germ. Since we cannot be sure that we have accounted for all the relevant factors in a situation, we cannot be sure that we have an identity of all factors except one. An illustration or two may be helpful here. Suppose we try to isolate a single factor to explain why X defeated Y in an election. Let us assume that X favored capital punishment and Y did not. If it is claimed that this was the reason for his victory, we must look to see whether all other factors were substantially the same. Were the men equally able, and did they agree pretty much on all other issues? Did they have the same number of influential friends and enemies; did they have equal campaign funds, etc? The attitude toward capital punishment may have been a factor in X's victory, but not necessarily the cause, or even the most important factor.

Let us now examine a sociological problem. Consider the difficulties in applying the differential method to the problem of juvenile delinquency. Can we find two individuals, one a delinquent and one a good junior citizen, in whom all characteristics except one are alike? It is unlikely that their antecedents and experience will differ in one respect and one respect only. No two individuals are identical except for a single difference. Similar considerations apply to such problems as finding the causes of divorce or war or dope addiction. But many people apply the Method of Difference carelessly, forgetting that other conditions are not the same. The next time someone tells you that he knows "the cause" of war or juvenile delinquency or some other social evil, check to see whether he has isolated a single causal factor, all others remaining the same. If not, has he used one of the other causal methods to be described shortly?

These difficulties have led some social scientists to abandon the search for the "causes" of social behavior and to limit themselves to a search for "tendencies" stated in statistical form. Children from broken homes, for example, may show a greater tendency toward delinquency than those from stable families. If so, that would be useful knowledge. A tendency, however, is simply a modest way of indicating a possible causal connection.

When appropriate, and when properly applied, the method of difference, or the differential method, is the most convincing possible kind of proof that we have found the cause. But sometimes it is impossible to apply this method.

"The method of variations"
We turn now to a second method for determining the causes of events. This method is called the "method of variations," known in statistics as the "method of correlations." Examples: A manufacturer of cosmetics uses newspaper advertising to sell his product. Each ad contains a coupon offering the reader a free sample. The advertiser finds that an increase in the lineage of the ad brings in a larger number of coupons; a decrease brings a smaller number. An increase in the crop of oranges (other things 'being equal") is followed by lower prices; a smaller crop by higher prices. In other words, if two kinds of factors vary "directly," so that an increase in one factor is always followed by an increase in a second, and a decrease by a decrease (the cosmetics example), or if they vary "inversely," so that an increase is always followed by a decrease, and a decrease by an increase (the orange crop), then there is a good reason to suspect a causal relation between the two factors. They vary together, or concomitantly.

There are, of course, possibilities of error in applying this method. It has been pointed out that there is a tendency for women's dresses to be shortened during periods of "prosperity" (the 1920s, the period of the Second World War, and, as we know, they reached new heights in the unparalleled boom of the '60s) and lengthened in periods of depression (1932). But the correlation may be an "accidental" one. It was once discovered that over a period of time the number of storks in Sweden varied in precise proportion to the number of human births in the United States, but this does not prove a causal relation between the two factors. We should also seek other kinds of evidence which make it reasonable to believe that a causal relation does in fact exist.

"The method of agreement"
A third method, called the "method of agreement," seeks a common factor in the conditions which precede the effect that interests us. If a single common factor is discovered, this often gives us some probability that we have found the cause. For example, the public health authorities in a small town were confronted with an outbreak of typhoid fever. The authorities investigated the food and beverages consumed by the victims just prior to the outbreak of the fever and found that all of the victims had just one thing in common: they had all attended a picnic and had drunk from the water in a well at the picnic grounds. Since this was the only common factor, it was a reasonable inference that the well was contaminated. Laboratory tests showed that the water contained the typhoid bacillus.

The method of agreement, or "common factor method," is not so convincing as the differential method, but it is useful when the more precise method cannot be applied. The method of agreement also is subject to careless applications. As an example of the dangers involved in the use of this method, consider the anecdote concerning the man who wished to find the cause of his becoming intoxicated every time he attended a cocktail party. This man had heard of the success of the public health authorities in using the "common factor method," and he decided to emulate them. He looked for the single common factor in the beverages he had imbibed on each occasion when he became intoxicated. After making an exhaustive study of the matter, he found that intoxication followed after drinking bourbon and soda, Scotch and soda, rye and soda, rum and soda, brandy and soda, and vodka and soda. Since soda was the only common factor in every instance, he concluded that soda was the cause, and at the next party he attended he insisted on drinking his whiskey straight!

"Negative" tests for hypotheses
Thus far we have examined methods which tell us how to search for causes and how to discover them. It is also well to remember two negative tests which must be "passed" by hypotheses which assert a causal connection. The tests are these: (1) Nothing can be the cause if the effect fails to occur in its presence, and (2) Nothing can be the cause if the effect occurs in its absence. Let us illustrate.

Dr. Goldberger eliminated microbes as a possible cause of pellagra, for, although microbes are always carried in the bloodstream, pellagra failed to occur when the blood of victims was injected into the blood-streams of the investigators. Microbes are not the cause of pellagra, then, for the effect (pellagra) failed to occur in their presence. An ancient Roman, Pliny the Elder (A.D. 23-79), once disproved the claims of the astrologers by. using the same test' "If a man's destiny is caused by the star under which he is born," Pliny wrote, "then all men born under the same star should have the same fortune. But masters and slaves, kings and beggars, are born under the same star at the same time." In other words, the star cannot be the cause of a particular kind of destiny, for in its presence that particular kind of destiny fails to occur. One more example: the belief that a "broken home" is the cause of juvenile delinquency. But not all children from broken homes become delinquents. There may be some connection, of course, but we cannot yet speak in terms of cause and effect"

The second negative test may also be illustrated by the delinquency case. Children from stable homes sometimes become delinquents. The effect has occurred in the absence of the factor "broken home," so this cannot be the "cause" of delinquency. All too often we forget to apply the negative test' We often jump to the conclusion that one thing is the cause of another because we forget that there may be negative evidence. This happens most frequently when our emotions cause us to try to prove what our hearts desire. Sir Francis Bacon called this tendency to ignore evidence that does not suit our purposes an "Idol of the Tribe," by which he meant a faulty habit of thinking, common to the human race. In Aphorism 45 of his Novum Organum he gave a striking example of this error:

And therefore it was a good answer that was made by one who, when they showed him hanging in a temple a picture of those who had paid their vows and then escaped shipwreck, and would have him say whether he did not now acknowledge the power of the gods-"Aye," asked he, 'but where are they painted that were drowned after their vows?" And such is the way of all superstition, whether in astrology,, dreams, omens, divine judgments, or the like; wherein men, having a delight in such vanities mark the events where they are fulfilled, but where they fail, though this happens much oftener, neglect and pass them by.

Fallacies in causal reasoning

The search for causes, as we have seen, is beset with numerous forms of fallacious reasoning. Perhaps the most important of the general fallacies in causal reasoning is the "post hoc," an abbreviation for the ##Latin expression "post hoc, ergo propter hoc": "after this, therefore because of this." This means: The fact that one thing follows upon another is no proof that the first is the cause of the second. For example, I have a pain in my shoulder and take a pink pill. A little later the pain disappears. I say that the pill was the cause of the disappearance of the pain. Why? Because I took the pill and then the pain disappeared. But the mere fact that one thing follows another is no proof that the first is the cause of the second. To prove a causal connection we must use one of the procedures discussed earlier. It may be that the pain would have disappeared even if I had not taken the pill. An effect, of course, always follows the cause. When one thing follows another the first may be the cause, but more proof must be forthcoming before we can say "proved." Mere succession in time is not proof. However, succession of events may involve a causal connection, as in the sharp increase in births in New York City following upon the Great Blackout of October, 1965. The fact that the increase came nine months later indicates the likelihood of some connection. In this case the inference is based on more than mere succession in time.

And here are some more post hocs: Old man Jones celebrates his one hundredth birthday, and the newspaper reporters, as usual, are on hand, curious to know just how he did it; "Well," says old man Jones, "I drink a pint of beer every day." He drank a pint of beer, and he lived another day, and he did this again and again. But there are teetotalers who live until one hundred, and some beer drinkers have been known to die young. Post hoc reasoning is common also at baseball parks. A fan yells, "Hit a home run!" and the batter hits one over the fence. It will be difficult indeed to convince this fan that his yell was not the cause of the home run. Baseball players are notoriously superstitious, and their superstitions are based on post hoc reasoning. Their managers are superstitious too. If a manager fails to shave on the day when his team ends a losing streak, he will probably assume that his failure to shave was the cause of the victory, and allow his' beard to grow until his team loses in. A recent example of post hoc reasoning: In February, 1971, sixty-five people were killed in a southern California earthquake. Prior to the earthquake a fundamentalist preacher had warned that southern Californians were sinners and that God planned to punish them. This was sufficient evidence for many people. They were absolutely convinced that the earthquake was God's punishment for their wicked ways.

Another error in causal reasoning is that of reversing the connection between the cause and the effect. This error is often called "over-simplification." An English writer once argued that since those among the English poor who had cows were the most industrious, the way to make the others industrious was to give them cows. Though the writer's reasoning was fallacious, he may, nevertheless, have a point. Is it not possible that, with cows to care for, many people would become more industrious and more concerned with their own welfare? The next example is a clearer example of the fallacy of reversing cause and effect: We find that students who major in mathematics generally rank high scholastically. This is considered proof that the study of mathematics makes students bright, but perhaps only bright students major in mathematics.

The error of reversing cause and effect puts the cart before the horse. An interesting application of this reasoning occurs in the frequent controversies concerning the "10w cultural level" of television programming. We accuse the television industry of debasing the taste of the public by a "vast wasteland" of low-quality shows. The industry retorts that it gives the public what it wants; if it aims higher, it will lose money. The industry claims that it is the low taste of the public that is responsible for a situation that they, too, deplore. This situation is one of reciprocal causation. The low taste affects the industry, and the industry may further debase an already low taste. This is the familiar "vicious" circle, but there is also a 'beneficial" circle. If the programs improve, taste will improve, and then the spiral will be reversed.

One final point. Suppose we find that students who smoke heavily are less successful in their studies than students who don't smoke. It would be a mistake to conclude from this that smoking is the cause of low grades. It may be so, but the facts cited are not proof that it is, for it may be that both the heavy smoking and the low grades are due to other factors, such as personality traits, lack of academic interests, or extracurricular activities.

A summary of this chapter may be welcome. We defined a "cause" as the set of necessary and sufficient conditions of an event. When we say that X is the cause of Y, we mean: (1) If X occurs, Y will always occur; that is, X is the sufficient condition of Y, and (2) if X does not occur, Y will not occur; that is, X is the necessary condition of Y. This is the meaning of "cause" in the ideal sense of the term; in practice, scientists are often satisfied when they know either the sufficient or the necessary conditions.

We then noted three methods used by scientists in discovering causes: the methods of difference, variations, and agreement. The first is the most rigorous of these methods, but it demands complete control over all of the factors in a situation, and so is difficult to apply outside of the physical and biological sciences. Two situations must be exactly alike except for the presence or absence of a single factor, the effect occurring when this single factor is present.

These three methods tell us what the cause is. There are also methods for eliminating "false causes," which tells us what the cause is not: "Nothing can be the cause in whose absence the effect occurs." "Nothing can be the cause in whose presence the effect fails to occur."

We noted some of the errors or fallacies which result from the careless use or application of the methods, and we concluded with a discussion of some of the fallacies with which the search for causes is beset The major fallacy is called the "post hoc": the assumption that if Y follows X, this in itself is sufficient proof that X is the cause of Y. We also reverse causes and effects, and we sometimes assume that when two things are associated with each other, one must be the cause of the other, when actually both may have a common cause.


FOR DISCUSSION AND WRITING

1. In the following statements, causes and results are described. In each, decide whether the cause was necessary, or sufficient, or both.
a. The life expectancy of American men ranks twenty-fourth in the world. The reason is that American men tend to overdo things. They consume too much in the way of calories, cholesterol, nicotine, and alcohol.
b. Beverly found herself allergic to candy wafers. A chemist friend analyzed the contents of the candy and found six elements: sugar, corn syrup, retsyn, copper gluconate, vanilla, and a coloring agent. By isolating elements and testing Beverly's reactions, it was found that vanilla was the cause of the allergy.
c. After years of testing, medical researchers have isolated the gumbuba fly as the sole cause of the rare tropical disease synchromose.
d. Frank smoked cigarettes for twenty-seven years. He died of lung cancer last fall
e. Because of the extreme rainfall this spring, floods occurred throughout the South.

2. Analyze the following evidence and conclusion. How reliable is the conclusion?
a. New York and New Jersey are heavily industrialized.
b. Sixty percent of the population of those two states are directly involved in some way with light or heavy industry.
c. Eighty-five percent of the industries surveyed are housed in plants in which a residual noxious fume of some kind is emitted.
d. A recent survey shows that the rate of cancer in New York and New Jersey is hi her,, on a per capita basis, than in any other state. Conclusion: The high cancer rate in New York and New Jersey is the result of industrial pollution.

3. Which of the following statements represent logical cause-effect reasoning?
a. Man has always gone to war, and therefore it is impossible for us to ever end war.
b. The 1974 Buick must be well constructed, because Buick has a reputation for quality.
c. The United States is a major world power because of the strong moral fiber of its citizens.
d. I failed my English exam today because I stayed out late last night and neglected to review.
e. A copper bracelet cured my rash.

4. Select one of the following statements and write a paragraph in which you analyze the relationships between the causes and the effects:
a The man who reads fast is the man who is on his way up in this world. In a recent survey, 500 men and women were tested for reading speed. Those who were capable of reading 1,000 words a minute or faster had an average income of $15,000. Would you like to make $15,000? Write for our brochure on the Hogins Speed Reading Course.
b. The spring of 1973 was one of the coldest in recorded history. This unusual cold snap followed immediately upon the French testing of a new hydrogen bomb in the Pacific Ocean. The editorial pages of American newspapers were filled with the cries of "Stop the French bomb testing, or it will destroy the balance of nature!"
c. At a midwestern university 1,000 students were interviewed concerning their dating habits. Ninety-one percent of those who were going steady were in the bottom half of their class. Those who went out on dates with two or more persons were in the top half of their class. Those who seldom dated were in the top 15 percent of their class. The conclusion is inescapable: because of the obligations of steady dating, the student who dates one person solely will suffer scholastically. Indeed, there seems to be a direct relationship between the number of dates and one's academic standing.

5. This chapter describes three methods for determining cause and effect: the "method of difference," the "method of variations," and the "method of agreement." For each of these methods write a paragraph in which you apply the method to an effect and an alleged cause.

---

CHAPTER 10
Are All Generalizations False?

We begin with a generalization: human beings are great generalizers. Every race has its proverbs, and proverbs are generalizations. "It never rains but it pours." "Faint heart never won fair lady." "Familiarity breeds contempt." Sometimes, of course, these proverbs are incompatible with each other, as in "Absence makes the heart grow fonder," and "Out of sight, Out of mind." Listen attentively to those around you, and note the generalizations that float into every conversation: Women drivers are the most careless. Professors are absentminded. The Irish are alcoholics. Gentlemen prefer blondes. Politicians are crooks. The French are great lovers. People on welfare don't want to work. And so on. After more of the same we may be tempted to agree with Justice Holmes that "the chief end of man is to frame general propositions, and no general proposition is worth a damn,"

Our awareness of the inadequacy of "sweeping generalizations" may lead us to say that all generalizations are false. But this is truly a sweeping generalization! And worse: if it is true, then the witticism that "all generalizations are false, including this one" would appear to be justified. But this will not do either, for this generalization asserts that it itself is false, from which it follows that it is not the case that all generalizations are false. Or perhaps we should say that "all generalizations are half-truths-including this one"? But this is not much better. The fact of the matter is that some generalizations are true, others are false, and still others are uncertain or doubtful. The deadliness of this platitude may be forgiven because of its truth.

By a "generalization" is meant a general law or principle which is inferred from particular facts. As a sample of the way in which we arrive at such generalizations, consider the following: Some years ago I saw my first Italian movie. The directing, the acting, the dialogue, the lighting-all were superior. Encouraged by this initial experience, I saw another Italian movie. It, too, was enjoyable. I saw other Italian movies, always with the same results-comedies, dramas, "Westerns," thrillers. I generalized: All Italian movies are enjoyable.

A generalization is a statement that goes beyond what is actually observed, to a rule or law covering both the observed cases and those that have not as yet been observed. This going-beyond is called the "inductive leap." An inductive leap is a "1eap in the dark," for the generalization may not be true, even though the observations on which it is based are true. Thus, there may be a bad Italian movie-happily I have not seen it-but if so, then I should not say that all are good.

A generalization involves an "inductive leap." The word induction, from Latin roots meaning "to lead in," means that we examine particular cases and "lead in" to a generalization. Induction is the method we use when we learn lessons from our experience; we generalize from particular cases. Deduction, on the other hand, refers to the process of "drawing out" the logical consequences of what we already know (or assume) to be true. By induction we learn that Italian movies are enjoyable. If a friend tells us that he saw a bad movie, then by deduction we know that he did not see an Italian movie. Both induction and deduction are essential characteristics of rational thinking.

A generalization is a statement of the form: "All A's are B's." "All" means exactly what it says: all without exception. A single exception overthrows a generalization of this kind. Before we proceed further we must first dispose of a popular confusion concerning the expression, "The exception proves the rule." This is a sensible statement when properly interpreted, but it is sometimes understood in a manner that makes it nonsense. If I say that "all A's are B's," a single exception will make my statement false. Now, suppose that someone says, 'The fact that there is a bad Italian movie proves that all are good because it is an exception, and the exception proves the rule" Does a wicked woman prove that all worn-en are saints? The sensible interpretation of the expression, 'The exception proves the rule," is this: When we say that a certain case is an 4'ex-ception," we imply that there is a rule which generally holds. When a mother tells her daughter, "Have a good time at the prom, and, for tonight, you have my permission to stay out until 3 A.M.," she implies that this is an exception to the rule which requires earlier reporting. A statement that creates an exception implies a rule for all nonexceptional cases, but a generalization that is stated as a rule without exceptions (all A's are B's) would be overthrown by a single exception.

Scientific laws, stated in the form "All A's are B's," or some variation thereof, are never "violated." When an exception to a law is definitely established, the law in its previous form is abandoned, but it may be p05sible to revise it to exclude the "exception" as a special case because of special circumstances. The revised law: "All A's, under such and such conditions, are B's." Water freezes at 320 F. at sea level.

All too often "general propositions are not worth a damn," as Holmes remarked. This is because we generalize too hastily on the basis of insufficient evidence.. The fallacy called the 'hasty generalization" simply refers to the fact that we jump too quickly to conclusions concerning "all." For example, we see a woman driving carelessly, and generalize:

"All women are poor drivers." We see a car weaving in and out of traffic, and note that it has a California license: "Wouldn't you know," we say. "A California driver. That's the way they all drive out there." Anita LOOS'S gay heroine thought that gentlemen preferred blondes because she was a blonde and men were attracted to her.

We learn that Napoleon got along on five hours of sleep. From this we may conclude that 'five hours of sleep is all that anybody really needs." Our assumption is that what Napoleon could do, anybody can do, until we learn that we are not Napoleons. (If we don't learn this eventually, we aren't permitted to circulate freely.) The next example is undoubtedly the worst example of generalizing ever committed: A man declared that all Indians walk in single file. When challenged for his evidence, he replied, "How do I know that? I once saw an Indian walk that way.

Hasty generalizing is perhaps the most important of popular vices in thinking. It is interesting to speculate on some of the reasons for this kind of bad thinking. One important factor is prejudice. If we are already prejudiced against unions or businessmen or lawyers or doctors or Jews or Negroes or whites or gentiles, then one or two instances of bad conduct

by members of these groups will give us the unshakable conviction that "they're all like that." It is very difficult for a prejudiced person to say, "Some are, and some aren't." A prejudice is a judgment formed before examining the evidence.

A psychological reason for asserting "wild" generalizations is exhibitionism: The exhibitionist desires to attract attention to himself. No one pays much attention to such undramatic statements as "Some women are fickle," or "Some politicians are no better than they ought to be." But when one says that "all men are liars," this immediately attracts notice. Goethe once said that it is easy to appear brilliant if one respects nothing, not even the truth.

Let us avoid careless and hasty generalizing. The fault of bad generalizing, however, need not make us take refuge in the opposite error-the refusal to generalize. This error is illustrated in the anecdote concerning the student who wrote an essay on labor relations, in which he argued for equal pay for women. Women, he wrote, work hard; they need the money; they are the foundation of the family; and, above all, they are the mothers of most of the human race! There is an old anecdote about the cautious man whose friend pointed to a flock of sheep with the remark, "Those sheep seem to have been sheared recently." "Yes," said the cautious man, "at least on this side."

Generalizations are dangerous, but we must generalize. To quote Justice Holmes once more: he said that he welcomed "anything that will discourage men from believing general propositions." But, he added, he welcomed that "only less than he welcomed anything that would en-courage men to make such propositions"! For generalizations are indispensable guides. One of the values of knowledge lies in its predictive power-its power to predict the future. Such knowledge is stated in generalizations. It is of little help to me to know that water froze at 320 F. yesterday unless this information serves as a warning to put antifreeze in my car radiator before winter comes. History, in the "pure" sense of this term, merely tells us what has happened in the past, but science furnishes us with general laws, and general laws tell us what always happens under certain specified conditions.

Science is interested in the general, rather than in the particular or individual. When Newton saw an apple fall from a tree in his orchard-even if this story is a fable, and therefore false in a literal sense, it is true in its insight-he was not interested in the size and shape of the apple. Its fall suggested an abstract law to him, the law of gravity. He framed this law in general terms: Every particle of matter attracts every other particle of matter with a force directly proportional to the product of their masses and inversely proportional to the square of their distances. Chemists seek general laws concerning the behavior of matter. The physician wants to know the general characteristics of the disease called myxedema, so that when he has a case he will recognize it and know exactly how to treat it. The finding of general laws, then, is the aim of all science-including history insofar as it is a science.

The problem of the scientist is one of achieving sound generalizations. The scientist is careful not to make assertions which outrun his evidence, and he refuses to outtalk his information. He generalizes, but recognizes that no generalization can be more than probable, for we can never be certain that all the evidence is in, nor can the future be guaranteed absolutely-not even future eclipses of the sun and moon. But the scientist knows that certain laws have a very high degree of probability.

Let us look at the logic involved in forming sound generalizations. The number of cases investigated in the course of formulating a scientific law is a factor in establishing the truth of the law, but it is by no means the most important one. Obviously, if we observed one hundred swans, all of which are white, our generalization that "all swans are white" does not have the same probability it would have if we observed one thousand swans. But no matter how great the number of specimens involved in this type of observation, no more than a moderately high degree of probability is ever established. Countless numbers of white swans were observed throughout the ages (without any exceptions), and then in the nineteenth century black swans were observed in Australia.

The weakness of the method of "induction by simple enumeration of cases" is amusingly illustrated by Bertrand Russell's parable in his History of Western Philosophy:

There was once upon a time a census officer who had to record the names of all householders in a certain Welsh village. The first that he questioned was called William Williams; so were the second, third, fourth. . . . At last he said to himself. "This is tedious; evidently they are all called William Williams. I shall put them down so and take a holiday." But he w as wrong; there was just one whose name was John Jones."

Scientific generalizations based on other types of evidence than simple enumeration often acquire a much higher degree of probability after only a few observations. When a chemist finds that pure sulphur melts at 125 degrees C. in an experiment in which every factor is accurately analyzed and controlled, the law concerning the melting point of sulphur achieves as great a degree of certainty as is humanly attainable. Accurate control of every element of one case, then, is more important in establishing probabilities than is mere enumeration of many cases.

A single carefully controlled experiment, such as the sulphur experiment, can give us a much higher degree of probability than the mere observation of thousands of swans. The reason is that we also know that no chemical element thus far observed has a variable melting point under conditions of constant pressure. The chemical law is thus consistent with and is borne out by the rest of chemical knowledge, whereas the 'law" holding that all swans are white was based on an "accidental factor. Or consider the generalization concerning the mortality of mankind. This law is based not merely on the fact that countless numbers of human beings have died in the past, but also on the fact that all living beings must, by reason of physiological limitations, die, and that all matter wears out in time. So the harmony of a particular generalization with the rest of our knowledge is also a factor in giving it a high degree of probability.

So much for the logical analysis of generalizations. Thus far, we have been concerned with "uniform" generalizations, which take the form: "All A's are B's." A generalization, we have seen, is a statement that says something about "all" of a group, the evidence consisting of observations of items in which we always find a single characteristic. The observed cases are taken as a sample of the whole group or population with which we are concerned. We observe a number of swans and take these as a sample of all swans, past, present, and future. We find that all are white, and make the inductive leap: Swans are always white, everywhere.

"Statistical" statements
We shall now examine "statistical" statements. Statistical statements give us information, not about characteristics possessed by all of a group or population, but about those possessed by a definite proportion (or most) of the group or population, as when we say, "Most A's are B's," or "Sixty-five percent of all A's are B's." The first thing to note here is that statistical statements may, in fact, be generalizations and thus involve the notion of "all." This point involves very important (and common) misunderstandings.

In order to make this point clear, let us reinterpret our "uniform" generalizations. We say: "The sample is so-and-so (all observed swans are uniformly white) -therefore, the whole population of swans is uniformly white." Now, we do the same sort of thing in statistical generalizations. We say, "In the sample of redheads we examined, 53 percent were hot-tempered; therefore, 53 percent of all redheads are hot-tempered." (Or 53 percent of the whole population of redheads is hot-tempered.) Logically, both examples, uniform and statistical, are of the same type, for in each we make the inductive leap from the sample to the whole population. The only difference between them is that in the one case we assert a uniform character in the whole population, while in the other we assert that a characteristic holds in a certain proportion in the whole population.

This fundamental point will help us to evaluate the degree of probability of a statistical generalization. We saw earlier that uniform generalizations can never be absolutely certain-though for practical purposes we often consider them so, especially in the physical sciences. The probability of a generalization depends especially on the quality and also on the quantity of the cases that constitute the sample. The same holds for statistical generalizations, which may have a high probability, depending on the character of the evidence. Though the inductive leap is involved m all generalizations, in some cases the leap is justified. Let us examine the criteria of justification for the leap.

Before we proceed we shall discuss an important distinction: that between the sample and the inference we draw from it. It is one thing to describe a sample accurately and quite another to draw an accurate inference. If I say, "I have observed ten swans (the sample) and all were white," we may assume that the sample is accurately described. But if I now go on to generalize (that is, draw the inference) concerning all swans, my inference may not be a good one. A generalization always involves a leap in the dark," sometimes justified and sometimes not Similarly, if I say, "I have talked to ten friends concerning their income, and six [60 percent] told me that they earned more than $20,000 a year," the description of the sample may be accepted as true. But suppose I now go on to make the following inference: "Therefore, 60 percent of all Americans earn more than $20,000 a year." This would be a hasty generalization indeed.

We distinguish, then, between the sample and the inference. A study of I. Q. scores of 18,782 Mr. Force enlisted men revealed that those who were accountants in their civilian lives had the highest median (128.1), while those who were miners were among the very lowest (92.0). Now, these figures involve no inferences. They simply describe the actual facts in the sample. We draw an inference, on the other hand, when we assume that all accountants and all miners in the United States would have shown the same kinds of averages as the sample. In our discussion, henceforth, we shall be concerned only with the logical problems involved in statistical inferences.

Suppose that a public opinion poll was recently taken. The polling I organization tells us that 58 percent of the American people approve of the record of the present administration in Washington. How do they know this? Let us examine the evidence on which this finding is based. Obviously not everyone was consulted. A sample was taken. There were 3,000 interviews. Since there are approximately 150 million adults in the United States, each individual in this sample is taken as representative of 50,000 adults. Further, in the sample, 1,000 persons said that they had no opinion." Eleven hundred and sixty said that they "approved," and 840 said they did not. Thus 58 percent of those with opinions approved, and this means, we are told) that approximately 87 million Americans approve. The pollsters assume that the undecided individuals will probably divide in the same proportion as the others when they make up their minds.

Now, we are not raising any questions concerning the truth of the report made of the sample. But is the inductive leap from the sample to the generalization concerning 150 million people justified? It may be. It all depends upon the reliability of the sample. What makes a sample reliable? It must be fair, unbiased, and representative of the whole. But the crucial problem is to determine whether or not it has these characteristics.

The size of the sample is obviously important. A sample of 100 would not be so reliable as one of 1,000, and 1,000 would not be so reliable as one of a million. But large numbers in themselves may not be the most important factor in establishing the reliability of generalizations or inferences.

The unimportance of large numbers as such is best illustrated by the ill-fated Literary Digest presidential election poll in 1936. The magazine sent pre-election ballots to 10 million persons and received over 2 million responses. The responses showed Landon running ahead of Roosevelt. In the election in November, however, Roosevelt got about 28 million votes, Landon around 18 million.

The reason for this colossal failure was the unrepresentative character of the sample. The Digest took names "at random" from telephone directories and lists of registered owners of automobiles. These were relatively well-to-do folk. The lower income groups, however, were completely, or almost completely, unrepresented.

An ideal sample is one taken "at random" from the entire population, and not from a selected portion of the population being studied. The Gallup poll, for example, uses a special kind of random sampling, and, barring a spectacular failure in 1948, has been far more successful than the Literary Digest poll. Let us see how the Gallup poll operates. A sample of 3,000 individuals is taken, but with great care to make the sample representative. The population is classified into subgroups by geographic regions, by rural or urban residence, economic status, age, education, and declared politics. In 1948, for example, Gallup estimated that 28 percent of the American people lived in the Middle Atlantic states, 10 percent on the West Coast; that 34 percent lived in cities of over 100,000 population; that 23 percent were of an "average" economic station; that 43 percent were between the ages of thirty and forty-nine; that 42 percent had gone to high school; and that 38 percent called themselves Democrats, 36 percent Republicans, and 26 percent independents or members of smaller parties. The 3,000 interviews in the sample were distributed so that each geographic area, each economic group, etc., would be represented in its appropriate numerical strength.

Individuals are then chosen "at random," rather than by selection, from within each subgroup, and the resulting sample is highly representative of the whole population. The Gallup poll enjoys a successful record, on the whole, except for 1948. In other words, the method works, and one must respect its findings. But no poll can ever eliminate the possibility of error or guarantee accuracy except within a margin of error of several percentage points. And in a presidential election forecast the pollster is either completely right or completely wrong in predicting who will win. Odds of ten to one against a candidate of one of the major parties are probably not justified even if all the polls are confidently unanimous as to the final results. These were the odds against Harry Truman in the presidential election of 1948!

An election prediction can be judged by the election results, and a long series of successful predictions gives us confidence in the methods of the pollsters. This check cannot be made on polls which tabulate public opinion on issues of the day, for the whole population is never counted. Similarly for polls which rate television shows, for the whole audience is not counted. Such polls, of course, also generalize on the basis of samples. To illustrate the logical problems in assessing the reliability of a statistical study of the "public opinion poll" type we shall comment on Sexual Behavior in the Human Female, by Alfred C. Kinsey and his staff.

Kinsey's study, published in 1950, tabulates and classifies data concerning 5,940 white American females, ages two to ninety. He did not claim that his averages necessarily apply to all human females, despite the title of his book, nor even to all American women, of whom there were approximately seventy million in 1950. It is inevitable, however, that such inferences will be drawn, and our question is: Are such inferences justified? This depends entirely on the representativeness of Kinsey's sample.

Critics of Kinsey's report have emphasized the unrepresentiveness of his sample. His subjects were not distributed proportionately in geographic areas; most were from Illinois, Florida, and California. They were more highly educated than a representative cross-section of the population; 75 percent of his subjects went to college, as compared with a national average of 13 percent. Three percent of his women did not go beyond grade school as compared with the national average of 37 percent. A larger than average proportion were from middle and upper economic groups. Very few of the women were Roman Catholics or orthodox Jews.

Critics have also argued that the very nature of the study involves a kind of bias, for many women will refuse to discuss matters of such "delicate privacy" with interviewers, so that his volunteers must be unrepresentative of women in general. And there is also the problem of credibility. Critics have said that people who like to talk about such things tend to understate or overstate, and even to embroider a little.

Kinsey, of course, recognized the limitations and incompleteness of his sample, and, as noted, did not claim that it was representative of the whole population. But it will be interpreted in this way, and if Kinsey wished to avoid such interpretations, he should have called his study "Sexual Behavior of 5,940 Women." Inferences would probably be drawn, however, even if he had so titled his study.

The elements of distortion in Kinsey's sample detract from its reliability as a basis for generalizing. On the other hand, as a review of the book in Life put it, though the statistics are not perfect, they are, at any rate, "the only statistics in town." His study is by no means worthless as an index of sexual behavior. We must not use an "all or nothing" approach here. The reliability of his sample with respect to university women as a single group, for example, is certainly much higher than that for the female population as a whole. But we cannot conclude that the whole female population resembles the sample since the sample is not a representative one.

Generalizations in statistics, then, are judged by the same logical criteria we use in judging any generalizations. Fallacies, however, are more common in statistical than they are in uniform generalizations. For it is easier to check on the reliability of a uniform generalization: one exception overthrows the general rule or 'law." In statistics, however, since nothing is said about any specific individual, an "exception" is a meaning-less term. An exceptional individual does not disprove an "average." But there is, as we have already noted, a method for checking the reliability of a statistical generalization concerning a population, and that is to count the whole voting population in an election. But even a test of this kind is not conclusive, for many of the voters do not vote on election day, because of laziness, overconfidence, or some other reason.

Errors of inference in statistics are frequently overlooked because of the mathematical language in which statistics are presented. The spell which numbers weave often prevents us from seeing errors in arguments--errors which would be obvious were they not clothed in mathematical garb. And many dishonest reasoners take advantage of this fact and present highly selected data for purposes of propaganda rather than information. Misuses of the science of statistics have resulted in such jibes as, "Figures don't lie, but liars figure," and there are three kinds of lies: ordinary lies, damnable lies, and statistics." But these cynical remarks should not be taken as criticisms of statistics. The fault never lies with the figures, or with the science, but with their careless use. It is simply not the case that "you can prove anything with figures" (or statistics), just as it is never the case that "you can prove anything by logic." To the uninitiated, it just seems that you can.


FOR DISCUSSION AND WRITING

1. Below are some generalizations followed by the samples on which they are based. In each, explain why the sample may not be typical of the class being investigated.
a. Generalization: Water always makes me ill. Sample: During the last week I drank a glass of water after drinking a highball, a glass of wine, a six-pack of beer, a shot of vodka, and a half-pint of whiskey.
b. Generalization: Child geniuses usually turn out to be disappointing when they grow up. Sample: A recent article in Reader's Digest described seven cases in which children rated "genius" became alcoholics, irresponsible, or mentally ill.
c. Generalization: Watching television causes lower grades. Sam-pie: Two students in my history class have failed all of their tests, and blame their long hours of television viewing, rather than studying, for their low performance.
d. Generalization: Smoking marijuana is harmful. Sample: Captain Fleming of the police department says that the use of marijuana was admitted by seven suspects recently captured in a heroin ring.
e. Generalization: If you want to avoid tooth trouble, you should avoid candy. Sample: My sister stayed away from candy for three years, and during that time she had no cavities.

2. Read each of the following paragraphs and follow the directions that accompany them.
a. In a survey covering twenty-three states, the research team of a Democratic candidate for the presidency discovered the following statistics. Of the registered Democrats in those states, 43 percent had salaries below $7,000; 37 percent had salaries between $7,000 and $10,000; and 20 percent had salaries above $10,000. Imagine that you are the candidate's adviser. What advice would you give him on his campaign strategy with respect to economic issues, in the light of these statistics. What factors must be considered in your analysis'?
b. In a recent survey, English majors at a northeastern university were studied. In the years 1950-1970, 42 percent took jobs as teachers, either in junior high or senior high schools; 18 percent found jobs with publishers or in publisher-related industries; 3 percent either found no job at all or were engaged in occupations for which English skills had no particular value;' and 37 percent continued their education by entering graduate school. If you were given the opportunity to redesign the English curriculum at that university, what use would you make of those statistics? What conclusions could you draw concerning the present curriculum in terms of its meeting the needs of English majors?
c. A university committee on student affairs examined class attendance during the academic year 1972-1973. They found that class attendance on Mondays was 80 percent of the enrollment; on Tuesdays, 84 percent; on Wednesdays, 91 percent; on Thursdays, 83 percent; and on Fridays, 71 percent. What generalization could the committee make about the tendencies of college students to cut classes?

3. Most-if not all-of us repeat and accept generalizations in all areas of our lives. List several that you find yourself repeating without usually stopping to examine them carefully. How valid are they? How often do you use them as the basis of action?

---

CHAPTER 11
On Matters of Taste and Opinion

We have been engaged in analyzing matters of logic and science. Science, we all agree, statements must be proved by evidence. In physics and chemistry, certainly, scientific laws are not just "matters of opinion." A successful. laboratory test is something quite definite and convincing. The law court, too, requires proof. In a criminal trial the evidence might prove a man guilty 'beyond a reasonable doubt" There may be differences of opinion concerning some verdicts-miscarriages of justice occurring in acquittals as well as in convictions-but the rule is that it is the evidence that counts. The "reasonable man's judgment" is the ultimate criterion for questions of fact in a law court, and reasonable men the world over would probably reach the same verdict in a given dispute. In the social sciences, too, we try to get beyond matters of mere opinion. Carefully tabulated statistics tell w what percentage of paroled convicts will probably "go straight" thereafter, within a given "margin of error."

Value judgments
We agree that logic is relevant in questions concerning facts, for these questions involve evidence and proof. There is an important class of statements, however, which we have not yet examined and which you will often be called upon to write in essays, reviews, and reports in college. These are "value judgments," which many people regard as exempt from the requirement of proof or as incapable of proof. Examples: "I'. S. Eliot was a great poet." "The wartime bombing of cities is morally wrong." "The Beatles were the greatest pop music group of the 1960s." A value judgment, as we shall use the term in this discussion, is an assertion that something is either good or bad in an aesthetic or a moral sense. This restriction means that we shall exclude the purely technical sense of good (or bad) in this discussion, as when we speak of "a good automobile tire" (one that will run a long distance) or "a good repair job." Value judgments, then, are statements such as "X is beautiful" (possessing aesthetic excellence) or "X is morally right." Value judgments obviously refer also to statements that things are ugly or that actions are morally wrong.

Value judgments are usually contrasted with "factual statements," which make assertions about events that can be observed in the world of space and time. By a factual statement we do not necessarily mean a true statement. A factual statement, in the sense in which we use this term, refers to one that is about facts. Factual statements are true or false, for they may describe the facts correctly or incorrectly. "The Gateway Arch in Saint Louis is higher than the Leaning Tower of Pisa" is a factual type of statement. So also is "The Leaning Tower is higher than the Gateway Arch." "The Gateway Arch is more beautiful than the Leaning Tower," on the other hand, is a value judgment.

"It is illegal to grow and possess marijuana in the state of California" is a statement of fact, which happens to be true. We can verify this statement by looking up the law. "One ought not to smoke marijuana" is a value judgment which asserts that such conduct is morally wrong. Now, there is general agreement that it is possible to specify the kind of evidence which would prove factual statements true or false, but many people think that value judgments are incapable of proof. Value judgments, it is said, are "mere matters of opinion." It is important to note the precise sense m which this ambiguous phrase is meant. An "opinion" sometimes means a judgment that has a certain measure of probability, but not certainty, as when a man expresses the opinion that higher taxes are necessary to stave off inflation. When the evidence is conflicting, as in a case of this kind, we may speak of 'legitimate differences of opinion." But the "opinion" which asserts that higher taxes are necessary may be true, and the opposite opinion may be false. When people say that value judgments are mere matters of opinion, on the other hand, they usually mean "a matter of personal feeling or preference" or a "matter of taste." When we qualify the word "opinion" by mere in this discussion, this is the view to which we refer. According to this view, value judgments are incapable of proof and thus outside the realm of logical or scientific criticism.

If the "mere opinion" point of view is correct, then one value judgment is "as good as another," and proof is not only impossible but irrelevant If one value judgment is as incapable of justification as another, then reason and intelligence are irrelevant in the discussion of such matters. In this chapter we shall endeavor to show that logic is relevant in the discussion of value judgments as well as in the realm of scientific or "factual" statements. We shall first discuss two theories which hold that value judgements are incapable of logical or scientific justification. One of these we shall call the "taste" theory; the other the "approval" theory. We shall then discuss the sense in which logic is relevant in value judgments.

The "taste" theory
There is an ancient adage which tells us that "of matters of taste there is no disputing." This is sensible advice. If you prefer red wine and I prefer white, this establishes a basis for harmony such as prevailed in the famous Spratt family, and it would seem fruitless to argue the question as to which really tastes better. We may grant that it is impossible to prove that the taste of black caviar is superior to that of red, even though most "epicures" prefer the former. Some people may argue that it is a fact, not a "matter of taste," that French cuisine is superior to English cooking, but mankind has wisely decided that these are matters that ought not to be disputed. But how far, and to what kinds of things, can this principle be applied? "Matter of taste" is frequently used for things other than gustatory flavors. It often covers individual preferences and likes or dislikes in the arts as well, and it is sometimes held that ethical judgments are matters of taste. Let us examine the "taste theory" as applied in the fields of aesthetics and ethics.

No logical problem arises when one says that he prefers the sound piano to that of a violin or when he says "I prefer Tchaikovsky to Brahms." He is merely describing his personal taste. The interesting problem for logic arises when he says, "Tchaikovsky is a better composer than Brahms," or "Tchaikovsky's symphonies are more beautiful than those of -Brahms." Most literary critics regard Kahlil Gibran, author of The Prophet, as a second-rate writer, despite his phenomenal popularity all over the world? and perhaps an equal number say that Rod McKuen

writes bad poetry. Are these judgments true, or false, or neither? The "taste" theory holds that they are neither. When we assert value judgments of this kind, the taste theory tells us, the judgments merely express the preferences of the speaker, so that "second-rate," "bad," etc., are not words with any 6bjective reference. "X is beautiful," this theory tells us, means nothing more than "I like X," and "X is better than Y" means only "I prefer X to Y." If we grant that the speaker is telling the truth, and not lying about his actual preference, that is the end of the matter.

Is there no disputing "matters of taste" in the arts? In practice, of course, we do dispute such matters. The word "taste" is often used in a sense other than "personal preference"-for "keenness of discernment, or insight." Immanuel Kant, in one of his rare flashes of humor, once played on the double meaning of "taste" when he said "Art is a matter of taste, but there is no point in arguing matters of taste with the tasteless." Even when a person says that he prefers Tchaikovsky to Brahms, a Brahmsian is apt to accuse him of having a perverse taste. The Brahmsian believes that a person of "genuine" taste will prefer Brahms to Tchaikovsky. When some people tell us what they prefer, it thus appears, they regard their preferences as authoritative!

In practice, moreover, most persons will place limits on the taste theory. They may believe that one's preference for Brahms or Tchaikovsky is a matter of taste, but they will hesitate to say that the judgment Most male opera singers at the Met sing more beautifully than some truck drivers" is simply a matter of opinion, incapable of justification. Most of us do believe that there are shared standards of merit which go beyond merely individual liking and disliking. Semantically, then, it would appear that "I like X" is an inadequate translation of "X is beautiful." We never ask others to justify their feelings, as when they say, "I just happen not to like Picasso's Guernica." But we do think it is reasonable to ask them why they think a painting lacks aesthetic merit.

We have been discussing the taste theory in aesthetics. It may also be applied to ethical judgments. "X is morally right" (or "ought to be done") is held to be translatable into "I like X," and such judgments are also held to be mere matters of opinion, hence unarguable. But here again it seems that no one who understood the meaning of the words could fail to agree with the judgment "A teacher ought not to flunk a student in English 101 on the sole ground that he wears a beard." This sentence means a good deal more than -"I dislike teachers who do such things." Consciously or unconsciously we carry in our minds a definition of "wrong," or a standard of justice, and when we say the action of the teacher is wrong, we have classified it under our conception of wrongful or unjust actions. Again, to say that this judgment is a "matter of taste" and to attempt its translation into "I dislike X" is inadequate.

The "approval" theory
The second theory which holds that value judgments. cannot be logically supported or justified is the "approval theory." This theory tells us that judgments concerning right and wrong can be translated into, "My group approves or disapproves." This view is usually associated with the doctrine of "ethical relativism."15 The relativist holds that nothing is always right or wrong but that these terms are relative to time, place, and circumstance. The relativist is impressed by the great variety of contrary customs in different parts of the world. Monogamy is customary in western countries, polygamy in Arab countries, and in some places polyandry is the custom. The customary, the relativist notes, is considered morally right; the uncustomary morally wr6ng. And so what one country considers morally right, another considers wrong. And their customs of today may not be their? customs of tomorrow.

Thus far the relativist merely cites commonplaces known since the beginning of tourism. He now makes his distinctive contribution: What a group of people consider right, he says, is right. Thus, monogamy is right for Americans, and polygamy is right for Saudi Arabians. Right, of course, means right for them. When we say that polygamy is wrong in the United States, the relativist says, we mean only that Americans disapprove of this matrimonial system. Right and wrong are relative to group customs and group approvals.

But we noted earlier that the mere fact that people believe something is not sufficient to make it true, and similarly the fact that people practice certain customs doesn't make them right. Let us sum up the relativist position in a quotation from Pascal:

Three degrees of latitude reverse all jurisprudence; a meridian decides the truth. Fundamental laws change after a few years of possession; right has its epochs. . . . A strange justice that is bounded by a riven Truth on this side of the Pyrenees, error on the other side.

This theory has an appealing plausibility when we consider the variable customs throughout the world. It appears presumptuous for one nation to tell another that its customs are "immoral." It would be too easy to return the compliment. But the group-approval theory also seems quite inadequate as a reflection of what we mean by right and wrong. If group approval makes an action right, then it would be nonsensical, if not meaningless, to say: "I think this action is wrong, but I am in a minority." For "wrong" can have no meaning in this remark if the approval of the majority makes the action right. But does anyone seriously believe that the majority can never be mistaken in its judgment about right and wrong?

Let us now consider some of the implications of the two "translations" of aesthetic and moral judgments that we have been discussing. The taste and approval theories deny the possibility of genuine disagreements over values. What appear to be disagreements, they hold, are merely "verbal disputes." Consider a difference over moral values. When one man says that selfishness is morally wrong and another preaches that it is a virtue, if the first means "I dislike selfishness" and the second "I like it," then they are not really disagreeing with each other. They are merely uttering confessions about their feelings or making disguised autobiographical statements. It is as if I say I like tennis and you say you don't. Similarly, when one man says that polygamy is wrong, and another says it is right, if what each means is that his group disapproves on the one hand and approves on the other, then each is reporting a sociological fact about his society, and one statement does not contradict the other.

The approval theory also makes moral discussion impossible within a given group. If a moral problem arises, say over the denial of civil rights to a minority group, all the approval theory can tell us is: Take a "Gallup poll"; find out who has the votes. If 51 percent vote "right," then it is right. But "This has the vote of the majority" seems quite different from saying, "This is morally right." The majority have the power to decide what the civil laws shall be, but legality and morality are not the same; they sometimes diverge from each other. Would the abolition of the freedom of religious worship in the United States be right if 51 percent approved? Is the morality of such a matter determined by making an accurate count of the votes?

The "value standard" theory
We shall now discuss a new approach to the problem of values, the "value standard" theory, which makes logical analysis relevant in matters of values. We shall illustrate the theory by applying it to a problem in social ethics: Ought we to legalize gambling? My first reaction may be, "Yes, I don't see why not. Let people gamble if they wish to," or "No gambling is wrong." But there is a counterpart of the "1aw of rationality" m the field of values. The law of rationality tells us that we ought to justify our beliefs by evidence and reasons, instead of asserting them dogmatically. Similarly, if I am a reflective person, I will seek to justify my value judgments. (A value judgment which gives no reasons can scarcely be called a judgment.) I will think about my reasons for approving or disapproving of gambling, instead of saying, "I like (or dislike) gambling," or "The community approves (or disapproves) of gambling." I ought to consider the logical consequences of my choice. My thinking may proceed along the following lines: "It is impossible to suppress the human desire to gamble. If this desire is denied a legal outlet, it will find an illegal one. Illegal gambling funnels vast sums of money into the hands 6f undesirable elements in the community and gives these elements great power. They corrupt the police force and may even control the political machines in our great cities." These considerations make me lean toward legalization. Now I consider the other side: "If gambling is legalized, it may become respectable to gamble, and many more people may take to this vice. It is also possible that undesirable elements may manage to obtain control of legalized gambling."

Implicit in my thinking about this problem is the notion that the "public good" ought to be served. If I finally decide that, considering all the consequences, we ought to legalize gambling, I may justify my decision in some such fashion as this: "We ought to promote the general welfare. This duty requires us to choose the greater of two possible goods, or the lesser of two evils. The consequences that will follow from the legalization of gambling will, on the whole, diminish the general welfare less than the alternatives. Therefore we ought to legalize gambling." This "is a logical proof" of a value judgment. My major premise is a value standard which I assume as a basis for value decisions. My minor premise is a factual assertion to the effect that legalized gambling will diminish the general welfare least. ("General welfare," of course, is a term whose minimal meaning connotes freedom and respect for law, and I must define it in any actual discussion.) I then arrive at my conclusion.

Similarly, if I say that "government regulation of television is wrong," I ought to define what I mean by "wrong" in terms of a value standard. I should make my value standard explicit-whatever that value standard may be. I should then show how my standard applies to the matter of government regulation. If by "wrong" I mean "that which diminishes the general welfare," I must show how regulation will have that effect.

The value-standard analysis of value judgments makes logical discussion possible. A discussion of the legalization of gambling is not only a legitimate procedure but very useful when each disputant accepts the same value standard. Public discussion of such questions helps to clarify the consequences that follow from one or another course of action and thus leads to a thoughtful decision. Most value disputes, actually, involve disagreements concerning the appropriate means to mutually accepted ends. This is, of course, not necessarily the case: Some may have standards other than that of the general good, and not everyone who says that he seeks the general good actually does. Some may pay lip service to this standard, but belie their acceptance by their actual behavior. "Hypocrisy," as La Rochefoucauld said, "is the homage which vice pays to virtue."

A similar type of value standard analysis may be applied to aesthetic judgments. When I say that a novel is "good"-not merely that I like it-I must justify my judgment. I should define what I mean by "good," and I thus presuppose an aesthetic or critical standard of excellence in fiction. Perhaps my standard is that a good novel should have characters who are "real people," a significant human conflict situation properly developed and resolved, and an interesting story. One may quarrel with my standard, or with its application to a particular work, but such questions are not entirely outside the realm of rational discussion, as the taste theory presupposes. The thoughtful application of a standard is quite different from a snap-judgment made on the basis of surface liking or disliking, such as "This is a good novel; I always like to read stories about artists."

We have presented some examples of the way in which the value standard theory is used to justify value judgments. This approach is quite different from that of the taste theory. We shall now contrast the value-standard type of analysis with the type of analysis based on the theory of moral relativism.

The "untouchable" taboo in India is probably approved by the majority of the Indian people. According to the approval theory, if the people of a foreign country approve of a custom, then that custom must be right for them. It is thus not only improper for us to criticize the Indians for their caste system, but it must also be wrong for Indians to criticize, for the mere fact of approval makes the system morally right in India. The present government in India, however, seeks to abolish this system. This government obviously does not accept the approval theory. It believes that the system is wrong in India because it believes that the outcast system diminishes the general welfare of the Indians, or because it violates the principle that every human being should be treated as an end-in-himself and never as a mere means to an end.

To many, it would seem outrageous to condemn the Indian government for seeking to abolish untouchability. If we believe that the Indian government is right, then we do not accept the approval theory. And if Indian critics of untouchability are at least logically justified in their criticism of the system, then it is logically permissible for Americans to join in these criticisms, at least on the same grounds as those used by the Indian government. It is, of course, an extremely complex and difficult problem to decide whether a given custom is right or wrong. In judging untouchability we must consider the history of India, its traditions, religion, and customs of life, and we should also consider the adequacy of the standard we employ in making our value judgment. But these are matters open to logical discussion.

Ethical relativism has the insight that different customs may be equally right, depending on the history, traditions, and circumstances of life in different countries, and that we ought not to judge the customs of other countries by a blind application of the customs that happen to prevail in our own. Time, place, and circumstance do alter cases. But the relativist theory fails to distinguish customs and local rules, on the one hand, from standards and principles on the other. The principle that every human being ought to be treated as an end-in-himself is not on the same level as that of a marriage custom. And the relativists also deny that a custom may be wrong, as when a practice violates a standard that may be used to judge customs. In the United States, for example, the principle of equality has been violated in the custom of racial segregation.

Approval theorists are, of course, seldom consistent in applying their theory. This inconsistency indicates that most approval theorists do not themselves really believe that it is approval alone that makes an action right. They, too, assume that there are reasons for value judgments. They may say that polygamy must be right in a particular community "because the people approve of this custom." But if we ask them: "Why do the people approve?" we may get an answer something like this: "Large numbers of men were killed off in constant wars, or in dangerous occupations like seal-hunting, and this resulted in a surplus of women." A reason of this kind, however, indicates that this community judged polygamy right because it was thought desirable for the general good, and this is the use of a standard. Anthropologists, indeed, tell us that most customs are based on the necessity (or the presumed necessity) for particular kinds of behavior. If a group is to cope successfully with its environment, it must, of course, adapt its customs to that environment. Obviously approval has its basis in reasons, and the reasons are conceived as those which make the action right.

We have been seeking to justify the value standard theory in ethical and aesthetic value judgments. Reflection, we believe, will reveal that most of us do apply standards even though we may not be conscious of doing so.

But we must now deal with a question which may have been in the mind of the reader throughout this discussion: Can value standards themselves be justified by logical reasoning? This is indeed a difficult question, and its difficulty accounts for the continuing popularity of the taste and approval theories.

We have been justifying value judgments by the use of value standards without attempting to prove these standards. Value standards, of course, are neither true nor false in a literal sense, for they tell us what ought to be" and are not mere descriptions of facts. We have simply been assuming these standards in our proofs. At this point a more sophisticated 'taste" theorist may say, 'That is my real point. I grant that we can use logic in proving that the means are efficacious or not efficacious in achieving a given end, as in the discussion of legalized gambling, or in a marriage custom in relation to human happiness, but the end, as formulated in the standard, must simply be assumed. Whether we accept it or not is a matter of taste, unprovable."

This sophisticated version of the taste theory has a wide following among philosophers. But it is a long step beyond the simple form of the theory, for it synthesizes tastes and the use of standards. We may make the following comments on this version of the theory which asserts that the acceptance of standards is in the end an arbitrary matter:

1. Many, if not most, differences over values are differences with respect to the ways in which a certain end may be achieved, not differences with respect to the standards, and so discussion may turn disagreement into agreement. This is also true when we disagree over the application of a commonly accepted standard.

2. Some standards are more basic than others. We may say that abortion is wrong, because it involves the taking of a life. This indicates that we have taken the right to live as our standard, rather than the mother's right to terminate her pregnancy. But if someone were' to challenge our acceptance of this standard, we may wish to justify it by a more basic standard, such as "the natural law," believing that abortion is a violation of that law.

And further, we ought never to consider even our basic standards exempt from discussion) for no standard can be known to be absolutely final. As we develop in maturity, we may see beyond our present "ultimates."

3. The thoughtful person chooses his standards, not by mere liking," but by a personal commitment after surveying the consequences of his commitment with respect to everything he desires out of life. It is a question of how we really want to live. If this is in the end an arbitrary choice, it is one based on consideration of all relevant factors. It is not arbitrary in the way in which likes and dislikes may be.

Before we close this discussion, we shall briefly examine a different kind of philosophical theory of ethics which seeks to prove that some standards ought to be accepted and others not. It seeks to use reason and rational thinking to justify standards.

The philosopher Immanuel Kant believed that there was a supreme moral principle which could be established by reason itself. Men are rational beings, said Kant, and the essence of rationality is consistency and the avoidance of self-contradiction. A rational principle, furthermore, must be capable of universalization, that is, it must apply to all persons in all situations.

His basic principle is that "we should act only according to a maxim which can be universalized." This basic principle resembles the Golden Rule, which, as we know, tells us to "Do unto others as we would have them do unto us." Kant shows how the violation of this principle involves inconsistency.

Consider the person who makes a lying promise, that is, one who makes a promise that he does not intend to keep. The "maxim" of his act is that "one may tell a lie when it is convenient to do so." Can this maxim be universalized? Can the liar say: Let everyone do what I am doing? No, says Kant. The liar "gets away" with his lie because others believe him. But if everyone lied when it was convenient to do so, then mutual trust would disappear and no one would believe anyone, so that a liar could derive no benefit from his lie.

Do not lie or steal, says Kant, not because "Honesty is the best policy" (though it probably is), or because "Crime doesn't pay" (though it usually doesn't), but because such conduct is morally wrong. Morality demands that we universalize our conduct, that we give others the privileges we claim for ourselves (insofar as they are similar to us in relevant characteristics and in similar situations), and the liar and thief cannot do this without self-contradiction. For lying and stealing, if universalized, would frustrate the goals of the liar and thief. The one would not be believed, and the other would find that the rights of property had disappeared, so that he could not enjoy his ill-gotten gains in peace.

Slavery (or any form of coercion except for crime) is wrong. A slave is treated as if he were a purposeless thing, and his dignity as a fellow human being is disregarded. Can anyone be willing for everyone to be treated in like fashion, including himself? The contradiction is manifest, for no one can wish that his own wishes be frustrated. Slavery is wrong, says Kant, even though a minority of slaves may contribute to "the greatest happiness of the greatest number," for nothing that violates the basic principle of morality or justice can be right. And the rightness of these principles, he argues, is a necessary consequence of rational thinking. If the universalization of one's act leads to the defeat of one's own purposes, Kant holds, the act involves inconsistencies and self-contradiction, and the maxim of such an action cannot be an acceptable principle to a rational man.

In this chapter we have tried to show the relevance of logic in value judgments. The law of rationality is relevant here. We ought to justify our value judgments by reasons. We have shown how value standards may be used in making such judgments, and we have discussed the possibility of justifying the value standards by logic. Whether such justification can finally be achieved or not, the acceptance of such standards is far from being an arbitrary matter. We shall now sum up the different ways in which logic helps us to clarify our standards in the field of ethics.

1. Logic clarifies what we mean by right and wrong, in terms of our moral standards.

2. Logic helps us to determine whether a particular judgment makes sense in terms of these standards. We ought to work out the consequences that follow from our choices. We may find that we are defeating our own ends. Many people have supported the mandatory death penalty for kidnapping on the ground that it will effectively discourage the commission of this crime. But one should also consider some of the undesirable consequences that result from a law of this kind. The death penalty-for kidnapping as such-eliminates any inducement to the kidnapper to bring his victim back alive, and juries will also be reluctant to render a verdict of guilty in cases not involving murder, for a guilty verdict will mean the death penalty. If we want victims brought back alive, and if we want juries to do their duty, we may wish to revise our approval of the mandatory death penalty for kidnapping.

3. Logical analysis may also show us that we are striving for the wrong ends, as many prohibitionists in the 1920s discovered when they saw that the Volstead Act did not bring about the elimination of the problems associated with the consumption of alcoholic beverages. In fact, they discovered that prohibition created new problems and magnified others.

4. Logical analysis may also reveal self-consistencies in our own thin king. Few of us, perhaps, are as inconsistent as was the Mexican Society for the Prevention of Cruelty to Animals when it raised a consider-able sum of money in a benefit performance at the bull ring. But we are sometimes as inconsistent as was King David-though in other contexts-when he condemned the rich man (in Nathan's parable) for taking away the poor man's little ewe lamb. David himself had sent Uriah to his death in order to possess his beautiful wife, Bathsheba.

We see, then, that logic is relevant in all human problems-in problems of values as well as those of a scientific nature. To dismiss judgments of value as mere matters of opinion, and not subject to discussion, is to invite an irrationalist attitude, an attitude which dismisses the criteria of logical analysis and the possibility of testing of our opinions by experience.

Reason is opposed to dogmatism, fanaticism, and obscurantism. The rational way of life offers no panaceas; the rational individual recognizes the complexities of human problems and the difficulties in proving one side or the other in controversial issues. But most human problems are soluble, and if we try hard enough, using the best methods that the human race has so far devised for thinking about these problems, we are justified in the faith that we shall solve them, one by one. The best methods, as we have seen, require working out the probable consequences of our ideas and testing them in experience.


FOR DISCUSSION AND WRITING

1. Classify each of the following statements as a "value judgment" or as a "factual statement.
a. Cavelier pipe tobacco has a smooth taste.
b. The New York Times has more subscribers than any other newspaper.
c. Heavy petting before marriage is sinful.
d. Maxwell House coffee is made from a blend of ten varieties of coffee beans.
e. The new El Dorado is the finest Cadillac made.
f. White's toothpaste is better than Score toothpaste.

2. As this chapter makes clear, value judgments should be recognized for what they are--neither true nor false in a literal sense. With this in mind, write a 500-word essay in which you set forth a value judgment and support it with factual statements.

3. Read the following statement carefully. Then, in a 300-word essay, analyze its conclusion in the light of the following questions: What is the major premise? What is the value standard underlying the premise? What terms must be defined?

Capital offenders, it has been observed, are often the victims of their environment, and to execute them is a cruel and unusual punishment Therefore, the death penalty should be abolished.

---

Glossary

Ad hominem argument
An argument directed "to the man" rather than to the issue. Ad hominem arguments frequently attack the character, religion, or nationality of an opponent.

Ad ignorantiam argument
An argument based on ignorance, or an appeal to our lack of knowledge.

Ambiguity
The state of having two or more possible meanings.

Argument
Discourse that contains at least two asserted statements and the claim that one statement ought to be believed because another is true.

Begging the question
An error in reasoning In which we pretend to prove something when actually we assume In the "proof' that which we are supposed to prove.

Cause
The necessary and the sufficient conditions for the occurrence of an event.

Conclusion
In a syllogism, the proposition that must follow from the major and minor premises; a deduction.

Deduction
The process of "drawing out" the logical consequences of what we already know or assume to be true.

Deductive reasoning
The application of generalizations and premises to particular Instances.

Definition
A verbal equivalent.

Enthymeme
An incompletely stated syllogism, that is, one with a missing premise.

Generalization
A general law or principle which is inferred from particular facts.

Hasty generalization
A generalization based on insufficient or inadequate evidence.

Hypothesis
A tentative answer to a problem to test its truth.

Induction
The process of formulating a conclusion by going from specific instances or examples already known.

Inductive reasoning
The formulation of generalizations or conclusions as a result of individual examples or instances.

Law of contradiction
The principle of logic which says that if we have asserted two contradictory statements, we must abandon one or abandon all claims of rationality.

Law of excluded middle
The principle of logic which tells us that a thing either has a particular characteristic or does not have that particular characteristic.

Law of identity
The principle of logic which states that A is A, that is, that "anything is liself."

Laws of thought
The principles of identity, excluded middle, and contradiction.

Logic
The science of proof, the process by which statements are supported with adequate proof by being tested against the right amount and kind of evidence.

Major premise
The premise in a syllogism which contains the major term.

Middle term
A term in the syllogism which appeals in both premises but not In the conclusion.

Minor premise
The premise in a syllogism which contains the minor term.

Post hoc fallacy
An error in reasoning in which an effect is incorrectly attributed to a cause. solely on the basis of chronological order.

Proof
Sufficient evidence to justify a conclusion.

Semantics
The study of meaning; in logic, the study of the relationships between signs and symbols and what they represent.

Stipulation
A specific or particular sense or way in which a word will be used.

Syllogism
An argument consisting of two premises and a conclusion, viz.,

- All Armenians are left-handed;
- Joe is Armenian.
- Joe is left-handed.

Tautology
A statement that is true because it provides for all logical possibilities, and is valid as demonstrated by logical analysis.

Truth
A statement proven to be or accepted as true and in correspondence with fact, actuality, or logic.

Valid argument
An argument in which the conclusion is necessitated by the premises.

Value judgment
An assertion that something is either good or bad in an aesthetic or a moral sense.