CHAPTER 8 Bessie, the Cow How We Know THE UNIVERSE IS in a per motion, growing, cooling, e mountains are being worn What We Know are deepening. All life is al decay, and death. Even wl tables and stones-is not in level, they are whirls of ele much as it did yesterday or it has not changed, but be The crucial point to be considered in a study of language behavior is the relation. ship of language and reality, between words and not-words. Except as we under- coarse perceptions . To modern science, th stand this relationship, we run the grave risk of straining the delicate connection to us, it does so only bec between words and facts, of permitting our words to go wild, and so of creating for ourselves fabrications of fantasy and delusion. felt. It is solid only in the or a rapidly spinning top WENDELL JOHNSON ited, so that we constant telescopes, speedometers cord occurrences that ou in which we happen to of our nervous system. dren know today with t cannot hear. It is absur thing "as it really is." Inadequate as our a great deal. The disco scope has given us a m or feel radio waves, b pose. Most of our con istry, and in medicine kind or another to inc life, our unaided sen We cannot even obey out mechanical To retur for, let us organism forming sending puscles mode entire wha thaHAPTER 8 BESSIE, THE COW 83 We Know Bessie, the Cow We Know THE UNIVERSE IS in a perpetual state of flux. The stars are in constant motion, growing, cooling, exploding. The earth itself is not unchanging; mountains are being worn away, rivers are altering their channels, valleys are deepening. All life is also a process of change, through birth, growth, decay, and death. Even what we used to call "inert matter"-chairs and tables and stones-is not inert, as we now know, for, at the submicroscopic level, they are whirls of electrons and protons. If a table looks today very much as it did yesterday or as it did a hundred years ago, it is not because a study of language behavior is the relation- it has not changed, but because the changes have been too minute for our words and not-words. Except as we under- coarse perceptions. ave risk of straining the delicate connection To modern science, there is no "solid matter." If matter looks "solid" d delusion. g our words to go wild, and so of creating to us, it does so only because its motion is too rapid or too minute to be felt. It is solid only in the sense that a rapidly rotating color chart is "white" WENDELL JOHNSON or a rapidly spinning top is "standing still." Our senses are extremely lim- ited, so that we constantly have to use instruments such as microscopes, telescopes, speedometers, stethoscopes, and seismographs to detect and re- cord occurrences that our senses are not able to record directly. The way in which we happen to see and feel things is the result of the peculiarities of our nervous system. There are "sights" we cannot see, and, as even chil- dren know today with their high-frequency dog whistles, "sounds" that we cannot hear. It is absurd, therefore, to imagine that we ever perceive any- thing "as it really is." Inadequate as our senses are, with the help of instruments they tell us a great deal. The discovery of microorganisms with the use of the micro- scope has given us a measure of control over bacteria; we cannot see, hear, or feel radio waves, but we can create and transform them to useful pur- pose. Most of our conquest of the external world, in engineering, in chem- istry, and in medicine, is due to our use of mechanical contrivances of one kind or another to increase the capacity of our nervous systems. In modern life, our unaided senses are not half enough to get us about in the world. We cannot even obey speed laws or compute our gas and electric bills with- out mechanical aids to perception. To return, then, to the relations between words and what they stand for, let us say that there is before us "Bessie," a cow. Bessie is a living organism, constantly changing, constantly ingesting food and air, trans- forming it, getting rid of it again. Her blood is circulating, her nerves are sending messages. Viewed microscopically, she is a mass of variegated cor- puscles, cells, and bacterial organisms; viewed from the point of view of modern physics, she is a perpetual dance of electrons. What she is in her entirety, we can never know; even if we could at any precise moment say what she was, at the next moment she would have changed enough so that our description would no longer be accurate. It is impossible to sayTHE PROC HOW WE KNOW WHAT WE KNOW ABSTRACTION completely what Bessie or anything else really is. Bessie is no stable " Start reading from ject," but a dynamic process. The Bessie that we experience, however, is something else agaly experience only a small fraction of the total Bessie: the lights and she Wa of her exterior, her motions, her general configuration, the noise makes, and the sensations she presents to our sense of touch. And beck 8. "wealth" Of our previous experience, we observe resemblances in her to certain of animals to which, in the past, we have applied the word "cow." The Process of Abstracting 7. "asset" The "object" of our experience, then, is not the "thing in itself," but a interaction between our nervous systems (with all their imperfections) and something outside them. Bessie is unique-there is nothing else in the uni . "farm assets" verse exactly like her in all respects. But our nervous systems, automatically abstracting or selecting from the Bessie-in-process those features of hers in which she resembles other animals of like shape, functions, and habits classify her as "cow." When we say, then, that "Bessie is a cow," we are only noting the process-Bessie's resemblances to other "cows" and ignoring difference. 5. "livestock" What is more, we are leaping a huge chasm: from the dynamic process Bessie, a whirl of electrochemico-neural eventfulness, to a relatively static "idea," "concept," or word, "cow". The reader is referred to the diagram entitled "The Abstraction Ladder" on page 85.' 4. "Cow" As the diagram illustrates, the "object" we see is an abstraction of the lowest level, but it is still an abstraction, since it leaves out characteristics of the process that is the real Bessie. The word "Bessie" (cow,) is the lowest verbal level of abstraction, leaving out further characteristics-the differ. ences between Bessie yesterday and Bessie today, between Bessie today and Bessie tomorrow-and selecting only the similarities. The word "cow" se- 3. "Bessie" lects only the similarities between Bessie (cow,), Daisy (cows), Rosie (cow3), and so on, and therefore leaves out still more about Bessie. The word "livestock" selects or abstracts only the features that Bessie has in common with pigs, chickens, goats, and sheep. The term "farm asset" ab- stracts only the features Bessie has in common with barns, fences, live stock, furniture, generating plants, and tractors, and is therefore on a very high level of abstraction. Our concern here with the process of abstracting may seem strange, since the study of language is all too often restricted to matters of pronunci- The "Abstraction Ladder" is based on the "Structural Differential," a diagram orig nated by Alfred Korzybski to explain the process of abstracting. For a further explanation both of the diagram and of the process it illustrates, see h es, see his Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics (1933), especially Chapter 25.E KNOW Ise really is. Bessie is no static y THE PROCESS OF ABSTRACTING 85 vever, is something else again total Bessie: the lights and shack ABSTRACTION LADDER Start reading from the bottom up eral configuration, the nolee to our sense of touch. And beca 8. The word "wealth" is at an esemblances in her to certain of extremely high level of ab- applied the word "cow." "wealth" straction, omitting almost al reference to the characteris- tics of Bessie. 7. When Bessie is referred to as not the "thing in itself," but a an "asset," still more of her char- "asset" acteristics are left out. (with all their imperfections) and -there is nothing else in the un r nervous systems, automaticaly 6. When Bessie Is Included among "farm assets," reference is made -process those features of her in 6. "farm assets" only to what she has in common with e shape, functions, and habit all other salable items on the farm. cow," we are only noting the ows" and ignoring difference. 5. When Bessie is referred to as "live- stock," only those characteristics she sm: from the dynamic process. 5. "livestock" has in common with pigs, chickens, entfulness, to a relatively static goats, etc., are referred to. 85.1 ader is referred to the diagram 4. The word "cow" stands for the charac- we see is an abstraction of the teristics we have abstracted as common to 4. "cow" COW,, COW2, COW3 . . . COW. Characteristics ace it leaves out characteristics peculiar to specific cows are left out. d "Bessie" (cow1) is the lowest er characteristics-the differ- day, between Bessie today and 3. The word "Bessie" (cow,) is the name we ilarities. The word "cow" se 3. "Bessie" give to the object of perception of level 2. The name is not the object; it merely stands for (cowl), Daisy (cow;), Rosie the object and omits reference to many of the still more about Bessie. The characteristics of the object. e features that Bessie has in p. The term "farm asset" ab- on with barns, fences, live- 2. The cow we perceive is not the word, but the object of experience, that which our nervous sys- rs, and is therefore on a very tem abstracts (selects) from the totality that con- stitutes the process-cow. Many of the characteristics of the process-cow are left out. stracting may seem strange, icted to matters of pronunci- 1. The cow ultimately consists of atoms, electrons, etc., accord- al Differential," a diagram orig- ing to present-day scientific inference. Characteristics (repre ing. For a further explanation both sented by circles) are infinite at this level and ever-changing. ace and Sanity: An Introduction to This is the process level. specially Chapter 25.36 HOW WE KNOW WHAT WE KNOW ation, spelling, vocabulary, grammar, and.such. The methods by whit Composition and oratory are taught in many school systems seem to largely responsible for this widespread notion that the is to concentrate one's attention exclusively on words. at the way to study words But as we know from everyday experience, learning language is not simply a matter of learning words; it is a matter of correctly relating our words to the things and happenings for which they stand. We learn the language of baseball by playing or watching the game and studying when goes on. It is not enough for children to learn to say "cookie" or "dog they must be able to use these words in their proper relationship to nonver. bal cookies and nonverbal dogs before we can grant that they are learning the language. As Wendell Johnson has said, "The study of language begins properly with a study of what language is about." Once we begin to concern ourselves with what language is about, we "Work," too, is a shorthan tion, for a characteristic th are at once thrown into a consideration of how the human nervous system works. When we call Beau (the Boston terrier), Pedro (the chihuahua), to navigation to running an Snuffles (the English bulldog), and Shane (the Irish wolfhound)-creatures in common. The special m derived from abstracting that differ greatly in size, shape, appearance, and behavior-by the same kinds of work. ("A transfer name, "dog," our nervous system has obviously gone to work abstracting ing in the motion or displa what is common to them all, ignoring for the time being the differences of the acting force and ag among them. College Dictionary.) The indispensability Why We Must Abstract trated by what we do wh nates from the Latin wo This process of abstracting, of leaving characteristics out, is an indispens- present meaning from suc able convenience. To illustrate by still another example, suppose that we for each sheep as it left t live in an isolated village of four families, each owning a house. A's house sheep returning at night is referred to as maga; B's house is biyo; C's is kata, and D's is pelel. This Primitive as this example is quite satisfactory for ordinary purposes of communication in the village, unless a discussion arises about building a new house-a spare one, let us matics works. Each pebb the "oneness" of each she say. We cannot refer to the projected house by any one of the four words stracting from extension we have for the existing houses, since each of these has too specific a mean- ing. We must find a general term, at a higher level of abstraction, that ciples, the numerical fac means "something that has certain characteristics in common with maga, circumstances, numerica matical symbols a biyo, kata, and pelel, and yet is not A's, B's, C's, or D's. Since this is much therefore abstra too complicated to say each time, an abbreviation must be invented. So we choose the noise, house. Out of such needs do our words come-they g occurrence are a form of shorthand. The invention of a new abstraction is a great step ons proper forward, since it makes discussion possible-as, in this case, not only the 1 world, th discussion of a fifth house, but of all future houses we may build or see mi foreseen cit our travels or dream about. A producer of educational films once remarked to me that it is impossi ble to make a shot of "work. " You can shoot Joe hoeing potatoes, Susan, Definition polishing her car, Bill spraying paint on a barn, but never just "work." itions, con only descHAT WE KNOW ammar, and.such. The methods by which ON DEFINITIONS 87 caught in many school systems seem to be n exclusively on words. spread notion that the way to study word yday experience, learning language is not rds; it is a matter of correctly relating our ings for which they stand. We learn the or watching the game and studying what ildren to learn to say "cookie" or "dog"; ds in their proper relationship to nonver- efore we can grant that they are learning Words defining words n has said, "The study of language begins nguage is about." selves with what language is about, we "Work," too, is a shorthand term, standing, at a higher level of abstract ation of how the human nervous system tion, for a characteristic that a multitude of activities, from dishwashing Boston terrier), Pedro (the chihuahua), to navigation to running an advertising agency to governing a nation, have Shane (the Irish wolfhound)-creatures in common. The special meaning that "work" has in physics is also clearly pearance, and behavior-by the same derived from abstracting the common characteristics of many different as obviously gone to work abstracting kinds of work. ("A transference of energy from one body to another, result- ring for the time being the differences ing in the motion or displacement of the body acted upon, in the direction of the acting force and against resistance." Funk and Wagnalls, Standard College Dictionary.) The indispensability of this process of abstracting can again be illus- trated by what we do when we "calculate." The word "calculate" origi- nates from the Latin word calculus, meaning "pebble," and derives its g characteristics out, is an indispens- present meaning from such ancient practices as putting a pebble into a box ll another example, suppose that we for each sheep as it left the fold, so that one could tell, by checking the ilies, each owning a house. A's house sheep returning at night against the pebbles, whether any had been lost. yo; C's is kata, and D's is pelel. This Primitive as this example of calculation is, it will serve to show why mathe- ses of communication in the village, matics works. Each pebble is, in this example, an abstraction representing ng a new house-a spare one, let us the "oneness" of each sheep-its numerical value. And because we are ab- house by any one of the four words stracting from extensional events on clearly understood and uniform prin- ach of these has too specific a mean- ciples, the numerical facts about the pebbles are also, barring unforeseen a higher level of abstraction, that circumstances, numerical facts about the sheep. Our x's and y's and other racteristics in common with maga, mathematical symbols are abstractions made from numerical abstractions, , B's, C's, or D's. Since this is much and are therefore abstractions of still higher level. And they are useful in bbreviation must be invented. So predicting occurrences and in getting work done because, since they are h needs do our words come-they abstractions properly and uniformly made from starting points in the ex- tensional world, the relations revealed by the symbols will be, again bar- of a new abstraction is a great step ring unforeseen circumstances, relations existing in the extensional world. ble-as, in this case, not only the ure houses we may build or see in remarked to me that it is impossi- On Definitions Definitions, contrary to popular opinion, tell us nothing about things. shoot Joe hoeing potatoes, Susan They only describe people's linguistic habits; that is, they tell us what a barn, but never just "work."OPE 88 HOW WE KNOW WHAT WE KNOW words people use under what conditions. Definitions should be understood as statements about language. are asked the meaning of a word, mazes; we will tend to "have our are talking about. This habit disp House. This word, at the next higher level of abstraction, can be substituted for the more cumbersome expression, "Something that has characteristics in com "What is meant by the word red mon with Bill's bungalow, Jordan's cottage, Mrs. Smith's guest home, Dr. Jones, "Well, the next time you see som mansion . . ." traffic light facing them. Also, you Red. A feature that rubies, roses, ripe tomatoes, robins' breasts, uncooked their trucks are painted." abstraction. beef, and lipsticks have in common is abstracted, and this word expresses that Thus, some of the most helpf Kangaroo. Where the biologist would say "herbivorous mammal, a marsupial ples with which readers or liste of the family Macropodidae," ordinary people say "kangaroo." details-verbal images, example more thoroughly, so readers wil Now it will be observed that while the definitions of "house" and "red" experiences to what they read. given here point down the abstraction ladder (see the charts) to lower levels of abstraction, the definition of "kangaroo" remains at the same level. That is to say, in the case of "house," we could if necessary go and look at "Let's Define Our Terms Bill's bungalow, Jordan's cottage, Mrs. Smith's guest home, and Dr. Jones's An extremely widespread insta mansion, and figure out for ourselves what features they seem to have in stitious) attitude toward defin common; in this way, we might begin to understand under what condi- scription, "Let's define our to tions to use the word "house." But all we know about "kangaroo" from the talking about." As we have above is that where some people say one thing, other people say another. golfer, for example, cannot That is, when we stay at the same level of abstraction in giving a defini- cannot understand and use t tion, we do not give any information, unless, of course, the listener or a large number of words is reader is already sufficiently familiar with the defining words to work operations they stand for in down the abstraction ladder. Dictionaries, in order to save space, have to people often believe that som assume in many cases such familiarity with the language on the part of the reader. But where the assumption is unwarranted, definitions at the same ignoring the fact that the w level of abstraction are worse than useless. Looking up "indifference" in serious confusions and amb some pocket dictionaries, we find it defined as "apathy": we look up "apa- to discover this fact and tr thy" and find it defined as "indifference." words, and then, finding o Even more useless, however, are the definitions that go up the abstrac words in the definitions of tion ladder to higher levels of abstraction-the kind most of us tend to in a hopeless snarl. The on friend: make automatically. Try the following experiment on an unsuspecting to a minimum and to poi writing and speaking, this talki- ut. "What is meant by the word red?" "It's a color." "What's a color?" al "Why, it's a quality things have." "What's a quality?" "Say, what are you trying to do, anyway?" he s You have pushed him into the clouds. If, on the other hand, we habitu- ally go down the abstraction ladder to lower levels of abstraction when we the conE KNOW s. Definitions should be understood OPERATIONAL DEFINITIONS 89 are asked the meaning of a word, we are less likely to get lost in verbal vel of abstraction, can be substitute mazes; we will tend to "have our feet on the ground" and know what we thing that has characteristics in com are talking about. This habit displays itself in an answer such as this: Mrs. Smith's guest home, Dr. Jones's "What is meant by the word red?" tomatoes, robins' breasts, uncooked Well, the next time you see some cars stopped at an intersection, look at the acted, and this word expresses that traffic light facing them. Also, you might go to the fire department and see how their trucks are painted." "herbivorous mammal, a marsupial say "kangaroo." Thus, some of the most helpful definitions are those that include exam- ples with which readers or listeners may be familiar. A writer provides definitions of "house" and "red" details-verbal images, examples, illustrations-so readers can understand er (see the charts) to lower levels more thoroughly, so readers will have more opportunities to link their own o" remains at the same level. experiences to what they read. ould if necessary go and look at h's guest home, and Dr. Jones's "Let's Define Our Terms" features they seem to have in An extremely widespread instance of an unrealistic (and ultimately super- nderstand under what condi- stitious) attitude toward definitions is found in the common academic pre- w about "kangaroo" from the scription, "Let's define our terms so that we shall all know what we are ng, other people say another. talking about." As we have already seen in Chapter 4, the fact that a bstraction in giving a defini- golfer, for example, cannot define golfing terms is no indication that he ss, of course, the listener or cannot understand and use them. Conversely, the fact that we can define the defining words to work a large number of words is no guarantee that we know what objects or order to save space, have to operations they stand for in concrete situations. Having defined a word, language on the part of the people often believe that some kind of understanding has been established, ted, definitions at the same ignoring the fact that the words in the definition often conceal even more oking up "indifference" in serious confusions and ambiguities than the word defined. If we happen "apathy": we look up "apa- to discover this fact and try to remedy matters by defining the defining words, and then, finding ourselves still confused, we go on to define the ons that go up the abstrac words in the definitions of the defining words, we quickly find ourselves e kind most of us tend to in a hopeless snarl. The only way to avoid this snarl is to keep definitions to a minimum and to point to extensional levels wherever necessary; in ment on an unsuspecting writing and speaking, this means giving specific examples of what we are talking about. Operational Definitions Another way to keep extensional levels in mind, when definitions are called for, is to use what physicist P. W. Bridgman called "operational defini- tions." As he says: other hand, we habitu- To find the length of an object, we have to perform certain physical opera- of abstraction when we tions. The concept of length is therefore fixed when the operations by whichHOW WE KNOW WHAT WE KNOW CHASING O 90 length is measured are fixed. . . . In general, we mean by any concept nothing more than a set of operations; the concept is synonymous with the corresponding Chasing Oneself in Verba set of operations. 2 In other words, the kind of " that which never leaves the The operational definition, then, as Anatol Rapoport explains, is one that never points down the ab that tells you "what to do and what to observe in order to bring the thing and from there to the extensi defined or its effects within the range of one's experience." He gives the following simple example of how to define "weight": go to a railroad "What do you mean by dem station or drugstore, look for a scale, stand on it, put in a penny, read the "Democracy means the pres number at which the pointer comes to rest. That is your weight. But sup "What do you mean by rig pose different scales give different readings? Then your weight can be said "By rights I mean those priv to be within the range of, say, 140 to 145 pounds. Wit unds. With more accurate ent privileges." scales you might get closer readings, such as 142 pounds plus-or-minus one. "Such as?' "Liberty, for example." But there is no "property" called weight that exists apart from the opera- What do you mean by lit tions of measuring it. As Rapoport says, "If the only way we can be aware "Religious and political f of the amount of weight is by means of the scale, then the very definition "And what does that mea of weight has to be in terms of the scale."3 "Religious and political fr Such, then, is the scientific, or "operational" point of view toward def- inition-one that attempts rigidly to exclude non-extensional, non-sense Of course it is possible t statements. We can extend this idea from science to the problems of every- son and Lincoln have done day life and thought. Just as there is no such thing as "length" apart from tier in American History, the operations by which length is measured, there is likewise no "democ- Its Enemies, as Robert I racy" apart from the sum total of democratic practices, such as universal franchise, freedom of speech, equality before the law, and so on. Similarly, States; Conflict and Con there is no such thing as "brotherhood" apart from brotherly behavior, nor mind. Speakers who neve "charity" apart from charitable actions. may fail to notice when The operational point of view does much to keep our words meaning- This is by no means ful. When people say things like "Let's have no more of progressive meth- sionally meaningless no ods in our schools," "Let's get back to sound business principles in running talk about the future, our county government," "Let's try to do the Christian thing," "Let's re- conversation, and when store family values," we are entitled to ask, "What do you mean-exten- that have no extensin sionally speaking?" To ask this question often-of ourselves as well as of ghest ratiocinat others-is to do our bit toward reducing the vast amount of non-sense that bols are inde is written, spoken, and shouted in this incredibly garrulous world. go quickly f The best examples in everyday life of operational definitions are to be ned peas" t found in cookbooks, which describe the operations by means of which the manipulate s entity defined may be extensionally experienced. Thus: "Steak Diane. Slice ipulated ("I tenderloin beef very thin and give it a few whacks with a meat mallet to flatten it even more; sprinkle with salt and pepper to taste. Have your pan in one lon very hot . . . " (The Sunset Cook Book). Writers and speakers would do 1, even if well to study cookbooks occasionally to increase the clarity and verifiability of their utterances. s and not ar le, often ple 2 The Logic of Modern Physics (1927), p. 5. Operational Philosophy (1953), p. 25. find out And pure natical often