19

Definition and Understanding

We have seen that metaphor pervades our normal con­ceptual system. Because so many of the concepts that are important to us are either abstract or not clearly delineated in our experience (the emotions, ideas, time, etc.), we need to get a grasp on them by means of other concepts that we understand in clearer terms (spatial orientations, objects, etc.). This need leads to metaphorical definition in our con­ceptual system. We have tried with examples to give some indication of just how extensive a role metaphor plays in the way we function, the way we conceptualize our experience, and the way we speak.

Most of our evidence has come from language---from the meanings of words and phrases and from the way humans make sense of their experiences. Yet students of meaning and dictionary makers have not found it important to try to give a general account of how people understand normal concepts in terms of systematic metaphors like love is a JOURNEY, ARGUMENT IS WAR, TIME IS MONEY, etc. For example, if you look in a dictionary under "love," you find entries that mention affection, fondness, devotion, infatua­tion, and even sexual desire, but there is no mention of the way in which we comprehend love by means of metaphors like LOVE IS A JOURNEY, LOVE IS MADNESS, LOVE IS WAR, etc. If we take expressions like "Look how far we've come" or "Where are we now?" there would be no way to tell from a standard dictionary or any other standard ac­count of meaning that these expressions are normal ways of talking about the experience of love in our culture. Hints of

p.116 

the existence of such general metaphors may be given in the secondary or tertiary senses of other words. For instance, a hint of the love is madness metaphor may show up in a tertiary sense of the word "crazy" (= "immoderately fond, infatuated"), but this hint shows up as part of the definition of "crazy" rather than as part of the definition of "love."

What this suggests to us is that dictionary makers and other students of meaning have different concerns than we do. We are concerned primarily with how people under­stand their experiences. We view language as providing data that can lead to general principles of understanding. The general principles involve whole systems of concepts rather than individual words or individual concepts. We have found that such principles are often metaphoric in nature and involve understanding one kind of experience in terms of another kind of experience.

Bearing this in mind, we can see the main difference be­tween our enterprise and that of dictionary makers and other students of meaning. It would be very strange in a dictionary to see "madness" or "journeying" as senses of "love." They are not senses of "love," any more than "food" is one of the senses of "idea." Definitions for a concept are seen as characterizing the things that are inher­ent in the concept itself. We, on the other hand, are con­cerned with how human beings get a handle on the concept---how they understand it and function in terms of it. Madness and journeys give us handles on the concept of love, and food gives us a handle on the concept of an idea.

Such a concern for how we comprehend experience re­quires a very different concept of definition from the stan­dard one. The principal issue for such an account of defini­tion is what gets defined and what does the defining. That is the issue we turn to next.

p.117

The Objects of Metaphorical Definition:

Natural Kinds of Experience

We have found that metaphors allow us to understand one domain of experience in terms of another. This suggests that understanding takes place in terms of entire domains of experience and not in terms of isolated concepts. The fact that we have been led to hypothesize metaphors like love IS A JOURNEY, TIME IS MONEY, and ARGUMENT IS WAR suggests to us that the focus of definition is at the level of basic domains of experience like love, time, and argument. These experiences are then conceptualized and defined in terms of other basic domains of experience like journeys, money, and war. The definition of subconcepts, like BUDGETING TIME and ATTACKING A CLAIM, should fall out as consequences of defining the more general concepts (time, argument, etc.) in metaphorical terms.

This raises a fundamental question: What constitutes a "basic domain of experience"? Each such domain is a structured whole within our experience that is con­ceptualized as what we have called an experiential gestalt. Such gestalts are experientially basic because they charac­terize structured wholes within recurrent human experi­ences. They represent coherent organizations of our ex­periences in terms of natural dimensions (parts, stages, causes, etc.). Domains of experience that are organized as gestalts in terms of such natural dimensions seem to us to be natural kinds of experience.

They are natural in the following sense: These kinds of experiences are a product of

Our bodies (perceptual and motor apparatus, mental capacities, emotional makeup, etc.)

Our interactions with our physical environment (moving, manipulating objects, eating, etc.)

Our interactions with other people within our culture (in terms of social, political, economic, and religious in­stitutions)

p.118

In other words, these "natural" kinds of experience are products of human nature. Some may be universal, while others will vary from culture to culture.

We are proposing that the concepts that occur in metaphorical definitions are those that correspond to natu­ral kinds of experience. Judging by the concepts that are defined by the metaphors we have uncovered so far, the following would be examples of concepts for natural kinds of experience in our culture: love, time, ideas, under­standing, ARGUMENTS, LABOR, HAPPINESS, HEALTH, control, status, morality, etc. These are concepts that require metaphorical definition, since they are not clearly enough delineated in their own terms to satisfy the purposes of our day-to-day functioning.

Similarly, we would suggest that concepts that are used in metaphorical definitions to define other concepts also correspond to natural kinds of experience. Examples are PHYSICAL ORIENTATIONS, OBJECTS, SUBSTANCES, SEEING, JOURNEYS, WAR, MADNESS, FOOD, BUILDINGS, etc. These concepts for natural kinds of experience and objects are structured clearly enough and with enough of the right kind of internal structure to do the job of defining other con­cepts. That is, they provide the right kind of structure to allow us to get a handle on those natural kinds of experience that are less concrete or less clearly delineated in their own terms.

It follows from this that some natural kinds of experience are partly metaphorical in nature, since metaphor plays an essential role in characterizing the structure of the experi­ence. Argument is an obvious example, since experiencing certain activities of talking and listening as an argument partly requires the structure given to the concept argu­ment by the argument is war metaphor. The experience of time is a natural kind of experience that is understood almost entirely in metaphorical terms (via the spatialization of time and the time is a moving object and time is money metaphors). Similarly, all of the concepts (e.g.,

p.119

control, status, happiness) that are oriented by up-down and other spatialization concepts are grounded in natural kinds of experience that are partly understood in metaphorical terms.

Interactional Properties

We have seen that our conceptual system is grounded in our experiences in the world. Both directly emergent concepts (like up-down, object, and direct manipulation) and metaphors (like happy is up, events are objects, argu­ment is war) are grounded in our constant interaction with our physical and cultural environments. Likewise, the di­mensions in terms of which we structure our experience (e.g., parts, stages, purposes) emerge naturally from our activity in the world. The kind of conceptual system we have is a product of the kind of beings we are and the way we interact with our physical and cultural environments.

Our concern with the way we understand our experience has led us to a view of definition that is very different from the standard view. The standard view seeks to be "objec­tive," and it assumes that experiences and objects have inherent properties and that human beings understand them solely in terms of these properties. Definition for the objectivist is a matter of saying what those inherent properties are by giving necessary and sufficient conditions for the application of the concept. "Love," on the objectivist view, has various senses, each of which can be defined in terms of such inherent properties as fondness, affection, sexual desire, etc. Against this view, we would claim that we comprehend love only partly in terms of such inherent properties. For the most part, our comprehension of love is metaphorical, and we understand it primarily in terms of concepts for other natural kinds of experience: journeys, madness, war, health, etc. Because defining concepts (journeys, madness, war, health) emerge from our interactions with one another and with the world, the con-

p.120

cept they metaphorically define (e.g., love) will be under­stood in terms of what we will call interactional properties.

In order to get a clearer idea of what interactional prop­erties are in general, let us look at the interactional prop­erties of an object. Take the concept gun. You might think that such a concept could be characterized entirely in terms of inherent properties of the object itself, for example, its shape, its weight, how its parts are put together, etc. But our concept gun goes beyond this in ways that can be seen when we apply various modifiers to the concept. For exam­ple, take the difference between the modifiers black and pake as applied to gun . The principal difference for objec-tivist accounts of definition is that a black gun is a gun, while a fake gun is not a gun. black is seen as adding an additional property to gun, while fake is seen as applying to the concept gun to yield another concept that is not a subcategory of gun. This is about all that is said on the objectivist view. It will allow the entailments:

This is a black gun.     and    .   This is a fake gun.

Therefore, this is a gun.        Therefore, this is not a gun.

What such an account does not do is to say what a fake gun is. It does not account for entailments like:

This is a fake gun._______

Therefore, this is not a giraffe.

This is a fake gun.

Therefore, this is not a bowl of bean-sauce noodles.

And on and on ...

To account for such an indefinitely long list of entail­ments, we need a detailed account of just how fake modifies the concept gun. A fake gun has to look enough like a gun for the purpose at hand. That is, it has to have the contextually appropriate perceptual properties of a gun. You have to be able to perform enough of the appropriate physical manipulations that you would with a real gun (e.g.,

p.  121

hold it a certain way). In other words, a fake gun has to maintain what we might call motor-activity properties of a gun. Moreover, the point of having a fake gun is that it will serve certain of the purposes that a real gun could serve (threatening, being on display, etc.). What makes a fake gun fake is that it cannot function like a gun. If it can shoot you, it is a real gun, not a fake gun. Finally, it cannot originally have been made to function like a gun: a broken or inoper­able gun is not a fake gun.

Thus, the modifier fake preserves certain kinds of the properties of guns and negates others. To summarize:

pake preserves: Perceptual properties (a fake gun looks like a gun)

             Motor-activity properties (you handle it like a gun)

             Purposive properties (it serves some of the purposes of a gun)

fake negates:  Functional properties (a fake gun doesn't shoot)

             History of function (if it was made to be a real gun, then it's not a

             fake)

This account of how fake affects the concept of gun in­dicates that the concept gun has at least five dimensions, three of which are preserved by fake and two of which are negated. This suggests that we conceptualize a gun in terms of a multidimensional gestalt of properties where the di­mensions are perceptual, motor activity, purposive, FUNCTIONAL, etc.

If we look at what perceptual, motor-activity, and pur­posive properties are, we see that they are not inherent in guns themselves. Instead, they have to do with the way we interact with guns. This indicates that the concept gun, as people actually understand it, is at least partly defined by interactional properties having to do with perception, motor activity, purpose, function, etc. Thus we find that our con­cepts of objects, like our concepts of events and activities, p.122

are characterizable as multidimensional gestalts whose di­mensions emerge naturally from our experience in the world.

Categorization

On the standard objectivist view, we can understand (and hence define) an object entirely in terms of a set of its inher­ent properties. But, as we have just seen, at least some of the properties that characterize our concept of an object are interactional. In addition, the properties do not merely form a set but rather a structured gestalt, with dimensions that emerge naturally from our experience.

The objectivist account of definition is inadequate to ac­count for understanding in another way as well. On the objectivist view, a category is defined in terms of set theory: it is characterized by a set of inherent properties of the entities in the category. Everything in the universe is either inside or outside the category. The things that are in the category are those that have all the requisite inherent properties. Anything that fails to have one or more of the inherent properties falls outside the category.

This set-theoretical concept of a category does not ac­cord with the way people categorize things and experi­ences. For human beings, categorization is primarily a means of comprehending the world, and as such it must serve that purpose in a sufficiently flexible way. Set-theoretical categorization, as a model for human categori­zation, misses the following:

1. As Rosch (1977) has established, we categorize things in terms of prototypes. A prototypical chair, for us, has a well-defined back, seat, four legs, and (optionally) two armrests. But there are nonprototypical chairs as well: beanbag chairs, hanging chairs, swivel chairs, contour chairs, barber chairs, etc. We understand the nonprototypi­cal chairs as being chairs, not just on their own terms, but by virtue of their relation to a prototypical chair.

p.123

2. We understand beanbag chairs, barber chairs, and contour chairs as being chairs, not because they share some fixed set of defining properties with the prototype, but rather because they bear a sufficient family resemblance to the prototype. A beanbag chair may resemble a pro­totypical chair in a different way than a barber chair does. There need be no fixed core of properties of prototypical chairs that are shared by both beanbag and barber chairs. Yet they are both chairs because each, in its different way, is sufficiently close to the prototype.

3. Interactional properties are prominent among the kinds of properties that count in determining sufficient fam­ily resemblance. Chairs share with stools and other kinds of seats the purposive property of allowing us to sit. But the range of motor activities permitted by chairs is usually different from stools and other seats. Thus the interactional properties relevant to our comprehension of chairs will in­clude perceptual properties (the way they look, feel, etc.), functional properties (allowing us to sit), motor-activity properties (what we do with our bodies in getting in and out of them and while we're in them), and purposive properties (relaxing, eating, writing letters, etc.).

4. Categories can be systematically extended in various ways for various purposes. There are modifiers, called hedges (see Lakoff 1975), that pick out the prototype for a category and that define various kinds of relationships to it. Here are a few examples:

par excellence: This picks out prototypical members of a category. For example, a robin is a bird par excellence, but chickens, ostriches, and penguins are not birds par excel­lence.

strictly speaking: This picks out the nonprototypical cases that ordinarily fall within the category. Strictly speak­ing, chickens, ostriches, and penguins are birds even though they are not birds par excellence. Sharks, blowfish, catfish, and goldfish are not fish par excellence, but they are fish, strictly speaking.

p.124

loosely speaking: This picks out things that are not ordi­narily in the category because they lack some central prop­erty but which share enough properties so that for certain purposes it could make sense to consider them category members. Strictly speaking, a whale is not a fish, though, loosely speaking, it may be considered one in certain con­texts. Strictly speaking, a moped is not a motorcycle, though, loosely speaking, mopeds could be included among motorcycles.

technically: This circumscribes a category relative to some technical purpose. Whether something is technically in the category or not will depend on what the purpose in clas­sifying it is. For the purpose of insurance, a moped is techni­cally not a motorcycle, though for purposes of bridge tolls it technically is.

Some other hedges include in an important sense, to all in­tents and purposes, a regular..., a veritable ..., to the ex­tent that..., in certain respects, and many, many more. These various hedges allow us to place objects, events, and experiences under a wide variety of categories for various purposes, e.g., to draw practical distinctions in sensible ways, to provide new perspectives, and to make sense of apparently disparate phenomena.

5. Categories are open-ended. Metaphorical definitions can give us a handle on things and experiences we have already categorized, or they may lead to a recategorization. For example, viewing love as war may make sense of certain experiences that you took as love experiences of some kind or other but that you could not fit together in any meaningful way. The love is war metaphor may also lead you to categorize certain experiences as love experiences that you had previously not viewed as such. Hedges also reveal the open-ended nature of our categories; that is, an object may often be seen as being in a category or not, depending on our purposes in classifying it. Though categories are open-ended, categorization is not random, since both metaphors and hedges define (or redefine) categories in systematic ways.

p.125

Summary

We have argued that an account of how people understand their experiences requires a view of definition very different from the standard account. An experiential theory of def­inition has a different notion of what needs to be defined and what does the defining. On our account, individual con­cepts are not defined in an isolated fashion, but rather in terms of their roles in natural kinds of experiences. Con­cepts are not defined solely in terms of inherent properties;

instead, they are defined primarily in terms of interactional properties. Finally, definition is not a matter of giving some fixed set of necessary and sufficient conditions for the ap­plication of a concept (though this may be possible in cer­tain special cases, such as in science or other technical disciplines, though even there it is not always possible); instead, concepts are defined by prototypes and by types of relations to prototypes. Rather than being rigidly defined, concepts arising from our experience are open-ended. Meta­phors and hedges are systematic devices for further defining a concept and for changing its range of applicability.