p.126

                         

              How Metaphor Can Give Meaning to Form

We speak in linear order; in a sentence, we say some words earlier and others later. Since speaking is correlated with time and time is metaphorically conceptualized in terms of space, it is natural for us to conceptualize language metaphorically in terms of space. Our writing system re­inforces this conceptualization. Writing a sentence down allows us to conceptualize it even more readily as a spatial object with words in a linear order. Thus our spatial con­cepts naturally apply to linguistic expressions. We know which word occupies the first position in the sentence, whether two words are close to each other or far apart, whether a word is relatively long or short.

Because we conceptualize linguistic form in spatial terms, it is possible for certain spatial metaphors to apply directly to the form of a sentence, as we conceive of it spatially. This can provide automatic direct links between form and content, based on general metaphors in our con­ceptual system. Such links make the relationship between form and content anything but arbitrary, and some of the meaning of a sentence can be due to the precise form the sentence takes. Thus, as Dwight Bolinger (1977) has claimed, exact paraphrases are usually impossible because the so-called paraphrases are expressed in different forms. We can now offer an explanation for this:

---We spatialize linguistic form.

---Spatial metaphors apply to linguistic form as it is spatialized.

---Linguistic forms are themselves endowed with content by vir­tue of spatialization metaphors.

p.127

More of Form Is More of Content

For example, the conduit metaphor defines a spatial re­lationship between form and content: linguistic expres­sions are containers, and their meanings are the content of those containers. When we see actual containers that are small, we expect their contents to be small. When we see actual containers that are large, we normally expect their contents to be large. Applying this to the conduit metaphor, we get the expectation:

   MORE OF FORM IS MORE OF CONTENT.

As we shall see, this is a very general principle that seems to occur naturally throughout the world's languages. Though the conduit metaphor is widespread, we do not know yet whether it is universal. We would expect, how­ever, that some metaphorical spatialization of language would occur in every language and, whatever the details, it would not be surprising to find such correlations of amount.

An English example of more of form is more of con­tent is iteration:

He ran and ran and ran and ran.

which indicates more running than just

He ran.

Similarly,

He is very very very tall.

indicates that he is taller than

He is very tall.

does. Extended lengthening of a vowel can have the same effect. Saying

He is bi-i-i-i-ig!

indicates that he is bigger than you indicate when you say just

p.128

He is big.

Many languages of the world use the morphological device of reduplication, that is, the repetition of one or two sylla­bles of a word, or of the whole word, in this way. To our knowledge, all cases of reduplication in the languages of the world are instances where more of form stands for more of content. The most typical devices are:

Reduplication applied to noun turns singular to plural or col­lective.

Reduplication applied to verb indicates continuation or com­pletion.

Reduplication applied to adjective indicates intensification or increase.

Reduplication applied to a word for something small in­dicates diminution.

The generalization is as follows:

A noun stands for an object of a certain kind.

More of the noun stands for more objects of that kind.

A verb stands for an action.

   More of the verb stands for more of the action (perhaps until completion).

An adjective stands for a property.

More of the adjective stands for more of the property.

A word stands for something small.

More of the word stands for something smaller.

Closeness Is Strength of Effect

A much subtler example of the way metaphor gives mean­ing to form occurs in English (and possibly in other lan­guages as well, though detailed studies have not been done). English has the conventional metaphor

CLOSENESS IS STRENGTH OF EFFECT.

Thus, the sentence

p.129

Who are the men closest to Khomeini?

means

    Who are the men who have the strongest effect on Kho­meini?

Here the metaphor has a purely semantic effect. It has to do with the meaning of the word "close." However, the metaphor can also apply to the syntactic form of a sen­tence. The reason is that one of the things the syntax of the sentence indicates is how close two expressions are to each other. The closeness is one of form.

This metaphor can apply to the relation between form and meaning in the following way:

If the meaning of form A affects the meaning of form B, then, the closer form A is to form B, the stronger will be the effect of the meaning of A on the meaning of B.

For example, a sentential negative like not has the effect of negating a predicate, as in

John won't leave until tomorrow.

   The form n't has the effect of negation on the predicate with the form leave.

There is a rule in English, sometimes called negative trans­portation, which has the effect of placing the negative further away from the predicate it logically negates; for example,

Mary doesn’t think he'll leave until tomorrow.

Here n't logically negates leave rather than think. This sentence has roughly the same meaning as

Mary thinks he won'/ leave until tomorrow.

except that in the first sentence, where the negative is further away from leave, it has a weaker negative force. In the second sentence, where the negative is closer, the force of negation is stronger.

p.130

Karl Zimmer (personal communication) has observed that the same principle governs differences like

Harry is not happy

  versus

 Harry is unhappy.

The negative prefix un- is closer to the adjective happy than is the separate word not. The negative has a stronger effect in Harry is unhappy than in Harry is not happy. Unhappy means sad, while not happy is open to the interpretation of being neutralneither happy nor sad, but in between. This is typical of the difference between negatives and negative affixes, both in English and in other languages.

The same metaphor can be seen at work in the following examples:

I taught Greek to Harry.

 I taught Harry Greek.

In the second sentence, where taught and Harry are closer, there is more of a suggestion that Harry actually learned what was taught himthat is, that the teaching had an ef­fect on him. The following examples are even subtler:

I found that the chair was comfortable.

I found the chair comfortable.

The second sentence indicates that I found out that the chair was comfortable by direct experience---by sitting in it. The first sentence leaves open the possibility that I found it out indirectly---say, by asking people or taking a survey. In the second sentence, the form I is closer to the forms the chair and comfortable. The syntax of the sentence indicates the directness of the experience with the chair by which I found that the chair was comfortable. The closer the form I is to the forms the chair and comfortable, the more direct is the experience that is indicated. Here the effect of the syntax is to indicate the directness of the experience, and closeness indicates the strength of that effect. This

p.131

phenomenon in English is verified in detail by Borkin (in press). The same metaphor can be seen at work in examples like:

Sam killed Harry.

Sam caused Harry to die.

If the cause is a single event, as in the first sentence, the causation is more direct. The second sentence indicates indirect or remote causation---two separate events, Harry's death and what Sam did to cause it. If one wants to indicate causation that is even more indirect, one can say:

Sam brought it about that Harry died.

The effect that the syntax has in these sentences is to in­dicate how direct the causal link is between what Sam did and what happened to Harry. The principle at work is this:

The closer the form indicating causation is to the form indicating the effect, the stronger the causal link is.

In Sam killed Harry, there is only a single form---the word kill---to indicate both the causation and the effect (death). The forms for this meaning are as close as they can be: one word includes them both. This indicates that the causal link is as strong as it could be: a single event. In Sam caused Harry to die, there are two separate words---cause and dieindicating cause and effect. This indicates that the link between the cause and the effect is not as strong as it could bethe cause and the effect are not part of the same event. In Sam brought it about that Harry died, there are two separate clauses: Sam brought it about and that Harry died, which indicates a still weaker causal link.

In summary, in all of these cases a difference in form indicates a subtle difference in meaning. Just what the sub­tle differences are is given by the metaphor closeness is strength of effect, where closeness applies to ele­ments of the syntax of the sentence, while strength of effect applies to the meaning of the sentence. The close-

p.132

ness has to do with form, while the strength of effect has to do with meaning. Thus the metaphor closeness is strength of effect, which is part of our normal con­ceptual system, can work either in purely semantic terms, as in the sentence "Who are the men closest to Khomeini?," or it can link form to meaning, since closeness can indicate a relation holding between two forms in a sen­tence. The subtle shades of meaning that we see in the examples given above are thus the consequences not of special rules of English but of a metaphor that is in our conceptual system applying naturally to the form of the language.

The me-first Orientation

Cooper and Ross (1975) observe that our culture's view of what a prototypical member of our culture is like de­termines an orientation of concepts within our conceptual system. The canonical person forms a conceptual reference point, and an enormous number of concepts in our con­ceptual system are oriented with respect to whether or not they are similar to the properties of the prototypical person. Since people typically function in an upright position, see and move frontward, spend most of their time performing actions, and view themselves as being basically good, we have a basis in our experience for viewing ourselves as more up than down, more front than back, more active than passive, more good than bad. Since we are where we are and exist in the present, we conceive of ourselves as being here rather than there, and now rather than then. This determines what Cooper and Ross call the me-first orientation: up, front, active, good, here, and now are all oriented toward the canonical person; down, back­ward, passive, bad, there, and then are all oriented away from the canonical person.

This cultural orientation correlates with the fact that in

p.133

English certain orders of words are more normal than others:

More Normal up and down front and back active and passive good and bad here and there now and then

Less Normal down and up back and front passive and active bad and good there and here then and now

The general principle is: Relative to the properties of the prototypical person, the word whose meaning is nearest comes first.

This principle states a correlation between form and content. Like the other principles that we have seen so far, it is a consequence of a metaphor in our normal conceptual system: nearest is first. For example, suppose you are pointing out someone in a picture. If you say

         The first person on Bill's left is Sam.

you mean

The person who is on Bill's left and nearest to him is Sam.

To summarize: Since we speak in linear order, we con­stantly have to choose which words to put first. Given an otherwise random choice between up and down and down and up, we automatically choose up and down. Of the two concepts up and down, up is oriented nearest to the prototypical speaker. Since nearest is first is part of our conceptual system, we place the word whose meaning is nearest (namely, up) in first position. The word order up and down is thus more coherent with our conceptual system than the order down and up.

For a detailed account of this phenomenon and a discus­sion of apparent counterexamples, see Cooper and Ross (1975).

p.134

Metaphorical Coherence in Grammar

An Instrument Is a Companion

It is common for a child playing with a toy to act toward it as if it were a companion, talking to it, putting it on his pillow next to him at night, etc. Dolls are toys designed especially for this purpose. Behavior like this occurs in adults, who treat certain significant instruments like cars and guns as companions, giving them names, talking to them, etc. Likewise, in our conceptual system, there is the conventional metaphor an instrument is a companion, which is reflected in the following examples:

AN INSTRUMENT IS A COMPANION

Me and my old Chevy have seen a lot of the country to­gether.

Q: Who's gonna stop me?

A: Me and old Betsy here [said by the cowboy reaching for his gun].

Domenico is going on tour with his priceless Stradivarius.

Sleezo the Magician and his Magic Harmonica will be per­forming tonight at the Rialto.

Why With Indicates Both instrumentality

and ACCOMPANIMENT

The word with indicates accompaniment in English, as in:

I went to the movies with Sally. (companion)

The fact that it is with and not some other word that in­dicates accompaniment is an arbitrary convention of En­glish. In other languages, other words (or grammatical de­vices like case endings) indicate accompaniment (e.g., avec in French). But given the fact that with indicates ac­companiment in English, it is no accident that with also indicates instrumentality, as in:

I sliced the salami with a knife. (instrument)

p.135

The reason that this is not arbitrary is that our conceptual system is structured by the metaphor an instrument is a companion. It is a systematic, not an accidental, fact about English that the same word that indicates accompani­ment also indicates instrumentality. This grammatical fact about English is coherent with the conceptual system of English.

As it happens, this is not merely a fact about English. With few exceptions, the following principle holds in all the languages of the world:

The word or grammatical device that indicates accompani­ment also indicates instrumentality.

Since the experiences on which the metaphor an instru­ment is a companion are based are likely to be universal, it is natural that this grammatical principle holds in most languages. Those languages where the principle holds are coherent with the metaphor; those languages where the principle does not hold are not coherent with this metaphor. Where the instrument is a companion coherence does not appear in a language, it is common for some other con­ceptual coherence to appear in its place. Thus, there are languages where instrument is indicated by a form of the verb use or where accompaniment is indicated by the word for and. These are other, nonmetaphorical, ways in which form may be coherent with content.

The "Logic" of a Language

The use of the same word to indicate instrumentality as well as accompaniment makes sense. It makes such form-content links coherent with the conceptual system of the language. Similarly, the use of spatial words like in and at for time expressions (e.g., in an hour, at ten o'clock) makes sense given that time is metaphorically con­ceptualized in terms of space. Metaphors in the conceptual system indicate coherent and systematic relationships be­tween concepts. The use of the same words and grammati-

p.136

cal devices for concepts with systematic metaphorical cor­respondences (like time and space) is one of the ways in which the correspondences between form and meaning in a language are "logical" rather than arbitrary.

Conclusion

Subtle Variations in Meaning

Is paraphrase possible? Can two different sentences ever mean exactly the same thing? Dwight Bolinger has spent most of his career showing that this is virtually impossible and that almost any change in a sentencewhether a change in word order, vocabulary, intonation, or grammati­cal constructionwill alter the sentence's meaning, though often in a subtle way. We are now in a position to see why this should be so.

We conceptualize sentences metaphorically in spatial terms, with elements of linguistic form bearing spatial prop­erties (like length) and relations (like closeness). Therefore, the spatial metaphors inherent in our conceptual system (like closeness is strength of effect) will automatically structure relationships between form and content. While some aspects of the meaning of a sentence are conse­quences of certain relatively arbitrary conventions of the language, other aspects of meaning arise by virtue of our natural attempt to make what we say coherent with our conceptual system. This includes the form that we say things in, since that form is conceptualized in spatial terms.

Regularities of Linguistic Form

We have seen that metaphors play an important role in characterizing regularities of linguistic form. One such reg­ularity is the use of the same word to indicate both accom­paniment and instrumentality. This regularity is coherent with the conceptual metaphor instruments are compan­ions. Many of what we perceive as "natural" regularities

p.137

of linguistic form are regularities that are coherent with metaphors in our conceptual system. Take, for example, the fact that questions typically end in what we perceive as a "rising" intonation, while statements typically end in what we perceive as a "falling" intonation.

This is coherent with the orientational metaphor un­known is up; known is down. This conceptual metaphor can be seen in examples like:

That's still up in the air.

I'd like to raise some questions about that.

That settles the question.

It's still up for grabs.

Let's bring it up for discussion.

And the reason that the verb come is used in come up with an answer is that the answer is conceptualized as starting out down and ending where we are, namely, up.

Questions typically indicate what is unknown. The use of rising intonation in questions is therefore coherent with unknown is up. The use of falling intonations with state­ments is therefore coherent with known is down. In fact, questions with falling intonation are understood not as real questions but as rhetorical questions indicating statements. For example, "Will you ever learn?" said with falling in­tonation is a way of saying, indirectly, "You'll never learn." Similarly, statements with rising intonation indicate uncertainty or inability to make sense of something. For example, "Your name's Fred" said with rising intonation indicates that you're not sure and want confirmation. "The Giants traded Madlock" said with rising intonation in­dicates an inability to make sense of something---that it doesn't fit with what you know. These are all examples of the use of rising and falling intonation coherently with the unknown is up, known is down metaphor.

Incidentally, WH-questions in English have falling in­tonation, for example, "Who did John see yesterday?" Our guess as to the reason for this is that most of the content of

p.138

WH-questions is known, and only a single piece of in­formation is taken to be unknown. For instance, "Who did John see yesterday?" presupposes that John saw someone yesterday. As might be expected, tone languages generally do not use intonation to mark questions at all, usually making use of question particles. On the whole, where in­tonation signals the difference between questions and statements, rising intonation goes with the unknown (yes-no) questions and falling intonation with the known (statements).

Examples like this indicate that regularities of linguistic form cannot be explained in formal terms alone. Many such regularities make sense only when they are seen in terms of the application of conceptual metaphors to our spatial con­ceptualization of linguistic form. In other words, syntax is not independent of meaning, especially metaphorical as­pects of meaning. The "logic" of a language is based on the coherences between the spatialized form of the language and the conceptual system, especially the metaphorical as­pects of the conceptual system.