COGNITIVE SEMANTICS: ARGUMENT IS WAR




For the scientist, truth is objective and absolute. The standard theory of truth is the "Correspondence Theory": a statement has an objective meaning, specifying the conditions under which it is true. Truth consists of a direct correspondence between a statement and some state of affairs in the world.

A theory of truth entails a theory of Meaning. Linguistics, a social science, seeks to apply the theoretical and methodological principles used by the natural sciences (with physics as the model). These also called empirical sciences depend on the formal sciences (mathematics and logic), which contribute to build the theory of truth already mentioned.

According to different theories of meaning, the objective meaning of a sentence is described as the conditions under which this is true or false. Semantics is the study of how linguistic expressions can fit the world directly, without the intervention of human understanding. Meanings are assigned to words without referring to particular contexts of use. The world is made up of building blocks, i.e., definable objects and clearly delineated inherent properties and relations.

The problem of meaning is directly related to the bi-directional influence between subject and object. Konrad Lorenz, the famous naturalist, considers this a crucial matter; he states that in order to avoid subjective criteria (prejudice and emotions) and to achieve universally valid judgments and evaluations we need to penetrate into the cognitive processes of the perceiving subject.

"The process of cognition and the properties of the object of knowledge will only be able to be investigated when they are undertaken simultaneously. The object of knowledge and the instrument of knowledge cannot legitimately be separated, but must be taken together as a whole".

Terms like 'metaphor' came to traditional grammar from the long-lasting discussion held by Classical Greek scholars on the subject of the origin of the rules which govern language. For some of them, language was a matter of convention, a tacit agreement among members of a community, reinforced through custom and tradition. For others, language has its origins, as Lyons puts it, "in eternal and immutable principles outside man himself," principles which are therefore inviolable. The relevant issue in this discussion is that of the nature of the meaning-form relation of words. For the naturalist school, the relation between a word and its meaning is a 'natural' one, in the sense that a word either resembles its referent (onomatopoeia) or is suggestive or imitative of some qualities or activities of things (sound-symbolism). There are however many more cases of words whose natural meaning is extended through a series of processes such as addition, deletion, substitution, and transposition of sounds, on the one hand, and metaphor, on the other. For the Greeks, metaphor was based on the discernible connection observed between the shape or function of the primary and secondary referents. In this way, naturalists explained "metaphorical extensions" such as the mouth of a river or the leg of a table. These early attempts on a theory of language meaning proved to be very useful for the developments of etymological studies. But, as Palmer remarks,

"It may seem obvious that foot is appropriate to mountains, or eye to needles, but a glance at other languages shows that it is not. In French the needle does not have an 'eye', and in many languages (e. g. the Ethiopian languages or some of those of North America) the mountain does not have a 'foot'. Moreover, in English eye is used with a variety of other meanings, e. g. the centre of a hurricane or a spring of water, which are not so obviously related semantically to the organ of sight, yet it is not used for the centre of a flower or an indentation, though these might seem intuitively to be reasonable candidates for the extension of the meaning".

Linguistic tradition has recognized, to a great extent, if not totally, the literary authority over the subject. Therefore, it is of no surprise that for most linguists the concept of metaphor differs only slightly from that of the literary tradition. It is generally described as an exceptional event among the rest of linguistic phenomena, as a kind of marginal meaning (Bloomfield). The first approach to metaphor linguists seemed to have had was that it is a case of semantic violation, a deviation from the 'normal', 'logical' meanings of utterances, what Leech describes as "the process whereby literal absurdity leads the mind to comprehension on a figurative plane". The distinction made was, in general, between a literal meaning and a metaphorical one. The metaphorical meaning was achieved through processes of non-logical meaning production accounted for in terms of ambiguity, paradox, contradiction, etc. The production and understanding of metaphors in these terms depend on the creative aspect of language, which differs from Chomsky's competence in that it refers to voluntary, conscious creativeness, i.e., the one required in poetical works. Traditional and structuralist linguistic trends have, at best, developed a way to make apparent the kind of 'anomaly' present in metaphors, whatever this anomaly might be.

Other authors consider metaphor a process more useful for historical linguistics. To illustrate this approach, let us take Robins' definition of metaphor, who conceives of it as a means whereby the meaning of a word changes through time.

"A very extensive type of semantic widening consists in metaphorical uses, wherein on the basis of some similarity of the meanings a word is used in different sorts of contexts and in reference to different sorts of features, usually of a more abstract nature, than once was the case. Metaphorical extension of meaning... is too well known to require exemplification".

Similar examples of the little attention paid to metaphor is seen in language comprehensive works, like those of Quirk et al. and Palmer, who simply relate it to figurative, non-literal, and transferred meaning. Greimas groups images, symbols, and metaphorical definitions as 'figures' without a practical function in logical language, and therefore only interesting to literary studies. Figurative meanings replace literal meanings as 'bricolage' only to serve to 'something else', which is the poetic communication itself.

An early understanding of metaphor as a crucial aspect of language function and thought can be found in Ogden and Richards' Symbolism, i.e., the study of the influence of language upon thought. They state that language meaning operates within a universe of discourse, "a collection of occasions on which we communicate by means of symbols."

"When we say... "The sun is trying to come out," or "The mountain rises," we may clearly be making no different references than if we were to give a scientific description of the situation, but we may mean these assertions to be taken 'literally'... interpreting our symbols... as names used with a reference fixed by a given universe of discourse".

When we do not have a symbol at hand we can choose one whose referent is analogous to our referent and transfer this symbol. Then if the speaker fails to see that such symbol is metaphorical, i.e., if he takes it literally, "falsity arises, namely the correct symbolization of a false reference by which the interpreter could be misled". Furthermore, whenever a term is taken outside the universe of discourse for which it has been defined, it "becomes a metaphor, and may be in need of fresh definition".

"...reference... cannot be formed simply and directly by one grouping of experience, but it is the result of varied groupings of experiences whose very difference enables their common elements to survive in isolation. This process of selection and elimination is always at work in the acquisition of a vocabulary and the development of thought. It is rare for words to be formed into contexts with non-symbolic experience directly, for as a rule they are learnt only through other words. We early begin to use language in order to learn language, but since it is no mere matter of the acquisition of synonyms or alternative locutions, the same stressing of similarities between references and elimination of their differences through conflict is required. By these means we develop references of greater and greater abstractness, and metaphor, the primitive symbolization of abstraction, becomes possible. Metaphor, in the most general sense, is the use of one reference to a group of things between which a given relation holds, for the purpose of facilitating the discrimination of an analogous relation in another group".

The use of metaphor, according to these authors, involves the same kind of contexts as abstract thought; the important point is that the members of "a group of things" will only possess the relevant feature in common, and that irrelevant or accidental features will cancel one another.

"The metaphorical aspects of the greater part of language, and the ease with which any word may be used metaphorically, further indicate the degree to which... words have gained contexts through other words".

Here what matters is the reference process itself and its discursive purpose; the structure is irrelevant. As can be seen, as far back as 1923, Ogden and Richards stressed the basic relationship among thought, language, and metaphor; an approach that has only recently been taken into account.

Within generative semantics, there is a large amount of different approaches to metaphor in relation to a semantically-based linguistic analysis of anomalous sentences and their contribution to the elucidation of semantic selectional features (Galmiche). An extreme example of this wide range of conceptions is that of Thorne: he tries to explain the existence of 'ungrammatical', 'deviant' sentences in poetry by postulating the poet's creation of "a new language (or dialect)" in which the semantic selectional restrictions are different from those in the normal language. However, he never takes the concept of metaphor, nor even that of connotative meaning, into account.

Lakoff proposes a theory which analyses sentences in terms of their logical structure, surface structure,  context, and conveyed meanings. According to this author, the "literal meaning" of a sentence is a function of its logical structure and its derivation into surface structure. The conveyed meaning of a sentence is explained in terms of the relevant aspects of context. Consequently, two different lines of investigation can be adopted in  semantics, one dealing with language structure and the other, with language use, that is, with Pragmatics. Nevertheless,

"Most grammars have been concerned exclusively with literal meaning, and this kind of analysis is necessary before a meaningful study of pragmatics can begin. ... in comparing a system like generative semantics to that of case grammar, it is the analysis of literal meanings that must be compared, not the conveyed meanings...pragmatics left aside". (Cook)  

A sociolinguist approach to metaphor is that of Grice, who suggests a general CO-OPERATIVE PRINCIPLE between speaker and hearer, including the 'maxim' of quality: "Try to make your contribution one that is true." Grice points out that this maxim "...is flouted by irony and metaphor (You're the cream in my coffee) -and the hearer has to work out what it is that the speaker is trying to convey."

A pragmatic approach to metaphor is that of Searle, who postulates that the phenomenon of metaphor is related to the difference between sentence/ utterance meaning and speaker's meaning. In a metaphorical expression, both the literal (sentence) and the metaphorical (speaker's) meanings are present, the former being the explicit message and the latter what the hearer must comprehend through searching something the speaker utters that is not expressed when the sentence is interpreted literally. What is uttered is an attribution of some features of the second term to the first one, and both components of metaphors are related so that both must be taken into account if utterance meaning is to be determined. 'X' restricts the range of meaning that can be recognized in the metaphorical sentence and in the speaker's utterance. The hearer or reader understands the metaphor's utterance meaning (which makes sense of the original sentence, or the metaphor) attending to these restrictions. In Searle's own terms, "the combination of S and P (subject terms and predicate terms of a metaphorical sentence) creates new R's (the predicate of the utterance meaning). We have a  specific set of associations with the 'P' terms, but different 'S' terms restrict the values of R differently". To illustrate this, he uses two metaphors:

Kant's second argument for the trascendental deduction is so much mud/gravel/sand paper.

Sam's voice is gravel.

Both metaphorical meanings are different because of the difference between Kant's argument and Sam's voice. The predicate gravel has different meanings in each sentence. We subscribe Searle's conception in the following respect: there is a selection of some features from all the possible features lying in the connotative depth. But the idea that the selection is based on the type of subject that fills X position seems to be rather structural. The meaning to be conveyed by the speaker depends on what the speaker himself wants to convey, together with the context in which the metaphorical utterance is performed. The hearer must therefore make use of a complex set of clues to decode the underlying meaning, including the specification of the subject defined (X), knowledge of the world shared by both participants, the personal information the hearer has about the speaker, the immediate context of situation, seriousness on the part of the speaker (who may be being ironical, e.g., saying Lucy is an angel when both participants know that she is not good or compassionate at all; in this sense, seriousness can be included in the shared knowledge or even in the immediate context, so that there seems to be considerable overlapping), and many others. Searle says that in cases like Sally is a block of ice we just perceive a connection which is the basis of the interpretation; these connections, according to Lakoff, coincide with the system of Conceptual Metaphors, a central part of synchronic linguistics. Nevertheless, in his account for literal meaning, Searle subscribes "the usual false assumptions that accompany that term." Everyday, conventional language is literal and not metaphorical in its nature, therefore metaphor is in the domain of principles of language use.

"...we can see why most philosophers of language have the range of views on metaphor that they have: They accept the traditional literal-figurative distinction. They... say that there is no metaphorical meaning, and most metaphorical utterances are either trivially true or trivially false. Or... they will assume that metaphor is in the realm of pragmatics, that is, that a metaphorical meaning is no more than the literal meaning of some other sentence which can be arrived at by some pragmatic principle. This is required, since the only real meaning for them is literal meaning, and pragmatic principles are those principles that allow one to say one thing (with a literal meaning) and mean something else (with a different, but nonetheless literal, meaning)".  

The term connotation is crucially important to delimit the approach to metaphor in semantics. Connotative content is distinguished from denotative content (also labelled, dictionary or central meaning by different authors), which overlaps with the notion of reference, a criterion for the correct use of a word provided the properties (semantic features) which define that word. Connotative meaning is the communicative value an expression has by virtue of what it refers to, over and above its purely conceptual content. This is why some authors considers connotation as a part of pragmatics (Baldinger). There is a multitude of additional non-criterial properties that we have learned to expect a referent of a word to possess, beside its 'neutral' denotation. Connotative meaning has been viewed as more attached to reality itself than to language proper, a peripheral phenomenon depending on the context of expression and, for that reason, unstable, depending on the particular cultural and historical period, and on the experience of an individual speaker. While the denotative, central meaning of an expression has the form of a finite set of discrete features of meaning, connotation is open ended, infinite as our knowledge and beliefs about the universe. An interesting example of this distinction is the one given by Leech when he lists the denotative features related to the word woman:
           
[+ human]
            [-  male]
            [+ adult]

Leech includes next physical, psychological, social characteristics, "typical concomitants of womanhood", and putative features:
           
            [+ biped]
            [+ having a womb]
            [+ gregarious]
            [+ subject to maternal instinct]
            [+ capable of speech]
            [+ experienced in cooking]
            [+ skirt or dress-wearing]

            [+ prone to tears]
            [+ frail]
            [+ cowardly]
            [+ emotional]
            [+ irrational]
            [+ inconstant]
            [+ gentle]
            [+ compassionate]
            [+ sensitive]
            [+ hard-working]
            [- trouser-wearing]

It is evident that such a list of constituents may be infinite and that they are highly dependant on cultural biases, as well as on sex, age, personal opinions, preferences, and so on. Whenever a sentence or a word is uttered, it maps both a connotative and a denotative meaning, since every expression of language is uttered by a given person in a given context. metaphors are forms of expressing this connotative meaning. The connection between the terms in a metaphorical relation lies in connotation. The denotative meaning is left aside while some of the numerous connotative meanings are selected as underlying semantic features of a word, e.g., Lucy is an angel. The denotations already stated for woman do not fit the ones we can attribute to the notion of angels (this is why linguists in general treat metaphors as semantic violations). But from all the connotations of woman only a few are mapped onto angel, which in turn has equally numerous connotations of its own. We reckon that woman's selected connotative features such as gentle, compassionate, sensitive, are mapped onto a group of some connotative features of angel,  such as generous, incapable of evil, suffering, and so on. The use of a metaphorical expression triggers an automatic, arbitrary, and totally new selection of underlying connotative meanings, which remain implicit in what we may call the underlying structure of denotative meanings, until they are mapped to the surface by a metaphor. Needless to say, when we utter Lucy is an angel we are by no means saying that she is not a human being, or that she is a religious entity, or 'non- specified' in terms of sex and age; what we are in fact expressing is a kind of 'non-neutral' definition of Lucy. This metaphor is, in fact, part of everyday language (a dead metaphor); it functions as an 'economical' device, one through which we can mean a great deal of information by using only one word. The indeterminancy of the semantic features the metaphor conflates (Lucy is a very good, generous, compassionate, and suffering woman, incapable of any evil act) can be viewed as a sort of implicit contract of meanings between the speaker and his/her interlocutor(s). The specification of such features is always limited; the metaphorical expression could be telling us more, or perhaps less, about Lucy than a 'literal' one. This issue will be dealt with in the next chapter. The amount of information to be included in a metaphorical expression can only be solved by the speakers of the language according to the conceptual world they share and the concrete situation they are in. Furthermore, in many cases the hearer has to decode from a complex of clues that which corresponds to a specific connotation selected by the speaker according to what he/she intends to express through a metaphor. When someone chooses the word angel instead of saint, he/she has searched for the most suitable word, in a particular situation, occasion, and time, and has undertaken this complex task automatically, almost as if he/she were using this word in its "neutral" meaning. As Widdowson puts it, "the effect of metaphor depends on avoiding the resolution of ambiguity: the user keeps two meanings in his mind at the same time." We can argue that an expression such as Angel is an available container of three levels of meaning that a speaker can make use of; according to Harman these levels are: thoughts (denotation/ connotation), communication of thoughts, and speech acts. We consider speech acts as constituents of the second level since they depend on a speaker's particular choice or intention, and the third level corresponds to the context of culture and  situation, in Malinowsky's terms. Suppose speaker A and speaker B and Lucy are all people living in the United States in 1993 and that they are acquaintances. Speaker A has a low opinion of Lucy and he says "Lucy is an angel". The actual meaning of Angel must be decoded along the following clues:



Obviously, the picture here proposed can be much further extended, including, at the first level, all the types of connotative meaning; at the second level, the diversity of speech acts that can be performed; at third level, nationality, social class, age, etc.

At this point, explicit metaphor comes into view as not being at all a peripheral phenomenon in language, but instead a very extended one in terms of use within the linguistic universe. The issue of conceptual (that is, implicit, not linguistic) metaphorical meaning in everyday language will be analysed in the next chapter.




Cognitive semantics, as a theoretical paradigm, claims that lexical concepts must be studied against the background of the human cognitive capacities at large. In other words, it holds that if language is one of the fundamental cognitive tools of man, it should not be studied autonomously, as if it contained a semantic structure that is independent of the broader cognitive organization of the human mind (which is the basic structuralist tenet). By contrast, cognitive semantics holds that there is no specifically linguistic-semantic organization of knowledge, separate from conceptual memory. Consequently, lexical semantic research should be conducted in close collaboration with other sciences that study the human mind and its working principles in order to find out how man comprehends, stores and retains information, and expresses human experience, be this specific to the individual or to his culture. This implies, for example, that attention should be paid to cultural differences in the metaphorical patterning of experience, as do a number of authors, in an effort to individualize semantic universals operating as general strategies for coping with part of human experience.

Gallagher, in the context of formal thought research, emphasizes the importance of the system of correspondences in Piaget's theory:

"Simple examples of correspondences (or morphisms in mathematics) are the places set at table for each invited guest or a test mark for each student. The relationship formed is called a mapping. In the familiar one-to-one correspondence -that is, isomorphism- each element of the first or original set has one, and only one, corresponding element from the second or image set".

Analogy and metaphor are, therefore, very important devices for structuring knowledge; they are also crucial in scientific reasoning because they provide richness and wider scope to arguments and descriptions, and supply a powerful and creative framework for scientific exploration, one that is "not possible with ordinary discourse or with propositional reasoning." According to Ortony,

"Metaphors are necessary as a communicative device because they allow the transfer of coherent chunks of characteristics -perceptual, cognitive, emotional and experiential- from a vehicle which is known to a topic which is less so. In so doing they circumvent the problem of specifying one by one each of the often unnameable and innumerable characteristics; they avoid discretizing the perceived continuity of experience and are thus closer to experience and consequently more vivid and memorable".
           
Cohen considers a metaphor as a mapping of the elements of one set on those of another.

A map is a system, and their elements are in correspondence with the mapped system;  this correspondence may take a large number of forms: "a map must be isomorphic with the mapped system with respect to some of its features." Gallagher concludes that metaphors are much more than the simplistic form A is B, and that they are complex comparisons, involving tension elimination: a shift from central meaning to marginal meaning.



THE THEORY OF CONCEPTUAL METAPHOR

The theory of Conceptual Metaphor, as a cognitive approach, is basically concerned with understanding. According to Lakoff and Johnson, the traditional scientific views cannot cope with human understanding in the long run: it emphasizes the fact that there are real things, existing independently of us, which constrain both how we interact with them and how we comprehend them. It focuses on truth and factual knowledge due to the importance they have for our successful functioning in our physical and cultural environments. It is well known that the eighteenth-century scientific world conception is an extension of common sense: reliance on the senses and empirical proof through technological advances.

According to Lakoff and Johnson, we all think and act according to what we assume to be true. “Absolute truth" cannot be achieved. This claim is based on the principle that the acquisition and use of truth depends on our understanding of the world we live in. Human understanding can be characterized in terms of categories emerging directly from our experiences as human beings. This categorization of the world is defined as follows:

... a natural way of identifying a "kind" of object or experience by highlighting certain properties, downplaying others, and hiding still others.

This is the origin of categories such as OBJECT, SUBSTANCE, orientational categories (IN-OUT, UP-DOWN, etc.) and many others according to which we define objects and situations. The properties highlighted, downplayed or hidden in the process of categorization are to be found within natural dimensions which make up gestalts in terms of which objects and situations are categorized. These natural dimensions range from perceptual and functional to purposive and causative ones. Early approaches to these notions were proposed by Lennenberg and Lorenz when trying to give an account of the natural and bodily dependence of human behaviour and understanding.

Categories are built according to a sequence that goes from CONCRETE to ABSTRACT (NON-CONCRETE), CLOSE to APART, PHYSICAL to NON-PHYSICAL, DEFINED to NON-DEFINED, etc. In other words, we primarily comprehend and know the world through this process, and thus we play "tricks" in conceptualizing whenever a matter of knowledge is abstract or undefined; we take another concept or object to concretize, come close, or define the former. Lakoff and Johnson examine the basic concept of Causation, taking into account the building-block theory, the Piagetian theory of manipulation, and the theory of prototypes. The concept of causation is based on the prototype of DIRECT MANIPULATION, which develops since we are children. The prototypical core of a concept is not "primitive", i.e., it is not an unanalizable semantic schema, but rather a group of elements perceived as a unity prior to its components, namely, a gestalt which consists of properties that naturally occur together in our daily experience, when we perform direct manipulations.

A good example of the way we categorize is that shown by a statement like I've invited a sexy blonde to our party. In this description, attention is focused on only a few dimensions and properties of the person being referred to, due to the fact that the purposes for which the expression is required have determined this focus of attention; had these purposes been different, the properties highlighted would certainly have been others (for instance,
the colour of her eyes).

Properties are not inherent to objects or situations but are rather the product of human interaction with the world around, and it is this interactional nature of properties which constitutes the basis for our notion of what is true and what is not: something will be true for us if it "makes sense only relative to human functioning." Thus, a true statement will involve the choice of categories of description, which in turn involves both our perceptions and purposes in a given situation; besides, it will leave out what has been downplayed by the categories used. That is why statements like I've invited a sexy blonde to our party and, let us say, I've invited a Marxist to our party can both be true about the same person referred to.

All this accounts pretty well for statements easily correlated with our daily experience, like I live in Chile, but what about statements which do not so clearly fit our knowledge of the world, for instance, the fog is in front of the mountain? The answer involves the concept of "projection": when our basic categories do not fit our reality, we project them onto objects and situations, and thus assign orientation to what does not have it inherently (a mountain) or entity structure to what is not clearly delineated as an entity (a mountain, the fog), etc. This is how we manage to understand everything in terms of a number of basic categories of understanding. This concept of projection is essential to see how we can understand some things in terms of others, that is, how we conceptualize metaphorically.

Although proposed by Lakoff and Johnson - respectively, a well known linguist (who has contributed to cognitive semantics since its earlier developments), and a philosopher of science- the theory of Conceptual Metaphor can be traced to Michael Reddy's paper "The Conduit Metaphor", where he analyses the way everyday English person conceptualizes the concept of communication: his contention is that we do not literally "get thoughts across" when we talk; rather, this suggests that communication transfers thought processes somehow bodily. Actually, no one receives anyone else's thoughts directly in their minds when they are using language. A speaker's feelings can be perceived directly only by him/ her; they do not really "come through to us" when he/she talks. Nor can anyone literally "give you an idea" -since these are inherently internal processes.

...[some examples] seem to involve the figurative assertion that language transfers human thoughts and feelings. ...if language transfers thoughts to others, then the logical container, or conveyer, for this thought is words, or word-groupings like phrases, sentences, paragraphs, and so on.

The logic... we are considering... called the conduit metaphor... would now lead us to the bizarre assertion that words have "insides" and "outsides." After all, if thoughts can be inserted, there must be a space "inside" wherein the meaning can reside. ..."content" is a term used almost synonymously with "ideas" and "meaning". ...recollection is quite meaning-full (sic). ...Numerous expressions make it clear that English does view words as containing  or failing to contain thoughts...

...of the entire metalingual apparatus of the English language, at least seventy percent is directly, visibly, and graphically based on the conduit metaphor.

Reddy uses this theory and a considerable amount of data to demonstrate that everyday language is largely metaphorical; that the locus of metaphor is thought, not language; that metaphor is a major and indispensable part of our ordinary, conventional way of conceptualizing the world; and that our everyday behaviour reflects our metaphorical understanding of experience. In Lakoff's words, Reddy "gave us a tiny glimpse of an enormous system of conceptual metaphor. Since its appearance, an entire branch of linguistics and cognitive science has developed to study systems of metaphorical thought that we use to reason, that we base our actions on, and that underlie a great deal of the structure of language."

The theory of conceptual metaphor used as theoretical framework for this study is the one proposed by Lakoff and Johnson in Metaphors We Live By. In essence, this theory holds that our conceptual system, and therefore our language, is largely metaphorical in nature. A large section of their book is devoted to demonstrate what it means for a concept to be metaphorical and the pervasiness this phenomenon has in language as a whole.

The most important claim we have made so far is that metaphor is not just a matter of language, that is, of mere words. We shall argue that, on the contrary, human thought processes are largely metaphorical. This is what we mean when we say that the human conceptual system is metaphorically structured and defined. Metaphors as linguistic expressions are possible precisely because  there are metaphors in a person's conceptual system. Therefore, whenever... we speak of metaphors... it should be understood that metaphor means metaphorical concept.

A Conceptual Metaphor (CM) is a mental phenomenon, the inner, unconscious, cognitive conceptualization of a domain of experience: entities in the world, actions, states, people, living beings, etc., perceived in terms of another domain of experience of an entirely different type. This conceptualization, applied to almost every domain, takes the form of an identification of the type X is Y, similar to the conventional metaphor in poetry, but it has no direct expression in the language; it is realized through instantiations, utterances which are often taken to be non-metaphorical, since they do not take the X is Y form, but are produced by concepts which are metaphorically built.

The concepts and thoughts -our conceptual system- according to which we function in our lives (of which we are, needless to say, unaware of), must be viewed in their relation to our actions and attitudes, as reflected in our daily activities.

We have found... that metaphor is pervasive in everyday life, not just in language but in thought and action. Our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature. ... Our conceptual system thus plays a central role in defining our everyday realities. If we are right in suggesting that our conceptual system is largely metaphorical, then the way we think, what we experience, and what we do every day is very much a matter of metaphor.

This unified way of conceptualizing a domain of experience metaphorically is realized in many linguistic expressions. For example, the way we conceptualize an argument is almost completely structured by the CM ARGUMENT IS WAR; in other words, our experiences of arguments depend basically (and perhaps exclusively) on the CM just mentioned. Briefly, this means that the way in which we conceive an act of war is partially used to understand (and therefore to act towards) arguments; we think of and experience arguments as if they were, in most aspects, armed conflicts. Language is a reflection of how our minds work in understanding things, and therefore it is not surprising that the evidence validating this proposal consists of hundreds of linguistic expressions which demonstrate the way we conceptualize metaphorically pieces of language that we try to understand or assume to be literal (or, in ontological terms, true), though no speaker of English may say that an argument is a war is a common saying, because this CM is not actually part of the language. There is plenty of examples of colloquial, everyday  linguistic expressions, used commonly throughout the world, which show a systematic way of talking about an argument as if it were a war situation:

Your claims are indefensible.
He attacked every weak point in my argument.
His criticisms were right on target.
I demolished his argument.
I've never won an argument with him.
You disagree? Okay, shoot!
If you use that strategy, he'll wipe you out.
He shot down all of my arguments.

The authors describe the metaphorical process as a mapping (in the mathematical sense) from a source domain (in this case, war) to a target domain (in this case, argument). This mapping is tightly structured, that is, there is a systematical ontological correspondence between the entities in the domain of argument (the participants, their claims, their ways of discussing) and the domain of war (the parties, their weapons, their strategies).

The generalizations governing... metaphorical expressions are not in language, but in thought: they are general mappings across conceptual domains. ...the locus  of metaphor is not in language at all, but in the way we conceptualize one mental domain in terms of another. The general theory of metaphor is given by characterizing such cross-domain mappings. And in the process, everyday abstract concepts like time, states, change, causation, and purpose also turn to be metaphorical. ...metaphor (that is, cross-domain mapping) is absolutely central to ordinary natural language semantics, and... the study of literary metaphor is an extension of the study of everyday metaphor... characterized by a huge system of thousands of cross-domains mappings.

Lakoff and Johnson adopted a mnemonic strategy for naming these mappings, e.g., TARGET-DOMAIN IS SOURCE DOMAIN. In this case, the name of the mapping is ARGUMENT IS WAR. When we speak of the ARGUMENT IS WAR metaphor, we are using a mnemonics for the set of ontological correspondences that characterize a mapping as THE ARGUMENT-IS-WAR mapping.

TARGET-DOMAIN              SOURCE-DOMAIN
                                                         
      ARGUMENT       IS           WAR

Two aspects must be borne in mind concerning metaphorical conceptualizations: they are partial and they are systematic. We have said that metaphorical concepts are characterized by the structuring of one concept in terms of another. This does not mean that we conceive of the elements involved in a metaphorical relation as being the same, but rather as being characterized by the same categories. What is essential  is that only some of the aspects structuring a concept are used to structure another. In the CM TIME IS MONEY, only the aspects of the concept of MONEY which define it as a unit of measure and as a quantifiable limited resource are taken into account (that is, highlighted); its other characteristics, such as "used to buy things", are hidden from the focus of our attention. Thus, metaphorical conceptual structuring is necessarily partial.

Metaphorical concepts are systematic: they conform a coherent system of metaphorical expressions, not random or isolated cases. We can perceive a certain pattern in our way of talking about arguments, time, and many other subject-matters; that is, we refer to them in some ways and not in others. The interesting fact is that in the metaphorical relation we can always distinguish between an element belonging to a well-defined domain of our experience which is used to refer to another concept from a less clearly defined conceptual domain. For instance, in the CM TIME IS MONEY, the vocabulary taken from the conceptualizations we have of money reveals that it is not clear to us the type of "thing" time is, and that in order to think and talk about it, we need to borrow the concepts and the corresponding vocabulary from another "thing", in this case, money. This process does not occur in isolated or random instances, but rather in systematic and coherent fashions of discourse.

So far we have been looking at CMs of a single kind, namely, Structural Metaphors. There are at least two other types of metaphorical conceptualizations, the ones called Orientational and Ontological Metaphors. The Orientational Metaphor implies the global organization of a conceptual system. Its name derives from the fact that most of them are related to concepts defining spatial categories, such as HAPPY IS UP or MORE IS UP. Ontological Metaphors are those whose bases are found in our experience with physical objects (especially our own bodies) and with objects and substances existing in our environments. Thus, in these cases, we will find projections of entity structure to things and events, or the categorization of non-substance elements as being substances. In effect, Lakoff and Johnson state that events and actions are conceptualized as objects, activities as substances, and states as containers. The most obvious case of Ontological Metaphor is that known as Personification, in which the object defined is specified as if it were a human entity. By viewing objects and experiences as human beings, we can give the former the numerous characteristics attached to the latter. The function or purpose we aim at when we use a concept is the relevant issue at this point: by using such a conceptual and linguistic device we can express a greater deal of information about the topic or about our inner experiences in relation to it. The CM INFLATION IS AN ENEMY, for instance, allows us to say we must attack inflation, we are defending ourselves from inflation, as if inflation were a human enemy involved in a war situation, such as a battle. We must distinguish this case of Personification from those of Metonymy (Met) or reference of one concept to another related to it, as when we say, for example, Mrs. Grundy frowns on blue jeans, in which blue jeans stands for people wearing blue jeans. We must, anyway, recognize that, the same as metaphors, metonymies are used to have a better understanding of objects and experiences that lack a clearly delimited or easily understandable conceptualization. Metonymies are also partially structured and present systematicities in their functioning as linguistic expressions, and they are very active in the cultural sense.

The Conceptual Metaphor approach also attempts to examine the nature of "folk theories," models of some aspects of reality that are most often taken as constituting "common sense". Folk theories, according to these authors, involve conceptual metaphors from which a chain of deduction emerges, e.g., from the CM SEXUALITY IS A PHYSICAL FORCE, a speaker may assume, as these authors put it, that

physical appearance is a physical force> a woman is responsible for her physical appearance...sexual emotions are part of human nature...sexual emotions are a response to being acted upon by a sexual force> a person who uses a force is responsible for the effect of that force> a woman with a sexy appearance is responsible for arousing a man's sexual emotions...sexual emotion naturally results in sexual actions> sexual action against someone's will is unacceptable> to act morally, we must avoid sexual action> avoiding sexual action requires inhibiting sexual emotions> to act morally, one must inhibit sexual emotions...sexual emotions are part of human nature> to inhibit sexual emotions is to be less than human> a woman with a sexy appearance makes a man who is acting morally less than human> to be less than human is to be injured (we have inserted the symbol > to indicate the logical chain)

These deductions are not explicit. Speakers do not follow a conscious chain of deduction since deductions have a logic and a structure that remains unconscious behind the reality the speaker takes for granted. According to Lakoff and Johnson, folk theories and CMs are easy to understand "because they are deeply engrained in [American] culture...they are largely held by people." They state that if metaphors and folk theories are readily available to us for use in understanding, they are "ours" in some sense, like the one in the example above, and that no theory of communication or understanding can pretend to be adequate if it does not account for the crucial role that CMs have on folk theories.




The theory of conceptual metaphor has given origin to a number of studies. A few of them are very briefly accounted for in what follows.

Kronfeld focuses his discussion about metaphor on its very nature: are metaphors semantically special phenomena? Is their meaning different from literal meaning? As she explains, most approaches agree in that the meaning of a metaphor is not a function of the meaning of its constituents, but instead it is a "new" and completely different meaning; in order to discover this new meaning we need a "construal" from the hearer/reader, i.e., a mechanism of some nature to actively construct (produce) and deconstruct (understand) the meaning of a metaphor. Kronfeld distinguishes two opposite approaches in relation to the idea of a construal. The non-constructivist group claims that meaning is only literal meaning and that metaphors have no meaning at all or are reducible to literal paraphrases, rejecting the existence of construals. The constructivist approach postulates the existence of a construal mechanism for understanding metaphors, but it extends this mechanism to both figurative and literal language. Among supporters of the last view we have several post-structuralist critics, such as Derrida, De Man, and "experientialist" semanticists like Lakoff and Johnson, for whom the puzzle of metaphor exists but it is not exclusive of metaphor.

Carbonell attempts to augment the power of a semantic knowledge base used for language analysis by means of metaphorical mappings. His central concern is the creation of a process model to encompass metaphor comprehension.

Understanding the metaphors used in language often proves to be a crucial process in establishing complete and accurate interpretations of linguistic utterances.

Carbonell states that, although there is a set of general (conceptual) metaphors in English, the majority of them are instantiated versions of a few general metaphors, which he calls "common" metaphors, as opposed to those which are "creative". According to this idea, the problem of understanding a large class of metaphors may be reduced from a reconstruction to a recognition task. The identification of an instance of one of the general metaphorical mappings is a much more tractable process than reconstructing the conceptual framework from the bottom up each time a new metaphor instance is encountered.

Each of the general metaphors contains not only mappings of the form "X is used to mean Y in context Z," but inference rules to enrich the understanding process by taking advantage of the reasons why the writer [or speaker] may have chosen the particular metaphor (rather than a different metaphor or a literal rendition).

Carbonell's most important contribution is the Implicit-Intention Component: what the speaker is contending when he/ she chooses to use a metaphorical expression instead of a literal one. The hypothesis holds that this component is the information about what the metaphor conveys that is absent from a literal expression of the same concept; in other words, there is much more information and expression intended in a CM than in a literal expression.

The issue of determining the Implicit-Intention Component is what Gibbs, in his own terms, tries to arrive at through a series of experiments concerning the semantic content and the presence of conceptual metaphors in idioms. He postulates that idioms have complex figurative interpretations that are not arbitrarily determined but are motivated by independently existing CMs that provide the basis for a great part of everyday thought and, therefore, language; idioms cannot be simply defined, as do dictionaries, as equivalent in meaning to simple literal phrases.

...literal paraphrases of these idioms [blow your stack, flip your lid, hit the ceiling] such as "to get very angry" do not convey the same inferences about the causes, intentionality, and manner in which someone experiences and expresses his or her anger. ...[literal paraphrases] are not by themselves motivated by single conceptual metaphors and therefore do not possess the kind of complex interpretations as do idiomatic phrases.

Gibbs' inquiries show that idiom use and comprehension vary depending on how discourse encodes information. This process is partially motivated by entailments of CMs. This also holds for Carbonell's instantiations of a CM, even though he only analyses them in single sentences. Both positions can be bridged at this point.




Comments

Popular posts from this blog

COGNITIVE SEMANTICS: LOVE IS A JOURNEY (LAKOFF & JOHNSON)