Linguists are impressed by what they call the "creative" aspect of language, a characteristic which they consider to exhibit itself every time people use language. To the teacher of rhetoric (or to other teachers, for that matter) this way of talking about commonplace everyday parlance often seems strange, and unjustified in the light of their (often) fruitless efforts to "improve" the speech and writing of their students. Let us examine in just what sense one can say that everyday language is creative.
One interpretation of the creative aspect of language is that it is constantly novel in the strict sense that is evident when comparing the sentences an individual utters and seeing that very few are exact repetitions of each other. The reason this fact is of interest can be shown by comparing the novelty inherent in sentence production with the strict variability that one observes in any complex sequence of motor activity. Thus, a minute analysis of a person's walk would no doubt reveal that the individual movements which are chosen as the unit of measurement constantly form new .sequences none of which are identical. But this kind of variability, and a host of others (such as the microscopic differences one can observe in different replications of the "same" objects), does not ordinarily make a difference. Only departures of a relatively gross sort, as in different types of gaits, indicate a meaningful difference. In discourse, however, when two sentences are different, they generally differ in meaning. Thus, the continual novelty of language is not an uninteresting fact.
There is another aspect to the creativity of language, one that is closer to the meaning of "creative writing" as teachers of rhetoric use this phrase. Here, the novelty involved refers to the manufacture of new metaphors and new juxtapositions of words in a manner that is in some way felicitous and pleasing to the hearer or reader. It is this kind of novelty whose presence often mystifies the .student in a poetry lesson and whose absence exasperates the composition teacher.
It is important to realize that the two aspects of creativity just outlined are consequences of the same inherent property of language, what I shall call its "indexical" feature. This characteristic operates at both the word level and the phrase or sentence level. Let us examine how this is true in both instances.
A moment's reflection would reveal that, contrary to popular belief, a word does not refer to a thing but to an infinite class of things that share certain essential properties Actually, because many or most words do not refer to concrete objects, it is more correct to say that words refer to concepts, a "concept" being defined as a set or class that has certain essential (or criterial) and non-essential attributes or properties. A dictionary definition is one way (not the only way) by which the attributes of a concept can be specified. The specification of the attributes of a concept is not a simple matter, and in fact, there are no known ways of generating such specifications in a completely adequate manner. The method that linguists use today is of an entirely different sort, known as "semantic feature" analysis, examples of this being the proposals of Katz and Fodor (1963) and of Weinreich (1966).
Thus, the illustration given for the concept "bachelor" by Katz and Fodor specifies the following features: 1. Human-Male-Never been married; 2. Human-Male Young-Knight; 3. Human -Who has the lowest academic degree; 4. Animal-Male-Young-Fur seal without mate during breeding season. Bolinger (1965), in a critique of this approach, has suggested some additional features, such as, for meaning 2 of "bachelor": Human-Male-Military-Hierarchic-Noble-Inferior-Dependent-Proximate-Young, and for meaning 4: Animal - Phocine - Hirsute - Male - Adult- Young - Nubile-Unmated. But even these additional features are incomplete (where are Solid, Organic, Pliable, etc.?) and, as Bolinger rightly argues, the list is potentially endless if it is to represent all that a speaker knows about the concept of "bachelor."
Similar difficulties beset the lexicographer's attempt to specify in full the attributes of a word. Thus, "table" can be defined as "An article of furniture with a flat horizontal top upheld by one or more supports" (let us, for the moment, omit its other meanings). But note that the class of objects falling in the category "furniture" cannot be specified: a large flat stone habitually used for placing food on it during meals could be a "table." And what if it is not quite flat? Or used only once? Or used to place flowers on it, even though one could eat meals at it? This is not a matter of legislating by authority or majority, since someone could use the word in such instances and it would then be less a question of correct reference than a recognition of how the user chooses to conceive of a particular object (cf. "I did not like the way the pantomimist came across when he pretended to sit down at the table to eat his imaginary meal.").
One might think, at first, that this kind of fuzziness of the meaning of words is a disadvantage. Quite the contrary is true. For it is this kind of indeterminacy which gives language such an unbounded power of expression, Putting it this way is not quite correct, for it is not the word that has the power of expression but the language user, Because words refer to concepts, and because concepts are indeterminate, people can use words as indexical devices to "home in" on ideas they wish to convey. But note that "indeterminate" does not mean "randomly variable," for in that case, communication would be impossible. Some constraints must operate. What those are cannot at the present time be specified, but the amount of constraint undoubtedly varies with such factors as the nature of the concept, the amount of information provided by the context (verbal and non-verbal), the inferential capacities o f the listener, and the degree of shared background experiences between the speaker and the listener.
Syntax is an efficient device for indexicality. Assigning words to syntactic categories (e.g., Noun, Verb, Adjective) are forms of constraint that specify indexical possibilities while the building of larger syntactic segments (e.g., noun phrase, verb phrase) and their various manipulations by means of transformations (e.g., nominalization, relativization, sentence embedding) provide means for expressing more complex conceptualizations (e.g., modification, intersection, sub-categorization, causation, modalization, etc.). The syntactic structure of a sentence is thus a reflection of the kind of conceptualization being expressed by the speaker 14 However, no one-to-one relation is to be expected between these two aspects, although limiting conditions probably obtain. For instance, simple conceptualizations can be expressed by means of complex syntactic structures, but there are limits on the degree of complexity of conceptualization that can be expressed by means of simple syntactic structures. When children first begin to speak, their conceptual (semantic) capacities are superior to their syntactic know-how (see Slobin, 1969), and it is doubtful that syntactic development could proceed without the prior conceptualizations that are developed through perceptual interaction with the environment (e.g., visual constancy, physical gradients of similarity, causation in the form of temporal contiguity, etc. See Osgood, in press.).
Once these minimal conceptualizations are mastered (during the first 18 months of life), they are sufflcient15 to insure syntactic development which then proceeds at a fast rate, quickly outstripping the child's conceptual development. He now has a very powerful device that is suitable for expressing far more complex reasoning than he is capable of (which may account for many-of the "cute" but nonsensical things a small child says). The availability of a powerful syntax can foster intellectual development in two ways. First, by stimulating conceptual analogues of syntactic operations: for example, relativization suggests differentiation ("The red ball which Daddy gave you"), quantifiers suggest delimitation ("some milk"), conjunctions suggest logical relations (e.g., causation: "You can't go out because it's raining"), and so on. Second, as the child's intellectual capacities grow, syntactic transformations provide a vehicle for the realization of inferential reasoning and continued explorations of more complex conceptualizations for which he can receive social support and help from others.16
The recognition of the indexical nature of language use, and its investigation, pave the way for the understanding of individual differences in communicative competence, a theme to be explored in the last two sections below. The creativity aspect of language can thus be seen to reside not only in the fact that the sentences people produce are characteristically novel (in the interesting sense of meaningful and intended variability), but also in the fact that linguistic utterances do not stand in a one-to-one relation to the conceptualizations that give rise to them. The speaker's creativity consists in the manufacture of syntactic constructions that facilitate, through their indexicality, just those inferential processes the hearer must use to recover the intended conceptualizations: through the use of analogies, converging ambiguities, the drawing of parallels, restatement and paraphrasing, and so on. The hearer's creativity resides in the quality of the guessing game he is called upon to perform as he analyzes the elliptical speech of the speaker and attempts to recover his intended conceptualizations: through inferential reasoning, through flexibility in interpreting figurative speech, through a wait-and-see game whereby he must be willing to tolerate ambiguity and suspend final conclusions at earlier stages of the speaker's utterances until later relevant information is provided, and so on. These processes are just as characteristic (commonplace, necessary) of ordinary speech as they are of "creative writing."
14. The fact that linguists choose terms reflecting specific types of cognitive processes to tag various transformation rules is an indication that they accept this relationship. Clearly, transformation theory is a theory of cognitive processes. See Chomsky, 1968.
15. The fact that such meager conceptual devices are sufficient to trigger and make possible the much more complex syntactic operations that follow in time is a strong indication of the presence of specific innate components governing syntactic development. (See Chomsky, 1965, Lenneberg, 1967, and McNeill, 1966).
16. The quality of the home and school environment are thus significant in both the conceptual and syntactic domains. In the latter case, exposure to and acquisition of certain complex operations (e.g., sentence embedding transformations) and their use for particular communicative purposes (e.g., discussion of abstract relations, description of feelings and interpersonal relationships) may provide the stimulus and the possibility for intellectual growth. See further discussion in subsequent sections.
Back to the Titlepage