The structuralist emphasis on language

The structuralist emphasis on language as “a system where everything holds together” is conceptually, at least, a direct outgrowth of the concern for organic form in Humboldtian linguistics.

For Humboldt, a language is not to be regarded as a mass of isolated phenomena – words, sounds, individual speech productions, etc. – but rather as an “organism” in which all parts are intercon- nected and the role of each element is determined by its relation to the gen- erative processes that constitute the underlying form. In modern linguistics, with its almost exclusive restriction of attention to inventories of elements and fixed “patterns” the scope of “organic form” is far more narrow than in the Humboldtian conception. But within this more narrow frame, the notion of “organic interconnection” was developed and applied to linguistic materials in a way that goes far beyond anything suggested in Humboldt. For modern struc- turalism, the dominant assumption is that “a phonological system [in particular] is not the mechanical sum of isolated phonemes, but an organic whole of which the phonemes are the members and of which the structure is subject to laws.”52 These further developments are familiar, and I will say nothing more about them here.

As noted above, the form of language, for Humboldt, embraces the rules of syntax and word formation as well as the sound system and the rules that determine the system of concepts that constitute the lexicon. He introduces a further distinction between the form of a language and what he calls its “character.” It seems to me that, as he employs this term, the character of a language is determined by the manner in which it is used, in particular, in poetry and philosophy; and the “inner character” (p. 208) of a language must be distinguished from its syntactic and semantic structure, which are matters of form, not use. “Without changing the language in its sounds, and still less in its

forms and laws, time, through a growing evolution of ideas, a heightened power of thought, and a more deeply penetrating capacity for feeling, will often bring into a language what it did not formerly possess” (p. 116; Humboldt 1999: 86–7). Thus a great writer or thinker can modify the character of the language and enrich its means of expression without affecting its grammatical structure. The character of a language is closely related to other elements of the national character and is a highly individual creation. For Humboldt, as for his Cartesian and romantic precursors, the normal use of language typically involves creative mental acts; but it is the character of a language rather than its form that reflects true “creativity” in a higher sense – in the sense that implies value as well as novelty. For all his concern with the creative aspect of language use and with form as generative process, Humboldt does not go on to face the substantive question: what is the precise character of “organic form” in language. He does not, so far as I can see, attempt to construct particular generative grammars or to determine the general character of any such system, the universal schema to which any particular grammar conforms. In this respect, his work in general linguistics does not reach the levels achieved by some of his predecessors, as we shall see directly. His work is also marred by unclarity regarding several fundamental questions, in particular, regarding the distinction between the rule-governed creativity which constitutes the normal use of language and which does not modify the form of the language at all and the kind of innovation that leads to a modification in the grammatical structure of the language. These defects have been recognized and, to some extent, overcome in more recent work. Furthermore, in his discussion of generative processes in language it is often unclear whether what he has in mind is underlying competence or performance – Aristotle’s first or second grade of actuality of form (De Anima, book II, chap. 1). This classical distinction has been reemphasized in modern work. See note 2, and references given there. The concept of generative grammar, in the modern sense, is a development of the Humboldtian notion of “form of language” only if the latter is understood as form in the sense of “possession of knowledge” rather than “actual exercise of knowledge,” in Aristotelian terms. (See note 39.) It should, incidentally, be noted that the failure to formulate rules of sentence construction in a precise way was not simply an oversight of Cartesian linguis- tics. To some extent it was a consequence of the express assumption that the sequence of words in a sentence corresponds directly to the flow of thought, at least in a “well-designed” language,53 and is therefore not properly studied as part of grammar. In the Grammaire générale et raisonnée it is maintained that, except for the figurative use of language, there is little to be said in grammar regarding rules of sentence construction (p. 145). In Lamy’s rhetoric, shortly after, omission of any discussion of “the order of words and the rules that must be observed in the composition of speech” is justified on the grounds that “the

natural light shows so vividly what must be done” that no further specification is necessary (p. 25).54 At about the same time, Bishop Wilkins55 distinguishes those constructions that are merely “customary” (take one’s heels and fly away, hedge a debt, be brought to heel, etc.) from those which follow the “natural sense and order of the words” and therefore need no special discussion (p. 354); for example, the arrangement of Subject, Verb, and Object, or Subject, Copula, and Adjective, or the ordering of “grammatical” and “transcendental” particles relative to the items they govern, etc. (p. 354). At the opposite pole from the belief in “natural order” is the view that each language contains an arbitrary collection of “patterns” learned through constant repetition (and “generalization”) and forming a set of “verbal habits” or “dis- positions.” The belief that language structure and language use can somehow be described in these terms underlies much of the modern study of language and verbal behavior, often coupled with a denial of the possibility of useful cross-linguistic generalizations in syntax (see pp. 57–58, above). Like the reliance on a presumed natural order, it has helped foster a neglect of the problem of specifying the “grammatical form” of particular languages or the general abstract schema to which each language must conform.56 In summary, one fundamental contribution of what we have been calling “Cartesian linguistics” is the observation that human language, in its normal use, is free from the control of independently identifiable external stimuli or internal states and is not restricted to any practical communicative function, in contrast, for example, to the pseudo language of animals. It is thus free to serve as an instrument of free thought and self-expression. The limitless possibilities of thought and imagination are reflected in the creative aspect of language use. Language provides finite means but infinite possibilities of expression con- strained only by rules of concept formation and sentence formation, these being in part particular and idiosyncratic but in part universal, a common human endowment. The finitely specifiable form of each language – in modern terms, its generative grammar (cf. note 39) – provides an “organic unity” interrelating its basic elements and underlying each of its individual manifestations, which are potentially infinite in number. The dominant view throughout this period is that “languages are the best mirror of the human mind.”57 This virtual identification of linguistic and mental processes is what motivates the Cartesian test for the existence of other minds, discussed above. It finds expression throughout the romantic period. For Friedrich Schlegel, “Mind and language are so inseparable, thought and word are so essentially one, that, just as certainly as thoughts are considered to be the characteristic privilege of humankind, we can call the word, in accordance with its inner meaning and dignity, the original essence of man.”58 We have already made reference to Humboldt’s conclusion that the force that generates lan- guage is indistinguishable from that which generates thought. Echoes of this

conclusion persist for some time,59 but they become less frequent as we enter the modern period.

The association of language and mind is regarded differently in the earlier and later phases of the period under review.

The earlier view is that the structure of language reflects the nature of thought so closely that “the science of language differs hardly at all from that of thought” (Beauzée, p. x)60; the creative aspect of language use is accounted for on the basis of this assumption.61 On the other hand, the observation that language serves as a medium of thought begins to be rephrased as the view that language has a constitutive function with respect to thought.

La Mettrie, for example, in discussing how the brain compares and relates the images that it discerns, concludes that its structure is such that, once the signs of objects and their differences “have been traced or imprinted on the brain, the soul necessarily examines their relations62 – an examination that would have been impossible without the discovery of signs or the invention of language” (op. cit., p. 105); prior to the discovery of language, things could only be perceived in a vague or superficial way. We have already referred to Humboldt’s view that “Man lives primarily with objects, indeed, since feeling and acting in him depend on his presentations [Vorstellungen], he actually does so exclusively, as language presents them to him” (op. cit., p. 74; Humboldt 1999: 60). Under the impact of the new relativism of the romantics, the conception of language as a con- stitutive medium for thought undergoes a significant modification, and the notion that language difference can lead to differences, even incomparability in mental processes, is explored.63 This development, however, is not part of our main theme; its modern elaboration is familiar, and I will discuss it no further here.