Out of Egypt:

"the world of signs and the world of things": philosophy of language final paper

Philosophy of Language final paper, in case anyone's interested. It's a (very preliminary) critique of Saussure. Comments appreciated, if anyone has free time. Send email since comments box is broken.

“A science that studies the life of signs within society is conceivable; it would be a part of social psychology and consequently of general psychology; I shall call it semiology.” With this sentence from the first section of his Course in General Linguistics, Ferdinand de Saussure launched what would later become the field of semiotics . This field, though Saussure describes it as a science, rests upon a questionable metaphysical assumption. In his view, the arbitrariness of the linguistic sign entails that a sign cannot be defined except in terms of other signs. The difficulties with this view, though Saussure’s initial work seems to neglect them, are clear: language, if understood only as a system of differences, is an isolated system, sealed off both from other natural languages and from the physical world. After briefly outlining Saussure’s theory of signs, this paper will argue that Saussure’s theory ignores a conceivable means by which linguistic meaning could be linked to the physical world, and therefore never adequately defines signification.

Saussure’s semiotic theory develops from his attempt to refound linguistics on more scientific terms. Though his program ends up denying that words have meaning in the traditional sense (which involves reference to objects external to language, and thus external to the human mind), its first principles are difficult to deny. “The thing that constitutes language,” Saussure writes, “is…unrelated to the phonic character of the linguistic sign.” Speech is merely the vehicle by which signs are communicated from one consciousness to another. Saussure’s linguistics – and the philosophy of language which develops from it – is not concerned with speech as a physical phenomenon, but rather with its psychological effects. Furthermore, he argues, speech, which is the articulation of a defined set of sounds combined in a particular (and arbitrary) fashion, has psychological effects only within a social context. Only “by virtue of a sort of contract signed by the members of a community” does speech become comprehensible as language. The contract to which he refers fixes the meaning of words as “associations which bear the stamp of collective approval.”

These principles are reasonable in themselves. The difficulty comes when Saussure works out their implications, as he begins to do in the section entitled “The Nature of the Linguistic Sign.” In this section, having already cleared away misconceptions about linguistics that were held by his predecessors, the grammarians and the philologists, he focuses in on the proper subject of linguistic study, the sign. This he describes as a union between two terms, “a concept and a sound-image” – a signified and its signifier. The link between these terms, he states, is arbitrary.
A linguistic sign is not like a visual symbol. In the latter case, there is something suggested by the image that would not be suggested equally well by another image. For example, he writes, “[t]he symbol of justice, a pair of scales, could not be replaced by just any other symbol, such as a chariot.” A linguistic sign, by contrast, is received either as a sequence of phonemes or a sequence of written characters. In neither case (except that of ideographic languages, such as Chinese) does this sequence, either in its individual units or collectively, bear a resemblance to the signified. Onomatopoeic words and interjections are the only possible exceptions which Saussure considers. He writes with regard to onomatopoeia that any perceived similarity is often an accidental result of phonetic evolution; true onomatopoeia is rare enough to be theoretically insignificant. With regard to interjections, he writes that a simple comparison of languages shows too much divergence for there to be a similarity between signifier and signified apart from that fixed by convention.

Saussure’s account of the linguistic sign as arbitrary is not objectionable in itself. Philosophical naturalists, such as Michael Devitt and Kim Sterelny, affirm this as well. However, they do not see how this affects reference. Simply because the word ‘tree’ is not “tree-like” does not make it less able to refer to the concept of trees. Physical trees, or any other natural object, fail to appear within the framework of Saussure’s theory, however. It is difficult to argue against the error of Saussure’s metaphysics, though, since it makes itself known through what he ignores.

Saussure takes the arbitrariness of the linguistic sign as a jumping-off point for his theory of difference as the determiner of meaning. “Language,” he writes, “is a system of interdependent terms in which the value of each term results solely from the simultaneous presence of the others.” This leads him to an even more radical assertion:

Whether we take the signified or the signifier, language has neither ideas nor sounds that existed before the linguistic system, but only conceptual and phonic differences that have issued from the system. The idea or phonic substance that a sign contains is of less importance than the other signs that surround it.

Saussure here shows his ignorance of a casual theory of names such as that offered by Devitt and Sterelny in the fourth chapter of their book. In such a theory, words have an intensional mention because they have been linked to extensional reality through a chain of reference that begins at a ‘dubbing’ of a particular physical entity.

There are admittedly problems with such a theory, as semotician Umberto Eco describes. In his Semiotics and the Philosophy of Language, he considers the project of creating a semiotic dictionary in which all word-concepts would be reduced to their simplest component elements. These elements he calls ‘primitives,’ and says that it is possible to regard them as “rooted in our world experience.” However, many concepts are not encountered in the typical individual’s world experience, and so Saussure’s point has some validity – such concepts must in fact be defined in terms of the concepts which one has encountered. Furthermore, there is a difficulty, which Devitt and Sterelny call the qua-problem , in determining the boundary of a particular concept. One must know that a given object is appropriately described by the concept which one seeks to associate with it – “It seems that the grounder must, in effect and at some level, ‘think of’ the sample as a member of a natural kind, and intend to apply the term to the sample as such a member.” For example, one who is observing various kinds of cats needs to be able to distinguish tigers from cheetahs, and both from lions. Saussure is correct insofar as this individual’s situation would be different if one of those terms did not exist in his language.

However, this individual would certainly not be bound by his language simply to call all of them ‘cat’ and then drop the issue. There is an observable, physical difference between these animals and if our hypothetical observer found this difference significant, he could incorporate it into his language by means of the dubbing process. Saussure’s discussion on page 120 of the interface between speech and thought shows that he had no concept of a language of thought, a “mentalese,” in which such distinctions could be made prior to their expression in public language. To him, language is a creative force, marking off “contiguous subdivisions…on both the indefinite plane of jumbled ideas and the equally vague plane of sounds.” By contrast, it seems possible that an individual’s thought can be just as definite as language, if not more, since thought is private (and therefore devoid of the ambiguity that results when one’s intents in speech are misunderstood) and arises, at least partially, from sense impressions.

All the conceptual distinctions that we possess today must have come into existence somehow. Saussure’s theory offers no explanation of how this might have occurred. If “what distinguishes one sign from the others is what constitutes it,” as he writes, there is nothing going on in the linguistic world except the interplay of signifiers. This is explicitly stated on the following page: “When isolated, neither Nacht [‘night’] nor Nächte [‘nights’] is anything: thus everything is opposition. Language, in a manner of speaking, is a type of algebra consisting solely of complex terms.” It may be true that such distinctions as that between singular and plural, or that between night and day, can be understood in terms of simple binary opposition, but not all language is like this. Our hypothetical taxonomist might conceivably recognize the fundamental similarity between a succession of tigers that he saw and thus refer to them all by the name ‘tiger.’ Upon encountering a cheetah for the first time, certain sense impression might activate primitive concepts in his mind, such as spottedness versus stripedness, and he might then call this animal by the new name ‘cheetah.’ This process could then continue as he encountered lions, pumas, and so forth. The opposition, upon which Saussure founds his whole theory, would no longer be binary. The hypothetical taxonomist could add new species to the genus of ‘cat’ without changing the meaning of earlier species terms, despite Saussure’s contention to the contrary.

The hypothetical scenario outlined above is an attempt to show how the signifier-signified relationship could be established in the first place. Saussure offers no corresponding account, since his diachronic linguistics concerns itself solely with the development of languages already in existence and the only mechanism of change it offers is phonological – changes in sound making words sound more or less like other words, and thus shifting their meaning. Perhaps this is the linguistic equivalent of the biologists’ explanation of apparent design in nature as the result of random mutation. By contrast, a causal theory of naming, with a corresponding theory of “mentalese” as the ground for public language, would make humans’ purposeful actions a driving force behind linguistic change and the principal means by which new terms are introduced into the language. Unlike Saussure’s theory, a causal theory strives to connect with the physical world, which is more certain than our speech about it could ever be.