Language

Plato's Cratylus examines the question whether names belong to things by nature or by convention. Cratylus maintains that names are natural signs of the things they signify, so that the correct name for each thing is the one that expresses its nature. Hermogenes takes the opposite view, that names are established by agreement and custom, and that any sound might serve as the name for any thing. Socrates examines both positions and finds that existing languages embody contradictory principles of symbolization, which suggests that human language does not originate as a gift from the gods, for if it did, signs would be perfectly adapted to the things signified.

The implication Plato draws is that names are not a reliable path to the knowledge of things. If the same thing bears different names in different languages, and if the principles by which names were originally assigned are uncertain, then the philosopher who relies on words rather than on direct examination of things will be led into error. In the Seventh Letter, Plato goes further, saying that the knowledge of "the highest matters and the first principles of things" does not admit of exposition in language as other branches of knowledge do. No man of intelligence, he writes, will venture to express his philosophical views in words that cannot be changed.

This suspicion of language is connected with the broader question, treated under the idea of Knowledge, of whether the objects of genuine understanding can be communicated at all through the medium of conventional signs. Plato suggests in the Cratylus that the hypothesis of a natural language, in which words would be perfect images or imitations of things, is not something men can hope to construct, but serves rather as a standard by which the imperfections of existing languages may be recognized. The distinction between natural and conventional language is one that Aristotle takes up and resolves on the side of convention.

Plato establishes the two questions that determine the course of the subsequent discussion of language: whether names are natural or conventional, and whether language can be trusted as an instrument of philosophical inquiry. Aristotle will answer the first by affirming that words are conventional signs of mental impressions; the second question remains open throughout the tradition.

Aristotle resolves the question left open in Plato's Cratylus by affirming that language is conventional. In the opening of De Interpretatione, he establishes the framework that dominates the subsequent discussion: spoken words are symbols of mental impressions, and written words are symbols of spoken words. The mental impressions themselves are natural and the same for all men, but the words that express them are conventional and differ from language to language. Every sentence is significant not as a natural instrument but by convention.

This framework gives language a determinate function in relation to thought and reality. Words do not reveal the essences of things directly, as Cratylus maintained, but neither are they without significance. They signify because a community has agreed to use them in certain ways, and they succeed in communication insofar as the mental impressions of speaker and hearer correspond. The question of how far this correspondence can be relied upon is treated more fully under the idea of Sign and Symbol, but Aristotle's own practice of distinguishing the multiple senses of common words before discussing any philosophical subject suggests that he regards ambiguity as a natural feature of language that must be managed rather than eliminated.

Aristotle also lays down the basic distinctions of linguistic analysis. A noun signifies without reference to time; a verb signifies with reference to time; a sentence is a combination capable of truth or falsehood. These distinctions provide the foundation for the liberal arts of grammar, logic, and rhetoric as separate disciplines, each establishing its own rules for the use of language by reference to a special standard of correctness. The Rhetoric adds to this the consideration of language as a tool of persuasion, concerned not only with the truth of what is said but with its effect upon the hearer.

Aristotle's theory that words are conventional signs of natural thoughts provides the common framework within which the subsequent tradition discusses the relation between language and knowledge. Aquinas will apply this framework to the problem of predication about God; Locke will press the question of whether words attach to thoughts as securely as the theory assumes.

Augustine's On Christian Doctrine is directly concerned with the art of reading and interpretation, and in the course of addressing this practical problem it develops the most complete theory of signs in the ancient tradition. A sign, Augustine writes, is a thing which, over and above the impression it makes on the senses, causes something else to come into the mind. Words are the most common class of signs, but the theory extends to gestures, images, and sacraments. The broader inquiry into signs and symbols is treated under the idea of Sign and Symbol, but Augustine's analysis bears directly on the philosophy of language by establishing that words do not convey meaning in themselves; they function as prompts that direct the mind toward what they signify.

In the Confessions, Augustine describes how he learned language as a child by observing the gestures and expressions of adults and gradually inferring what each word meant. But his deeper point, developed in On the Teacher, is that hearing words is not sufficient for understanding. The teacher can present signs, but the student grasps their meaning only through an interior consultation with truth itself. Words merely prompt us to look within, where the true teacher, whom Augustine identifies with Christ, illumines the mind. The connection between this doctrine and the theory of knowledge is discussed under the idea of Knowledge.

Augustine's treatment of language also addresses the problem of interpreting Scripture, in which the same text may bear both a literal and a figurative sense. The literal sense is what the words directly signify; the figurative sense is what the things signified by the words themselves signify. This distinction between levels of meaning becomes fundamental to Western biblical interpretation and, more broadly, to the theory of how texts may carry significance beyond the immediate reference of their words.

Augustine's analysis of the gap between sign and understanding raises a question that the modern philosophers of language will press with greater urgency: whether speakers can ever be confident that their words carry the same ideas for every hearer. Locke's sustained examination of the imperfections of language takes up this question in detail.

Aquinas adopts the Aristotelian framework in which words signify concepts and concepts signify things, but he applies it to a problem that Aristotle did not face: how human language, whose concepts are derived from the experience of finite, material things, can say anything true about God, who is infinite and immaterial. If the word "good" means the same thing when applied to God and to a creature, the application would be univocal, and the infinite difference between divine and created goodness would be denied. If the word means something entirely different in each case, the application would be equivocal, and theological discourse would be impossible.

Aquinas's solution is the doctrine of analogical predication. When we say that God is good, we mean that the goodness we know in creatures derives from God's goodness and imperfectly resembles it. The word is used neither univocally nor equivocally but analogically: it preserves a real relation between the divine and the human while acknowledging the infinite disproportion between them. "Whatever is said of God and creatures is said according to the relation of a creature to God as its principle and cause." This doctrine addresses the limits of language in relation to the highest objects of thought, and it has consequences for the theory of knowledge as well, since the possibility of natural theology depends upon the meaningfulness of analogical speech about God.

Aquinas's treatment of language also extends to the multiple senses of Scripture, following Augustine's distinction between literal and figurative meaning. The words of Scripture signify things, and the things signified by the words may themselves signify further things. This multiplicity of senses does not, for Aquinas, introduce confusion, because the literal sense always remains the foundation on which the other senses are built. The problem of interpreting inspired language is treated more fully under the idea of Theology.

Aquinas's doctrine of analogy represents the mediaeval tradition's most systematic attempt to determine the conditions under which language can be used meaningfully about objects that exceed the scope of ordinary experience. Hobbes and the modern empiricists will challenge this position by arguing that words which cannot be traced to specific impressions or definite ideas are without significance, and that analogical speech is precisely the kind of language that disguises its own emptiness.

Hobbes's treatment of language represents the modern turn toward a suspicion that words cause men unwittingly to deceive themselves as often as they enable one man intentionally to deceive another. In Leviathan, he defines words as "marks" that men set down to recall their own thoughts and "signs" by which they communicate those thoughts to others. Language is the foundation of science and civil society: without names, "there had been amongst men neither commonwealth, nor society, nor contract, nor peace, no more than amongst lions, bears, and wolves."

Yet the same power that makes language useful makes it dangerous. Hobbes gives particular attention to what he calls absurd, insignificant, or nonsensical speech, "whereby we conceive nothing but the sound." We cannot imagine anything infinite, he observes, and therefore a word like "infinite" is a form of absurd speech taken upon credit from "deceived philosophers and deceived, or deceiving, Schoolmen." Terms such as "infused virtue," "free will," and "immaterial substance" are, in Hobbes's view, examples of insignificant names that disguise the absence of any definite idea. "Words are wise men's counters; they do but reckon by them: but they are the money of fools." This criticism reduces what previous thinkers treated as disputes about the nature of things to disputes about the meaning of words.

Hobbes's remedy is strict definition. Every philosophical inquiry should begin by fixing the signification of its terms, and reasoning should proceed as a kind of computation, adding and subtracting names according to agreed definitions. When definitions are perspicuous and the computation correct, truth follows. When definitions are loose, the result is absurdity. The standard by which Hobbes criticizes the language of other philosophers depends, as Adler observes, upon what Hobbes holds to be true about the nature of things and of the mind, and his opponents may reply that his semantic criticism holds only if his own metaphysics and psychology are correct.

Hobbes's treatment of language inaugurates the modern attitude that ascribes to words a power to mislead that is independent of the speaker's intentions. Locke will develop this suspicion into a sustained examination of the imperfections and abuses of language, devoting the entire third book of his Essay to the subject.

Locke devotes the entire third book of his Essay Concerning Human Understanding to language, examining in detail both the imperfections and the abuses of words, and the remedies for them. His central thesis is that words, in their primary or immediate signification, stand for nothing but the ideas in the mind of the speaker. When different speakers use the same word, they may associate it with different ideas, and communication succeeds only insofar as the same idea is excited in the hearer's mind as stands in the speaker's. Locke is deeply skeptical that this condition is often met.

"Vague and insignificant forms of speech, and abuse of language," Locke writes, "have so long passed for mysteries of science; and hard or misapplied words with little or no meaning have, by prescription, such a right to be mistaken for deep learning and height of speculation, that it will not be easy to persuade either those who speak, or those who hear them, that they are but the covers of ignorance, and hinderance of true knowledge." Abstract terms such as "substance," "liberty," and "justice" are especially susceptible to this difficulty because the ideas they stand for are complex, variable, and poorly defined. Much of what passes for philosophical dispute is, in Locke's view, a disagreement about the meanings of words rather than about the nature of things.

Locke's remedy consists in tracing ideas back to the particular observations and experiences from which they arise, and in using words with constant awareness that they are signs of ideas, not of things themselves. When this discipline is observed, many apparent disagreements dissolve. The connection between Locke's philosophy of language and his theory of knowledge is treated under the idea of Knowledge, but it bears directly on the discussion of language by establishing that the imperfections of words are not merely inconveniences of communication but obstacles to the attainment of truth.

Locke's sustained examination of language makes the philosophy of language a branch of epistemology. Hume will sharpen Locke's principle into a criterion for distinguishing meaningful from meaningless speech; Kant will ask whether the categories of thought are independent of the linguistic forms in which they are expressed.

Rousseau introduces into the discussion of language the question of its origin, a problem that previous writers had treated incidentally or not at all. In the Discourse on Inequality, he observes that language and society seem to presuppose each other: language requires a community of speakers, yet the formation of a community seems to require language. This circularity, Rousseau acknowledges, makes the question of how language first arose one of the most difficult in philosophy.

His speculative account proposes that the first expressions were not abstract labels but cries of passion: exclamations prompted by pain, fear, desire, and surprise. The first language was figurative rather than literal, and the first speakers were, in this sense, poets rather than logicians. Men pointed at visible objects and used imitative sounds to signify audible ones, but these methods of expression, being insufficient to convey ideas about absent or future things, required the eventual invention of articulate speech and the institution of conventional signs. Rousseau observes that such an institution could only be made by common consent, which itself seems to require the very faculty of speech it is meant to explain.

This account challenges the view, implicit in Hobbes and Locke, that language is a practical instrument invented by rational adults for the communication of ideas. Rousseau insists that language is rooted in feeling, not in calculation, and that the analytical precision which Locke regards as the remedy for the imperfections of language is in fact a late development of a faculty that began in passion and imagination. The connection between the origin of language and the origin of human society is treated under the ideas of Custom and Convention and State.

Rousseau's contribution to the discussion of language lies in opening the historical question of how conventional languages arose from natural expression. His account connects the philosophy of language with the problem of the origin of man and society, a connection that Darwin and the modern linguists will develop from a different starting point.

Hume sharpens the empiricist critique of language into a principle of method: when we entertain any suspicion that a philosophical term is employed without meaning, we need only inquire from what impression the supposed idea is derived. If no impression can be produced, the term is without significance. This criterion, applied consistently, calls into question much of the vocabulary of traditional metaphysics, for words like "substance," "self," and "necessary connection," when examined, either lack corresponding impressions entirely or correspond to impressions quite different from what the words suggest.

On the question of general or abstract ideas, Hume sides with Berkeley against Locke. There is no such thing as a genuinely abstract idea, no mental image of "triangle in general" that is neither equilateral, scalene, nor isosceles. What we possess are particular ideas that become general through the attachment of a general term. The word "triangle" calls up a particular mental image, and the habit of using the word in connection with many different particular triangles gives that image a general function. Language, on this account, creates the appearance of abstraction that the mind cannot achieve on its own. The bearing of this analysis on the theory of ideas is discussed under the idea of Idea, but it is directly relevant to the philosophy of language because it suggests that the power of general names to mislead is greater than Locke supposed.

Hume's position is closely connected with the modern attitude, shared by Hobbes and Berkeley, that general or universal names signify nothing which can be perceived or imagined, and that men are deceived by the tendency of words to counterfeit a reality which does not exist. Many traditional philosophical problems, on Hume's view, arise from mistaking words for ideas, and the remedy is to apply the criterion of impressions with unflinching consistency.

Hume's criterion for meaningful speech provides the foundation on which later empiricists will build. Kant will respond by arguing that the categories of the understanding, though they cannot be traced to particular impressions, are nonetheless meaningful because they are the conditions under which any experience is possible at all.

Kant does not write a treatise on language as such, but his critical philosophy alters the terms in which the relation between language and thought is understood. The empiricists, from Locke through Hume, treated words as signs of ideas derived from impressions, and judged words to be meaningless when no corresponding impression could be produced. Kant argues that the mind brings to experience a structure of its own: the categories of the understanding, such as substance, causality, and unity, are not abstracted from sense data or created by linguistic convention but are preconditions of any experience whatsoever.

This has direct consequences for the philosophy of language. If words like "cause" and "substance" cannot be traced to particular impressions, the empiricist criterion would pronounce them meaningless. Kant's answer is that they are neither experiential in origin nor empty of content; they are transcendental conditions of experience, and their legitimate use consists in articulating the structure that the mind imposes on all possible objects of experience. In this sense, the categories vindicate a range of philosophical language that the empiricist critique would have eliminated. The relation between Kant's categories and the Aristotelian analysis of the basic forms of predication is treated under the idea of Idea.

Kant also draws a line between what can be meaningfully said and what cannot. The ideas of pure reason, such as God, the soul, and the totality of the world, exceed the conditions of possible experience and therefore cannot be the objects of theoretical knowledge. Language about the supersensible is not false but lacks the content that only sensible intuition can supply. The claims of traditional metaphysics are, on Kant's analysis, instances of transcendental illusion: the attempt to apply the categories of the understanding beyond the domain in which they have legitimate employment.

Kant shifts the discussion of language from the question of how words relate to impressions to the question of how the categories of thought relate to the structure of experience. Whether the categories he identifies are truly independent of the grammatical structures of particular languages, or whether they reflect the forms of predication that Greek, Latin, and German happen to share, is a question that later writers on language will raise.