Frege and Saussure – 9/4/07 Overview Last week we saw some views of language that characteristically exhibit three kinds of commitments: Expressivism: The purpose of words is to outwardly express „inner‟ thoughts, feelings, or inner experiences. Denotationism: The basic kind of meaning is the reference of (proper) names to objects Instrumentalism: Language is an instrument or tool for the expression or communication of ideas. In this respect, language is secondary to thought. We can think of Frege and Saussure, in different ways, as adding another commitment to the mix, one that will later have decisive and unsettling consequences for all 3 commitments: Structuralism: Language as a whole can be understood as a system or structure of signs standing in regular synchronic and diachronic relations. The principles of this structure are responsible for what we ordinarily refer to as the “meaning” or “significance” of terms. We can understand the principles of this structure and use them to criticize ordinary linguistic usage. We see a commitment to structuralism in various places in Frege and Saussure. Despite great differences, they share a view of language that will be definitive of the twentieth century. Saussure This structuralist picture defined by Saussure in the Course in General Linguistics formed the basis for the analytic work of “structuralist” philosophers, anthropologists, and psychologists throughout the middle part of the twentieth century, including the literary theorist Roland Barthes, the anthropologist Claude Levi-Strauss, the linguists Roman Jackobsen and Emile Benveniste, and the Freudian psychoanalyst and philosopher Jacques Lacan. The “archeology of knowledge” undertaken by the philosopher Michel Foucault itself applies some “structuralist” terms of analysis, while also seeking to problematize and extend them. 1. The sign The sign is, for Saussure, the unity of two elements, what he calls the sound-image (or signifier) and the concept (or signified). The sound-image is “sensory”, but it is not directly material. Instead, it is the “psychological imprint of the sound” on our senses. The concept is the psychological idea or thought. The sign requires both elements, in coordination, to be the sign that it is. But the connection between signifier and signified is arbitrary. One and the same signified, for instance the concept tree, can be signified by any of the following signifiers: “tree”; “baum”; “arbor”; etc. The connections are, Saussure says, fixed in the case of each language by convention. 2. The Chain of Signifiers In speech (parole), signifiers unfold over time in sequence. In this sequence, the chain of auditory signifiers runs along in parallel with the chain of signifieds. (We can think of the two chains as like two sides of a sheet of paper, or two parallel streams (compare the picture on p. 649). These are at first like “two parallel, shapeless masses” but they are given articulation by being broken up into discrete “thought-sounds” or signs. 3. Language is a system of differences without positive terms “Language is a system of interdependent terms in which the value of each term results solely from the simultaneous presence of the others.” (p. 650). The connection between the particular signifier and the particular signified is itself completely arbitrary, as is shown by the variety of words with which I can communicate the same concept in different languages. The system of language itself, though, is a system of the relations of signifiers. If I explain the meaning of a word to you, or explain to you what I mean by a particular word, I can only give you more signifiers; I can never guarantee that they evoke the same signified. Thus language, as it is spoken, is a linear chain of signifiers, each one following from the last in the order spoken and each relating to others as a matter of definition and context. Language itself, Saussure suggests, can be thought of as this purely relational system of differences. Saussure compares the system of differences to that of economic exchange. In economic exchange, a particular unit of money (e.g. a dollar) has the value it does because it can be: 1) exchanged for something of value (e.g. a loaf of bread) or 2) exchanged for another instrument of value (e.g. a Euro). In the same way, signifiers can be exchanged for signifieds (thus we cross the bar of signification) or they can be exchanged for other signifiers. To this system of exchange, all that matters is the differences in value between the terms. (e.g., it doesn‟t matter if a dollar is printed on paper or is made of something else – all that matters is what it can be exchanged for). The properties of signification are defined by the differences between significant elements, not by these elements themselves. “Everything that has been said up to this point boils down to this: in language there are only differences. Even more important: a difference generally implies positive terms between which the difference is set up; but in language there are only differences without positive terms. Whether we take the signified or the signifier, language has neither idea nor sounds that existed before the linguistic system, but only conceptual and phonic differences that have issued from the system.” (pp. 652-53). 4. Syntagmatic and Associative relationships We can distinguish between two sorts of relationships characteristic of the system of language as a whole: syntagmatic (or diachronic) and associative (or synchronic). Syntagmatic relationships characterize terms that regularly follow one another. Associative relationships characterize terms that are regularly associated with one another without following one another. For instance, we can think of the ordinary English expressions “it‟s time for a change” and “change is good” as governed by syntagmatic relations; whereas the relation between “change” and “exchange” is associative. Frege In the context of the history of analytic philosophy, Frege‟s greatest innovation was to suggest the idea of – and partially work out – a clarified language of thought or logical notation, what he called Begriffschrift. Its aim was to remedy the errors of language and demonstrate the true structure of logical inference by way of a unified symbolic calculus. “If it is a task of philosophy to break the power of words over the human mind, by uncovering illusions that through the use of language often almost unavoidably arise concerning the relations of concepts, by freeing thought from the taint of ordinary linguistic means of expression, then my Begriffschrift, further developed for these purposes, can become a useful tool for philosophers.” (pp. 50-51) Frege‟s initial concern was with the logic of mathematical judgments: he wanted to make this logic rigorous, in particular, in order to more clearly distinguish real proofs from pretenders and clarify the logical basis of mathematics (he was one of the first advocates of the program of logicism, which sought to reduce mathematics to logic). He quickly found that ordinary language, and the “subject-predicate” logic he inherited from Aristotle, was not sufficient to the task of showing the “real logical relations” between mathematical expressions. Thus he turned to the notion of a Begriffschrift or concept- script to do so. The idea had antecedents in the philosophy of Leibniz; like Leibniz, he compared it to a special instrument for producing clarity, like a microscope for thought. The Begriffschrift and the logical unity of language it revealed led him to the concept of logical contents of thought, or senses. Senses are: 1) To be distinguished from symbols or signs; 2) To be distinguished from the references, meanings, or objects of symbols or signs; 3) Strongly objective (and so to be distinguished from any subjective thought or experience) 4) The logically relevant part of a judgment; the part that plays a role in logical inference; the part that determines a truth-value (true or false). In encountering the article, we are considering the questions: “What is the meaning of a word?” and “What is the meaning of a sentence?” We might give various answers. We might think, for instance, that the meaning of a word is the object that it stands for: for instance, the meaning of “The moon” would just be the moon. But there are lots of words that seem to be meaningful, but don‟t stand for anything that we can find in the world or point to. Consider, for instance, “Pegasus” and “two.” Clearly, if we want to say that part of the meaning of a word is the thing it stands for, we have to say more as well: we have to supplement our account to handle terms that do not obviously stand for anything but still have a meaning. Another reason for supplementing, or dividing, our account of meaning comes from considerations about judgments of equality or identity. Language is full of judgments of identity; we very often say that one thing is another, and it can be important to find out that this is true. Consider, for instance, the judgment: A: Clark Kent is Superman. This is clearly not an obvious truth; when somebody finds it out, they know something that they didn‟t know before. But if we thought that the meaning of a word is just the thing that it stands for, this would be mysterious. The judgment “Clark Kent is Superman” would mean the same as B: Superman is Superman. Since both “Clark Kent” and “Superman” mean the same person, the judgment “Clark Kent is Superman” would seem to just say that that person is that person. But this is something that everybody knows! If we want to explain how (A) can be a different judgment from B, we have to distinguish something else in the meaning of a word, besides just what the word stands for. We have to distinguish, in other words, between the reference (Bedeutung) of a word – what it stands for – and the sense (Sinn). Both “Superman” and “Clark Kent” have the same reference: they refer to the same individual. But they have different senses. When we make judgment (A), we are referring to the same thing using two different senses. When we find out that (A) is true, we find out that these two senses do indeed have the same reference. But we might not know this initially; if we just know the terms “Clark Kent” and “Superman,” we know how to refer to this person two different ways, but we don‟t know that it‟s the same person we‟re referring to. With the sense/reference distinction drawn, we can also handle cases of senses with no reference. For instance the terms “Pegasus” and “The King of France” have senses – we know what they mean – but they don‟t refer to anything. We can specify the sense, and even say things about what would be true of the reference if there were one, even though in fact there is not. In fact, Frege thinks that cases of sense without reference are a drawback of ordinary language that we could avoid if we could change it to a “logically perfect” language. “A logically perfect language (Begriffsschrift) should satisfy the conditions, that every expression grammatically well constructed as a proper name out of signs already introduced shall in fact designate an object, and that no new sign shall be introduced as a proper name without being secured a Bedeutung.” (p. 632).” The sense of the term “Pegasus,” for instance, isn‟t just what you or I decide to mean by it: it‟s what the word means for all of us. We can agree or disagree about what it means, and there is a right and wrong answer about what it means, even though it doesn‟t have a reference. In “Thought” Frege distinguishes among “three worlds” or realms. The first world is the physical world of physical events or objects. The second world is the inner, subjective world of private contents of thought, what vary from person to person in idiosyncratic fashion and can never be compared. (This is what Frege usually treats as the realm of psychology). But senses are not subjective, since it is essential to communication that they can be shared. Senses, therefore, must be in the third world or realm: the realm of objective contents. Like the thoughts behind sentences, they are eternal and unchanging, and our “grasping” of them explains our ability to use words meaningfully. Thinking about senses might help us to get clear on the nature of thoughts and constituents of the third realm generally. Frege holds that sense determines reference: for every sense, there is at most one reference. We can have two senses that pick out the same reference, but never two references from the same sense. So we can say that understanding a word – knowing its sense – means knowing how to go from the word to a reference: knowing which thing it picks out, if it picks out any. Frege thinks something similar is true, as well, of the senses of whole sentences. The senses of whole sentences are the thoughts or contents they express, and we can think of the sense of a sentence as being made up of the senses of the terms within it. But what is the reference of a sentence? It doesn‟t seem that the reference of a sentence – say, “the cat is on the mat” – is any object; the sentence doesn‟t refer to the cat, or the mat, or to the way that they‟re connected, but really to all of these. In thinking about the reference of a sentence, though, Frege again holds to the model that sense determines reference. The sense of the sentence determines, in particular, whether the sentence as a whole is true or false. Accordingly, Frege holds that the reference of a sentence is one of two “truth-values,” the True and the False: “We are therefore driven into accepting the truth value of a sentence as constituting what it means. By the truth value of a sentence I understand the circumstance that it is true or false. There are no further truth values. For brevity I call the one the True, the other the False. Every assertoric sentence concerned with what its words mean is therefore to be regarded as a proper name, and its Bedeutung, if it has one, is either the True or the False.” (p. 10). Just as understanding a word is going from its sense to its reference, judging a sentence is going from its sense to its truth-value, finding out whether it is true or false. In both kinds of case, the reference is completely determined by the sense: we could not have another reference, given the particular sense that the word or sentence has. Indeed, Frege thinks that logical relations among sentential senses are, in an important sense, prior to logical relations among terms or their senses. This priority is expressed by Frege‟s famous context principle: “Only in the context of a sentence does a term have meaning.” The idea is that we have to look at the logically relevant relations – which are relations between sentential senses rather than between the senses of terms, and make possible judgments – in order to determine content even before we start thinking about the relationships between individual words and their references. Senses can take part in logical inferences and so are governed by logical laws. It was important to Frege to distinguish these laws from any psychological laws: they are not dependent on the particular constitution of our minds. Nor are they dependent on any intuitive or psychological accompaniments that go on in our minds when we make judgments. The judgments do the same things – have the same roles – and exhibit the same relations to truth-values regardless of what goes on in our minds. For the truth or falsity of a judgment is completely independent of whether I (or anyone) think it‟s true or false. Putting the picture together, then, Frege seems to hold that it is essential to our ability to speak meaningful language at all that we be able to grasp senses, that we be in constant relation to the Third Realm of ideal, objective contents. These contents, remember, do not change at all according to whether we grasp them or not: they retain an austere indifference to our own purposes and needs. Nevertheless, there is an interesting ambiguity in Frege‟s discussion of the nature of senses and thoughts. Even though the senses of words never change and have nothing to do with what we do, we are supposed to understand grasping the sense as having a certain kind of ability: the ability to go from the word to the reference (in the case of word-senses) or from the sentence to the truth-value (in the case of sentence-senses). So it looks like even though senses themselves have nothing to do with what we can do, grasping a sense means, precisely, being able to do something. Some Questions to Think About: 1) Does Saussure‟s structuralism fall into “psychologism” in Frege‟s sense? 2) What assumptions about the relationship between thought and language govern Saussure‟s system? What about Frege‟s? 3) If language is a structure of signs, what can we hope to gain by understanding this structure for each philosopher? 4) In what ways to Frege and Saussure both presuppose – and in what ways do they displace – traditional metaphysical distinctions between the body (sensible) and the mind, soul or spirit (supersensible)? 5) In contrast to Saussure, Frege identifies a “third realm” of objective contents of thought. Why the difference? What is at stake for Frege in the identification of a Platonistic, strongly objective realm of thought? 6) How might each philosopher think about what is involved in learning a first language? In knowing one? In translating from one language to another?