Insight and Error in Wittgenstein

1. Introduction

Wittgenstein, in my opinion, is the greatest philosopher of the 20th century. I express this opinion not because I want to develop or defend it but solely to situate my assessment of his work within my overall admiration for it. I want the criticisms I make to be understood against this background. Also, to understand this article, it is important to try to recall the enormous influence that Wittgenstein had on philosophy at the time I was educated in the subject. In the middle decades of the 20th century, he was by far the most influential philosopher in the English-speaking world. Although he had both followers and critics, he could not be ignored. No philosopher today has anything like the stature or influence that Wittgenstein had in the 20th century, especially after the publication of the Philosophical Investigations in 1953. For me, personally, his philosophy poses the greatest challenge: if Wittgenstein is right, the sort of philosophy I am attempting to do is impossible. Wittgenstein argued powerfully that there can be no such thing as a general philosophical theory of language, mind, consciousness, society, and so on. I wanted and still do want to do precisely that: to present a general philosophical theory of language, mind, consciousness, society, and so on.

2. Theory and Method in the Investigations

I will argue in this article that Wittgenstein’s philosophy of language is mistaken and that the overall conception of philosophy he presents is too limited. However, within its limitations, there are a rather large number of important successes, and I will mention some of those. I will first give a brief sketch of Wittgenstein’s ideas and then go into more detail. The project, described in the Philosophical Investigations (Wittgenstein 2009), is best understood as a reaction against the limitations of his first book, the Tractatus (Wittgenstein 1922). I will say something about both books, starting with the Investigations. It seems to me it contains a certain conception of language and a consequent conception of philosophy. The conception of language is that we should think of using a language as a form of activity or a series of activities. On this view, using a language is like playing a series of games, which Wittgenstein called “language-games.” The analogy with games is meant to emphasize several aspects. One is that language in use, language in action, is what counts, and we should not think of meanings as sets of introspectable entities passing through our minds when we speak, nor as sets of objects for which words stand. Rather, we should think of words as pieces in a game. No one asks in chess “What is the meaning of a rook, really?” They just understand its function in the game and its role in playing the game. There is no further question about what it really means, what it stands for, and what psychological facts are associated with it. And so we should think of words. He says we should think of how the words are used in the language, and there is no further question about a deeper meaning or the underlying psychological reality that extends meaning to the words. It is their use in language that matters.

However, what about the referential and descriptive roles of language? Wittgenstein of course acknowledges that many words in a language are used to refer and that often we use language to describe states of affairs in the world; but, it is a mistake to suppose that that these are the only important functions of language and that meaning is entirely, or even primarily, a matter of reference and truth. For Wittgenstein, an important type of language game is the one in which we describe objects and states of affairs in the world. In fact, there are a number of language games in which we do this. However, the mistake is to think that that is all we ever do, or even the main thing we do, with words. The mistake, according to Wittgenstein, is to suppose that the describing function of language—the language game of giving and exchanging descriptions—is the essence of language, and all other uses of language are to be determined by, or are in some way, derivations or departures from that.

Now, correlated with his conception of language is a certain conception of philosophy: philosophical problems typically arise when language is no longer functioning in its normal way, when we prize it apart from its normal use and examine it in abstract. For example, when we are actually using the word “cause,” there is no problem about the concept of causation. There may be factual disputes about the cause of the current recession, or the cause of a car accident, but the notion of causation is not itself problematic. We find it easy to use the word “cause” in describing the cause of an accident, or the cause of a disease, or the cause of the economic downturn. In real life, there is no special philosophical problem. The problem arises when we take the word “cause” apart from the language game where it has its home and try to find the essence of causation. We try to find it in “necessary connection,” or “constant conjunction,” to use the Humean terminology, but the mistake is to suppose that there is some essence we are looking for. The way to resolve the problem about causation, and so on with most and perhaps all the other philosophical problems, is to see how we actually use these words in practice. When we examine them in practice, we do not address and solve the traditional philosophical problem. We simply dissolve it. We simply remove it by discovering the actual use of the word “cause” and how it functions in actual human discourse. We remove the urge to think there must be some underlying essence, some essential feature of causation which underlies our use and gives the word “cause” its real meaning. There is no real meaning beyond the use; just look at language in action.

An immediate difficulty with the analogy between games and language is that with very few exceptions, games require language and, hence, cannot be used to explain language. Only a tribe that already has a language can have chess, football, rugby, baseball, poker, and other card games, because the elements of these games are linguistically constituted. I will come back to this point.

The Investigations, then, contains both a philosophy of language and a philosophy of philosophy. The conception of language generates a conception of philosophy. The conception of language can be put, leaving out various qualifications and details, by saying that we should not think of the meaning of the word as the object for which it stands. Nor should we think of it as some mental entity that flows through the mind when we use the word meaningfully. Rather, we should think of it as a matter of the use that it has in the language of which it is a part. This is put in the slogan “meaning is use.” The conception of philosophy that this generates is that the philosophical problems arise when we fail to understand the use of the corresponding words in the language. So, if we understood the use of the psychological words such as believe, think, know, guess, suppose, wonder whether, intend, and so on, the philosophical problems about the mind would simply dissolve. The solution to the philosophical problems consists in giving an accurate description of the language games in which these words are being used. You do not ask, “What is an intention, really?” Rather, you ask, “What are we doing when we say that somebody intends to do such and such?” and that is really a way of asking, “What is the language game we play with ‘intend,’ ‘intention,’ and other related words? What do we do with the words in question?”

We can illustrate the distinctive features of Wittgenstein’s method by considering the analysis of intention. Anscombe’s (1963) book on intention, a book that is an application of Wittgenstein’s methods, is a good example. The traditional philosopher, and I am one for the purposes of this discussion, would analyze intention by asking, “What sort of thing is an intention, anyway? What sort of phenomenon does the noun ‘intention’ stand for? And what sort of phenomena does the verb ‘intend’ stand for?” The traditional philosophical task is ontological. Wittgenstein’s task is not, in that way, ontological. It is linguistic in the sense that it tries to examine the language game in which we use the word “intention,” and this will remove, or so his assumption goes, certain important philosophical mistakes. So, he asks, “How do we use the word ‘intention’?” And according to Anscombe, statements of intention give us an answer to a certain sense of the question, “Why?” If you ask, “Why did you do it?” or “Why did such and such occur?” there is one type of answer that is given by saying, “I intended such and such,” and examinations of these answers will tell us what we wanted to know about intention.

I will later on argue that the whole conception that Wittgenstein had is seriously limited. In this particular example, the fact that specifying intention can answer (one sense of) the question “Why did he do it?” is the beginning, and not the end, of the investigation of intention. It has to follow from the nature of intention that specifying the intention can answer the question, “Why?” but Anscombe’s application of Wittgenstein’s theory does not, so far, explain the nature of intention. I will say more about this later. However, though I think Wittgenstein’s theory is mistaken, it is not unintelligent. It is very powerful, and it cannot be removed swiftly. It follows from Wittgenstein’s conception of language that there is no essence of language. There are just countless uses of words and sentences, and we should not look for something that they all have in common which constitutes the essence of language. Consequently, we will not find any essential single method in philosophy, but rather families of therapeutic methods, by which we remove the urge to philosophical mistake by giving an accurate characterization of the use of the words in question. It is essential though to see that Wittgenstein thinks that the traditional philosopher’s obsession with representation, an obsession that he himself had when he wrote the Tractatus, reveals a misconception of how language functions. There is no essence of language, and representing or describing is just one kind of language game among countless others.

3. From the Tractatus to the Investigations

I now want to explain in a little more detail the development of Wittgenstein’s thought from the Tractatus to the Philosophical Investigations.

A key sentence in understanding the Tractatus, for me, is the following:

We must not say, “The complex sign ‘aRb’ says ‘a stands in relation R to b ’”; but we must say, “That ‘a’ stands in a certain relation to ‘b’ says that aRb” (Wittgenstein 1922, 3.1432)

What this means is that the sentence can only succeed in representing a fact if it mirrors the fact, if it has an isomorphic structure to the fact. A sentence of the form aRb can represent the fact that object a stands in a certain relation to object b, only because a stands in a certain relation to b, namely, they both flank the letter R. And the sentence, for example, “The book is on top of the table,” can mirror the relationship of the book and the table because the actual relation to being on top of is conventionally pictured by the expression “is on top of.” And that relation is a structural relation in the actual fact, so the structure of the sentence mirrors the structure of the fact. Of course, I have distorted Wittgenstein for the purposes of exposition. He would never give an example like, “The book is on top of the table,” because he thought that if we analyze such sentences, we would get the basic terms of the language and the basic facts in the world, and they would never be about such things as books and tables. He is never explicit about what they might be, but in any case, I just give this example as an illustration. Representation in language, and consequently in thought, is only possible because of an isomorphic relationship between the arrangement of the names in the sentence, the Satz, and the arrangement of objects in the fact, the Tatsache. Now, there are a lot of interesting relationships between the Tatsache and the Sachverhalt, but I will not go into these, because that is not my main interest here. Let us now turn to the Investigations.

For me, a key sentence in the Investigations is this one: “But how many kinds of sentence are there? Say assertion, question, and command?—There are countless (unzählige) kinds: countless different kinds of use of what we call ‘symbols,’ ‘words,’ ‘sentences,’ . . . ” (Wittgenstein 2009, §23). He then gives a list of several different uses of language, which I will come back to later.

The analogy between using a word and using a piece in a game, or between speaking and playing a game, is fatally flawed for the following reason: all the paradigm cases that we think of as examples of games, such games as football, baseball, soccer, chess, hockey, and card games, require language. You cannot explain language by appealing to games, because the games in question require language. You cannot understand the games if you do not understand language, because the constitutive rules of the game only exist if they can be linguistically expressed. The game can only exist if its elements are linguistically represented. You cannot have an element such as a touchdown, checkmate, or a goal unless you have some way of linguistically representing these phenomena and so on with the other crucial elements of games. The exceptions that one can think of to this are such simple things as to be marginal cases of games. If a small child is throwing a ball and catching it, is that a game? Let us concede that Wittgenstein is right with his “family resemblance” concept of a game and say that yes, the child’s activity is a sort of game. But, all the same, we cannot explain language with an appeal to such simple examples because the examples lack the sort of meaning structures that are typical of the use of words; I will come back to this point later. So, I think there is a fatal flaw in Wittgenstein’s basic conception. It may be that we are misled as to Wittgenstein’s intentions, and perhaps he was himself misled by assuming that “game” is a translation of the German Spiel. Spiel can, of course, mean simply “play.” I do not know how much the inadequacy of translating Spiel as “game” affected Wittgenstein’s thinking, but in any case, the point that I am making remains the same. If you think of paradigmatic games such as chess or football, these cannot provide a model for language, because the games themselves presuppose language. You cannot have such games without a language. Furthermore, the moves in such games do not typically represent anything, but I will argue that in the use of language, representation is a pervasive and essential feature. If, as I have been claiming, Wittgenstein’s overall conception of language is inadequate, what is a more adequate conception?

4. An Alternative Conception of Language

I have given an all too quick summary of some features of Wittgenstein’s philosophy of language, and I now want to present an alternative that I think is superior to either of his conceptions.

The key to understanding language and meaning is to see that human language is an extension of basic biological forms of intentionality, especially perception, belief, desire, intention, and memory. Linguistic meaning is derived intentionality. So, to explain the nature of language and meaning, first I have to describe pre-linguistic biological intentionality. ¹ A rather small number of crucial notions are essential to understanding biological intentionality. All these notions carry over to linguistic intentionality, so I will explain them briefly.

All these are features of biological intentionality, and they are common to the experiences of both ourselves and other higher animals. My dog has all these forms of intentionality. He can see that it is raining, believe that it is raining, and wish to go outside. So far then, you need the following notions that are crucial for language: psychological mode and propositional content, conditions of satisfaction, and direction of fit. Some, not all, intentional states have a special causal component in the conditions of satisfaction. They are, what I call, causally self-reflexive. So, the intention to raise my arm will be satisfied only if that very intention causes the movement of the arm, and my memory that I went on a picnic yesterday will be satisfied only if the fact that I went on the picnic causes the memory. The notion of causation in question is intentional causation. Furthermore, many intentional states are complex. So, for example, I do not just intend to pull the trigger; I intend to fire the gun by means of pulling the trigger. I do not just intend to raise my arm; I intend to vote by way of raising my arm. There are only two forms of complexity in the structure of actions. These are the causal by means of relation—I fire the gun by means of pulling the trigger and the constitutive by way of relation—I vote for the motion by way of raising my arm.

We need all these notions to explain human language. I am going to assume that many species of animals, as part of their biological nature, have intentional states with propositional contents and psychological modes, they have conditions of satisfaction and direction of fit, and they have causally self-reflexive intentional states, where there is a representing relation between the intentional state and its conditions of satisfaction. The question is how do we get from pre-linguistic intentionality of the sort I have described to linguistic meaning?

We need to distinguish between the intrinsic or observer-independent intentionality of actual mental states, such as perceptions, beliefs, and desires, and the derived or imposed intentionality that humans impose on sounds, marks, and so on. Humans have the capacity to impose intentionality on entities that do not have it intrinsically, such as sounds or marks. This observer-dependent, or derived intentionality, is created by conscious, original, observer-independent intentionality. The sentence has intentionality, but it is entirely derived or observer-relative. How does that happen? My question is not the evolutionary question of how it in fact happened in history, but the philosophical question: what is the logical structure of the move from pre-linguistic, intrinsic, observer-independent intentionality to observer-relative, linguistic intentionality, commonly thought of as linguistic meaning?

Let us suppose that we have beasts, more or less like ourselves, that have all these features of intentionality. How do they get meaning? What is the difference between raising your arm just to be raising your arm and raising your arm as a signal, let us say a signal to someone that they are to come toward you? This question is the essential question in getting an account of meaning that shows how meaning is a natural development from pre-linguistic forms of intentionality. Our question is, what fact about noises and bodily movements makes them meaningful in the semantic sense?

The answer to the question is this: the movement of the arm is a condition of satisfaction of the intention to raise it. However, if the movement is meaningful, the movement itself now has additional conditions of satisfaction. It has the condition of satisfaction that someone, your hearers or observers, should come toward you. This, I believe, is the essence of speaker meaning in general. Meaning is the intentional imposition of conditions of satisfaction on conditions of satisfaction. The speech act consists of the production of a sound, remark, or gesture, all of which are intentional, and if it is a meaningful speech act, then the intentional production that has conditions of satisfaction will now produce an entity that has further conditions of satisfaction. In the case of statements, the conditions of satisfaction are truth conditions. In the case of orders, they are obedience conditions. In the case of promising, they are fulfillment conditions.

So, now we have explained how from intentionality we get to speaker meaning. Speaker meaning is the intentional imposition of conditions of satisfaction on conditions of satisfaction. So far, we do not have anything as rich as a natural language, but we do have the first step in the development of a natural language. The next steps are to introduce the ideas of the communication of meaning and conventional meaning. Typically, when a speaker performs a speech act, such as in my example of raising the arm, he intends not merely that the movement of the arm has conditions of satisfaction, but he intends to communicate this speech act to the observer or hearer. His intention to communicate consists in the intention that the meaning intention should be recognized as such and that it is intended to be recognized. So, we get from meaning to communication by the common sense observation that what is communicated is meaning, and the intention to communicate is the intention that the meaning should be recognized. Furthermore, if the animals we are considering develop a practice of regularly using certain sounds or marks with a certain speaker meaning, the regularity will produce expectations of the sort that is characteristic of linguistic meaning. If the speaker produces a sound that has a conventional set of conditions of satisfaction, the hearer can reasonably assume that the sound is uttered with that meaning.

We get to the crucial notions of sentences and word meaning if we add something to the notion of speaker meaning. First, we add the notion of convention; the people communicating evolve conventional devices that have conditions of satisfaction. So far, these might be very simple conventional devices that were in effect one word sentences such as, “DANGER!” “FOOD!” and “RAIN!” However, a second point is crucial, and it is that the animal should be able to produce structures that articulate the capacity we have to distinguish objects and their properties. We can distinguish particular objects and say things about particular objects; “This one is on the left,” “That one is in front of me,” “That one is on the right,” and so on. In these cases then, we have something corresponding to reference and predication. And that is to say that when we introduce expressions for reference and predication, we have the capacity to impose meaning, not just on expressions, marks, or sounds that, like sentences, represent entire states of affairs, but on components of the sentences that correspond to references and predication. The introduction of an internal syntax of language means that we are off and running toward having something corresponding to actual human languages, where you have not just meaningful utterances, but you have meaningful utterances with an inner syntactical structure. So far, a preliminary account of language has contained the notions of speaker meaning, the communication of speaker meaning, linguistic convention, and the inner syntactical structure of sentences.

The conception of language that I am describing is a conception where the essential notion is that of a certain type of speech act that Austin baptized as “illocutionary acts,” and one that is quite contrary to Wittgenstein’s conception of language. On Wittgenstein’s conception, there are countless, or unzählige, uses of language. On the conception that I am beginning to describe, the possibilities are limited in number, and the limitations are set by the possibilities of meaningful utterances and sentence meaning. Meaning serves to relate utterances to conditions of satisfaction in various ways, and the various possible ways have to do with such things as the point of the utterance, the direction of fit of the utterance, and the intentional state expressed in its sincerity condition. So, for example, the point of a statement is to represent how things are. That is its illocutionary point. The direction of fit is word-to-world, and its expressed psychological state is belief. This is the sincerity condition. Given these three notions—illocutionary point, direction of fit, and expressed psychological state—it seems to me we can discover there are a very limited number of possible types of illocutionary acts, and here they are.

I am leaving out here lots of complexities for the sake of a brief and simple exposition. If, for example, you adjourn the meeting by saying, “The meeting is adjourned,” the sincerity condition includes the belief that the utterance adjourns the meeting, the desire to adjourn the meeting, and the intention to adjourn the meeting. For reasons that go beyond the scope of this article, I think Declarations are the representational device by which we create the primary elements of human civilization—money, property, government, marriage, and so on.

I am making a very strong claim. In the illocutionary act line of business, there are five and only five possibilities, and these are determined by illocutionary point, expressed psychological state, and direction of fit. They are Assertives, Directives, Commissives, Expressives, and Declarations. These are determined by the nature of mind and meaning. In addition to these five categories, there are other things we do with language that are parasitic or derived from these basic cases. The most famous examples of these are fictional cases where you go through the motion of performing a speech act, typically an Assertive, but you do not have the normal commitments of an Assertive. When you begin the story with, “Once upon a time, there was a girl named Little Red Riding Hood,” we know that you are not actually making an Assertion, but you are making a fictional Assertion. That is, you are going through the motions or pretending to make an Assertion. This is not a separate kind of speech act, but a variation on the previously existing kinds.

5. Answering Wittgenstein’s Challenge

With this simple apparatus in mind, let us answer Wittgenstein’s challenge that we cannot give a general theory of language. I will argue that all his apparently puzzling examples can easily be assimilated to the account that I have just provided. The crucial passage (Wittgenstein 2009, §23) reads as follows:

But how many kinds of sentences are there? Say assertion, question, and command?—There are countless kinds: countless different kinds of use of what we call “symbols,” “words,” “sentences.” And this multiplicity is not something fixed, given once for all; but new types of language, new language-games, as we may say, come into existence, and others become obsolete and get forgotten. (We can get a rough picture of this from the changes in mathematics.)

Here, the term “language-game” is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life.

Review the multiplicity of language-games in the following examples, and in others:

Giving orders, and obeying them—

Describing the appearance of an object, or giving its measurements—

Constructing an object from a description (a drawing)—

Reporting an event—

Speculating about an event—

Forming and testing a hypothesis—

Presenting the results of an experiment in tables and diagrams—

Making up a story; and reading it—

Play-acting—

Singing catches—

Guessing riddles—

Making a joke; telling it—

Solving a problem in practical arithmetic—

Translating from one language into another—

Asking, thanking, cursing, greeting, praying.

—It is interesting to compare the multiplicity of the tools in language and of the ways they are used, the multiplicity of kinds of word and sentence, with what logicians have said about the structure of language. (Including the author of the Tractatus Logico-Philosophicus.)

I think it is clear from the text that Wittgenstein regards this paragraph as decisive. It refutes the idea that there can be any simple, general account of language of the sort I provided, because the multiplicity of language is endless. We can endlessly describe existing language games, and we can endlessly invent new language games.

Wittgenstein presents this with the idea that, somehow or other, it is impossible to get a rational taxonomy of different uses of language or different types of speech acts. I do not think this shows that at all. Also, a priori, I think we should be suspicious. Why should language be so taxonomically recalcitrant? No one would say you cannot give a classification of different types of political parties, or different types of religions, and I think similarly, we can give a classification of different types of speech acts. Wittgenstein speaks as if it is easy to invent a new kind of language game, but I do not think it is. I used to ask my students to invent a new type of illocutionary act or a new language game, and typically the things they invented were variations on familiar types.

I am now going to accept Wittgenstein’s challenge and show how we can explain each of his examples. He presents this list as if it were decisive against any sort of account of language of the sort I have presented. I propose to go through each of his examples one at a time and show how they can be answered.

So, I do not think there is anything impossible about the challenge that Wittgenstein presents to us, and some of his examples are a bit like saying, “Well you cannot classify animals, because once you have all the animals classified, what would you say about teddy bears or glass animals or stuffed animals?” The answer is that those do not raise a special problem. You see them as derivative or parasitic on the standard case. The teddy bear is not a bear. It is a representation of a bear. Moreover, the fictional assertion is not an assertion, it is a pretended, or as if, or play-acting assertion. What is the problem supposed to be? I think if you adopt a naturalistic biological view of language of the sort that I have been advocating, then Wittgenstein’s skepticism about the possibility of a general account of language poses no insuperable difficulties. Each and every one of his examples can be answered.

6. The Analysis of Intention: A Test Case

In this section, I want to consider the discussion of intention to contrast my approach to philosophical problems with Wittgenstein’s. To begin with, I think there are certain things to be learned by the Wittgensteinian approach. I do not think Anscombe remarks on this, but it is an important fact about the verb “intend” that it does not naturally take the imperative mood. ² You cannot say to someone, “Intend to be kind to your mother.” The closest you could get would be, “Form the intention to be kind to your mother,” but then the activity here is forming the intention and not intending. This suggests what I think is true, that “intending” is not the name of an action. In general, actions are something you can be ordered to do, and the corresponding verbs take the imperative mood. However, “intend” does not take the imperative mood. That is a deep point, and I will come back to it later on. The rival view that I would present is as follows: intention is a type of intentionality. And specifically, there are two types of intentions, which I call “intentions-in-action” and “prior intentions.” The distinction is between my intention to do something prior to doing it—such intentions are typically plans—and then there is the intention I have in which I am actually doing it, what I call an “intention-in-action.” The closest to ordinary English for the intention-in-action is “trying.” The proof that there must be a distinction between the intention-in-action and the prior intention is that the conditions of satisfaction are different. If I have a prior intention to raise my arm, a condition of satisfaction is that the prior intention causes the whole action of raising my arm. If I have an intention-in-action of raising my arm, then the conditions of satisfaction are that the intention should cause the arm to go up. The whole action consists of an intention-in-action plus a bodily movement, where the bodily movement, as caused by the intention-in-action, is the condition of satisfaction of the intention-in-action. The prior intention represents the whole action, and the intention-in-action presents the bodily movement. So, if I intend to raise my arm, then I have, so to speak, a plan to raise my arm. However, when I am actually raising my arm, and I do so intentionally, the fact that corresponds to the adverb “intentionally” is the execution of a plan, the actual case of trying to do what I intended to do. It is possible to have a prior intention that you do not carry out: you do not try to do the thing you intended to do. And, of course, it is possible to have a lot of intentional actions for which there is no prior intention. Furthermore, intentions are causally self-reflexive in that the intention is satisfied only if it functions causally by way of intentional causation in producing the conditions of satisfaction. Moreover, as I mentioned briefly earlier, human intentions are often complex. I do not just raise my arm, but I vote for the motion by way of raising my arm. I do not just pull the trigger, but I fire the gun by means of pulling the trigger.

Now, let us return to Anscombe’s discussion. The reason that by specifying a person’s intention we can often answer a certain sense of the question “Why?” derives from the nature of intentions. In the case of a complex intention, the complex will typically be such as to achieve a certain effect, by means of or by way of bodily movements. So, for example, if we ask “Why is that man moving around like that?” one answer might be “He is sharpening an axe.” In stating his intention to sharpen an axe, we have given an explanation of his behavior as to why he is making those movements. The complex intention is “I intend to sharpen an axe by means of making these axe-sharpening movements.” The reason that giving the intention answers the question “Why is he doing it?” is that the purpose of making the movements is to bring about the sharpening of the axe, and the question “Why?” asks for a statement of the intended purpose.

7. Why the Project Failed

Now that seems to me to be a much deeper analysis of intention than you get by the Anscombe language game analysis, but I mentioned all this because I now want to ask the question of why the project failed. I want to say that the reason the project failed is that though Wittgenstein is right in that describing and representing is only one type of language game among others, the distinctive feature of the biological approach to language is, essentially, representational content. That is, a propositional content occurs in almost every type of speech act. There are a few speech acts where it does not occur, as in when you greet someone or you issue an exclamation—“Hi!” “Ouch!” “Damn!” and so on—but the other cases of asserting, describing, explaining, ordering, commanding, requesting, apologizing, thanking, congratulating, and even declaring war and pronouncing someone husband and wife all have propositional contents. Why is this important? Because it means that the representational content, or propositional content, is a feature of just about every type of speech act. There is no separating speech acts from representing. It is not that there is just one speech act of representing, and then a whole lot of other speech acts, but just about every type of speech act involves representation. In the classification of speech acts that I gave, with the exception of some Expressives, every speech act involves a representational or propositional content. In Directives, you have to represent what it is you are trying to get the hearer to do. In Commissives, you have to represent what you are committing yourself to doing. In Assertives, you have to represent what it is you are claiming is the case. In Declarations, you have to represent the state of affairs that you are bringing about by declaring it to exist.

Now with this in mind, it seems to me that we can see a limitation in both of Wittgenstein’s philosophy of language and his conception of philosophy. As far as his philosophy of language is concerned, his account is impoverished. We do not get a general theory of language and its functioning on a Wittgensteinian account. We assemble, as he says, reminders for specific purposes, but as far as theories are concerned, I want something more than a reminder. I want a general account, and he provides no reason as to why we cannot have one. The limitations of his philosophical method are that they do not go far enough. In the end, we want answers to the ontological questions: what is the nature of causation, what is the nature of knowledge, what is the nature of truth, and so on. Wittgenstein gives us important preliminary steps in answering these questions, but he does not go all the way.

This account has no doubt been too swift and too brief, but the general features, I think, do admit of a reasonably clear summary. Wittgenstein’s account of language is inadequate because he fails to see that language is basically an extension of pre-linguistic forms of biological intentionality. Given language, intentionality becomes vastly more powerful than it can be in pre-linguistic forms. My dog can think there is someone at the door, but he cannot think that there are 17 people at the door and wish that they would all come back next week. Now, once we see that language is an extension of pre-linguistic forms of intentionality, then it is possible to state a general theory of language. (I have tried to do this in various writings [Searle 1969; 1979].) Once we state that then we can see that it is simply not the case that there are an indefinite or infinite number of different types of speech acts. If you take the illocutionary point of the speech act as the basic phenomenon, then there are a very limited number of things we can do with language. There are five and only five basic types of speech act, and this limitation, to repeat, is set by the possibilities of meaning, both speaker and sentence meaning, and those possibilities are themselves set by the nature of intentionality.

The limitations in Wittgenstein’s philosophy of language produce limitations in his conception of philosophy. On his account, we should not think of the philosophically troubling words, such as “meaning,” “truth,” “causation,” and “intention,” as representing phenomena in the world. We should think of them rather as pieces in language games, where we understand the piece when we understand the sort of moves it can be used to make. I am arguing, on the contrary, that all the examples I gave of the generally puzzling philosophical phenomena represent real phenomena in the real world, and we cannot evade the philosophical responsibility of trying to analyze those phenomena just by citing the sorts of moves we make in the language games with the corresponding words. Once you have said that specifying an intention answers a certain sense of the question “Why?” then you need to go on and explain exactly how it works. What exactly is it about the phenomenon of intention that makes the specification of that phenomenon explanatory of certain forms of human behavior? That seems to me to be the next step.

As a rival conception of philosophy, I want to propose the following: there is a single overriding question in contemporary philosophy concerning how we can make the human reality consistent with, and indeed a natural consequence of, the more basic reality as described by such sciences as physics, chemistry, and evolutionary biology. We know that the world consists entirely of physical particles in fields of force and that these are typically organized into systems, where the boundaries of the system are set by causal relations. However, in addition to that we have a human reality of consciousness, intentionality, language, ethics, and aesthetics, not to mention such social phenomena as government, property, money, and marriage. The fundamental philosophical question today is how do you get from the basic reality of physics and chemistry to the human reality? Moreover, it is not enough for us to show that the two are consistent. We have to show that the human reality is an extension or a natural consequence of the basic reality. Once you have electrons and protons, it is natural that the systems of which they are parts will eventually evolve elections and presidents. The account of how the human reality is a development of the basic reality seems to me to be the philosophical project that, in one way or another, a very large number of us are engaged in. And, indeed, I have been engaged in this project all my philosophical life, for several decades, even before I became aware that that was what I was doing. In this project, many of the traditional philosophical problems remain, but they are now treated naturalistically: consciousness, rationality, ethics, and political obligations are all parts of nature like digestion. I think Wittgenstein gave us an important preliminary to undertaking this project, but he did not himself address the project, and it seems to me that in that respect, his philosophy is essentially limited.

8. Some Long-Term Effects of Wittgenstein’s Philosophy

The long-term effects of the work of a major philosopher do not always lie on the surface in the form of footnote references and citations and explicit arguments for and against. Often, the effects of a philosopher or even a philosophical movement have to do with changes in the whole mode of philosophical sensibility. I think Wittgenstein in particular, but indeed the whole approach to philosophy of which he was the single greatest influence, has produced long-term effects in philosophy. For example, when I travelled on the European continent in the 1950s, a lot of the philosophical discussions then prevalent seemed to me like so much hot air. So, for example, in Germany, I was taught by a man referred to as Herr Professor Doktor Heyde. He was primarily concerned to prove that Bewusstsein ist Freiheit (consciousness is freedom) and other such abstract generalizations. His advice to his students was to constantly search for Allgemeine Wahrheit (Universal Truth). And about Allgemeine Wahrheit, he cautioned us vielleicht nicht finden, aber immer suchen (maybe not to find but always to search). And in France, typical philosophical questions were posed in very abstract forms even when they were supposed to concern practical issues. A typical question was, “Is it possible to have a socialism without a humanism?” Such approaches to philosophy became out of the question in English-speaking countries. And even in France and Germany, I do not hear this sort of talk much anymore. Wittgenstein did a great deal to bring a sense of reality to philosophical discussion.

One of the organizers of the conference, Nigel Pleasants, pointed out to me that some of my own work is very much in the Wittgensteinian spirit. So, for example, my reconciliation of the minimalist theory of truth with the correspondence theory of truth (Searle 1995) effectively consists in showing that the language game we play with disquotation, “‘Snow is white’ is true if and only if snow is white,” and the language game we play with correspondence, “‘Snow is white’ is true if and only if it corresponds to the fact that snow is white,” are not inconsistent with each other. Disquotation and correspondence do not express rival theories of truth; this can be seen if you examine the actual uses of the words in question. This dissolution of the problem is very much in the spirit of Wittgenstein’s project of showing that the philosophical problems dissolve if you understand the functioning of the language-games. I was not thinking of Wittgenstein when I did this analysis, but I think it is, as Nigel Pleasants says, very much in the spirit of Wittgensteinian philosophical method.