DRAFT
Logic, meaning, action and speech
Peter Widell
Department of Communication and Culture, Aarhus University
1. Introduction
In this paper I will look at the speech act concept departing from the study of the concept of instrumental action and the concept of social institution. It is my contention that the phenomenon covered by these three concepts are intimately related, and that their interrelation is crucial for the constitution of the basic speech act types. Nevertheless, they must not be intermingled.
In a previous article (Widell 2001), I have stated that there exists only one speech act of illocutionary type, namely the assertive speech act. This statement I still adhere to. But where, in Widell (2001), I have focused on the speech act itself to make my point, I will in this article more closely look at the speech act as a construct on the basis of simpler constituents. If you do not understand these simpler constituents and their special composition, you do not really know what a speech act is.
One of my main motives for my approach is the desire to reach a decent ‐ nominalistically thought out ‐ meaning theory for expressed thoughts or propositions, a meaning theory that is able to reflect the underlying logic behind action and speech in general. To get it in place, however, we have to know a little bit about how logical analysis and meaning theory interact. In the following, I will, therefore, starting from a common, but erroneous view of what logic and meaning consists in, work my way towards a semantic elucidation of logic (sections 2 and 3) and then locate the true place of logic in thought as it is embodied in human perception and instrumental action (section 4). Based on this pragmatically coined concept of truth and meaning, I will, then, construct the basic speech act types: First and foremost, assertive as a meeting place between the world of instrumental action and the social world (section 5). Next, I will look upon the two necessary ramifications of the assertive speech act, normative and constative (section 6), the perlocutionary acts commissive, directive and informative (section 7), and finally the metaassertive (section 8).
2. Logic, sentence and truth
One of the great changes in philosophy in the 20th century has been the showdown with the classical conception of conceptual meaning as we know it from e.g. Plato, where semantic content has been regarded as a kind of forms or ideas in a self‐subsisting realm beyond the physical world. The confrontation with this concept of concept‐ or meaningrealism, as the position is often called, has usually taken place within the framework of a nominalistic conception of the meaning‐concept, where meaning is defined in terms of the far more well‐understood concept of truth. Here it is believed that knowing the meaning of a sentence is the same thing as knowing the conditions in the world that must exist if the sentence has to be true.
This attempt to understand meaning in terms of truth, however, has often been treated from a false or misleading understanding of logic, where logic is seen exclusively as inference. This is exemplified in formalism or meta‐mathematics, where there has been attempts to understand inferences in for instance geometry as purely syntactic manipulations of sentences according to certain calculation rules (Hilbert 1899). Thus, a sentence q is true according to formalists, not so much because it has anything to do with the world, but because qʹs truth can be calculated or proven within the framework of a comprehensive deductive system from a limited number of obvious or otherwise redeemed axioms. This syntactical solution is for several reasons unsatisfactory. If nothing else, because it neglects both the intuition of truth as correspondence between sentence and the world and the concept of the sense of a sentence as its truth conditions, that is conditions in the world, which will make the sentence true. Opinions here are rather seen as a place in a system that implicitly defined sense: The form contains the meaning.1
Against this syntactical conception of logic stands, as we have seen, the semantic conception of logic as build on truth as correspondence between thought and situations in the world. According to this view the syntactical response to the meaning problem is a pseudo‐response. Meaning is, of course, not syntax, and it cannot be syntactically defined. On the other hand, the challenge is admittedly still to avoid meaning realism. But for that you do not need syntacticism. Here it suffices to find the linguistic meaning in the relationship between a sentence and some actual situation in the world.
The semantic logic of thought as it appears in perception and action has an equally long history as formalism. The pioneer here is Gottlob Frege (2002 (1892)), and this conception of meaning reaches its preliminary peak with Alfred Tarski (1932, 1944) and his attempt to define truth and meaning for a given language through a formal recursive enumeration of all pairings of the given languageʹs sentences with their truth conditions under the general form: ʺʹpʹ is true, iff pʺ.2 This form Tarski baptized Convention T. A crucial point – showing the superiority of the semantic starting point – is that although the point of departure is not the inference as in the syntactic definition of logic, what is considered as one of the greatest achievements in syntacticism, first order predicate calculus, can easily be obtained from starting to pair sentences with their truth conditions. You get the same result as in syntacticism, in spite of this crucial difference in starting point and approach: While logic within the syntactic conception of logic – qua syntax – is seen as just one realization of the possibilities a syntactic base enables and therefore is given a conventionalistic interpretation – see Rudolf Carnapʹs famous dictum ʺIn logic there is no moralityʺ (Carnap 1934, section 17) – the semantic conception of logic leads to the opposite: here logic appear as compelling. Or, as Wittgenstein, who inherits the semantic viewpoint on logic from Frege, notes: ʺLogic is transcendental.ʺ (Wittgenstein 1921: 6.13) Although you can represent logical inference by a syntactical calculus, it will only gain status as a genuine inference – and not just a syntactically simulated inference – when it is assigned sets of truth values, i.e. a semantic: logic is a semantically defined structure.3
3. Assertion, proposition and truth
Semantics, as defined by Tarski, has also its problems: It is true that we can ‐ with some important reservations Tarski himself discusses ‐ recursively enumerate all sentences in the language of the form ʺʹpʹ is true iff pʺ (and thus also indirectly all meaningful sentences in the language). But we are, however, only able to do it for what Tarski calls ʺopenʺ languages, i.e. languages that does not include the language that the words in the form “’p’ is true iff p” itself comes from, namely in the last instance the form of natural language. Here we must know the meaning in other ways. However, you can go in another direction. This Donald Davidson has tried to do (Davidson 1967). Instead of seeing Tarski’s Convention T as the basis for a formal recursive enumeration of all sentences in an artificially delimited language, Davidson accepts that the words ʺis trueʺ in Convention T cannot be defined a la Tarski, but rather must be assumed to be known in beforehand by the language user. We have, thus, a pragmatic starting point for recursive enumeration of all language sentences: Convention T is now seen as an implicit recipe for the construction of true (and false4) sentences. It is crucial that language will no longer be understood simply as true or false sentences, but as sentences used in specific situations to give us a truth value and, accordingly, a meaning and a logic.
Such situated sentences are called propositions: the sentence ʺIʹm sick,ʺ said on a given occasion to Hans by Grete expresses the same proposition as the sentence ʺYouʹre sick,ʺ said on the same occasion to Grete by Hans.
4. From proposition to thought: the logic of instrumental action
Perhaps we should dig a little bit deeper. Understanding is both within the syntactic and the semantic understanding of logic tied to a system of expressions, which in the first place enables us to talk about ‐ a syntactic or semantic ‐ understanding. However, it remains to be answered whether logic is a logic of language at all. Whether a thought attached to a sentence need such a sentence in order to be revealed. In a late age Frege becomes more and more convinced that the medium of thought is not language, but rather perception and action. According to Frege (1997 (1918‐1919): 329), we must carefully distinguish between ʺ(...) (1) the grasp of a thought ‐ thinking, (2) the acknowledgement of the truth of a thought – the act of judgment, (3) the manifestation of this judgment ‐ assertion.ʺ Or to give an example: We must distinguish between (1) the general thought of a glass of water on a table, (2) the awareness in a particular situation that there is a glass of water on a table, and (3) the indication in front of an interlocutor that there is a glass of water on the table by saying ʺThere is a glass of water on the table.ʺ (3) involves language. But according to Frege, we already find logic in (1) and (2), in (1) as an inference and in (2) as a semantic assignment of a truth value.
This idea of extending logic to include perception and action may very well be said to be more or less hidden assumption in modern speech act theory (Searle 1983, 2001, Bratman 1987, 1999; Velleman 1989, 2000), cognitive science (Thagard 2007a, 2007b) and implicature theory (Grice 2001 (1975)). It has also been one of the incentives behind the so‐called neo‐fregeanism (McDowell 1977, 1984, 1986; Evans 1982), which I will dwell on below.
Unfortunately, (1), that is the general thought that there is glass of water on a table, represents a disregard for the nominalistic ambition to overcome meaning realism: (1) creates the impression that thought can exist in pure form regardless of its situation anchorage, and therefore without any specific truth value. Precisely this point has neofregeans sought to correct by claiming that statements without basis in the objects they are trying to fixate, are meaningless. Actually, they are attributing the viewpoint to Frege himself (Evans 1982, Chapter 1; McDowell 1984; 1986). However, in a meticulous Frege-exegesis Geirson (2002) tries to demonstrate that it cannot be Fregeʹs view, and that Frege is right against the neo‐fregeans. I do not think Geirson is right, however, at least not without having my reservations. Geirson discusses objectless thoughts only in connection with language use. Here I think he has ‐ at least partly ‐ right in his criticism, which I will elaborate in the next section. However, I agree with the neo‐fregeans that there are no object‐free thoughts.
Let us look at our example about drinking a glass of water: Even a simple action like drinking a glass of water involves inferences. First and foremost, the action involves the following premises: (A) a desire or goal, here to drink a glass of water, (B) a perception of something in the situation, namely that there is a glass of water on the table (C) general knowledge about which means you should use to achieve your goal, in this case that you can quench your thirst. By taking the glass and bring it to the mouth. In addition, we also have (D) the conclusion that is the use of the means to achieve the goal mentioned in premise (A), in our example the bringing of your glass to your mouth. Most importantly in this context is that we do not have any linguistic formulation to make: the inference is action‐immanent, embodied. But it also applies to the thoughts it involves.
The important thing is the thought in (B), where we have the general grasp of perceiving something in the situation incorporated in a judgment: Here we have the equivalent of supplying a sentence with a truth value: the agents perceiving of the glass on the table corresponds to a situation where the agent, if he had wanted to formulate this part of his action in words, would have assigned the proposition ʺThe glass on the tableʺ the truth value true. But notice again that nothing has to be formulated linguistically in the execution of the action described by (A) (B) (C) and (D). All thoughts are embodied. Special attention should be paid to the fact that the condition for having these thoughts is that they are directed towards objects which can be identified in time and space. It may be that the agent is hallucinating or confuses things in the situation, or that he merely feigns his action. In such cases, the action should be characterized as a failure. But this simply corresponds to the falsity of the phrase ʺthe glass is on the tableʺ. It does not alter the fact that action ‐ and the thought contained in it ‐ only finds existence if it seeks an object that exists in the world and if necessary can be identified by others.
Therefore, neo‐fregeans are right when it comes to instrumental actions. In connection with instrumental action the thought necessarily implies an object: here, thought has to point towards truth.5
5. Speech: Thought and its social manifestation
Something quite different, however, apply when we are dealing with language. Here, in connection with language I can only agree with Geirson when he says that thought can manifest itself in objectless form. But what is language? Of course, we have already seen some suggestions above: Language is syntactic structure (Hilbert), language is semantic structure (Tarski) and language is an implicit recipe for producing true (and false) sentences (Davidson). But there are other suggestions. First and foremost, the suggestion we are familiar with from speech act theory. Here, language is a particular form of social action.
Central in this suggestion stands the illumination of the so‐called performative action, exemplified by acts such as to baptize, to bet, to close a meeting, to declare war and similar acts (Austin 1971a; Austin 1971b (1958)). The special feature of this type of behaviour is that it rather than relate to antecedently existing facts themselves creates facts: To say ʺthe meeting is closedʺ is not so much to relate to an antecedently existing meeting as it is to create the fact of the closing of a meeting (a fact you then, obviously, can relate to as a fact in another speech act, e.g. “I/he opened the meeting five minutes ago”). These facts, Searle (1995) has dubbed social facts. The characteristic of these facts is that they are based on what Searle calls collective intentionality6 and on the possibility for each and every individual in a social group to create and to act upon set of construction recipes or ‐ as Searle inspired by John Rawls and Immanuel Kant (Searle, 1971 (1965): note (c)) have called them – constitutive rules of the form ʺʹX counts as Y in context C ʺ, i.e. rules where any expression ʹXʹ amongst the members of the social community has been provided with some content, Y. Money transactions, marriages, coronations, baptizing acts, bets and signatures are all social facts governed by such rules. This concept of a rule‐supported social fact (Searle 1965; 1969) Searle has since sought to deepen and clarify. As could be seen, the form of the constitutive rule has a certain similarity with Tarski’s Convention T. It is no coincidence, as we will see.
From Searle (1995) and onward Searle talks about endowing an expression ʹXʹ with a content Y as endowing it with a certain social status, backed by a set of sanctions. Consequently, Searle speaks of them as status functions. They would then, according to Searle, consist in imposing on an arbitrary artefact or an arbitrary action (created by a primary intention) a secondary intention: A bank note is an artefact upon which we in an economic transaction impose the status of being exchangeable with a certain amount of products. A handful of water poured over a child’s head is an act upon which we impose the status of being an act of inaugurating the child in the Christian faith, etc.
But here we find the crux that some of these secondary intentions just might be thought contents from the sphere of instrumental action. Therefore, it is possible immediately to connect Searle’s constitutive rules with Tarski’s Convention T and more accurately state what a language is: A language is a set of arbitrarily coined artefacts7 which by virtue of a set of constitutive rules is used in certain situations to carry action‐immanent thoughts. Thus, the phrase ʺthe glass is on the tableʺ is an artefact that is able to carry the action‐immanent thought that there is a glass on the table, such as it is embodied in the act of drinking a glass of water. But itʹs the same as indicating the relation between the sentence and its truth conditions through the application of Convention T: ʺThe glass is on the tableʺ is true if (we can see that) the glass on the table.
That’s why we can now state what the making of speech acts consists in: it consists in binding action‐immanent thoughts to arbitrarily chosen artefacts through the use of social or communicative versions of Convention T. The fact that we, as Searle formulates it, are able to ʺimpose intentionality on intentionalityʺ, makes it possible for us to speak about objectless thoughts as Geirson wants it. Through a first time binding of a particular action to a particular artefact ‐ which requires that the act must not be devoid of purpose ‐ we can ensure that we in next‐situations can make ourselves clear about those acts and its components, although there aren’t any instantiations of the act in the speech situation. For example, it is now possible for me to use the phrase ʺthe glass is on the tableʺ meaningful, although there are no objects in the immediate situation that makes the sentence true. That does not mean that the meaning would then no longer be bound to the speech situation. But it does mean that it only has meaning in virtue of the fact that there has been a baptizing (Kripke 1980 (1972)), inauguration (cf. Kamlah & Lorenzen 1978) or norming situation (for expression‐bound meaning) prior to the speech situation where the sentence ʺthe glasses are on the tableʺ has been made situation independent bearer of the thought that some glasses is standing on a table.
6. The logic behind speech: normative and constative
The distinction between norming situation and use situation is crucial for the existence of object‐free thoughts, whether they are encoded in a homogeneous medium or not. The norming situation is the needleʹs eye through which we transform material which can only be understood in the instrumental act world into something that can also be understood in the speech act world. And the transformer is the constitutive rule form, ʺ‘X‘ counts as Y in context Cʺ. But it will in fact say: Concrete semantic rules are conditional imperatives of the form: ʺTo understand utterances in the speech community you must use a specific ʺXʺ in the same way in use situations like you did in the baptizing or norming situation.ʺ8
Only to the extent this communication ethical norm is respected, linguistic understanding is possible.9
This gives us the following constructive derivation of the first general speech act categories: THE FIRST AND MOST IMPORTANT SPEECH ACT is the creation of an artefact – and when it is language we are dealing with, we call it, of course, a linguistic expression – through which the speaker is making it clear for the hearer that he assigns a truth value to a specific thought attached to a piece of instrumental action. Is such a communicative intention present, we will talk about the truth ascribing speech act as an assertive speech act, or for short: an assertive. Because the assertive contains (i) a true proposition in accordance with (ii) a constitutive rule, borne aloft by (iii) a communication ethical norm, I have elsewhere proposed the following definition or analysis of the assertive: To assert p is the same as declaring yourself (cf. ii) committed (cf. iii) to the truth of p (cf. i).
In connection with the first assertive, we cannot presuppose the existence of a propositionally articulated language supported by constitutional rules. But we always have to presuppose the assertive speech act itself. It’s a transcendental condition. Now, one could in principle be content with a language consisting only of ʺoccasion meaningsʺ (Grice 1971 (1957)), i.e. meaningful content limited in scope to the speech situation. But in such cases meaning would degenerate to truth. If, however, we aspire to exceed the physical limitations of the speech situation with the ambition to speak with understanding also about the absent and the mere counterfactual, we have to know the distinction between norming and using. Or as we also can say it: what we must know is a distinction between the normative where we can determine the sentence meanings, and the constative where we can make use of already prescribed sentence meanings. But this is THE NEXT TWO AND SECOND MOST IMPORTANT SPEECH ACTS in the language. Here it is in a certain sense as meaningless to talk about what is the more important of the two. The normative must, of course, be the first act. But the normative gets its importance only through the subsequent existence of the constative. In other words: both are equally important for building linguistic semantic contents.
As can be seen, the concept of propositional meaning is introduced in compliance with all nominalistic precautions. But let us take a look upon how we have succeeded with regard to our ambition to avoid meaning realism?
It is clear that in the space between normative and constative there must necessarily exist a concept of meaning identity: the truth condition, we associate with ʺThis is a glassʺ in a use situation have to look like the truth condition, we associate with ʺThis is a glassʺ in a norming situation. But that does not mean that there is a concept of ʹglassʹ behind the two situations that could give some metaphysical guarantee a la Plato for this truth conditional identity. The concept ʹglassʹ is known only through its application instances. This means that the concept is only as stable as our continued use of the word can provide and guarantee. Stability is pragmatically founded. What is central is that we will always have a norming situation to relate to if there is a dispute about a particular semantic content. But this situation is a singularity, not a universal. The concept ʹGlassʹ is an extract of the empirical world.10
There will always be features in the norming and the use situations that transcend these situations, for instance our capacity to make this distinction itself11 – and besides that everything we have elucidated about truth, meaning, assertive etc. The unsurpassable existence of these features in speech contributes to point out that neither millianistic nominalism12 nor empiricism is viable options if we are to understand how language works: language is a normative context for synthesizing all the various situations in the world it relates to. Speech acts are principle‐ and maxim‐guided entities, not empirical facts.
7. The set of perlocutionary speech acts: commissive, directive and informative
As a segregation of the assertive, the normative and the constative are the two most basic speech acts.14 And it is so far the only set of speech act types which on the one hand satisfies the logical aspects of language and on the other hand meets its social aspects.
Classical speech act theory normally enumerates five disjunctively related, basic types of speech act, in Searleʹs 1975‐version the types assertive, commissive, directive and expressive and, finally, as a special type the declaration type of speech act (Searle, 1979 (1975): passim). This, I mean, is a major mistake.
For a start, the declaration is – as Searle himself later notes (Searle 1989:79) – not an independent speech act type, but an ingredient in any non‐verbal institutional act, an observation I have strengthened further by making it a part of any institutional act what so ever, including every illocutionary act. Further, if we put the expressive aside (cf. note 13), we are left with three types of speech act from the classical list of speech act types: the assertive, the commissive and the directive.
Concerning those acts, the mistake I am alluding to are not so much the neglect of not seeing all those speech acts as kinds of declarations. To treat speech acts in general as sorts of declarations was, on the contrary, one of Austin’s motivations behind moving from what he called the special theory of the performative to the general theory of the illocutionary act (Austin (1971 (1962):147). And as to this move, I can completely agree. No, the fault lies in considering the above three types of speech acts as lying on the same level – to consider them as disjunctive categories. They are not disjunctive. While in using assertives15 we seek an understanding built on knowledge of an underlying semantic rule to perform commissives and directives do not demand such a rule. These acts are not illocutionary acts at all, as is assumed in classical speech act theory. They are in fact instrumental actions.
In his general theory, Austin tried to define the illocutionary act in contradistinction to another type of linguistic act, which he called the perlocutionary act. Because of an unclear distinction between illocutionary and perlocutionary acts and because of a residual problem of definition of the performative/declaration, to which I will return in the next section, Searleʹs three speech act types are all wrongly classified as illocutionary acts. To my knowledge, they are perlocutionary acts, as I am now going to show.
An important ingredient in Austinʹs distinction between illocutionary and perlocutionary acts is that while illocutionary acts ʺhave some conventional conditional power ʺ (127), i.e. rely on some semantic conventions that must be understood by hearer, before the operation is successful, the successfulness of perlocutionary acts is more robustly dependent on the acceptance of what the speaker wants with him in the situation. We can also formulate it like this: While the illocutionary act belongs to the field of social action, perlocutionary acts belong to the field of instrumental action. But precisely this is not respected, if you ‐ like Austin and Searle ‐ believes that commissives and directives are illocutionary acts. Let us look, first, at the directive: A directive targeted against B is an act from Aʹs side with the intention to get B to do something, such as boiling water for coffee. Since its conditions of satisfaction can arise only if B actually does what A wants him to do, it is out of the question to see the act just as an understanding‐oriented act. Here it is not enough for A that B understands Aʹs intention. B must also accept to perform the desired action, i.e. draw the conclusions A wants him to draw. Therefore, a directive act has to be a perlocutionary act. But the same goes for the commissive act, e.g. a promise: If A promises to do something to B, it is not enough that B understands that A makes a promise. If Aʹs promise is to be successfully executed, it must include the acceptance of B to do something in the future. That is: the commissive must be a perlocutionary act.
For the so called directive and commissive illocutionary acts we therefore have to conclude that they are both, actually, perlocutionary acts.
But the analysis also indicates that there still has to be another perlocutionary act besides the commissive and the directive. We have, in fact, already prepared a place for it. The motive for executing an assertive speech act is normally not just that you want to obtain understanding. You also want to inform your interlocutor. That is, you also want your interlocutor to believe that what you are telling him is true. But that demands acceptance from the part of your interlocutor. Therefore, we must in addition to the assertive have an information‐conveying perlocutionary act, which we will, appropriately, call an informative.
This is then THE THREE THIRD MOST IMPORTANT SPEECH ACTS: The commissive, the directive and the informative. While the commissive and the directive are action‐guiding perlocutionary acts – the commissive aiming at having the hearer to accept an execution of a future act from the side of the speaker and the directive aiming at having the hearer to accept an execution of a future act from the side of the hearer ‐ the sole aim of the informative is to make the hearer accept a proposition as a piece of information. This is the options if you want to perform a speech act at this level.
8. Performative and metaassertive
If we look at the reason for putting the commissive and the directive in line with assertive in classical speech act theory, weʹll mainly find it in an erroneous understanding of the role of the performative/declaration in language. To declare an ʺXʺ for a Y is, on the one hand, a universal feature of social action. But only the assertive – in either it’s normative or constative form – is, on the other hand, able to accommodate this type of declaration. That will never be spotted in classical speech act theory. Here, it is assumed that the commissive and the directive each create a space for their own conditions of satisfaction different from the truth conditions of the assertive. If A says to B ʺI must ask you to leaveʺ as a devise for making a promise, the device is not, according to the classical speech act theory, used to assert that a promise is made. Rather, the promise is created. Therefore, the device has a different status as the device for making an assertive speech act. However, I believe this analysis is wrong. In line with among others Davidson (1979), I would rather see the relationship between proposition and the performative expression as given by reference: In ʺYou go. This is a requestʺ is this used in the second sentence to refer to the first sentence as it should be understood in its pragmatic context.16 Performative expressions can be realized in many ways in a sentence, but whatever form the performative expression takes it can as far as I am concerned only be an assertive or ‐ as the proposition, the assertive houses, not are about the world directly, but about the proposition, the sentence expresses in the speech situation – a metaassertive. Whether I am saying ʺI beg you to goʺ, using a so called performative sentence, or I am saying ʺgo!ʺ, using an imperative, I have – according to my analysis ‐ issued a metaassertive utterance.
It’s not impossible to imagine language without metaassertives. But why, then, use these kinds of speech acts? The answer seems to be quite simple: We have metaassertives because we want to avoid ambiguities in our propositions, not about the world, but about how we want what we say, rooted in the world of instrumental action. To build metaassertives it is always necessary in beforehand to have a stock of assertive utterances containing first order propositions about things and situations in the world. But as soon as we have built up a reservoir of such first order propositions, we can then start to build metaassertives to help us clarifying what speech act we intends in more complex situations: ʺThe door, my young man, will be closed in five seconds. This is an order. ʺ
9. Conclusion
The above account of the speech act area is an attempt to give the speech act a central place in human communication as mediator between the instrumentally anchored thought and the socially embedded proposition. I have tried to show that these two dimensions in human communication meets at a certain point, namely the assertion. Here, we find the necessary and sufficient means to build any language: Here we find logic, truth, meaning, action and speech.
Notes
1 The concept of implicitly defined meaning originates from Hilbert (1899, 1918), who defines it in a completely controlled and precise manner. In a vaguer sense syntacticism can be traced generally in twentieth century philosophy and science. Thus, we know it from Saussureʹs structuralism and from its misguided attack on what is called the nomenclature theory of meaning (Saussure 1983 (1916):16, 65), and we know it form a number of structuralist, post‐structuralist and social constructivist theories. What characterizes all of these attitudes to the syntax is that they see syntax as more or less constitutive of meaning. It is misconceived.
Another thing is that syntax through its indirect characterization of the semantics of logic can be an excellent means of keeping track of the semantics, as it is exemplified in computers. But syntax can never constitute or define the semantic content. For example, it is possible to build an admittedly primitive, but still in a semantic and pragmatic sense, fully functional language solely through the use of holophrastic sentences.
2 ”Iff” stands for ”if and only if".
3 Here the statement ʺpʺ is simply a name for the semantic tableau ʺ{S, R}ʺ and ʺp & qʺ a name for the semantic tableau ʺ{S, F, F, F}ʺ, gained from the input‐sets “{S,S,F,F}” (for p) and “{S,F,S,F}” (for q) respectively.
4 In logic ʺtruthʺ and ʺfalsehoodʺ are of equal significance. It is, however, not the case when we deal with action, perception and language. Here, ʺtruthʺ prevails because ʺtruthʺ pragmatically means information about the state of the world: it is the state of the world we are interested in when we draw inferences as agents. Falsity, we cannot really use.
In this paper, we will look only at truth with this pragmatic proviso amended.
5 Some might argue that we can only perceive the action of drinking a glass of water because we can deal with the phenomenon linguistically. Davidson has put forward such an argument (Davidson 2001 (1997)). I think Davidson is wrong. In many situations we can easily identify instrumental actions without intervention of language. The actions themselves supply us with all the relevant identification criteria (see Searle 2002:61‐76).
An interesting question is whether more complex actions can do with object‐free thoughts alone. Thinking about drinking a glass of water tomorrow is a more complex context than just drinking a glass of water. The usual argument is that it would be difficult to understand the first action without some form of linguistic mediation. Here, I can only agree. However, this does not shake the assumption that more basic forms of action are embodied in the sense here illuminated and thus does not poses the possibility of thinking without relating to publicly accessible objects.
6 This shall be understood emphatically: According to Searle, all attempts, including Grice’s (1957, 1968), to arrive at a notion of collective intentionality through a concept of pure instrumental action have failed. Space does not allow to dwell on this discussion, so I will simply refer to Meggle (2002) for further information. Unfortunately, we also have to skip other important features of the complex concept of a social fact.
7 As the artifacts are concerned, they must not be too transparent dependent on random features of those situations where we make use of them. You must have a relatively easy access to them. Or, as it formulated by Saussure, the linguistic sign is arbitrary (cf. Saussure 1983 (1916):67, 73).
8 This can be said to be a correspondence theoretical rule for (social) meaning the same way as Convention T is an even more fundamental correspondence theoretical rule for linking sentences to their meanings as truth conditions. Neither the first or second rule is ‐ as Austin have pointed out (Austin 1939) ‐ depending on any isomorphism between linguistic expressions and the world. This is pure correlations, so we can maintain a realistic epistemology (demanding an independently existing object for thought) as Frege and the neofregeans demand, and at the same time ensure a nominalistic language theory where the linguistic expressions ‐ pace Saussure – gets status of labels.
9 The claim that one should make use of (1) true statements and (2) meaningful expression corresponds to two of Grice’s four maxims of rational communication (Grice 1975), namely the maxim of truthfulness (quality) and the (communication ethically supported) maxim of correctness (Manner). What the other two maxims are concerned, they help to control the speech, but qua maxims they belong exclusively to the area of instrumental action.
10 The distinction is obviously expressing an idealization. In practice, you can, off cause, not recall or trace norming or baptizing situations for the words and propositions you have learned. But it is not the point either. The point is that in case of communication difficulties it is always possible to initiate a more reflective and thoroughgoing re‐norming of an already known word or to create a whole new word: ʺForm now on we will call this [pointing] ʺglass ʺ.
Scientific practice is largely characterized by an awareness of this distinction between norming situation and use situation. That goes, for instance, for Galileo’s and Einsteinʹs clarification of the distinction for crucial concepts in physics. (cf. e.g. Einsteinʹs reflections on the concept of simultaneity in the theory of special relativity). But also in everyday life you are now and then re‐norming language more or less explicitly along the distinction. Language adapts to situations, and occasionally leads to language change.
11 Being able to distinguish between norming situation and use situation are the same as being able to distinguish between analytic and synthetic statements. In practice, it may be difficult to distinguish between those two situations, so here the distinction reflects an idealization. The discussion of the distinction between analytic and synthetic ‐ or analyticity ‐ is one of the most important debates in contemporary philosophy of language (see Juhl & Loomis 2009). I have discussed the problem elsewhere (Widell 2008). Here I can merely say that I agree with the critics of the distinction that there is much to say about its ʺhardnessʺ, but it is not correct when one of the sharpest critics of the principle of analyticity Quine (1951) considers the principle to be without meaning. It is too fanatic to say that. On the contrary: it could easily be demonstrated, that the principle is vital and constantly active as a condition for keeping up language as a system of semantic intuitions. For instance, you are not in doubt that “Julie is the sun” can’t be true according to the meaning of the words.
12 Millianistic nominalism is the assumption, attributed to John Stuart Mill, that the meaning of a proper name is its bearer, and that predicates only have meaning by denoting the set of objects they relate to.
13 The fact that one person at a given moment performs a specific speech act is a fact, but the actual speech act qua speech act is not. Here the speech act is a normative guided phenomenon. The primary mode of the human language faculty is not to be an empirical fact. It is ‐ to express it as Kant did ‐ rather a condition of the possibility and validity of speech and language überhaupt (Kant 1999 (1781)).
14 In other versions of the speech act theory presented here, I operate with a further speech act type ‐ the expressive ‐ as an example of an assertive speech act. Here, I will simply refer to my discussion of the speech act type in Widell (2008).
15 They can of course be split into two. However, we will ignore this in this context.
16 This also implies that we can avoid an operator analysis of performative expressions. The metaassertive is an assertive and thus presupposes, accordingly, that what it concerns exists independently of whether the metaassertive is used for it or not, unlike an operator which operates on ‐ and thereby modifies ‐ what it concerns.
This analysis implies that we still will be able to find ourselves inside the framework of first order predicate logic ‐ and thus avoid meaning realism. Our assertives is META‐, but they are nevertheless about specific speech episodes in the world.
Literature
Austin, John L. (1939) Are There A Priori Concepts? Proceedings of the Aristotelian Society 18:82‐105.
Austin, John L. (1971a (1958)) Performative‐Constative. The Philosophy of Language: Oxford Readings in Philosophy.
John R. Searle (ed.). Oxford: Oxford University Press:13‐22.
Austin, John L. (1971b (1962)) How to Do Things with Words: The William James Lectures delivered at Harvard University in 1955. Oxford: Oxford University Press.
Bratman, Michael E. (1987) Intentions, Plans and Practical Reason. Cambridge, MA: Harvard University Press.
Bratman, Michael E. (1999) Faces of Intention: Selected Essays on Intention and Agency. Cambridge: Cambridge University Press.
Carnap, Rudolf (1937; 1. Edition in German 1934) The Logical Syntax of Language. London: Routledge & Keagan Paul.
Davidson, Donald (1967) Truth and Meaning. Synthese 27, 3:304‐323.
Davidson, Donald (1997) Mood and Performances. A. Margalit (ed.) Meaning and Use, Dordrecht‐Holland: D. Reidel Publishing Company:9‐20.
Davidson, Donald (2001 (1997)) The Emergence of Thought. Subjective, Intersubjective, Objective. Donald Davidson (ed.) Oxford: Clarendon Press:123‐134.
Evans, Gareth (1982) The Varieties of Reference. Oxford: Clarendon Press.
Frege, Gottlob (1952 (1. edition in German 1892)) On Sense and Reference (Translated by Max Black). A. W. Moore (ed.) Meaning and Reference. Oxford: Oxford University Press:56‐78
Frege, Gottlob (1967 (1. edition in German 1879)) Begriffschrift. J. van Heijenoort (ed.) From Frege to Gödel: A Source Book in mathematical Logic 1879‐131:
Frege, Gottlob (1956 (1. edition in German 1918‐19)) The Thought. Mind, No. 259:289‐311.
Geirsson, Heimir (2002) Frege and Object Dependent Propositions. Dialectica 56: 299‐314.
Grice, Herbert Paul (1989 (1948, 1957) Meaning. Herbert Paul Grice Studies in the Way of Words. Harvard: Harvard University Press:213‐223.
Grice, Herbert Paul (1989 (1975)) Logic and Conversation. Herbert Paul Grice Studies in the Way of Words. Harvard: Harvard University Press:22‐41.
´Hilbert, David (1968; 1. edition in German 1899) Grundlagen der Geometrie. Stuttgart: Teubner.
Hilbert, David (1918) Axiomatisches Denken. Mathematisches Annalen 78:405‐415.
Juhl, Cory & Eric Loomis (2009) Analyticity. London: Routledge.
Kamlah, W. & P. Lorenzen (1973) Logische Propädeutik: Vorschule des vernünftigen Redens. Manheim/Wien/Zürich.
Kant, Immanuel (1999 (1781)) Critique of Pure Reason. Cambridge: Cambridge University Press.
Kripke, Saul (1980 (1972)) Naming and Necessity. Oxford: Basil Blackwell.
McDowell, John (1977) On the Sense and Reference of a Proper Name. Mind 84:159‐85.
McDowell, John (1984) De Re Senses. Frege: Tradition and Influence. Crispin Wright (ed.). Oxford: Blackwell:98‐109.
McDowell, John (1986) Singular Thought and the Extent of Inner Space. Subject, Thought and Context. Philip Petit & John McDowell (eds.). Oxford: Clarendon Press:137‐168.
Meggle, Georg (ed.) (2002) Social Facts and Collective Intentionality. Dr. Hansel‐Hohenhausen AG.
Quine, Willard van Orman (1951) Two Dogmas of Empiricism. Philosophical Review 60:20‐43 (Reprinted in Willard van Orman Quine (1999 (1953)) From a Logical Point of View: Nine Logico‐Philosophical Essays. Cambridge, MA: Havard University Press:20‐46).
Saussure, Ferdinand de (1983 (1916)) Course in General Linguistics. Peru, IL. Open Court Press.
Searle, John R. (1969) Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press.
Searle, John R. (1971 (1965)) What Is a Speech Act? The Philosophy of Language. John R. Searle (ed.). Oxford: Oxford University Press:39‐53.
Searle, John R. (1979 (1975)) A Taxonomy of Illocutionary Acts. Expression and Meaning. Cambridge: Cambridge University Press:1‐29.
Searle, John R. (1983) Intentionality. Cambridge: Cambridge University Press. Searle, John R. (1989) How Performative Works, Linguistics and Philosophy 12.
Searle, John R. (1995) The Construction of Social Reality. London: Allen Lane, The Penguin Press.
Searle, John R. (2001) Rationality in Action. Cambridge, MA: The MIT Press.
Searle, John R. (2002) Animal Minds. John R. Searle Consciousness and Language. Cambridge: Cambridge University Press:61‐76.
Tarski, Alfred (1932) Der Wahrheitsbergriff in den Sprachen der deduktiven Disziplinen. Anzeiger Akademie der Wissenschaft Wien, Math.‐nat. Kl. 69:23‐25.
Tarski, Alfred (1944) The Semantic Conception of Truth. Philosophy and Phenomenological Research ‐ A Quarterly Journal 4, 3:341‐375.
Thagard, Paul (2007a) Cognitive Science. Stanford Encyclopedia of Philosophy. Edward C. Zalta (ed.). Stanford: http://plato.stanford.edu/entries/cognitive‐science/.
Thagard, Paul (ed.) (2007b) Philosophy of Psychology and Cognitive Science. Amsterdam: Elsevier.
Velleman, J. David (1989) Practical Reflection. Princeton, NJ: Princeton University Press.
Velleman, J. David (2000) The Possibility of Practical Reason. Oxford: Oxford University Press.
Widell, Peter (2001) Illokutioner er assertiver: Et bidrag til belysning af talehandlingens struktur og taksonomi. Augias 56‐59:15‐64.
Widell, Peter (2008) Talehandlingsteori. Unpublished manuscript.
Widell, Peter (2009) A Plea for a Simple Solution: What a Performative Theory of Illocutions Really Looks Like. Proceedings of the 22nd Scandinavian Conference of Linguistics, Aalborg, Denmark, June 19‐22, 2006. Hans Götzsche et al. (ed.).
Wittgenstein, Ludwig (2000 (1923)) Tractatus Logico‐Philosophicus. London, New York: Routledge.