Нов български университет

Anita Kasabova

 

The problem of speech acts as we know it owes its present state to Austin (1962) and Searle (1968, 1969, 1983) and concerns the performance of illocutionary acts and their rules. An illocutionary act is a linguistic act performed in speaking and Austin coined the notion of performatives in his analysis of the relation between linguistic statements and linguistic and/or non-linguistic actions. On his view, a speech act or performative such as a promise involves, on one hand, a securing of uptake and, on the other, an obligation for the speaker (and hence entails conventions). He considers illocutionary acts as “utterances which have a certain (conventional) force” (1962:109). On Searle’s view, the problem of speech acts is related to, or even based on, the assumption that linguistic conventions underwrite the rules enabling or ‘realizing’ the performance of illocutionary acts (1968:40). I think that these familiar accounts presuppose philosophical and linguistic theories that are less familiar today. This is why I propose to re-examine the relation between linguistic action and linguistic structure following the linguist, philosopher of language and Gestalt psychologist Karl Bühler (1934). In addition, I briefly present a speech act theory by the Munich phenomenologist Adolf Reinach (1913) who calls speech acts social acts, for in such acts an utterance is not only intentionally expressed and performed by a subject, but the act is directed towards another person or persons (fremdpersonal) (1913:39). In addition, this act needs to be grasped by an addressee (vernehmungsbedürftig), so that the achievement of a speech act involves a tendency to be grasped or directedness which Searle calls a relation of fit between a speaker and an interlocutor (1983:10).

1. SPEECH ACTS AS LINGUISTIC PRODUCTS

Why return to old theories? Surely not even a historiographic interest is adequately accounted for by showing that a certain problem has a history of 70 years or more? Rather, one might object, we should keep up to date with the latest developments in the field. But the question is, how is the present interest best served? In philosophy, the answer does not necessarily lie in the most recent upshots of a theory and an examination of a theory’s roots may turn out new insights and a better understanding of the problem. This method of inferring back from the consequences to the grounds is also applicable to the analysis of speech acts themselves. Austin, it seems, considered the non-linguistic aspect of speech act undervalued by linguistic theories at his time and hence emphasized the importance of what speech acts do. Nearly half a century later, today’s theories emphasize the pragmatic aspects of speech acts, such as the influence of context on declarative utterances and their informational effects, whilst disregarding the linguistic nature of these utterances (and discussing the non-sentential nature of the speech act). The linguistic nature of speech acts has in its turn become the object of increasingly specialized semantic theories concerned not only with linguistic meaning but with the truth-conditional relevance of declarative utterances, their logical form and syntactic matter, as well as their context-invariant features. The result is a dispute between pragmatics and semantics (examined in the collection of essays edited by Zoltan Gendler Szabó, 2005) increasing the divide between ‘what is said and/or done’ and ‘what is meant’. My hunch is that a sound understanding of the phenomenon of speech acts can be gained by examining some theoretical results obtained by Bühler in the 1930’s which, apart from a brief revival in the mid-80’s, seem to have been largely forgotten.

Rather than considering speech acts as disguised assertions or as a premise for inferences about conventions or non-linguistic actions, I follow Bühler’s methodology in considering speech acts as linguistic products or dependent parts of a complex linguistic structure. I argue that the performing statement which expresses and achieves the speech act is not a disguised assertion but a dependent moment or accident of the act. This moment can be distinguished (but not separated) from the speech act by drawing an inference from it to the linguistic action which produced it. In other words, I propose to analyze the speech act as the consequence of a linguistic action rather than its ground, by means of a regressive inference, from the product to the action.

1.1. THE GRAMMATICAL STRUCTURE OF SPEECH ACTS

But first a grammatical point should be noted, since speech acts are linguistic acts and hence have a grammatical structure: their components stand in morphological, syntactic and semantic dependence relations (as well as phonetic and phonological relations). Not surprisingly, Austin distinguishes between constatives or statements declaring something to be the case and explicit performatives or statements which are also actions. He then, however, withdraws that distinction (1962:121), so it would seem that constatives and performatives are both speech acts. If that is so, the notion of speech acts covers a variety of linguistic expressions to the point of ambiguity. Indeed, the statement: ‘there is a bull in the field’ can be intended and/or taken either as a warning or as a mere piece of information. But this ambiguity is a result of determining performative utterances without distinguishing between the speech act, the performing statement in which the speech act is brought about, the position of the speaker and his intentions, as well as the relations between speaker and interlocutor. Reinach explains that the social act (speech act) is not divided into an independent act-accomplishment and an accidental observation (Konstatierung) about it; rather, it forms an intimate unity of voluntary act and voluntary utterance. In this case the experience is not possible without the utterance, since the utterance is necessary for achieving the act’s indicative function (kundgebende Funktion) (1913:41).

Besides, from a grammatical point of view, there is a distinction between speech acts expressed by verba dicendi on one hand, and verba cogitandi and sentiendi, on the other. For the grammarian, as Bühler points out, speech acts are verbal expressions or declarative utterances performed by means of the accusative combined with the infinitive and verbs referring to speech and speaking (verba dicendi) such as: promising, commanding, decreeing, requesting, questioning, ordering, imparting information or accepting a promise. A classical example is Cicero’s ordering Carthage to be destroyed (Delenda Carthago!).[1] The verbal utterance ‘I order C to be destroyed’ combines the verb with an infinitive and an accusative object (C). The infinitive combined with the simple present expresses a determined temporal relation: a speech act occurs in the present tense, or at the time the speaker (the person of the verb) is speaking. So the present tense refers to the time at which the speaker is speaking. The speaker is momentarily indicated by the personal pronoun ‘I’ which refers to the speech act in which it is pronounced. If the infinitive (to destroy) is modified by a predicate, as in: ‘I order that C be destroyed’, the accusative case expresses the subject of the action (C).

1.2. THE ACT-MODE OR ILLOCUTIONARY FORCE OF SPEECH ACTS

So, how do we analyze speech acts? Austin emphasizes the non-linguistic aspects of speech acts in his distinction between rhetic (acts which have a linguistic meaning), illocutionary and perlocutionary acts (which have further effects intended by the speaker’s utterance). An illocutionary act has a certain force. It is both saying and doing something: the speaker intends to do something in making an utterance. A rhetic act has a meaning but not the force and only a perlocutionary act is effective, for it brings about a change in the world by saying something (1962:121). One could object (as Reinach would) that illocutionary acts such as promising are also effective, i.e. they achieve consequences by saying something, as well as in saying something.[2] A promise not only carries with it an obligation for a speaker and needs to be grasped by an interlocutor, but in accomplishing this act, something is changed in the world (1913:45, 60). Similarly, commanding not only has an illocutionary force but is effective (wirksam) and requires uptake, since it gives rise to an obligation for the interlocutor to do something. Although a speech act may have direct physical effects, as when speaking a password for a door to open, it achieves these effects because it confers a meaning and not only because it has an illocutionary or sense-conferring force. I think Austin’s distinction between rhetic, illocutionary and perlocutionary acts concerns different aspects of speech acts rather than different kinds of linguistic acts.

Pace Austin, the difference between goal-directed linguistic acts is expressed in Bühler’s distinction between a linguistic action (Sprechhandlung) as a product on the level of praxis and creation (poesis) on one hand and a speech act (Sprechakt) on the level of conferring meaning by positing signs (Zeichensetzung), on the other. A speech action is doing or achieving something by talking: it can be enacted in a child’s play or in resolving a problem by discussing it, whereas a speech act has the quality of a sense-conferring act (sinnverleihender Akt) (1934:53-68).[3] In addition, by emphasizing the non-linguistic aspects of speech acts, their specificity is obscured; namely, that they are achieved by means of linguistic signs: they bring about obligations and claims for the speaker and the addressee through manifesting and expressing an intention on one hand and an appellative function on the other.

2. THREE SIGN-FUNCTIONS IN SPEECH ACTS

On Bühler’s view the sign-functions are threefold: as expression (Ausdruck), presentation (Darstellung) and appeal (Appell) which he also refers to as a triggering function (Auslösung) (1934:28). The sign is a symptom of the speaker’s expression, it presents an object or state-of-affairs to be fulfilled or achieved and it triggers a signal for the interlocutor. The sign indicates and manifests the intentions of the speaker and it functions as symbol of the object or state of affairs it presents: the sign bears information about an object or state of affairs and, in a linguistic act, it is coordinated with the object or state of affairs it presents or designates. The symbolic function of the sign enables the speaker to produce the conditions for making his utterance understood. Bühler writes:

The object named by a name is intentionally aimed at and also more or less intentionally reached in concrete speech-experiences; this is the case every time a member of a linguistic community himself uses the name meaningfully (sinnvoll) and correctly as sender, or correctly understands it as the receiver of a verbal message in which it is used. (1934: 164).

The interlocutor’s behaviour is steered by the sign functioning as a signal. According to Reinach, the achievement or constituting of a command or promise is an essential feature of speech acts or social acts. This feature distinguishes them from other statements. If A says to B: “I promise you I’ll do p”, this statement brings about (self-imposed) obligations for the promiser and claims for the promisee, unlike the statement: “1 x 1 is 1”. Reinach’s distinction foreshadows Austin’s distinction between locutionary and illocutionary acts (1962:101-3): “1 x 1 is 1” is a locution or declarative uttering a certain meaning, whereas “I’ll promise you I’ll do p” performs a speech act. Two necessary conditions for accomplishing a speech act are the illocutionary force of the utterance (Kundgabe) and its appellative function or need of uptake (vernehmungsbedürftig). The first condition accounts for expressing a performative statement, whereas the second condition underwrites its effectivity (Wirksamkeit) in bringing about a change in the world (1913:45).

A speech act depends on what Bühler calls steering (Steuerung) or direction of fit between the illocutionary force of an utterance and its uptake. A command occurs when it is manifested by the speaker who commands that p with the tendency or direction of fit to be taken up by an interlocutor who grasps the content of that p. The signs expressed by the speaker function as a signal for the addressee by steering his behaviour. If the command is not grasped, it has not attained its goal and failed its task, as a thrown spear which falls without having attained its target.[4] For example, I command you to close the door but you don’t. A failure of uptake and/or failure of grasping my intention and/or unwillingness to carry out the action entail a failure of the command because the latter is a relational moment or Aristotelian accident depending on both the force of the utterance and its uptake. For a promise is a moment fitting or linking an utterance with its uptake. Likewise, a promise is an accident depending of the relation of obligation it creates between speaker and addressee.[5]

Reinach illustrates this essential feature of promises as a dependent particular or accident in a binary dependence relation between promiser and promisee, by using the example of slaves in ancient Rome who were unable to make or receive promises because they were legally incompetent (1913: 85-6). Although their legal incompetence is invalid on grounds of natural law, Reinach argues that such grounds are insufficient for asserting that slaves are able to impose obligations and acquire claims. In performing a promise, obligations are imposed by the speaker and claims are acquired by the addressee, independently of who the promiser and promisee are – according to Reinach, they may be slaves or free men, angels or devils (1913: 85) - as long as a promise is accomplished by them. He notes that social acts are representative, for they can be performed by proxy: a promise of one person can be carried out in the name of another, as when a lawyer carries out the will of his deceased client.  Yet one could object that the effectivity of a speech act is not independent of who is in the role of the performer (1913:53). If I say to my audience: “attack the Russian army”, my command will lack effectivity, but if a general says to his soldiers: “attack the Russian army”, there is little doubt that his command will be grasped by interlocutors who will carry out the required action and hence bring about the command or realize the state of affairs.

3. THE EFFECTIVITY AND CONTENT OF SPEECH ACTS

In addition, the effectivity of a speech act depends on the strength of the addressees’ conviction that what is expressed is true, as is the case with curses and magical acts. If a priest or sorcerer says: “I hereby command a plague to strike you”, the speech act is effective through the addressee’s fear and belief that this event will occur. The command may obtain or not, but according to Reinach the satisfaction-condition of the speech act is a possible state-of-affairs. A plague should strike the addressee. Besides, a magical act may obtain at some point, as when rain is conjured (accompanied by various rituals). But the effectivity or power of such speech acts is subjective: it lies in the conviction with which their content is held to be true. On one hand, as Reinach points out, stating a social act presupposes the speaker’s conviction (Überzeugung) of the stated content (Mitteilungsinhalt) (1913:43). This conviction is a mental state or second-order intention grounding the speaker’s manifest intention: he commands that p and he wants that p to be grasped by the addressee. In addition, the speaker may have a third-order intention that his second-order intention is not grasped by the addressee (and that the latter responds to the first-order intention only). On the other hand, the addressee grasps or understands that p and holds, with a certain degree of conviction, that p is true (his degree of conviction is his response to the speaker’s second- or third-order intentions). Conviction or the belief that p is true is a mental attitude of the speaker and the addressee and it determines both the force or act-quality and the effectivity of a speech act.

What is the content of a speech act? It seems that content refers to three components of a speech act: a state-of-affairs, an intension and a sense. First, the command or promise is constituted by the utterance of the speech act: I promise you that p. This content c1 is a state of affairs which is posited: it either obtains or not, but it should obtain. Reinach says that it is an essential condition of satisfaction or fulfilment of a social act that it posits a state of affairs (1913:46). If the state of affairs obtains and the commanded event occurs, the act is accomplished (Vollzug). The claim that p is the content c2, the informational content or act-matter carried by the act. The act-matter is what is asserted, promised or commanded – the intension or meaning that p which is produced by the act and given by the clause in the complement of the verb. The content c3 is what the addressee understands or grasps: the sense, or what the addressee who understands the utterance understands to be said.

Since speech acts express statements (c2), it follows, pace Austin, that they can be true or false, for something is asserted in them which is either true or false once the speech act and its conditions have been fulfilled, so the content (c2) is truth-conditional relative to its context. But in the speech act, act and utterance are intimately linked. Thus, saying “I do” before the registrar or altar is either true or false and this utterance designates an act of accepting someone as their lawfully wedded husband or wife. This declarative utterance is true or false relative to its religious and legal context. It seems that the speech act at least partly constitutes that context for, as Reinach points out, the acceptance of a promise such as: “I hereby accept (to take you as my lawfully wedded husband/wife)”, is an act designating itself. The adverb “hereby” functions as an indexical: it refers to an event happening just now, together with the performance of the act. – ‘Hereby’ designates the ‘accepting’ which here, as it were, refers to itself (1913:57). In addition (although neither Reinach nor Bühler assert this explicitly), it seems that the sense-relation of the performative is governed by the accusative case, for the declarative utterance is performed as the ‘hereby’ is combined with a verb in the present tense, an infinitive and an accusative object. This ‘combination’ is a linguistic structure (Sprachgebilde), composed of a case-governed agentive relation between words in a sentence. This structure is modified by the speech act because it is not a separate entity but intimately related to it, as its modifier or meaning-conferring character. This intimate relation or dependence of the speech act-character and the meaning of the sentence is shown by the accusative case-structure.

Bühler, following Reinach’s teacher Husserl, analyses statements as dependent moments of speech acts (1934: 57). ‘What is meant’ depends on ‘what is said (and done in the saying)’ because the two are interdependent. That is why Husserl claims that meanings (Bedeutungen) are syncategorematic or dependent parts of the act-character (LUIV, § 7). The linguistic structure (the statement) is modified by speech acts because the latter are act-characters belonging to the domain of meanings (Bedeutungen): speech acts are modifications of meaning because they are meaning-correlates (1934:67). Reinach examines this semantic feature of speech acts in considering modifications of speech acts or third-order intentions such as lying, deceiving or simulating.

3.1. MODIFICATIONS OF SOCIAL ACTS

Modifications of social acts can be qualitative modifications of the act or conditional modifications of content. Lying, deceiving or simulations are qualitative modifications or third-order intentions which Reinach calls pseudo-performances (Scheinvollzug). A speech act is performed but it is not grounded on a genuine intention. Deceiving is a pale, bloodless performing – the shadow, as it were, next to the bodily thing (1913:44). However, these pseudo-performances have real effects. To unravel this apparent paradox, let us first consider intentions, for the distinction between what someone says and what he means is modified if his assertion deviates from what his overt behaviour suggests he intends. An intention is a mental directedness towards an object. A first-order intention is an act in which an object is conceived and named: the command that p is applicable to a possible object, so the command posits an object, that p. A second-order intention is the speaker’s mental state: he wants that p is grasped by the speaker. The second-order intention represents the speaker’s first-order intention. A third-order or nested intention does not represent the speaker’s first-order intention, yet it is communicated to the addressee. I say: ‘I want you to do p’ but a qualitative alteration in the act of meaning signals to the addressee that I really don’t want you to p. Here the speech act is qualitatively modified so that it does not posit an object, although it appears to do so. Irony is a telling example of pseudo-performance, since the expression of a meaning is modified by a language use signifying the opposite. Husserl calls this alteration a suppositio materialis in the terminology of medieval logic, that is, a word functions as its own name, as mention and not as use (LUIV, § 11)[6]. Consider an English person saying ‘interesting’: the meaning of the adjective is modified because ‘interesting’ is used as an attribute and not as a predicate. The success of such a speech act does not necessarily depend on the uptake of the addressee, who may fail to recognize the counter-sensical meaning of ‘interesting’, but rather on the uptake of the audience. As Reinach says, social acts are collective acts and a speaker can have many addressees or be directed towards numerous subjects. (Richtungssubjekte) (1913: 47).

In addition, by qualifying lying and deceiving as deviations from genuine usage, Reinach refers to an old distinction between standard or regular and non-standard or irregular cases: deceiving is a non-standard instance of asserting. The standard/non-standard distinction leads back to the distinction between speech acts and linguistic structures in that it concerns the distinction between sense, grammatical nonsense (Unsinn) and absurdity or counter-sense (Widersinn) (Husserl: LUIV, § 12). The qualitative modification of the act is a modification of the meaning. The modified character of a speech act appeals to the addressee’s or the audience’s linguistic knowledge and the sense-conferring modification is performed by a speaker who knows the standard use. For a counter-sense is identified and recognized as such in regard to a standard or usual sense, which is why the meaning-transfer works. The qualitative alteration of what is said modifies the usual meaning of an utterance by producing a counter-sense which is an irregular instance of a regular case. Meaning-transfer is possible because linguistic presentation (sprachliche Darstellung) allows for leeways (Spielräume) of indeterminacy, as Bühler puts it (1934:66).

Another kind of modification of social acts is the conditional clause. These are acts with a conditional content, in which the content or act-matter modifies the force of the act, such as the command to do p in the event that q: ‘I command you to shoot if someone enters the building’. The force of the act is modified by the ‘if-clause’, so that the obligation to do p is conditioned by the occurring of the event. A conditional command brings about an obligation requiring an action of the addressees by positing a possible event. This obligation becomes actual if the event occurs (1913:46).

Reinach also considers collective social acts as a kind of act-modification: one speech act is performed by several people (Handeln im Verein) (1913: 47-8). Individual performances form a single unitary performance by carrying out a command. Consider military or religious songs which are performed in unison, by soldiers and schoolchildren alike. An officer commands his army or a teacher commands her pupils to sing: “we are warriors and we are marching on. Left right, left right, we are marching on.” Or: “onwards Christian soldiers, marching as to war”. In this act, the individuals use the personal pronoun we in the sense of ‘me and the other me’s’: we (and not I) momentarily, at the time at which the speakers are speaking, refers to their performance, thereby designating a collective speech act. A characteristic feature of that performance is that it is not affected by changes of context: it is accomplished in the same way by different groups of people, such as soldiers, churchgoers and children. The singing group performs a single act of several subjects in what Reinach calls a curious interrelation (Zusammenhang) which is irreducible to identity of content or addressee and even less to the deliberate simultaneous performance of the act (1913:47). Each addressee performs as a participant in a collective speech act and together they effect one obligation. In addition, Reinach explains the penal concept of criminal complicity in terms of this unitary structure of a social act carried out by several addressees (1913:48).

An analysis of the speech act by means of an inference from the linguistic product back to the linguistic action shows that what someone says and what he means are dependent moments of a speech act in which a linguistic structure is achieved. This structure is a verbal statement whose parts stand in an agentive relation governed by the accusative and it brings about a relation of obligation between a speaker and addressee. It also brings about various non-linguistic effects although, or maybe because, its semantic content is relatively variable.

обратно нагоре