How to Do Things With Words is a work by John Langshaw Austin , published in England in 1962 (posthumous edition) from a series of lectures given in 1955 at Harvard University. From this book was born the theory of speech acts , which will give rise to contemporary pragmatic linguistics and the philosophy of ordinary language.
With this book, JL Austin continues the studies of Émile Benveniste , Karl Bühler , Roman Jakobson , Charles Bally , Bronislaw Malinowski, and Ludwig Wittgenstein . With an innovative and didactic working method, he develops his theories of speech acts in twelve lectures, differentiating two types of statements: the constative statement and the performative (or performative) statement. A statement is performative when nothing is stated or described but an act is performed. The performative is subjected to conditions of "happiness", depending on a situational (or circumstantial) context.
Throughout his studies, he realizes that constants also depend on conditions, so he extends the performative criterion to all statements. Therefore, it elaborates a taxonomy of the different ways we can have of “doing” something when saying something, divided into three categories: the locutionary act (saying something is doing something), the illocutionary act (when saying something we are doing something) and the perlocutionary act (because we say something we are doing something). It recognizes in the illocutionary act the essential act of the word, therefore it tries to establish a taxonomy of the different values that the verbs of an illocutionary act can adopt.
Death prevents him from continuing his studies. John Searle , regarded as Austin's most direct successor, continues his investigations, re-examining the classification of illocutionary values. Austin's theories also receive two main criticisms: that of Dan Sperber and Deirdre Wilson, and that of John-Robert Ross , followed by George Lakoff .
Austin's method of work is innovative in every way. He calls his way of philosophizing "linguistic phenomenology." He does not consider the analysis of language as such, but as a philosophy that deals with language, in order to study phenomena. His method of analyzing language is based on trial and error, thoroughness, and concern for detail. It postulates observing language, distrusting it, disfiguring it, even disrupting it.
For this reason, the analyst must imagine new or unprecedented situations even when they are strange or extremely trivial. Taking the complete inventory of the possibilities of enunciations and exhaustive lists of speech acts allows to be in front of the “total situation”, apt to raise answers to the philosophical questions. You have unwavering and exclusive confidence in your method, even though it may seem laborious in the face of all the enunciation possibilities. For him, the uses of language are not infinite, therefore it is possible and necessary to establish a repertoire as wide as possible to arrive at an exact method. In that way, he opposes both philosophical abstraction and grammatical abstraction.
The distinction between constative and performative statements
Having observed that the grammatical category of "statements" is too broad, and is usually classified in the category of "pseudo-statements" - because they are not -, Austin concludes that philosophers often consider "that many expressions that they appear to be statements, or they are not formulated at all to record or provide direct information about the facts, or they serve that purpose only in part. 4 For this reason, it establishes differences between two types of statements: the constatative and the performative.
Verifying statements qualify statements that are only true or false, without "doing" anything. On the contrary, performative statements are neither true nor false, nor the expression of nonsense. They want to "do" something (gamble, get married, etc.) and often need to resort to an addition to the words themselves: this is what Austin calls the situational (or circumstantial) context.
When some circumstances are absent (for example, in the case of a marriage, if the statement "I declare you husband and wife" is uttered by anyone and not by the mayor), the performative statement does not become false, simply the act intended has not occurred. It may also have been fulfilled in bad faith (for example, if the person who is getting married says “Yes, I do”, when he does not want it), or it may not have appeared in its fullness.
When the action accomplished by a performative statement comes to fruition, it is called lucky. In the opposite case, it is called unlucky. Those concepts of lucky or unlucky performative statements are opposed to constative statements, which are true or false.
All the circumstances that cause the non-performance of the act are called "misfortunes." Austin classifies them according to the conditions to which they correspond.
Third and fourth Lectures
At this point, Austin realizes that constative statements also depend on circumstances, and are often the same as for performatives. Therefore, you must look for other criteria to differentiate the constants from the performatives.
Select the grammatical criterion: “We would be inclined to say that every expression that is actually a performative would have to be reducible, expandable or analysable in such a way that a form in the first person singular of the present indicative of the active voice is obtained ( grammatical). " 5
Sixth and seventh lectures
He realizes that the category of constants is very broad, and that, despite his attempts to differentiate them, many statements are both constants and performative, depending on the circumstances.
If there are no grammatical criteria, and no infallible tests that allow us to distinguish without any doubt the performatives from the constants, we must “start over”, posing the fundamental questions: “It is necessary that we reconsider in a more general way the meanings in that saying something can be doing something, or that by saying something we do something. "
He solves his previous statements by dividing speech acts into three categories: the locutory, illocutionary and perlocutionary act.
The locutory act
Austin proposes, as a starting point, that an enunciation consists at least of a phonetic act - producing certain sounds -, a phatic act - enunciating certain words with a certain construction and with a certain intonation -, and finally a rhetic act - use these words in a sense, with a specific reference -. Those three acts allow us to say something, it is what Austin baptizes as the locutory act.
The illocutionary act
The illocutionary act is a locutionary act, but it is an act performed by saying something. In other words, the locutionary act is only the act of saying something, while the illocutionary act can involve different uses of the same phrase, depending on how it is understood when it is pronounced (for example, depending on the context, saying "I'm cold" It can mean the wish that the interlocutor close the window, lend his coat to the speaker, be just information about my physical condition, etc.). It is the value attributed to the enunciation that founds the illocutionary act (order, statement, question, etc.)
This act must be produced in accordance with a convention. For example, if the act is a promise, the person who makes the statement "I promise you that" must be sincere, if not, the illocutionary act could not be carried out.
Austin realizes that almost all sentences can be considered as illocutionary, therefore he does not manage to define this category precisely. However, it recognizes in the illocutionary act the essential act of the word.
The perlocutionary act
The perlocutionary act is carried out by saying something, the act produces effects or consequences on the interlocutor. Thus, contrary to an illocutionary act, the functions of language that have an indirect effect on the interlocutor (flattering, pleasing, frightening, etc.) are called perlocutionary but that are not explicitly inscribed in the statement.
After having presented the three speech acts of the language, Austin summarizes the difference between the locutionary, illocutionary and perlocutionary act: “In a similar way we can distinguish the locutionary act 'said that ...', the illocutionary act 'held that ... ', and the perlocutionary act' convinced me that ... '”. The locutionary act has a meaning, the illocutionary gives a value to the fact of saying something, and the perlocutionary allows to obtain some effects thanks to the statement.
In this lecture, he underlines the need to distinguish the illocutionary act "by saying such a thing I was preventing it" from the perlocutive act "because I said such a thing I convinced him, or surprised him or restrained him." Then, he examines the consequences of the two acts, and deduces that the essential difference between them is that the illocutionary act is a conventional act, while the perlocutory act is not. Therefore, the distinction is blurred, because it can be difficult to define where conventions begin and end.
It takes into account two linguistic formulas: “when I said x I was doing y” or “I did y”, which corresponds to the illocutionary acts, and “because I said x I did y” or “I was doing y”, 9 which corresponds to the acts perlocutives. Although this test is not yet sufficient to distinguish the two acts, it can help.
At this point, he finds himself confronted with the problem of the initial distinction between constative and performative statements, which is no longer valid, because the one and the other do not take into account the total linguistic act.
It ends by establishing a taxonomy of the different values that the verbs of an illocutionary act can adopt: the judicial (acquit, convict, etc.), the exercise (demote, command, etc.), the compromisory (promise, vote for, etc.) etc.), behavioral (apologize, thank, etc.) and expository (affirm, deny, etc.). He stresses that he is not totally satisfied with this taxonomy. Austin cannot finish his work, death will prevent him from continuing it. Searle is going to resume his program. It is considered as the most direct successor to Austin's theory.