It is written: "In the beginning was the Word!"

Even now I balk. Can no one help?

I truly cannot rate the word so high.

I must translate otherwise.

I believe the Spirit has inspired me

and must write: "In the beginning there was Mind."

Think thoroughly on this first line,

hold back your pen from undue haste!

Is it mind that stirs and makes all things?

The text should state: "In the beginning there was Power!"

Yet while I am about to write this down,

something warns me I will not adhere to this.

The Spirit’s on my side! The answer is at hand:

Johann Wolfgang von Goethe, Faust

(Translation by Peter Salm, Faust, Part I New York: Bantam Books, 1985: 77).

As was intimated above, until the 1930s the study of sentence meaning was viewed from a truth-conditional perspective, which reduced language to a set of sentences that could be either true or false, while dismissing any other kind of discourse, ethical, aesthetic, and so on, as meaningless. It is this doctrine of logical positivism that Austin set about demolishing in his posthumously published book, How To Do Things With Words. He, like Wittgenstein before him, was committed to a pragmatic view of language, i.e., the view that language is used for doing things, and his theory gained ground through Searle’s Speech Act Theory (we will discuss Searle later on). In the 1940s Austin argued that there are some declarative sentences which, contrary to logical positivism, have no intention of making true or false statements. Rather, they seem to form a separate category or class, as in the following examples:

I bet you six pence it will snow tomorrow

I apologize

I promise I’ll be there on time

I hereby christen this ship the Mayflower

I warn you that thieves will be prosecuted

I sentence you to five years of hard labour

I declare war on France

According to Austin, these sentences do not merely describe states of affairs, but they do things, in that after you have declared war on France, or apologized, or sentenced someone to five years of hard labour, the world is no longer the same: France is no longer a "peaceful" country; you have made amends for your (aberrant?) behaviour; someone has been deprived of his freedom […] Furthermore, it is unreasonable to assert that any one of these sentences can be assessed in terms of truth-values, as is shown by the inconceivable and preposterous nature of the exchanges below:

A: I hereby christen this ship the Mayflower.

B: That’s false!

A: I bet you six pence it will rain tomorrow.

B: That’s certainly true.

These "strange" sentences, Austin termed performatives, in juxtaposition with all other types of sentences (assertions, statements, etc.) which he called constatives. It is noteworthy, though, that by the end of his book, Austin shows that performatives do not constitute a special class but, rather, all types of sentences can be said to perform acts. ‘Indeed what starts off as a theory about some special and peculiar utterances—performatives—ends up as a general theory that pertains to all kinds of utterances’ (Levinson, 1983: 231). In fact, Austin gradually shifts away from the view that performatives are a special class of sentences to the view that there is a class of performatives including explicit performatives and implicit performatives. Moreover, he seems to hold that performatives and constatives are actually sub-categories belonging to a general theory of illocutionary acts.

But let us resume our discussion of Austin’s theory by noting that, according to him, the study of performatives undoubtedly forecloses the possibility of assigning truth-conditions; yet, unlike constatives, performatives can go right or wrong. In this light, he outlined all the possible ways in which they can be ‘happy’ or felicitous, or ‘unhappy’ or infelicitous, as he put it. This typology of conditions that performatives must meet in order to be ‘happy’, he called felicity conditions, which comprise three categories (much of the discussion on Austin and Searle is based on Levinson, 1983 and Duranti, 1997).

1. (i) There must be a conventional procedure having a conventional effect.

(ii) The circumstances and persons must be appropriate, as specified in the procedure.

2. The procedure must be executed (i) correctly and (ii) completely.
3. Often, (i) the persons must have the requisite thoughts, feelings and intentions, as specified in the procedure and (ii) if consequent conduct is specified, then the relevant parties must so do.

In order to illustrate these felicity conditions, we should consider what happens when they are not met. Suppose that someone says I dub thee Sir Arthur Conan Doyle. It may be the case that he or she will fail to perform the act described if, for example, the act is taking place in the gym’s locker room or under a bridge—in which case felicity condition A (i) is flagrantly violated. Besides, the speaker may not be the appointed person, or there may be no witnesses, and so on, which definitely runs counter to condition A (ii). Furthermore, the act described may violate condition B if the speaker uses a fountain pen instead of a sword, or even utters Your name will be Sir Arthur Conan Doyle, and so forth.

Let us consider another example taken from Levinson (1983: 230).

Curate: Wilt thou have this woman to thy wedded wife…and, forsaking all others, keep thee only unto her, so long as ye both shall live?

Bridegroom: Yes

In a marriage ceremony, the bridegroom is expected to say I will in order for the "procedure" to be valid and ratified. The words uttered must be correct in order to complete the procedure (condition B). In other words, there has to be satisfactory uptake, i.e., ‘the role of the interlocutor in making a given illocutionary act successful’ (Duranti, 1997: 225). Pertaining to condition C, we could say that violating it is tantamount to being insincere, as in saying There is a garage open within walking distance, when you know that the garage is ten miles away and is closed (violation of condition C (i)), or I promise I’ll call you tonight, when you have no intention of doing so (violation of C (ii)).

For Austin, these violations are not all of the same kind. Violations of A and B conditions are called misfires, whilst violations of C conditions are abuses, ‘not so easily detected at the time of the utterance in question, with the consequence that the action is performed, but infelicitously or insincerely’ (Levinson, 1983: 230). As has been mentioned, these performative utterances actually do things

Austin also noted that performatives evince certain properties: they are always first person indicative active sentences in the simple present tense. As Levinson (1983: 231) observes, ‘[t]his is hardly surprising, since, if in uttering a performative the speaker is concurrently performing an action, we should expect just those properties’. It is true that of the following sentences, only the first one is a performative.

I bet you five pounds it’ll rain tomorrow.

I am betting you five pounds it’ll rain tomorrow.

I betted you five pounds it’ll rain tomorrow.

He bets you five pounds it’ll rain tomorrow.

(found in Levinson, 1983: 232)

Nevertheless, we could argue that there are many cases where first person indicative active sentences in the simple present do not constitute performative sentences. For example:

I now beat the eggs till fluffy

is a report of an action. So, other criteria should be sought to identify performatives. The adverb hereby is a good rule of thumb, since it only co-occurs with the performative usage, as shown below:

I hereby christen this ship the Mayflower

*I hereby now beat the eggs till fluffy

*I hereby go to school every day

In this light, christen is a performative verb, whereas beat and go are not. Yet, these criteria alone are not sufficient because performative sentences such as (a) below can be expressed equally well as (b), (c), or (d).

1. I accuse you of doing it on purpose
2. I find you guilty of doing it on purpose
3. You did it
4. Guilty!

As Austin himself later admitted, utterances can perform acts without having the form of explicit performatives. These implicit performatives, though, such as Go!, Guilty!, and so on, can be said to be ambiguous, in the absence of the explicit markers such as ask, declare, warn, and so forth. Thus, the utterance Go!, for example, may perform the giving of advice, or the issuance of an order, or daring etc. Similarly, the utterance Stay with me could be performing the act of giving advice, doing entreating, or even seducing!

So, we could claim that all utterances perform actions or have specific forces, as Austin was apt to say. He identified three types of acts that are simultaneously performed in saying something:

1. a locutionary act, i.e., the act of saying something, that is, the act of uttering specific sequences of sounds interpreted according to grammatical conventions, or subject to the assignment of truth values.
2. an illocutionary act, i.e., the making of a statement, promise, etc. in uttering a sentence, by virtue of the conventional force of the locutionary act. For example, you’re fired may be used in our society to signal one’s status from "employed" to "unemployed," etc.
3. a perlocutionary act, i.e., the act produced by the uttering of a locution, that is, the very consequences or effects of such locution irrespective of its conventional force. For instance, the locution you’re fired, uttered under the appropriate circumstances and by the right person to the right person (employer-employee), can also have the effect that the one who has lost his job may become depressed, suicidal, etc. or, conversely, liberated!

Certainly, it is the illocutionary act that Austin was interested in, in the first place. He argued that a) and b) are detachable, therefore, they can be studied independently; the dividing line between b) and c), though, seemed to him to be blurred. Imagine that someone utters:

Kill him!

Applying what we have discussed so far, we could say that this utterance, produced in appropriate circumstances, has the illocutionary force of ordering, advising, urging, or manipulating the addressee to kill him; however, the perlocutionary effect of the utterance is difficult to pin down, as it could be persuading, forcing or even frightening the addressee into killing him. (Or, it could have the perlocutionary effect of frightening him!)

Searle’s theory of speech acts is akin to Austin’s, albeit more systematised and rigidified. For Searle, felicity conditions are not just dimensions on which utterances can go wrong; they are actually constitutive of the illocutionary forces of the utterances. For example, if someone says, I promise to come tomorrow, then he should be performing each of the following conditions:

1. The speaker said he would perform a future action
2. He intends to do it
3. He believes he can do it
4. He thinks he wouldn’t do it anyway, in the normal course of action
5. He thinks the addressee wants him to do it (rather than not to do it)
6. He intends to place himself under an obligation to do it by producing the utterance U
7. Both speaker and addressee comprehend U
8. They are both conscious, normal human beings
9. They are both in normal circumstances—not e.g. acting in a play
10. The utterance U contains some IFID (illocutionary force indicating device) which is only properly uttered if all the appropriate conditions obtain

In this way, Austin’s felicity conditions can be used as a kind of yardstick against which to compare different speech acts. According to Searle, though, what would be more useful would be to ‘delimit the kinds of possible illocutionary force on principled grounds’ (Levinson, 1983: 239). To this end, he suggests that, in using language, we can do five things (see Duranti, 1997; Levinson, 1983):

1. talk about a state of affairs (assertives or representatives)
2. ask others to do something (directives)
3. express our feelings and air our views (expressives)
4. effect changes in a state of affairs, as in declaring war, etc. (declaratives or declarations)
5. commit ourselves to a future action (commissives)

Contrary to what Searle believed, this typology is far from principled or systematic, given the fact that there have been a great many other typologies (see Hancher, 1979; Allwood, 1976; Lyons, 1977a; Bach & Harnish, 1979). Human communication is not based on such mechanisms; rather, it is carried out by dint of communicative intention that speakers have, as Grice’s maxims have shown. Moreover, it seems that what people do with sentences is by no means determined or restricted by the surface form. In this light, one can say:

I want you to close the window

I’d be much obliged if you’d close the window

Can you close the window?

Are you able by any chance to close the window?

Would you close the window?

Would you please close the window?

Would you mind closing the window?

Hadn’t you better close the window?

May I ask you to close the window?

Did you forget the window?

How a bit less breeze?

It’s cold in here

Okay, love, what do people do in such cold weather?