Searle's 'Classic Speech Act Theory' (1969), a later approach (1979)

It is a foundational legend of speech act theory that in his Speech Acts (1969), Searle presents the first elaborate theory of speech acts--the truth is: The book does not provide any elaborate theory at all. What it does present is material which Searle gathered during his attempt to execute a three-step programme, the target of which is a "full dress analysis of the illocutionary act" (1969, 54). Of this programme, only the first step is completely executed. Furthermore, considering the results which the partial execution of the second step delivers we can clearly see that an execution of the programm along the lines Searle proposes never could lead to a "full dress analysis of the illocutionary act". Let us briefly review the three steps of Searle's programme.

The first step consists in the application of a rationale adopted from Alston's 'Linguistic Acts' (1964). In this paper, Alston pursues a semantical endeavour: He suggests using the conditions of the (say) 'flawless' performance of a certain act type for the derivation of semantical rules applying to semantic devices expressing its performance (e.g., the conditions under which a flawless request occurs for an analysis of 'I hereby request you to ...'). Searle transfers this rationale to his analysis of (what he believes is) a paradigmatic instance of an 'illocutionary act': promising. "In order to give an analysis of the illocutionary act of promising", he (1969, 54) explains, "I shall ask what conditions are necessary and sufficient for the act of promising to have been successfully and non-defectively performed in the utterance of a given sentence." On his view, an Alston-style analysis of this act type involves conditions of four categories: (1) 'propositional content'; (2) 'preparatory'; (3) 'sincerity'; (4) 'essential'. These categories, first applied in the execution of the analysis of promising, are supposed to be of crucial importance for the final aim, the analysis of 'illocutionary acts' in general.

The second step, apparently, involves the application of the schema applied in the first to (what Searle believes are) further instances of 'illocutionary acts', such as greeting, thanking and asking a question. The hope connected with the execution of this step is that the four categories applied in the analysis of promising 'carry over', that means, are applicable in general with analyses of (putative) illocutionary acts, thus enabling us to define the 'illocutionary act' as an act being subject to this bunch of categories. "If this analysis is of any general interest beyond the case of promising", Searle writes in the initial presentation of his approach (1965, 238-9; cf. 1969, 64), "then it would seem that these distinctions should carry over into other types of speech act". The problem is that in fact, the categories do not carry over. 

There is, for example, no particular propositional content condition to 'stating', and in the case of 'greeting', the application of a category such as 'propositional content' is simply nonsense. The case of 'greeting' also shows that sincerity is not a universal category, for the application of a sincerity condition to this act delivers nonsense, too (cf. Searle 1969, 66-7). This seems to leave Searle's programme with maximally two categories left: 'Essential' and 'Preparatory'. Yet reference to these will surely not enable us to define 'illocutionary acts' (in such a way as to distinguish them from non-illocutionary acts). Consider converting a penalty in football: one preparatory condition is that the penalty has been given by the referee, another that the player belongs to the team to which it has been given; the essential condition is that it counts as a goal ... -- Apparently, non-illocutionary acts are subject to conditions and categories like 'preparatory' and 'essential' quite as well as (putative) illocutionary acts. So, as it turns out, the execution of step 2 of Searle's programme actually reveals the failure of the programme, by showing that many 'illocutionary acts' actually are not subject to the bunch of categories Searle gained from step 1, and that, hence, the idea of defining it in terms of these categories is doomed to failure.

The third step of Searle's programme would be the full dress analysis of the illocutionary act". Searle never even starts realising this task. The apparent reason is that the partial execution of the second step showed that the means of executing step 3 are lacking. For in executing step 2 it turned out that the categories which, taken together, were supposed to distinguish the 'Illocutionary Act' from non-illocutionary acts, are not universally applicable.

The lack of a definition leaves the reader with a large host of open questions about the nature of those 'illocutionary Acts': Do 'Illocutionary Acts' always involve some of those "preparatory conditions" about which Searle makes so much fuss? (Cf. 1969, 63, "rule 5 applies only when rules 2 and 3 are satisfied".) Do they consistently involve a "sincerity condition"? (What about greeting? Cf. 1969, 66-7. What is the role of the revised version (1969, 62)?) Do they necessarily involve an audience's "understanding the utterance"? (Cf. the comments on 1969, 61, along with 1969, 62: "Condition 1 [applies] to all kinds of normal illocutionary acts".) Is the use of linguistic devices necessary for 'Illocutionary Act' performance? His account implies this strictly (the meaning of IFIDs is supposed to constitute those acts; cf. also the reference to a sentence within rule 5 of promising, many other elements  in Searle's exposition imply this assumption); yet Searle never explicitly confirms this view, and it is indeed rather strange, because is is perfectly clear that Searle accepts the non-verbal performance of 'illocutionary acts'.

Why does Searle not fulfill the promise made at the beginning, to provide a "full dress analysis of the illocutionary act"? You often encounter the view that Searle withdraws this promise when he emphasises, with reference to his analysis of promising, "the looseness of our concepts", by which "certain forms of analysis, especially analysis into necessary and sufficient conditions, are likely to involve [...] idealization" (1969, 55). Yet notice that Searle's conclusion is not the rejection of the task (but merely the acceptance of idealization). Furthermore, and more importantly, notice that the 'looseness of our concepts' only applies to natural terms, but not to technical terms: while it may thus apply to the analysis of promising, which is a natural term, it does not apply to the definition of "Illocutionary Act", which is a technical term. 

Apart from the lack of a definition, and a huge number of quite decisive questions about the identity of 'illocutionary acts' which remain unanswered, there is an even fiercer problem with his approach.  The numerous fundamental views concerning 'Illocutionary Acts', language and communication which Searle presents and develops along the way turn out to be, once viewed together, clearly mutually inconsistent. Let us restrict ourselves to one particularly fundamental case. From G.C.J. Midgley's paper "Linguistic Rules" (1959), Searle adapts (incidentally, without mentioning the source) a general view of linguistic rules and speech acts according to which 'Speech Acts' are constituted by rules of language (see, e.g., 1969, 38), where these rules are supposed to represent the meanings of linguistic expressions. This view presupposes that 'Illocutionary Acts' are institutional acts, constituted by conventions and performed in accordance with these conventions. Yet at the same time, Searle also postulates that 'Speech Acts' are "the basic or minimal units of linguistic communication" (e.g., 1969, 16). This clearly and obviously contradicts the presumption just introduced: If 'Speech Acts' are institutional acts, then they are not the minimal units of communication, and vice versa. Searle may here confound the two views of 'Speech Acts' isolated earlier by P.F. Strawson. 

To summarise, in Searle's classic approach the question what 'Speech Acts' are remains unanswerd. Searle's programme for developing a definition remains unexecuted. Indeed, the partial execution which Searle presents in (1969) reveals that the programme is really doomed to failure. Even worse, Searles characterisations of 'Speech Acts' is in various ways inconsistent.

In Expression and Meaning (1979), instead of clarifying the account adumbrated in Speech Acts, Searle changes over to another, completely different conception of 'Illocutionary Acts'. "The problem posed by indirect speech acts", he says there, for example, "is the problem of how it is possible for the speaker to say one thing and mean that but also to mean something else."  This passage and many others in the book make clear that Searle now rejects the assumption, earlier emphasised by him, that Illocutionary Acts' are institutional acts; now they rather seem to be cases of 'speaker meaning (what one says)', or attempts at (linguistic) communication, or cases of achieved (linguistic) communication. (Which of these options resembles his conception(s) in Expression and Meaning best remains unclear.) 

---------------------------------------------------------------------------------------------------

Searle, J.R. (1965), "What Is a Speech Act?", in Max Black (ed.), Philosophy in America. Ithaca: Cornell University Press, 221-239.

Searle, J.R. (1969), Speech Acts, Cambridge: Cambridge University Press.

Searle, J.R. (1979), Expression and Meaning, Cambridge: Cambridge University Press.