Semiosis & Sign Exchange

design for a subjective situationism, including conceptual grounds of business information modeling

Pieter Wisse

chapter 10

Writing acts on Searle

 

 

 

The axiomatic system John R. Searle applies to meaning is immediately recognizable in the second sentence of Speech Acts, an essay in the philosophy of language (1969, p 3):

[T]he speaker means something; the sounds he emits mean something; the hearer understands what is meant.

He draws on both traditional interpretations of meaning (see also § 7.3). The opening sentence of his book reads: “How do words relate to the world?” Words, he implies in that second sentence, relate to intentions of the speaker-in-the-world. It is one of meaning’s traditional interpretations. The other is that words also stand for some state of the world without regard for the speaker’s intentions. In this realist sense, words are then either true or false.

Searle excludes the hearer as an essentially structural element. For though not passive, the hearer only ‘acts’ as the outside receptor of the sounds from which he is believed to reconstruct what both the speaker’s intentions and the relevant state of – the rest of – the world are. So, meaning is given to him. How he handles the sign as a ‘gift,’ that is how he takes reception of it, unwraps it, uses it, etcetera, does not contribute to meaning. The accuracy of – the result of – his instant reconstruction (also read: interpretation) marks the level of understanding, it follows from Searle’s scheme. Understanding is all about correspondence between interpretations of speaker and hearer. The sounds and their meaning are supposed to be unambiguously instrumental for this purpose of agreement.

Of course I take grave liberties at starting to explain Searle’s axioms from no more than two sentences. Actually, I have so far reported on what I expect from his book only after I stop to think about just his beginning. For that is my interest in studying Speech Acts. Does Searle keep the axiomatic system of Austin, once his teacher at Oxford,1 basically intact? Again, it is what I expect to find. But then again, he might be unfolding an entirely different set of axioms. If so, are they similar to those I have pieced together myself? Does his work perhaps predate my anatomy of meaning, with every sign as a request for compliance? Going by his two opening sentences, however, I think he stays well within the tradition of the language philosophy of his teacher. But I want to make certain, and so I read on. I decide not to be bothered by Searle’s apparent restriction to speech. When necessary, I just substitute sign. As I read, I make notes. In this chapter, as a stylistic illustration of semiosis, I try to keep the flavor of my running commentary.

 

 

10.1 analytical authority of the speaker

The larger part of Searle’s introductory chapter of Speech Acts contains his attempt to make his method of investigation acceptable. Yes, it is a good example of a request for compliance. Apparently, his immediate environment is that of analytic philosophy. His introduction can even be read as an attack on philosophers who, according to Searle, define the concept of analytical statement too … analytically.2 Instead, he offers a projective procedure (p 6):

If you want to know if a statement is analytic ask yourself whether it is true by definition or in virtue of its meaning.

He avoids any serious discussion because, he claims, especially the concept of analytic deserves axiomatic status (p 7):

We could not embark on our investigation if we did not understand the concept, for it is only in virtue of that understanding that we could assess the adequacy of proposed criteria.

Now, I completely agree with Searle on the special status of what he calls assumptions. They are essentially without any ground themselves. Those boundary concepts are necessary as ground for the construction of systems of derived concepts. So, what Searle aims to get across is actually his metaphysics, with analytical philosophers annex logical positivists for his targeted audience.

I admire his brave attempt for it makes it at least possible to address fundamental issues in a discussion (though logical positivists would deny, as Searle accurately indicates about them, that such issues can be discussed). My major objection is, however, that his axiomatic system includes concepts, such as meaning and truth, he should not be explaining from but toward. It only leads to more circular reasoning when, for example, meaning ends up as a function of … meaning. And truth of … truth. To avoid pitfalls of premature contradiction thinkers like Schopenhauer and Peirce carefully choose different axioms. They thereby give themselves room to derive an essentially psychological, relativistic concept of meaning. It also constitutes an escape from an absolute concept of truth value. Searle is still captured by his unhappy assumptions.

Or is he? Does he actually inquire after integration of both axiomatic systems? What an individual person means is often absolutely true, he argues, for (p 11)

one knows such facts about language independently of any ability to provide criteria of the preferred kinds for such knowledge.

He of course refers to analytical criteria in the logical positivist sense of analytical that he criticizes. I agree with such criticism. What I find mistaken is the connotation that “one knows such facts about language” to be absolutely true. But Searle continues, on the basis of what I ‘analyze’ as contradictory assumptions, to state that (p 15)

the method of this book must seem naively simple. I am a native speaker of a language. I wish to offer certain characteristics and explanations of my use of elements of that language. The hypothesis on which I am proceeding is that my use of linguistic elements is underlain by certain rules. I shall therefore offer linguistic characterizations and then explain the data in those characterizations by formulating the underlying rules.

Effectively, Searle states that the “certain rules” are a priori and general, i.e., they guide all speakers of all languages. I don’t want to argue with it right now. Why not reason from that axiom, and see what it leads to? I object, however, when he offers this argument in support of his claim that absolute truth can be produced by an individual speaker. How does it follow?

 

 

10.2 rules: description versus prescription

I believe Searle misdirects himself even more, and in the process requires me to read like a detective, by his “naively simple” method when he proposes to derive underlying rules. My view is it is one or the other. He may of course assume a rule system, using it to derive – and thus explain – linguistic elements, whatever they are. Or he may assume a system of linguistic elements, using it to derive rules for producing them. He takes the latter approach. i.e., starting with classifying linguistic elements. But then it is a tautology to suggest that rules explain the linguistic elements they were derived from, in the first place. It is yet another zero-variable theory (Wicklund, 1990). Apparently Searle himself is already worried for he makes a preemptive strike at criticism (p 16):

There is nothing circular in this procedure, for I am using the hypothesis of language as rule-governed intentional behavior to explain the possibility of, not to provide evidence for, linguistic characterizations.

My interpretation is that he not only fails to provide a reasonable argument against circularity. He even makes it worse by adding a second cycle. The authority of the speaker to give evidence, i.e., to state the truth, now gives way again in favor of the possibility “to explain the possibility.” What is the difference?

Rather than establishing trust in his method, Searle succeeds in arousing my suspicion. He is stressing rules. But what precisely do those rules control? I suspect his vagueness covers a mistaken integration of concepts. At least I read from the last quotation that for his investigation he is “using the hypothesis of language as rule-governed intentional behavior.” Ordering these words in a reverse sequence, Searle first of all implies a distinction between intentional and non-intentional behavior. As I am still only studying his introductory chapter I appreciate his difficulty that he probably cannot make their difference completely clear, yet. But it is evidently important to him, so with special interest I am watching how he applies the terminology of intentionality. Next comes the term “rule-governed.” Searle not only assumes that behavioral regularities, or patterns, exist. He also indicates that behavior is generated by rules.

Yet another word forward informs he is not concerned with all regular behavior, but only with language. But does language equal behavior? He probably means language use. Or particular utterances, or specific speech acts, etcetera. If only he had written use, I could reason with more certainty about the nature of the rules that Searle assumes are governing the behavior that is language use. In some respects, an utterance may be taken as ruled by the language as system. But in what respects, more precisely? I am happy to concede language is partly characterized by application of particular sounds on a regular basis. Also their sequential configuration into strings are highly patterned. But that is all just statistics. I don’t believe at all those regularities are caused by rules that are mandatory for speakers. Rather, every speaker essentially applies his own methods. Those are methods that he has once learned (also read: constructed), continues to experiment with, keeps operational because he experiences most success with them, and changes for want of success. Again, they are essentially an individual person’s very own methods. For learning is not so much imitating, as it is individual integration or construction. As methods, he can change them. Or he can divert from them as he sees fit to enhance his success. The measure of success is of course how well his interests are served through requests for compliance (also read: conduct through – the causal mode of – sign engineering).

Making sense of Searle’s framework I suppose he simply forgets to distinguish language use from language system. I reach this conclusion on the authority of an earlier quotation where it says (p 15): “I wish to offer certain characteristics and explanations of my use of elements of that language.” But does this really help? What are those “elements”? Are the utterances the elements? Most likely not. Because with already completed utterances, the language is extensionally defined. There would be no need for rules to generate utterances, only for rules to choose among them.3 Or, by elements, does he mean the rules? It is difficult to imagine as his idea, too. For that makes his statement (p 15) “my use of linguistic elements is underlain by certain rules” empty. Or does he mean rules at different levels? But then, what are such levels? It makes me all the more curious about those “linguistic elements” that must necessarily lie somewhere between completed utterances and rules for their generation.

Though I disagree with the assumption of – formally linguistic – rule-based generation of utterances, above I have indicated I don’t want to argue over it too soon. This brings me back to the nature of the rules assumed by Searle. Is he really saying the language system provides all the rules? Is a rule ever violated? Probably, he is also applying a restriction. It would be much like I have done, i.e., by limiting the influence of the speaker’s methods of expression to some respects. From a Schopenhauerean perspective it is in fact simply the will of the speaker that ultimately controls his behavior. As a human being he has recourse to three modes of causation when attempting an exchange with another person (see also § 7.1): cause in the narrowest sense, stimulus, and motivational. With a sign an engineer characteristically aims at the motivation of the (intended) observer(s). That kind of ‘cause’ is in its turn ‘caused’ by the engineer’s interest(s), with surely a chain of intermediary ‘causes’ and ‘effects’ (in a mix of modes of causation, I presume; I speculate that most of its intrabody course, including the sign user’s intellect, is run on the basis of stimuli and effects, rather than signs and – motivationally induced – effects).

When the – attempt at – exchange is seen as a whole, the engineer’s sign is instrumental to his interests. And his sign production faculties – methods, rules, or whatever – are ‘only’ instrumental for the sign-as-instrument.

I fully realize that interests, at least what the sign user – consciously or unconsciously – knows about them as interpretants in his intellect, are intimately related to the sign user’s faculties, including those for sign engineering. I even speculate, after Peirce, that especially his faculties of sign engineering are crucial for interpretant development.

 

 

10.3 the overrated power of language

Suppose a particular intention results purely from, say, internal speech acts. With the engineering and observation of signs strictly rule-governed, it indeed results in intentions that are only and completely determined by the sign user’s language as system. In fact, the language system and intention production system would coincide.

It is obvious that especially speakers of the same language display some highly correlated intentions. I don’t believe, however, that their – experience of a – common language system is the causal factor. Fundamental in my opinion is that every person has interests. With Schopenhauer I believe interests are largely, say, embodied. That is nature. From nurture combined with nature every person also develops his interests. He does so while conducting his life in situations. Where different persons share situations, the effect is that (some) interests can converge. Another effect is that their methods for sign engineering converge in some respects. It is tempting to conclude identity from convergence. However identity, especially at the axiomatic level of conceptualization, almost instantaneously leads to contradictions. Axioms must therefore be designed to accommodate necessary and sufficient differences. Persons do engineer different signs, even when many are indeed similar. One and the same person may engineer one sign differently from another. In general, a rule system is only viable as a theory when it practically constitutes the requisite variety for explanation.4

Regardless of my own – strong and explicit – axiomatic preferences, my concern as a reviewer of Speech Acts is whether or not Searle expresses himself with requisite precision when he writes about “language as rule-governed intentional behavior.” My close reading so far and intermediary analysis lead me to the expectation of discovering, later on in his book, that he finds the rules governing intentional behavior somehow contained in language. Then my conclusion would be that Searle solidifies Austin’s misconceptions. For the relatively harmless formulas of Austin would have been systematized into rules for generating intentions and subsequent behavior. Searle claims an awesome power for language. According to my preliminary interpretation, he holds language, whatever it is, to cause all intentional behavior. Or is it not so awesome, after all? Does he later add severe constraints on what should be considered intentional, and what not?

 

 

10.4 beyond the speech actor

An important issue is how Searle relates the concepts of language and intention. He doesn’t mention intention when he states that (p 16)

speaking a language is performing speech acts[. … T]hese acts are in general made possible by and are performed in accordance with certain rules for the use of linguistic elements.

Where the “certain rules” originate from he still leaves open. His concept of intention soon returns, though, because the scope must be widened from – and here I substitute the more general term – the sign proper to include its production (p 17):

[N]ot only must I assume the noise or mark to have been produced as a result of intentional behavior, but I must also assume that the intentions are of a special kind peculiar to speech acts.

Now, that is interesting, indeed. Are speech acts ‘acted’ for expressing some intentions, and others not? Searle announces that (p 17)

a theory of language is part of a theory of action, simply because speaking is a rule-governed form of behavior.

I agree that sign engineering is behavior. It is action applying the mode of causation aimed at motivationally induced effects. An encompassing theory of behavior must include all three modes of causation. But is Searle saying that speaking is rule-governed, and other behavior is not? Or does he say that certain rules govern speech acts, and other rules govern other kinds of action?

I believe Searle is actually trying to differentiate between what I call the sign on the one hand, and the sign exchange on the other hand. For him it seems speaking is only about signs. Now Searle states that speaking (p 17), “being rule-governed, […] has formal features which admit of independent study.” In the next sentence he makes the distinction between “a study purely of those formal features” and “a study of their role in speech acts.” I suppose he considers speaking not yet a speech act. If so, his use of words is confusing. At face value I read “speaking” as the act of speaking, or being engaged in speech acts. It already becomes clearer with the terminology of sign and sign exchange. How he earlier defines (p 16) “the unit of linguistic communication” is then easily recognizable as another tautology. Of course, a communication unit is not a sign but a sign exchange. Still, Searle only extends the scope from sign to sign engineer. It explains his insistence on intentions. But the sign observer has so far not been included. Now that may be an unhappy result of the terminology of speech act. It emphasizes the actor, not the audience. And when an audience is considered, it behaves relatively passively. The terminology of sign exchange, however, immediately suggests active involvement by all participants, both sign engineer(s) and sign observer(s).

 

 

10.5 what’s in a game?

Searle’s actual scope is already wider than he indicates by including (p 16) “the production or issuance of the symbol or word or sentence in the performance of the speech act.” My impression is, rather, that formal rules are thought by Searle to govern speaking as the act of producing the speech. But those rules do not govern the acting-by-the-speech. Does he specify rules for such acting, too (p 17)?

A great deal can be said in the study of language without studying speech acts, but any such purely formal theory is necessarily incomplete.

This raises my expectation that Searle will complement the “purely formal theory” of language to arrive at a complete theory for speech acts. He also suggests that such a complement will not be purely formal. Isn’t every theory formal? Or does he use formal in the limited sense of rule-governed behavior? He draws the analogy with baseball to explain what the “formal” approach misses (p 17):

It would be as if baseball were studied only as a formal system of rules and not as a game.

It sounds as a neat argument but, instead, causes more confusion. It should only convince when speech act is sufficiently equivalent to “game.” I don’t believe this is clear to Searle. Is he talking about a particular game of baseball? That is, a match? Or about game at the type level, i.e., indicating the possibility of game instances? Or does he mean something like the baseball industry, including television rights, and all? What, precisely, contains his analogy?

Suppose actually playing baseball constitutes a configuration of discrete, say, moves. Exploring the analogy, then every such move would qualify as the equivalent of a speech act. And indeed there are baseball rules at the level of separate moves. The analogy should however have immediately alerted Searle to the practice of many players that they especially seek their advantage in attempting to transgress rules. It is too naive to assume that moves are rule-governed in the sense that moves are always in perfect accordance with rules. As I said, a player may intentionally subvert rules, or his execution of the intended correct move may be so clumsy as to result in a foul move.

Still actually playing, other rules of baseball apply to relationships between separate moves. Move a is all right when it occurs simultaneously with move b, but not with move c. Again, the game rules only provide a framework for behavior while playing the game.

Searle is, I agree, perfectly right if he were he to argue that the – instance of a – game is different from the rule specified at the level of game type. But then, game as type is by definition abstracted from game instances. The concept of rule is ‘typical’ for such abstractions. It is a convenient short-hand for describing regularities between instances. It is quite a different matter to have rules prescribing instances. The essence of a baseball game is determined by the players in action, not by the rules of baseball. Rules, and the referees for that matter, should support the essence to be expressed. From this perspective there are good rules and bad rules. But the development of a particular game that is, in fact, largely controlled by rules is immediately recognized as deterministic and, therefore, boring. Where players cannot make the essential difference, there is no real game in whatever the word’s ‘meaning.’

Another mistake is that Searle even overrates the importance of “purely formal” language rules for the production of speech. My idea is that those rules – when they exist at all in a prescriptive sense – have very little to do with the language game. What a language game is should be understood in the Wittgensteinean sense. Referring to subjective situationism I can make it more specific that the ‘rules’ for a particular language game reflect relevant persons, how they interpret their situations and corresponding interests. In order for it to be an optimally efficient tool it is even a necessity that the language every person uses bears testimony of his interpretations. And thus education of language as a system has a strong influence on interpretations of learners. They internalize, and perpetuate, the language game of their educators. Once internalized, the rules of language may seem to govern their behavior in the language game in question. Or even govern the language game in general. Of course, it does not help to avoid confusion to call a dynamical situation a language game. It directs attention to language rather than to the players of the game. But the nature of language is that of an instrument. It serves to engineer requests for compliance. Language is ‘only’ used to form-ulate, to sign-ify. A sign is engineered from the interests of the engineer, and observed from the interests of the observer.

 

 

10.6 a need for radical reorganization

After he points at the incompleteness of the “purely formal theory” of language for studying speech acts, Searle demonstrates another display of conceptual circularity. He first remarks (p 17):

It still might seem that my approach is simply, in Saussurian terms, a study of “parole” rather than “langue”.

Well, yes, that is exactly what I now expect him to pursue. I believe he is developing the complement of the “purely formal theory” by concentrating on ‘something’ outside language. Isn’t that the purpose of introducing the concept of speech act? Why then does Searle continue as follows?

I am arguing, however, that an adequate study of speech acts is a study of langue.

From my perspective, this would require the concept of langue5 to be considerably stretched. Even overstretched, I say. Is Searle implying langue is the rule system for parole as “a rule-governed form of behavior”? Is language the term he reserves to stand for the system controlling the speech part, and langue the term to stand for the system controlling the act part of speech act?

Another question is: How can I continue with my review? By now I have only progressed as far as the bottom of page 17 of Speech Acts. I am getting more and more confused. I appreciate that, as Searle does, writing an opening chapter on Methods and scope is fraught with risks. For it is impossible to explain in summary what actually the whole book has been written for. Maybe it is all clear to me after reading the book from cover to cover. With less than four pages of it to go I continue with my running commentary until I have finished studying Searle’s introductory chapter. Then I read the remainder of the book. Only after I have done so will I report any other comments.

So, how does Searle explain why speech acts can only be adequately studied as langue? The answer comes from a sentence that contains another axiom that is so very different from what I propose. Searle (p 17)

take[s] it to be an analytical truth about language that whatever can be meant can be said.

Later (p 19) he labels this position the “principle of expressibility” and announces it as “important for the subsequent argument.” He is actually declaring equivalent the two traditional uses of the term of meaning. His view has of course important conceptual consequences for consciousness of intentions, etcetera. However, I don’t try to outguess yet again what must undoubtedly follow. A quotation from the manifesto of the Significs Movement in the Netherlands, founded in 1922, merely establishes here that Searle’s assumption on expressibility is far from generally held (Mannoury, 1948, p 144, my translation from the Dutch):

The meaning of a language act for a speaker and for a hearer can be only partly determined from the words or [other] symbols that are exchanged. It can also be only deficiently expressed in different words.

W. Esch (1930) compares communication with stuttering, for “the ideal expression” is elusive. Searle seems to be comfortable with his assumption, however, and continues with his derivation of concepts (p 17):

There are, therefore, not two irreducibly distinct semantic studies, one a study of the meanings of sentences and one a study of the performances of speech acts.

As no two objects are irreducibly distinct, his statement may be taken as a tautology. However, I take the liberty of another interpretation. Searle is avoiding being too direct but he is actually stating that both studies are identical. Of course they are. But where I argue for upgrading semantics to pragmatics, he proposes to downgrade pragmatics to semantics. He doesn’t say anything about the study of sentences for that is the already well-known “purely formal theory.” Instead, he is not so much interested in what he sees as formal properties of sentences, but in what they mean. And meaning is next conceptualized as performance. With the meaning of a sentence thus equivalent to the performance of a speech act, Searle’s statement amounts to a tautology, too.6 Another matter is that it doesn’t need the preceding statement. For the equivalence of semantic studies – tautologically – holds whether or not “whatever can be meant can be said.” So, why does Searle state the particular axiom?

The conceptual confusion Searle creates originates from his idea that – in my words – a sign can be studied from the perspective of meaning without taking the particular sign exchange in account it is engineered for. But the meaning lies only in the exchange.

In a world that is interpreted as one single situation it is quite understandable that the fundamentally situational nature of meaning – up to a particular exchange – remains unrecognized, or atrophies. Without – the experience of – diversity all of the authority of meaning is mistakenly invested in the sentence, rather than in the sentence use. Searle is an example of a thinker who, like Austin before him, does not realize a radical shift is required to avoid contradictions. They conceptually juxtapose sentence and speech act, with meaning now something to be divided between these two concepts. But, again and again, every division runs into premature contradictions.

Multiple meanings can only be sorted out at a conceptual level that is more finely grained than that of the sign. It appears that necessary and sufficient details of variety can all be adequately explained at the level of sign exchange. But as a concept, sign exchange can only serve its explanatory purpose without premature contradictions when the earlier focus, i.e., on the sign, is given an altogether different position in the conceptual system. A sign is instrumental for a sign exchange, and a sign exchange is instrumental for meaning. The misunderstandings Austin and Searle generate are reducible to their inability to structurally reorganize their conceptual systems at their axiomatic levels. The concepts they introduce are, indeed, promising. But they cannot be just added onto an existing conceptual system. Especially Austin clearly shows how he struggles to fit one concept in, fails, tries another concept, fails again, etcetera. Searle doesn’t pause at his contradictions; he moves off in a particular direction, and continues.

 

 

10.7 theoretical convergence through reduction

Underlying the need for a concept such as speech act is, as I have pointed out above, the experience of more complexity. The transition from one situation, about whose singularity the sign user probably has no conscious knowledge, to many situations that are explicitly known is a paradigm shift. Austin and Searle are as yet unaware of such ontological aspects of the shift of emphasis from sentence to speech act.

Such is my explanation of their continued attempts at ‘fixing’ concepts from the perspective of their existing paradigm, but without any chance of success. Their conceptual ground just doesn’t permit the requisite variety. Searle’s opening chapter presents a case of compromising rather than fundamentally rethinking a conceptual system. Without clear, unambiguous directions from his axiomatic system it is impossible to arrive at equally clear and unambiguous interpretations of his derived concepts. I have only made this show of my extended running commentary to emphasize this point.

That multiple situations of language use are not yet an issue for Searle is borne out by a statement such as (p 18):

The speech act or acts performed in the utterance of a sentence are in general a function of the meaning of the sentence.

I read this as Searle saying that the situationless sign is still carrying most of the burden of meaning. With this emphasis, the required paradigm shift is of course far off. In fact, Searle retraces his steps in favor of all meaning residing in the situationless sign. He starts, from my point of view, promising enough by stating that (p 18)

[t]he meaning of a sentence does not in all cases uniquely determine what speech act is performed in a given utterance of that sentence[.]

I find it promising because Searle takes a step toward further distinguishing between, in my terminology, sign and sign exchange. Instead he returns to his axiom of “whatever can be meant can be said.” Not respecting the “real life” (Austin, 1962, p 143) of signs, Searle wants them to conform to what he believes is a “logical theory.” He negates the power of the concept of sign exchange, separate from the concept of sign, to accommodate multiple situations. The concept of speech act is in fact reduced again to that of sentence because (p 18)

it is in principle possible for every speech act one performs or could perform to be uniquely determined by a given sentence (or set of sentences), given the assumptions that the speaker is speaking literally and that the context is appropriate. And for these reasons a study of the meaning of sentences is not in principle distinct from a study of speech acts. Properly construed, they are the same study.

It seems an eminently rational approach. However, I believe the underlying assumptions about human nature to be extremely naive. What exactly Searle means by “speaking literally” remains unclear. Anyway, he can hardly have concluded from “real life” that every speaker always finds it in his interests to speak literally. Rather than building a theory on the assumption of what a speaker could literally do with language, Searle should first of all have build, or pointed at, a theory that explains why a speaker does not. In fact, a speaker even cannot speak literally when that restricts the determinants of the sign to what is rational, i.e., to what the sign engineer ‘holds’ within his faculty of reason. There are also preintellectual and perceptive determinants of sign engineering (see Figure 6.4.4). So, Searle is basing the convergence of (p 17) “semantic studies” on an idealization that contradicts what is already known about human behavior from Schopenhauer on the will, Peirce on pragmatism, and what has later been confirmed and elaborated upon by cognitive and social psychology.

What I also find peculiar at this stage of Searle’s text is why he supports the distinction between sentence and speech act in the first place when, after all, he reunites them. Is it the power of his present paradigm that forces him to retreat to the position where, as already quoted above, (p 18) “[t]he speech act or acts performed in the utterance of a sentence are in general a function of the meaning of the sentence.” What remains are (p 18)

two different points of view[, …] one which concentrates on the uses of expressions in speech situations and one which concentrates on the meaning of sentences.

I think that, at least in the light of his earlier reduction toward idealized literal sentences, again Searle is stating just one point of view, that is, of sign exchange. But, then again, he is not. A view different from sign exchange concerns the sign. Indeed, such views complement each other. That is the emphasis Searle places here. Those views are, however, qualitatively different. This point for example Austin misses when he attempts (1962, p 91) “to make a fresh start on the problem.” Such differences also don’t come out unambiguously with Searle, as it would first of all require a fundamental reorganization of his conceptual system, including axioms. I believe he is aware of situational variety but does not realize the importance of situation as a fundamental variable to help sort out differences in the “real life” of sign exchange.

A hammer is an instrument for hammering. In its turn, hammering is instrumental for driving in a nail. But starting from a hammer, is it responsible to argue for nailing as the one-to-one equivalent of hammering? The same hammer can actually be used in many different situations, for many different purposes. Of course, the hammer is an object constituting situations. Once established, it is nonsense to define a situation as a function of the hammer. But it could very well be conceived of as a particular use of the hammer.

 

 

10.8 reason in control

Searle concludes his opening chapter with further comments on his “principle of expressibility.” It becomes evident that important ground is missing. Does “whatever is meant can be said” imply that the sign engineer consciously knows what he means? And does he then also consciously control the engineering of the sign? And when he finds the sign lacking in literal content, does he know what is lacking? Does he know how to fix it? I understand very well how the “principle of expressibility” makes a subsequent theory of meaning much simpler. For it is then only required to deal with such idealizations. But what if that principle, as I believe it is, is totally unfounded? It casts serious doubt on derived concepts, too.

Running with his reductions, Searle even proposes a formal notation (p 20) with X for meaning, S for speaker, and E for expression. It looks like a mapping from a set of speaker’s meanings onto a set of expressions. It reminds me of what I have already commented upon as the nature of the rules. What Searle writes on p 15 started me wondering about what he thinks linguistic elements are. I now repeat my puzzlement. Is an expression such an element? If so, are the rules Searle insists upon equivalent to the relation of the set of X’s onto the set of E’s? I am aware my set theory is rusty, so I will have to be careful. But I still know enough not to be detracted by Searle’s notation. The results of symbolic logic are never better than what the axiomatic system already contains. In fact, they are often worse through misapplication. It may look clever, but leads nowhere. So, I will ignore Searle’s formalizations as long as his axioms confuse me.

But far from removing contradictions he surprises with another about-face. For suddenly Searle introduces it is necessary (p 20)

to distinguish what a speaker means from certain kinds of effects he intends to produce in his hearers.

Again, what is missing is a unifying concept such as the sign as a request for compliance. Does a speaker have two types of intentions, perhaps? Are intended effects not meanings? Or will they turn out to be of a different type? If so, what will the type look like that defines the ‘original’ meaning that is set apart from intended effect?

It is not at all inconsequential that Searle claims for his principle of expressibility (p 20)

to account for important features of Frege’s theory of sense and reference.

Gottlob Frege (1848-1925) paves the way for logical positivism and analytical philosophy.7 To philosophers such as Schopenhauer and Peirce, the concept of logic relates to all of the human intellect. Frege reduces logic to reason, and to specific operations on concepts, given symbolical expressions. The next step is the development of the belief that this ‘new’ concept of logic accurately describes the workings of human reason. Once again, the idea is that logic concerns all of the intellect but now with much reduced concepts of both logic and intellect. Together with the idea that the human intellect is nothing but reason, Fregean logic subsequently leads to the belief that logic can be exercised external to intellect, that is, by manipulating symbols. Actually, with his ars combinatoria Leibniz already advocates this view (H. Ishiguro, 1972). It is but another small step to the idea of artificial intelligence, in the sense of imitating and perhaps even exceeding the human intellect.

From a Schopenhauerean perspective the privileged status of modern logic is utterly undeserved. It amounts to a gross oversimplification, resulting in a host of premature contradictions. One of them is that Searle invokes the supposed authority of Frege. To “account for important features of Frege’s theory,” he argues, would have (p 20)

the consequence that cases where the speaker does not say exactly what he means […] are not theoretically essential to linguistic communication.

Perhaps some readers are uneasy about my criticism of Speech Acts. They may feel that I am too harsh. But am I, really? Is it not, rather, that only respect for a person’s interests can lead to a realistic anatomy of meaning? Searle’s theoretical foundation is seductively systematic. But I decline to comply. Unless sentence and speech act are equivalent, after all, it is nonsense (p 20)

to equate rules for performing speech acts with rules for uttering certain linguistic elements.

Even though it might be that (p 21)

for every possible speech act there is a possible linguistic element the meaning of which (given the context of the utterance) is sufficient to determine that its literal utterance is a performance of precisely that speech act[,]

this misses the point about sign exchanges. Essential is not that is it theoretically possible to speak one’s mind literally. It is essential that a person is interest-driven. A sign is not engineered to manifest oneself literally to some degree. The engineer produces a sign, and offers it for exchange, to promote his own interests. And he only initializes a particular sign exchange when he feels it in his interests to involve one or more sign observers to comply with them.

Another criticism is that Searle views the context as given, i.e., existing outside the speech act. I would say situation, of course, but that is conceptually irrelevant. Extremely relevant is that, in my view, situations are also subjectively interpreted. It is a personal variable, too. It must therefore be addressed within the scope of the anatomy of meaning. But then, Schopenhauer teaches that the rationality of the sign is limited (see Chapter 6, above). Or, as Gendlin puts it (1997, p 6)

situations are always wider than the existing language.

Searle will undoubtedly run into yet another, and especially far-reaching, contradiction when speaker and hearer have different interpretations of what they subsequently also act on as the relevant situation.

 

 

10.9 intentional literalism

I don’t extend my practice of running commentary beyond Searle’s opening chapter. Why even dedicate all of eight paragraphs to a text more or less equal in length? I retain my original reading notes, almost unedited, in this finished treatise because they exemplify the limits of rational discussion. For axiomatic systems are involved that are ‘fundamentally’ different. A special problem with interpreting Speech Acts is that, after all, Searle doesn’t make his ontology clear. So far I have done little else than guessing at his assumptions, trying to make sense out of them, and for the most part rejecting them in favor of interest-driven situationism.

It is no doubt a relief to the reader that I don’t have equally detailed comments on the remainder of Searle’s book. In fact, I did not read the last third all that closely. About half-way I finally managed to construct a somewhat coherent axiomatic system, as a framework for – subsequent – interpretation. I believe that intentional literalism is an apt name for Searle’s ontology.

What is literalism? I have combined literal and realism. Searle makes it clear he believes in objective reality. He is less dogmatic than hard-core logical positivists but still argues that (p 78)

we need to distinguish normal real world talk from parasitic forms of discourse such as fiction, play acting, etc.

Mannoury holds a different idea on “speaking the truth” (1925, p 76, my translation from the Dutch):

 “Normal” speech is ultimately always somewhat malevolent, or, let me say, expressive of a different will. It is a kind of wireless control of my fellow human beings in a direction which they would not have taken on their own accord.

It closely resembles some tenets of subjective situationism. Returning to Searle, his criterion for distinction is that (p 79)

in real world talk one can refer only to what exists.

Searle is actually stating that any person can unerringly tell reality from fiction. Not only that, every person is also able to be conscious about all real objects. He therefore assumes a mapping between real objects and, say, real meanings. Next, every person is capable of giving literal expression to his real meanings. So this involves another mapping, this time from a real meaning to an expression, that is, using a particular language. According to Searle’s principle of expressibility, a literal expression is always possible. What he probably means is that such an expression is unambiguous, complete, etcetera. And it is an honest expression, too, for (p 112)

[a] man is committed to the truth of whatever he asserts.

That is, I believe, not how it really works. Anyway, so much for the literalism part of the label. Searle’s variety is intentional in that he apparently considers intentions to exist objectively, too. I am not sure whether or not to accord special status to intentions. Is Searle saying that an intention, as a real object, immediately is a real meaning, too? Or is an intention also mapped onto a real meaning? At least, being a real object an intention can also be literally expressed.

Literalism implies that different persons can have identical meanings. In fact, as soon as a particular person ‘knows’ about a particular real object he holds the ‘standardized’ meaning about it. Literalism does not enable shared meaning. It does not explain how it might originate from differences. Instead, it even assumes that different persons hold identical real meanings once they have been ‘told’ about corresponding real objects.

Intentional literalism holds that intentions are a special kind of real objects. They involve – other – real objects. In particular, an intention is the real object that encompasses the relationship between the person and one or more other objects. One and the same person may have different relationships with identical objects; that accounts for different intentions.

A neutral statement about real objects, other than intentions, is impossible according to intentional literalism. A person who makes such a statement is supposed to perform an act of assertion. Thus is all “normal real world talk” intention-based.

Actually, all such talk is now assumed to be about communicating intentions as relevant real meanings (p 43):

In speaking I attempt to communicate certain things to my hearer by getting him to recognize my intention to communicate just those things. I achieve the intended effect on the hearer by getting him to recognize my intention to achieve that effect, and as soon as the hearer recognizes what it is my intention to achieve, it is in general achieved. He understands what I am saying as soon as he recognizes my intention in uttering what I utter as an intention to say that thing.

I think the key to arriving at what Searle is after lies in his last sentence. His label of intention refers only to “an intention to say that thing.” He limits the intention to the speaker saying it, for the implicitly underlying assumption appears that the doing of communication is about sharing meaning.

Why does the speaker really say something? I don’t want to confuse matters even more, but what is the interest behind the intention? For some time I have understood Searle’s terminology of intention as synonymous with my terminology of interest. Upon closer inspection there appears to be no serious psychological ground in Speech Acts, at all. The scope of its concept of meaning is extremely limited. An attempt at getting such an intention, including other real meanings, across is what Searle calls an illocutionary act (p 47):

In the case of illocutionary acts we succeed in doing what are trying to do by getting our audience to recognize what we are trying to do.

Again, I don’t think I “recognize” in Searle’s sense what he is “trying to do.” His sentence nevertheless makes it a little clearer why he insists on the difference between illocutionary and perlocutionary acts. With an illocutionary act (p 47),

the ‘effect’ on the hearer […] consists simply in the hearer understanding the utterance of the speaker. It is this effect that I have been calling the illocutionary effect. […] The characteristic intended effect of meaning is understanding.

Is this really what all the trouble is about? Is illocution “simply” another word for communication? In fact, Searle restricts his research area from communication in general to “normal real world talk.” He doesn’t want to be bothered by “parasitic forms of discourse.”

In “normal real world talk,” a person may be “trying to do” different things with respect to the other real objects that are, as it were, encapsulated by his attempts. Suppose that a copy of this treatise is the real object in question. A person may promise to give another person the particular copy. Or he may ask to borrow the copy. Or he may assert that it has been written. The nature of his relationship to the copy of this treatise, i.e., his intention, is expressed by promise, ask, assert, etcetera. Such an (p 30)

illocutionary force indicator shows how the proposition [about the other real objects] is to be taken, or to put it another way, what illocutionary force the utterance is to have; that is, what illocutionary act the speaker is performing in the utterance of the sentence.

I have not, however, finished my (re)construction of intentional literalism. The illocutionary force indicator seems to represent the most finely grained classification level on “how the proposition is to be taken.” Searle is giving examples of (p 54)

conditions [that] are necessary and sufficient for the [corresponding] act […] to have been successfully and non-defectively performed in the utterance of a given sentence.

In circular fashion Searle remarks that (p 54)

[i]f we get such a set of conditions we can extract from them a set of rules for the use of the illocutionary force indicating device.

And that is precisely the ‘result’ of his exercise! The rules thus ‘derived’ are merely restatements of conditions. And the conditions are so severe that, actually, any sentence would ‘do.’

The dream of Searle is clearly recognizable from the following statement. It is about the perfect language, perfect because it literally – whatever that is – maps real objects onto expressions. His ideal is to make sentences computable but (p 64)

[p]art of the answer to this question would depend on whether we can reduce all illocutionary acts to some very small number of basic illocutionary types.

Well, he makes a start by only taking “normal real world talk” seriously. Next, Searle confines the real world to such normal talk. My conclusion about the ontology of intentional literalism is that its real world is all illocution. Searle’s real world is fundamentally conventional. Though he gives the impression that language reflects the order of the real world, I believe he actually starts from his idealized picture of ordered language, called illocution, and structures his real world after that image. It is a fundamentalism in a narrow-minded sense.

Philosophy of language, as practiced by for example Austin and Searle, is actually still grounded in the idea that a word contains the essence of an object. And that a natural language,8 seen as rules governing configuration of words, contains the essence of the world of objects. It is enlightening to contrast this with De Saussure’s perspective on language (1916, p 68):

No one disputes the fact that linguistic signs are arbitrary. But it is easier to discover a truth than to assign it to its correct place.

In my opinion, the correct place is not to start from a language system. Nor is it to subsequently stay within the limits of language, when trying to explain human behavior. For it leads to an extremely limited scope for meaning, i.e., to “normal real world talk,” only. To declare illocution the norm – and everything else in sign exchange “parasitic forms of discourse” – only leads to an impoverished anatomy of meaning. The ontology of intentional literalism is barren. It takes language more seriously than language users. Moreover, the concept of illocution is a complex construct to explain something that is trivial when approached within a larger framework. And it does not even convince at explanation.

I don’t see any reason to relinquish the general concept of sign. A sign is a cause produced by a sign engineer for effect. The intended effect is not necessarily the understanding by the observer of the engineers intention. The effect essentially aimed at is a response by the sign observer, in whatever mode of causality. The engineer measures the success of the exchange of his sign after the extent to which the observer’s (re)actions comply with his interests. More often than not, the sign is more effective when it does not reveal the engineer’s interests ‘behind’ it. It may be considered morally regrettable, or even objectionable, but this is obvious from a psychological point of view. Ignoring it, language philosophers manage to take the ideal of intentional literalness seriously.

 

 

10.10 much ado about nothing

It is of course an appealingly nice idea. When speaking and acting are considered as different, shouldn’t there also be some overlap? Austin comes close to admitting that speaking is acting. Maybe he feels their relationship should be symmetrical. But not all acting is speaking. Speech is limited to just one of the three modes of causation.

Indeed, this obstacle is easily taken after different modes of causation are recognized. Speaking then, is acting in a particular causal mode. It is the mode of aiming at motivationally induced effects.

Language philosophers probably find it impossible to apply the conceptualization of cause and effect. Are they afraid the terminology associates with Newtonian billiard balls colliding? However, mechanics pertains to only one out of three causal modes, i.e., to “cause in its narrowest sense,” or physical phenomena.

Whatever reasons they have,9 such thinkers fail to acknowledge the opportunities for efficiency of signs over other causal modes. Suppose John wants to have one of Bill’s lower legs lifted again. How can he get Bill to comply with his interest, with himself spending the least effort? With three modes of causation to choose from, engineering a sign is often most efficient. But then, often enough it might not be. It depends on their relationship, and particular situation (both as understood subjectively, and thus separately, by each of them). Anyway, I want to make it clear that sign use can be perfectly logically integrated in an encompassing framework of action.

Because language philosophers such as Searle and Austin don’t allow themselves to theorize in terms of cause and effect, they are forced to struggle with other explanations (thereby invoking cause and effect, nonetheless). What Austin doesn’t grasp is that his initial problem statement already entails the conceptual scheme of cause and effect. Actually, somebody trained as an engineer will rightfully expect from the introduction of the concept of action that language use is going to be explained in terms of cause and effect. That is what I expected to find, anyway. On the contrary, Austin attempts to explain action in terms of what he thinks is essential of language use, and that is meaning. There is no integration. He misses it, I guess because he lacks a background perspective on the individual language user. For it is precisely the sign user who may modulate from one ‘mode’ of causation to another.

The irony is that the type of act Austin feels that does not really belong to his theory, is in fact the only relevant type seen from the wider perspective of different modes of causation. It is, of course, what he calls the perlocutionary act. He actually considers perlocution a nuisance for a consistent theory of meaning. My opinion is that he only could have arrived at a consistent theory by recognizing that the whole purpose of a specific sign is not something evasive such as shared meaning but getting the observer to comply with the interests of its engineer.

How to Do Things with Words is a great slogan, but it doesn’t really correspond to Austin’s theory of illocution. His start is still consistent with a focus on action. Though not systematically articulated, he is actually – but not originally – stating that there is more about statements than objective truth or falsity. Without absolute truth value, a statement expresses an opinion. (What I here call an opinion is what Austin calls an intention.) Then, what a speaker ‘does’ with words is communicating his opinion. Because a speaker can now communicate nothing but his intentions, actually every speech act is an illocutionary act. Please note that, at this stage of his book, the theme is almost reversed. His initial examples are not so much about individual opinions but concern the performance of rituals. The ritual, Austin argues, is ‘done’ by making the statement.

The orientation at ‘how to make intentions clear with words’ is further pursued by Searle. In the process he denies the relativity arrived at by Austin. Meaningful statements are true or false, once again. The illocutionary act consists of getting the speaker’s intention-as-meaning across to the hearer. However the hearer may act himself is considered perlocutionary, and therefore outside the scope of the speech act. As an engineer myself,10 with a differentiated understanding of cause and effect, I find it confusing that Searle titles his book Speech Acts.

Why do theories such as formulated by Austin and Searle find supportive audiences? My idea is that they allow a person to maintain a logical atomistic perspective on reality. Such axioms are even reinforced. For the objectively true statement has been preserved; a proposition informs about an object and its – relevant – properties. However, the speaker necessarily intends the hearer to interpret that proposition in a particular way, that is, according to the speaker’s – relevant – intention. That is why the, say, intentionally neutral proposition is enhanced with, say, instructions for interpretation. Those added instructions give the originally neutral proposition a particular illocutionary force.

It is an attractive theory, especially when so-called illocutionary forces are classified, even exhAustingly and stable by a simple, short list of labels. Any complete statement then becomes a composition of, first of all, a neutral proposition and, secondly, an illocutionary force indicator. Neutral propositions, in their turn, are composed following the rules of subject/predicate logic. The subject represents the object, and the predicates describe the object’s properties. And the illocutionary force indicator is simply selected from a list. It is easy to see why, for example, the artificial intelligence community enthusiastically welcomes speech act theory.11, 12

This combinatorial theory of meaning is elegant but, in the majority of situations, too simplistic. It breaks down because every sign that is engineered ultimately originates from the sign engineer’s will. Manifestations of his will are aided by his intellect as instrument. It makes a sign not only irrational to whatever extent, but intrinsically subjective, too. There is no such thing as a neutral proposition. Actually, sign and proposition may as well be taken as equivalents when the latter’s association with symbolic logic is dropped. Figures 7.5.6 and 8.1.1 outline aspects that are engineered into, and interpreted from, respectively, a sign/proposition. Again, concepts such as neutral proposition and illocution are just too simple for a general anatomy of meaning.

Authors like Austin and Searle ‘cause’ confusion with their concepts of speech act and illocution. Especially Searle reasons from a depersonalized, mechanistic worldview. Whatever theories of information modeling uncritically build from speech act theory and language action paradigm therefore lack conceptual grounds designed for requisite variety. But please, don’t just take my speech for it. Act on it yourself. My discussion hopefully inspires detailed study of their popular publications before applying their concepts.

 

 

notes

1. Lemma Speech Acts by A. Burkhardt in: Handbook of Metaphysics and Ontology, (pp 854-856; H. Burkhardt and B. Smith, editors).

2. In a footnote on p 9, commenting on publications by W. Quine and N. Goodman, Searle writes: “I am here concerned, however, not with the development of their thought as individual philosophers, but with a certain pattern of analysis in philosophy of which these two works are well-known and powerful examples.” It is not a convincing display of diplomacy, I would say. For he indicates that “these two works” are setting the pattern he rejects. What else can it mean than that Searle was, in fact, highly concerned about “the development of their thought.” About Austin and Searle I remark, in my turn, that I am concerned about the development of their thought as I know it from their works. I am especially concerned because, with their well-known and powerful publications, they have set a pattern of analysis in philosophy, and far beyond, that I find counterproductive for the purposes that persons apply it to.

3. An example of what feeds insecurity when reading Speech Acts is also that Searle mentions (p 10) “a given extensional criterion.” I understand extension as the absence of any criterion to define a set. When criteria apply for membership, that is intension. With extension, a set is determined by the very individual elements that are its members. Perhaps Searle uses extension differently. The most I can make out is that he may have mistaken extension for intension.

4. With requisite variety, I refer of course to Ashby (1956).

5. The distinction between langue and parole derives from De Saussure (1916). See the beginning of § 5.1 for my introduction of these linguistic concepts.

6. From the wider perspective of the anatomy of meaning presented in this treatise, the tautology is immediately recognizable. Meaning of a sign is the process of sign exchange.

7. See Frege, an introduction to the founder of modern analytic philosophy (1995) by A. Kenny.

8. Natural language, and its analysis, have become popular paradigms for design of information systems. Much as I understand the seductive power the promise of simplicity must hold, I regret this, especially when such methods are proposed as all-encompassing and guaranteeing success. That is nonsense. My criticism of Austin and Searle should inform that such approaches are only valid under strict conditions. So strict, even, to render ‘methods’ of natural language analysis trivial.
A fundamental objection that I have is that so-called natural use of language more often than not hides interests. My hypothesis is that, in general, the engineer does not directly elucidate his interests because that would make the observer, in complying with those, feel denying interests he himself has. Nor will the engineer invoke too openly the interests of the observer for that would make the observer come under obligation when complying with this particular request. Therefore, in “normal real world talk,” interests are not directly referred to in the sign. Instead, the engineer will invoke, say, an authority that is higher than both the observer and himself. When the observer recognizes that ‘objective power,’ the engineer secures compliance without losing any credits, on the contrary, for his future requests. What is suggested to hold such objective power may be different according to the situation. A religious person may think he obeys his god(s) for certain activities, and laws of physics for others, etcetera. Now my point is something like linguistic darwinism. The language a person uses is an instrument for engineering optimally effective and efficient signs, i.e., requests for compliance. When my idea about trying to place the observer, not under the engineer’s, but under their common outside and supposedly higher authority is relevant, it only follows that language develops as an instrument to suggest objectivity. Therefore, the nature of whatever language is a consequence of how one person attempts to organize compliance with his interest by others. It surely is not the origin of meaning. There is no higher authority than the will, as Schopenhauer says. This makes every person, as individual (also read: unique) objectification of the will, first and last his own authority. I do not believe that Schopenhauer has already been acknowledged as an early existentionalist but that is precisely what he is, too.
A more practical objection I hold against overemphasizing the use of natural language for analysis of information requirements is that such analysis does not address the essentially systemic character of the tool to be designed. A system is not merely an aggregate. From an inventory of requirements – and why not apply, among other methods, analysis of the use of natural language by stakeholders? – conceptual derivation, often involving abstraction, is necessary for creative synthesis. Analysis may be a science, but design is an art. Designing is not mechanical, at all. Suppose that it is my role to act as modeler. Rather than assuming that personal interests do not play an important part in business problems, my consistent experience is that it pays to honor them. But this means that I have to interpret beyond appearances. I have treated that theme in my book Informatiekundige ontwerpleer (1999). It clearly shows the dilemma of the professional modeler. He is mediating interests of others, but in the process he becomes involved with his own interests.
I find my design of an information system for financial accounting, documented in Aspecten en Fasen (1991) an excellent example of how superior results will probably even be obstructed by taking too seriously what stakeholders literally say. Nobody ever told me about phase and dimension. Yet, those are concepts central to my design. I invented them for that particular type of situation. I arrived at a synthesis that accounts for whatever stakeholders could articulate on financial accounting, precisely because I choose not to get distracted by what they might say. It was evident that a synthesis was needed for their interests. How could they be respected at an individual level, and yet be integrated? Their highly varied statements, and only a limited number, I considered as samples providing inspiration to design beyond them. Favorite sources of inspiration are also textbooks. A good textbook, too, is not an aggregation of what persons actually do, but a creative abstraction allowing/explaining generation of a wide variety of behavioral instances.

9. Typical is also H.P. Grice (see note 20 in Chapter 7). Interestingly enough, Searle takes argument – mistakenly, I believe, but that is not the real point here – with Grice. He attributes to Grice the opinion that (p 44) “[p]ut crudely, Grice in effect defines meaning in terms of intending to perform a perlocutionary act.” I do not read this from Grice’s article Meaning (1957). In general, Searle appears very polemical, but regretfully without much empathy. Apart from his theory being defective, his account of it is also not endearing.
What the quotation from Searle makes especially clear is the effort to keep what I consider normal reactions to actions out of the central theory of speech acts. A telling example is where he states that (p 46) “there is no associated perlocutionary effect of greeting.” Searle continues that a greeting is only about installing “in my hearer […] the knowledge that he is being greeted.” I can only find this an amazingly naive perspective. Especially greeting should have directed his attention beyond getting the hearer to understand the speaker’s intention. Greetings are exchanged in order to establish, maintain relationships. Suppose that somebody declines to greet you, consciously you assume. Now that is a sign, too, of course. But the absence of the expected sign will make you think about your relationship with the non-greeter. Actually, you may even be fully complying to the interest behind the non-greeting. These underlying behavioral aspects are all neglected by Searle as he is concentrating on (p 16) “ language as rule-governed intentional behavior.” I repeat that he does not succeed in making it clear to me what that is. Many variables that I consider relevant are missing from his theory.

10. An example of lack of consideration of system complexity (see also note 8, above) is Searle’s following explanation of his method of investigation (p 33): “Until we can get clear about the simple cases we are hardly to get clear about the more complicated ones.” Elsewhere it reads that he is (p 56) “going to deal only with a simple and idealized case.” I recognize it as a preferred method for teaching. But, then, students should be made aware of qualitative differences when more complexity is introduced. It will do for many purposes to assume that the earth is flat. When navigating large distances it is obviously necessary to acknowledge that the earth is a sphere. The rules of navigating flatness are quite different from dito rules for sphereness.
A theory must always consider the maximum relevant complexity. A good theory will also account for simples cases, often by ‘just’ fixing one or more variables. Searle is trying to reassure the reader that extrapolation from simple cases to complex cases will be unproblematic. It fits his principle of expressibility but “real life” needs a richer “logical theory” for proper explanation.
In § 7.5, I have reconstructed my development of the engineer-based sign structure by starting from what I consider the most complex case. How is it possible that even a single, short sound can serve as an unambiguous request for compliance? With such a case explained for, simpler cases often immediately fall into place.

11. The more conventional field of management information systems has also been afflicted. Illocutionary force indicator sounds impressive but, when modeling a management information system, I cannot think of any difference with a good-old status indicator. Different states allow an object to be tracked through a process. Of course, a particular actor attributes a particular state. The actor should therefore be referred to when such intentions need to be tracked, too.

12. While starting out from traditional artificial intelligence, H. Hendriks-Jansen has departed from it and developed a rich, general approach to understanding (human) behavior in Catching Ourselves in the Act: Situated Activity, Interactive Emergence, Evolution, and Human Thought (1996).

 

 

2002, web edition 2005 © Pieter Wisse

 

 

table of contents; previous chapter; corresponding prelude; next chapter.