ISSA Proceedings 2010 – The Rationality Of Rhetoric: How To Cope With Human Limitations
The Problem: Obeying rules of pragma-dialectical model in real life is unreasonable
Within the pragma-dialectical theory of argumentation (van Eemeren & Grootendorst 2004) discussants try to resolve a difference of opinion in a maximally rational way [i]. These rational agents are willing to engage in long-lasting and most complex discussions and sub-discussions when assessing the plausibility of standpoints. Other needs have to stand aside. In order to account for rhetorical moves, the concept of strategic manoeuvring has been added to the pragma-dialectical model (van Eemeren & Houtlosser 1999, 2006; van Eemeren 2010), with rational agents aiming for rhetorical effectiveness while still maintaining dialectical standards of reasonableness. However, the extended pragma-dialectical argumentation theory does not account for systematic interaction between rhetoric and dialectics. Rhetoric is a supplement that may be taken into account, a non-rational appendix to rational argumentation that has to subordinate to the demands of the dialectical rules (cf. a similar critique by Hohmann 2000).
A specific problem arising from the idealizations of the pragma-dialectical model is that it cannot be implemented in real life. As pointed out by van Eemeren (2010, p. 4), “the ideal of a critical discussion is by definition not a description of any kind of reality but sets a theoretical standard that can be used for heuristic, analytic and evaluative purposes”. The model establishes normative standards of reasonableness for criticizing arguments, but it does not provide rules for constructing rationally justified arguments in practice.
To illustrate this last point, let us see where the ideal model of pragma-dialectics takes us, if we strictly obey its rules, i.e. if we proceed in a strictly rational and dialectical manner. According to rules 7 and 8 of the ideal model (van Eemeren & Grootendorst 2004, pp. 147–151), all premises and justifications of an argument that were left implicit need to be reconstructed, in case of any doubt, by means of the intersubjective explicitization procedure and the intersubjective testing procedure. These procedures ensure a mutual understanding of the premises and argument schemata that have been used in an argument, and they test whether these premises and schemata are admissible and have been applied correctly. One can imagine how large the expenditure of time would be in real life if agents would follow these rules. Almost every argument contains one or the other implicit premise. The propositional content of statements is fuzzy and the formal shapes of argument schemata are far from clear. It may take hours for the discussants to agree on the precise content of a proposition or the shape of an argument scheme and its applicability. Usually, the validity or invalidity of an argument depends on just those formal and semantic particulars (cf. also Krabbe 2007 on the functional overload of the opening stage with such issues).
Perfectly rational agents, however, would never let this keep them from resolving their difference of opinion in a maximally rational way. Thus, they accept a rule that one would better not insist on in real life: The protagonist may at any time retract any speech act that he has performed (rule 12 in van Eemeren & Grootendorst 2004, pp. 153f.). This is to say that the antagonist has to accept that the protagonist puts forth claims consecutively, just to retract them one after the other. Expenditures of time carry no weight in the ideal model, after all. This course of action is rationally justified as long as the testing out of several claims serves rational objectives. There is only one thing that must not happen even without any time pressure: discussants must not end up with an infinite regress. That is why the following rule holds in the ideal model: The protagonist and the antagonist may perform the same speech act only once, and they must in turn make one move of speech acts (rule 13 in van Eemeren & Grootendorst 2004, p. 154).
The rules presented so far are normative. Any deviation from the rules counts as a fallacy, i.e. as a deficient move in argumentation (van Eemeren & Grootendorst 2004, pp. 174ff.), a derailment of strategic manoeuvring (van Eemeren & Houtlosser 2006, pp. 387f.; van Eemeren 2010, pp. 187ff.). If we compile a catalogue of those fallacies, we find quite useful moves on this list such as: presenting pros and cons in a systematic way is fallacious as you are allowed only one speech act in turn; at the same time, repetitions of speech acts, e.g. due to noise or misapprehensions, are not allowed in the discussion; premises that are taken as a matter of course must not be left implicit, but have to be made explicit as soon as an argument is challenged; the same holds for argument schemata, they have to be made explicit and be tested for their correct application.
The fact that these rules are hardly ever met in real life need not be of any concern to the ideal model, as it is absolutely legitimate to make idealized assumptions. The more so as these rules are not meant to be used for conducting real-life argumentation. What is astonishing, however, is that adhering to these very rules of the ideal model seems highly unreasonable in real life, although the rules should specify a rational course of action. Why is it, then, that not following the pragma-dialectical rules seems reasonable rather than irrational?
2. A problem analysis: Human constraints are not taken into account
Although the ideal model might work in an idealized world it would hardly be applicable in real life. And the reason for being so seems quite obvious: human beings are by nature subject to various constraints, and it is these constraints that make obedience to the rules seem irrational. Among the most important human constraints are the following.
(1) The limit of time: Humans do not live forever, and therefore they cannot discuss issues forever.
(2) The limit of information: Humans only have limited access to the information relevant to their decisions. Sometimes they have to argue on the basis of premises the applicability of which has to be assumed but just cannot be verified.
(3) The limitations of memory: Sooner or later, humans forget the things they hear. Most humans are not able to follow a discussion without losing one or the other information.
Humans cannot pursue the resolving of a difference of opinion in a perfectly rational way simply because they are not perfectly rational agents. Instead, they have emotions and intuitions, which they rely on in social contexts, and this is what they do within discussions, too.
By largely ignoring these limitations the pragma-dialectical model decreases its applicability in the real world. Nonetheless, a more applicable model can be derived from the ideal model by systematically taking into account the limitations of human beings.
3. The solution: A rhetorical model of argumentation
The question then is: If agents are aware of their limitations, how could they best deal with them? How may they arrive at a result that is as close to the ideal result as possible? Rhetoric offers answers to these questions by recommending well-proven, problem-oriented guidelines for discourse. Rhetorical considerations permit the effective composition of a speech. They cannot neutralize human constraints, but they can reduce the negative effects of these constraints.
A praxis model of rhetoric has to be put next to the ideal model of pragma-dialectics in order to understand the rationality of real-life argumentation. It is not idealized, perfectly rational homines dialectici that act within such a praxis model, but homines rhetorici with limited time, limited rationality and limited memory. Homo rhetoricus is quite aware of his limitations, and he tries to reach the best result under the given circumstances. He knows about his limited memory that makes him forget things. He knows that supposed premises may be false and that this could lead to false conclusions. He knows about his limited rationality that goes against rationally justified results. However, he tries to get the most out of the resources available. His objective is to persuade the recipient nonetheless. He merely succeeds in reaching a compromise between invested time and desired thoroughness, between logical complexity and logistic efforts, between plausibility and rationality (a similar idea can be found in Jacobs 2006). The sustainability of these compromises must prove in the course of time by success or failure of diverse rhetorical strategies and by their consequences in practical life.
3.1. Two simple examples: Alliteration and metaphor
Three examples (two simple ones and a more complex one) may illustrate the idea. The rather simple ones concern alliteration and metaphor. Below are given some well-known advertising slogans.
(1) Don’t dream it. Drive it. (Jaguar)
(2) Britain’s best business bank (Allied Irish Bank)
(3) Today Tomorrow Toyota (Toyota)
(4) Persil – washes whiter. (Persil)
Advertising slogans need to be short and memorable to be successful. And the memorability of the slogans just cited is established by alliterations. Those mnemonic sentences are imprinted not only on the speaker’s memory, but also on the hearers’ ones. With respect to a praxis model of rhetoric this means that figures of repetition, like alliteration, are a direct answer of rhetoric to a concrete problem that homo rhetoricus has, namely that of limited memory.
The second example concerns metaphors in science. I choose the Bohr Model of atoms. Bohr’s model depicted atoms as small, positively charged nuclei that are surrounded by electrons, and these electrons travel around the nucleus just like the planets move around the sun in our solar system. Although this model is obsolete in modern physics, the metaphor is still alive in modern theories that speak of atomic orbitals, electron clouds, and wave-like behaviour of particles. These metaphors acquired the function of names for abstract relations. It seems that metaphors like “orbit” or “path of an electron” are helpful, if not necessary, to envisage extremely abstract configurations. If I think of an “orbit” and the “path of an electron” I automatically think of small globules revolving around a central nucleus, i.e. I am transferring a concrete image in my mind to an abstract relation. This is an original rhetorical technique – with all its problems and dangers. Metaphors help to imagine abstract ideas. They transform abstract entities into concrete entities. And it is the concrete things that humans can best think about. Metaphors thus fill in linguistic gaps so that we may articulate concepts that we otherwise would not have been able to talk or even think about. With respect to my praxis model of rhetoric, this means that linguistic and cognitive limits of homo rhetoricus are compensated for by the rhetoric mean of metaphor.
3.2. A complex example: Usage declaratives
As a third and last example the rhetorical function of usage declaratives is to be analyzed. Usage declaratives are speech acts that explicate the usage of a word, for example definitions or paraphrases of a certain term. From the language economic point of view, paraphrases of a term (and the like) are violations of the commandment of brevity: “If you can say it with fewer words, then do so!” The use of more words than necessary is justified only (a.) if quality rises with quantity, that is: if you can say it more precisely by using more words. Or (b.) if it saves you words in the long run by introducing definitions.
The rational justification of the second possibility is quite straightforward. If one needs fewer words by introducing new terms, then the usage declarative indirectly meets the requirements of brevity. But what about the first possibility that quality rises with quantity? What is the rational justification from homo rhetoricus’ point of view? Does not the use of ambiguous terms offer rhetorical advantages, if you do it right? The solution proposed here goes as follows: Homines rhetorici are well aware of the fact that they do not have precise expressions for everything in their language. However, their limited rationality suffices to recognise that imprecise wordings may lead to misunderstandings. If discussants understand one and the same term in different ways, for example, they might think that they have a difference of opinion, though they both agree concerning the issue and only construed a term as different meanings. Or the other way round: they use the same term, but mean different things. It might appear as if they agree, although they diverge in substance.
But every speaker knows, at the same time, that his audience knows about the problem of vagueness. Homo rhetoricus anticipates this problem in his communication, and he tries to avoid any obscureness that could, from his point of view, become a problem. It is because of the available language, the limited rationality, the limited time for preparation, that he cannot avoid all ambiguity. It is not because he would act in bad faith.
3.3 The functionality of ethos
But why, then, should homo rhetoricus not deceive and mislead his listeners by vagueness? After all, he subordinates everything to the goal of persuading his audience. The reason is that there is a subsequent speech for every homo rhetoricus, when he has to step in front of an audience once more, and the audience again knows about the problem of vagueness. If in the meantime it should prove that he manipulated and misled his audience last time, then he would find it much more difficult to persuade his audience once again. (This is not the universal audience that Perelman & Olbrechts-Tyteca 1969 employed and which Tindale 2006 also relies on to ensure rationality. It rather is a particular audience consisting of imperfectly rational individuals).
Rhetoric introduced the technical term ethos, denoting the overall moral character of a person, his habits, his conducts, and his convictions. Every homo rhetoricus carries around with him such an ethos mark. Every convincing speech raises his ethos in the listeners’ view, if it proves of value in the long run. Every speech that turns out to be demagogic lowers his ethos in the listeners view. Ethos is a moral asset. Homo rhetoricus cannot afford to squander his credibility because his actions are geared towards long-term success. His arguments are always evaluated against the background of his credibility. On the one hand, the arguments of a notorious liar do not count. On the other hand, it is only with great effort, that the arguments of an acknowledged authority can be challenged.
If, for example, the sky diving instructor tells me to put on the harness this way around, as otherwise I should not be safe, then I would need very good reasons for rejecting his advice. In case of emergency, it does not occur to many of us to question the expert opinion and trust the lay assessment instead. The instructor has a self-interest in his customers’ reaching the ground safely. His reputation depends considerably on that. This is why he would not mislead us. But if an unknown skydiving pupil tried to convince me that it would be a better idea to put on the harness the other way round, then I have every reason not to let me be convinced. There is not enough deposit in his ethos account. Even if his arguments sound as plausible as possible, he still would not be able to compete with the instructor’s opinion.
Taking ethos into account, effects that in rhetoric the status of a person gains importance. Which is, from a pragma-dialectical point of view, a deviation from the rational course of action. But in practice we have to rely on the assertions of other people, as no one can know everything and verify everything. And this is why the accumulation of credibility – of ethos – is so important. Scrutinizing all proponents’ standpoints in conformity with the pragma-dialectical rules would impede not only all of our communication activities, but would impede most of our actions.
3.4. Rules within the praxis model
Certain rules hold within the praxis model of rhetoric, which are normative, just as the rules in the ideal model of pragma-dialectics are. In contrast to the dialectical ones, though, these rules have to be applicable in practice and have a chance to lead to results in real life. A normative persuasion rule is on top.
(1) Persuasion rule: “Try to maximize your success of persuasion in the long run!”
The main objective of homo rhetoricus is to win discussions. He wants to persuade others, not figure out the truth. The ethos mechanism acts as a counterbalance to this dangerously egocentric rule.
(2) Ethos mechanism: “Every conviction effected by the speaker that proves untenable, lowers the ethos of that speaker, and therewith the persuasive power of all consecutive contributions of the speaker who is made accountable for effecting the untenable conviction.”
As homo rhetoricus is to maximize his long-term success over a long sequence of contributions, he needs to take into account the ethos mechanism whenever he puts forward an argument, since the ethos account cannot be high enough for reaching the long-term success.
Regarding the disposition of a speech, I assume a normative rule of disposition.
(1) Disposition rule: “When speaking, take into account the constraints that you and your recipients are subject to.”
The constraints mentioned here regard the available time, language, memory etc. The use of various rhetorical means can be derived from this rule: shortening, amplification, repetition, and metaphor. These methods are permitted as long as they serve the resolution of a problem that arises from the limits of homo rhetoricus.
No more rules are needed to get the model started. The interaction of the ethos mechanism and the normative rules should result in the effect that it would be unreasonable and irrational for homo rhetoricus to pursue persuasive success by rhetorical tricks. The looming decline in ethos prohibits short-term thinking.
Limits of time, language, rationality, and so on prevent human beings from strictly obeying the rules of the ideal model. The most rational solution to this problem is to deviate from the rules. The rhetorical model offers a rational justification for a compromise between an ideal acceptability check and the constraints that apply in practice. This compromise is associated with both a cost and a promise. The cost consists of uncertainty whether the maximally rational solution has been reached. The promise is that no better solution could be reached under the given circumstances.
The optimality of rhetorical compromises can only be guaranteed over a whole series of discussions. Hence the most important rule within the rhetorical model is: “Try to maximize your success of persuasion in the long run!” The ethos mechanism acts as a counterbalance. It assures that every untenable conviction effected by the speaker lowers the ethos of that speaker. And this also lowers the persuasive power of his consecutive speech acts. Various rhetorical means can be derived from the rule of disposition. These figures are aimed at dealing rationally with the constraints of time, language, memory, and so forth.
i I would like to thank two anonymous reviewers for their helpful comments on an earlier draft of this paper. The remaining shortcomings are my own.
Eemeren, F. H. van (2010). Strategic maneuvering in argumentative discourse: Extending the pragma-dialectical theory of argumentation. Amsterdam: John Benjamins.
Eemeren, F. H. van, & Grootendorst, R. (2004). A systematic theory of argumentation: The pragma-dialectical approach. Cambridge: Cambridge University Press.
Eemeren, F. H. van, & Houtlosser, P. (1999). Strategic Manoeuvring in Argumentative Discourse. Discourse Studies 1 (4), 479–497.
Eemeren, F. H. van, & Houtlosser, P. (2006). Strategic Maneuvering: A Synthetic Recapitulation. Argumentation 20 (4), 381–392.
Hohmann, H. (2000). Rhetoric and Dialectic: Some Historical and Legal Perspectives. Argumentation 14 (3), 223–234.
Jacobs, S. (2006). Nonfallacious Rhetorical Strategies: Lyndon Johnson’s Daisy Ad. Argumentation 20 (4), 421–442.
Krabbe, E. C. W. (2007). On How to Get Beyond the Opening Stage. Argumentation 21 (3), 233–242.
Perelman, C., & Olbrechts-Tyteca, L. (1969). The new rhetoric: A treatise on argumentation. Notre Dame: University of Notre Dame Press.
Tindale, C. W. (2006). Constrained Maneuvering: Rhetoric as a Rational Enterprise. Argumentation 20 (4), 447–466.