ISSA Proceedings 2010 – The Enthymeme Between Persuasion And Argumentation
Currently, in the field of argumentation, distinct and even conflicting conceptions abound, one of the most widely debated of which is persuasion. For the epistemic tendency (Siegel and Biro 2008), persuasion and argumentation remain quite distinct for, even if it is allowed that persuasion may sometimes be the aim of argumentation, proponents of this point of view nevertheless consider that the validity of an argument must be evaluated on epistemic criteria alone. Basing himself on a different analysis, Marc Angenot arrived at the same conclusion in his latest book (Dialogue de sourds, 2008, p. 93-96): for him, argumentation rarely leads to persuasion, hence the two should be radically separated.
Argumentation has been distinguished from persuasion by pointing to the orator’s purpose: if the specific purpose is to obtain adherence from the addressee by all possible means, usually understood as including non-reasonable ones, then what is taking place is not argumentation but persuasion. The objective of obtaining adherence is also closely related to pathos and ethos, which are emotional elements seen, in this perspective, as being opposed to rational ones. According to this conception, persuasion, which is mainly identified with rhetoric, is a type of discourse whose priority is obtaining the addressee’s adherence: this activity is considered as being opposed to argumentation (unlike Perelman’s conception insofar as he considers adherence as the purpose of argumentation); for proponents of this tendency, which I seek to counter, persuasion is therefore considered as roughly equivalent to manipulation. I will come back to their distinction later since persuasion has been opposed to argumentation because it is frequently confused with manipulation.
Let us now examine in more detail which features of persuasion should always be present for us to be able to speak of the latter. If persuasion is based on rhetorical techniques aiming at producing a desired effect on the audience, it might be worthwhile to start by asking if an effect obtained by just any means should count as persuasion. For example, could we say that anybody has been persuaded when the persuader is playing on fear?
There are at least two types of things we fear: apart from the physical threats, there are all the disagreeable consequences that an orator can put forward in order to obtain the desired result from the addressee. Clearly we could not say that somebody has been persuaded when acting under threat of her life. But what about a warning to the hearer of the probable occurrence of bad consequences? In that case a distinction between a warning and a threat might be useful: if fulfilment of the bad consequences depends on arbitrary will and hence is under the control of the orator, we should speak of threat. But when these negative consequences are not due to the arbitrary will of the orator, we could sensibly say that the hearer has been persuaded by a warning of bad consequences. Furthermore, it would seem that it is acceptable to talk about persuasion, but only if the addressee comes to envisage as probable the realisation of these consequences.
Up to now the main elements of persuasion that have been underlined are: that the bad consequences presented as a means of persuading do not depend on the arbitrary will of the persuader, and that the addressee understands what the probabilities are that bad consequences would occur. But there is another important element that is required in order to obtain persuasion, i.e. a certain space for free will. Indeed, free will is a central element for distinguishing persuasion from manipulation. Some authors (Plantin forthcoming; Breton 2000, p. 66) hold that concealment of the speaker’s aim is the main element of manipulation. Nevertheless, the effect of such concealment is related to the obstruction of free will.
In this brief account of persuasion, manipulation is relevant only when considered as a borderline case of persuasion. In order to look more closely at the distinction between persuasion and manipulation, it might be useful to ask whether we should speak of persuasion or of manipulation when information is conveyed, for the conveying of information is purported to be objective. Emotional elements can, however, be present in objective discourse in which information is simply being conveyed.
For example, suppose that during a parents’ meeting the director of a school informs her audience about the consequences of smoking for lung cancer, adding as evidence a report on the distribution of the probabilities of getting lung cancer showing that the probability increases with the number of years of smoking and the number of cigarettes consumed. Is she engaging in a rational argumentation giving her audience true and reasonable elements? Yes. Could we say that she is trying to persuade the parents so as to induce them to stop smoking? Yes. Could we say that she is threatening her audience? No. If she further adds to this information a different sort of data about the performance of children who have suffered a traumatic experience when their parents were seriously sick during their childhood or adolescence, she is admittedly playing on their emotions so as to induce them to seriously consider the desirability of stopping smoking, and consequently to take the decision to do so. Could the director of the school be accused of manipulating the hearers? No. These two cases show that the degree of emotionality is not the crucial element in distinguishing argumentation from persuasion, or persuasion from manipulation. In both cases the orator is conveying a message containing good reasons for reaching a certain conclusion: firstly, smoking is a hazardous practice and bad for our health and secondly, falling sick may endanger our children’s future. Both could have been interpreted as acting on the subjectivity of the hearer by presenting the probable actualization of undesirable events. But there is no manipulation since the realization of those bad consequences is not under the control of the orator’s will, and the expected effect, the parents stopping smoking, does not represent a hidden benefit sought by the orator as her sole aim. All these features are important when considering the boundaries between persuasion and manipulation.
2. The path to persuasion
Now that I have tried to dissociate persuasion from manipulation, I will attempt to show in this section that, far from being opposed, persuasion and argumentation have much in common. In order to do so, it might be useful to bear in mind the elements required for the achievement of persuasion, which I would call the ‘path to persuasion’. To be persuaded the addressee needs to go through different steps that grosso modo imply a degree of consciousness. Firstly, a cognitive process is required, indeed the addressee must attain a minimum degree of understanding of the message/information the orator is conveying; secondly, she must also grasp what is at stake in order to be able, at a third stage, to assess and accept the claim whose outcome is a change in or reinforcement of current beliefs, opinions, attitudes or values. According to the conception of persuasion I am proposing, argumentation and persuasion share the same steps up to this point. But persuasion requires something else: fourthly, a change in the addressee’s disposition regarding an action or putting an end to a course of action. In other words, acceptance of the orator’s claim is only a necessary condition, for persuasion also requires a change of disposition not attainable by just any means.
According to O’Keefe persuasion is “a successful intentional effort at influencing another’s mental state through communication in a circumstance in which the persuadee has some measure of freedom.” (O’Keefe 2002, p.5). However in my view, influencing another’s mental state is too vague an outcome to consider that persuasion has been achieved. For as we saw in the example of the school director, it would not be sufficiently clear whether persuasion had been attained, even if the information was added with some kind of explanation as to what this information implied and the cognitive process was thereby brought about; in this case, the parents understand, consider and even fear the consequences, hence there is an influence on their mental state, but it is a changed disposition to act (or abstain) that, in my view, characterizes persuasive discourse. In other words, persuasion would be obtained only if the parents were in consequence disposed to stop smoking.
3. The Enthymeme
Let us now turn to the enthymeme and look at the role it plays for our understanding of the link between argumentation and persuasion. We shall briefly recall some earlier conceptions of the enthymeme which might be of interest in identifying those characteristics that were already present before Aristotle’s Rhetoric and are relevant to capturing the meaning and the role it plays in the present discussion.
In his “The Old Rhetoric” [L’Ancienne Rhétorique], Barthes introduces the enthymeme as one kind of argumenta, which, as opposed to inductive exempla, are modes of deduction. He says that originally the term argumenta, refers to the subject or argument of a play, seen as an articulated set of actions (Barthes 1970, p. 201). This is an interesting characteristic of great relevance to our concerns; for, as with the argument of a play, an argument is also articulated, and could not be conceived of as a simple set of ideas, since what makes of a set of ideas an argument is their articulation; it is only because of the type of relation between them that they constitute an argument of a particular kind.
Among argumenta, the enthymeme has a central place in rhetoric, as it is the proof (pistis) par excellence. But the term “enthymeme” belonged to the rhetoricians’ vocabulary long before Aristotle wrote his Rhetoric. Aristotle inherited from other rhetoricians, particularly Isocrates and Anaximenes, different but not contradictory concepts of the enthymeme that, as J. Walker suggests, must have contributed to inform that contained in his Rhetoric (Walker 1994, p.53). Isocrates, for example, emphasizes the search for arguments that are appropriate for the context (kairos), i.e. what is opportune for the argumentative situation, and also underlines the need to be aware of the importance of good style. For his part, Anaximenes in the Rhetoric to Alexander[i] underlines the adversarial character of enthymematic reasoning by recommending that enthymemes are to be invented through the analysis of the opponent’s discourse and emphasises the pertinence of brevity. The conception of the enthymeme presented in Aristotle’s Rhetoric comprises all these elements, as do today’s conceptions.
In the Aristotelian realm, rhetoric is the method of finding the means of persuasion; the enthymeme and the example are, as mentioned above, the two pisteis or proofs of the art through discourse.
Let me list the main features of enthymemes[ii]:
– they are deductive syllogisms, a form of reasoning in specific contexts;
– they are syllogisms founded not on true but on probable premises;
– their purpose is to state likelihood (eikos) and not necessary conclusions or outcomes;
– they admit of exceptions, counterexamples, counterarguments: roughly speaking, opposition; hence they are not final demonstrations;
– they start from endoxa or beliefs, attitudes and opinions generally accepted by the audience in order to reach, by inferences or deductions, a conclusion that follows on from the former;
– they are brief;
– they may lack a premise or conclusion.
Now the virtues of enthymematic reasoning are economy, efficiency, efficacy, its pragmatic character; this is so because:
– being short it can easily be followed by the audience; for this reason Aristotle recommends not starting the chain of arguments from a very distant point;
– starting from what is known and already taken for granted is important, because the addressee will in consequence be well-disposed towards what is said;
– the omission of a premise is also justified by reasons of economy, for if something is well known, says Aristotle, it is unnecessary to repeat it;
– the omission of the conclusion has advantages in terms of economy, but might also be beneficial to efficiency because if the audience comes to the conclusion by itself, its acceptance is more easily acquired;
– allowing members of the audience to come to a conclusion by themselves might be more respectful toward them.
I will come back later to the question of how these characteristics and virtues contribute to the link between argumentation and persuasion. Before that, it will be helpful to distinguish the two main fields of rational reasoning: demonstration and argumentation, for my aim is limited to showing the link between argumentation and persuasion. Therefore, I will very briefly recall the difference between demonstrative and argumentative enterprises.
4. The realm of rational reasoning: demonstration and argumentation
In the realm of rational reasoning, pride of place is given to demonstration as a path to a conclusion, proceeding step by step from a series of true premises using rules of inference, and thus forming a continuous chain of reasoning that should inevitably arrive at a conclusion. At the end of a demonstration we are obliged to accept the conclusion unless we can show the existence of a flaw. In the context of a formal logic demonstration, for example, it would not be acceptable to reject a step of the reasoning according to a certain rule, just because we have a hunch that it is not correct; we proceed by assuming that all the axioms and rules of the system are accepted and if we cannot prove an error it would be irrational to reject the conclusion (Plantin 1990, p. 160). Things are different, however, when we are in an argumentative context.
Conversely, argumentation is also a rational enterprise but its premises are not necessarily true: they are mainly probable and so is the conclusion. This is why argumentation is not about falsehood or truth, but about more or less fully convincing the addressee; it could be said that the core of argumentation is opposition between claims, and its aim is to advance reasons in favour of or against a claim, and not necessarily the resolution of the opposing thesis.
Now, to start focusing on the convergence of argumentation and rhetoric and especially to underline that the context, and therefore, also the audience are important for both the former, I would like to address an issue that might shed some light in our concerns.
Should we describe as argumentation a discourse that gives reasons for a conclusion which everybody in the audience agrees on?
If the core of argumentation is opposition between claims, it would seem out of place to ask this question. Nevertheless, in my view, it will be useful to do so, for it sheds light on an important issue concerning the role of enthymemes. So “yes”, I would describe a discourse that gives reasons for a conclusion as argumentation in at least two cases when everybody agrees: a) if we want to reinforce the degree of conviction, and b) if we want to be ready to face a further confrontation. Reflection on these two cases brings us to a crucial consideration: argumentation is always context-sensitive; the strict dialogical character of argumentation, requiring the exchange of two actually opposing theses, is a working hypothesis that, when qualified, may allow that an arguer is faced with other real or virtually present positions in a particular society (Bakhtin 1977, p. 105). It is, therefore, the context in a broad sense that makes it reasonable to label as argumentation a discourse that puts forward reasons leading to a conclusion which everybody in the audience agrees with, hence a discourse which, while not directly oppositional, nevertheless takes into account other opposing positions that are part of the social background. All these features of argumentative discourse will turn out to be central for the purpose of understanding the link between argumentation and persuasion.
5. The Hinging Role of Enthymemes
How, then, do enthymemes perform their hinging role between argumentation and persuasion? One of the most salient features of the enthymeme is that it is an incomplete syllogism. As we recalled earlier, it can lack either a premise or a conclusion. Let us analyze both possibilities in order to see, firstly, given the need to articulate the elements of an argument, whether these omissions compromise the idea of considering the enthymeme as an argument, and secondly, how they at the same time play a persuasive role.
Concerning the absence of a premise, Aristotle rightly says that it is not necessary to repeat what everybody knows. This could be interpreted solely as a criterion of economy, since we can save time and effort by omitting what it is not necessary to say. But it also has another related connection with the role enthymemes play in persuasion, namely the importance of taking the audience into account.
To be sure, when the orator fails to make explicit a premise in an enthymeme, he is not simply leaving a blank, since he does so basing himself on the audience’s knowledge, on its members’ opinions, attitudes, beliefs, values, etc., i.e. the endoxa. It is only because the omission is grounded on this knowledge, on the set of, in Sperber’s terms, public representations (Sperber 1996, p. 108), that it makes sense to omit the premise; it is not, therefore, a real omission. Thus the premise is, so to say, present. This pragmatic feature of enthymematic reasoning, namely the importance given to the audience, has at least three significant outcomes that link argumentation to persuasion. Firstly, of course, there is one of an utilitarian nature, namely efficiency, i.e. the fact of easily producing an effect on the audience; secondly, efficacy is also certainly a consequence, since the desired objective is obtained: persuasion. The third outcome of the absence of a premise in enthymematic reasoning is also linked to the role that endoxa plays. As I have already suggested, the elements of the latter are not only a set of ideas, they are interrelated and thus form different systems (Nettel 2009, p. 4) that the orator must take into account as a necessary condition for argumentation to take place, for it is a principle of reasonableness that the arguments advanced should be relevant to the hearer. This is so because argumentation in general needs a space of understanding: if the endoxa are not considered, it would be like talking to the wind. This is why the omission of a premise, instead of being a strategy of audience manipulation, is quite the contrary: a way of arguing in a perfectly acceptable manner. Seen in this light, the omission of a premise is not a hiatus, a lack of articulation among the ideas that form the enthymematic argument. For, I repeat, if the premise is not stated, it is nevertheless in a way present as part of the reservoir of a community’s public representations. The central role given by Aristotle to the importance of having the endoxa as a starting point when the goal is persuasion is confirmed by recent experiments that show that: “we seek to confirm it if we agree with it in the first place, and to disconfirm it if we don’t. This can hardly be sanctioned by a normative theory and is all the more disquieting in that it seems to be extremely widespread: ‘smart’ people do it (Stanovich and West 2007); open-minded people do it (Stanovich and West 2007); and physicians, judges and scientists do it (see Fugelsang and Dunbar 2005; Nickerson 1998; and references within)” (Mercier and Sperber forthcoming, p. 163). Is this a bias as some cognitivists suggest? Not to my way of thinking. It is simply a general tendency to coherence that might be inscribed in a genetically defined module!
What then of the omission of a conclusion? Given that the latter is the outcome of the articulation of the premises, the addressee might still reach by herself the orator’s anticipated conclusion, even if it is omitted. On the one hand this absence could certainly have a strategic purpose. But could this omission, then, be seen as a way of deceiving the addressee? Not necessarily. Rather it could be seen as the sign of a desire to avoid the responsibility of having to state a conclusion and, by the same token, of a desire to transfer this responsibility to the addressee. On the other hand, the lack of a conclusion could also be seen as a sign of respect for the audience’s autonomy, signalling a democratic attitude. Nevertheless, in both scenarios efficiency and efficacy are present, for it is easier to promote conviction on the part of the addressee if the latter reaches the conclusion by herself, and at the same time it is certainly efficacious in obtaining persuasion because working out a conclusion by our own means gives us the feeling that it is ours instead of the orator’s.
The conception of rhetoric as a discourse whose sole concern is to have an effect on the hearer is the fruit of a decontextualized interpretation of Aristotle’s definition, whereas he refers to rhetoric as the method of finding every possible means to persuade the hearer. However, “every possible means” must not be interpreted as signifying just any means, for the aim of persuading by the techné rhetoriké requires, as dialectic, a demonstration or pistis through the articulation of premises and the opposition of theses. In his Topics, Aristotle says that the rhetorical invention of enthymematic arguments starts by analyzing the pairs of oppositions that are opportune for the question at issue, given the situation and the occasion (kairos). As we recalled above, this is a central trait of dialectical argumentation, and the fact that a rhetorical discourse is often not an actual exchange between two opposite standpoints within a dialogue is not an obstacle; for even if the orator is the only person producing discourse, it is nevertheless opposing positions that are being discussed.
Argumentation is about finding convincing reasons for a claim and evaluating them (Mercier 2009, p. 9). Besides the steps that it shares with argumentation, persuasion also requires the achievement of a desired change of disposition on the part of the addressee; this change must nevertheless be based on the acceptance of a claim founded on the reasons advanced. Persuasion initially requires the conscious evaluation of these reasons and then acceptance of the claim that follows, i.e. what Mercier and Sperber call a “reflective inference” (Mercier and Sperber forthcoming); thus acceptance of the claim is only a necessary condition for the realization of persuasion. The enthymeme is an argument that, by virtue of its specific features, given a reasoned acceptance of a claim, also facilitates a change in the addressee’s disposition to act. This is why a mere mental influence on the addressee through subliminal information, for example, would not be enough for it to be considered that persuasion has been achieved: this would only be manipulation.
Argumentation and persuasion are intrinsically related because one cannot be persuaded of something that one does not consider fair, just, right, correct, true or plausible. In the above, I have tried to show not only that the enthymeme is a device that hinges between the reasonableness of argumentative reasoning and the persuasiveness of rhetorical reasoning, but also that it is called upon to play a technical, ethical, epistemological and ideological role in the theory and practice of human social discourse.
[i] Apparently there is no agreement as to whether it was written before or after Aristotle’s Rhetoric.
[ii] Most of the features contained in the following list are contested in Burnyeat 1994 and 1996. It is beyond the scope of this paper to discuss in detail Burnyeat’s position. If Aristotle’s Rhetoric is a response to Plato’s criticism of rhetoric, as has been considered (see Furley and Nehamas 1994, p. xii), then contrary to Burnyeat’s opinion it is likely that the terms “syllogism,” “enthymeme,” etc. in Aristotle’s Rhetoric were used in their technical meaning, i. e. as was the case in his previous works and not, as Burnyeat claims, in the sense they had in common language.
For the discussion of Aristotle’s conception of enthymeme as syllogism in its technical meaning, I lean on Grimaldi 1972, p. 87-91. See also Racionero’s notes in his Spanish translation of Rhetoric, about passages where Aristotle defines the enthymeme as syllogism or rhetorical syllogism: see Aristotle 1994, n. 16 p. 167-168 on Rhetoric 1355a 9; n. 54 p. 183-184 on Rhetoric 1357a 17; and n. 280 p. 417 on Rhetoric 1395b 24-26. See also Rubinelli 2003, p. 241, who maintains the syllogistic nature of enthymeme pace Burnyeat.
Angenot, M. (2008). Dialogue de sourds: Traité de rhétorique antilogique. Paris: Fayard.
Aristotle (1994). Retórica Q. Racionero (Ed.). Madrid: Gredos.
Bakhtin, M. (1977). Le Marxisme et la philosophie du langage. Paris: Minuit.
Barthes, R. (1970 ). L’ancienne rhétorique: Aide-mémoire. Communications 16, 172-229.
Breton, Ph. (2000). La parole manipulée. Paris: La Découverte.
Burnyeat, M. F. (1994). Enthymeme: Aristotle’s on the Logic of Persuasion. In Furley D. J. and Nehamas A. (Eds.), Aristotle’s Rhetoric (pp. 3-55), Princeton: Princeton University Press.
Burnyeat, M.F. (1996). Enthymeme. Aristotle on the Rationality of Rhetoric. In A. Oksenberg Rorty (Ed.), Essays on Aristotle’s Rhetoric (pp. 88-115). Berkeley, Los Angeles and London: University of California Press.
Furley D. J. and Nehamas A. (Eds.) (1994), Aristotle’s Rhetoric: Philosophical Essays. Princeton: Princeton University Press.
Grimaldi W. M. H. (1972). Studies in the Philosophy of Aristotle´s Rhetoric. Wiesbaden: Franz Steiner Verlag.
Mercier, H. (2009). La Théorie argumentative du raisonnement. PhD dissertation. Paris: EHESS.
Mercier, H. and Sperber, D. (forthcoming) Intuitive and Reflective Inferences. In Evans J. S. B. T. & Frankish K. (Eds.), Two Minds (pp. 149-170, Ch. 7). New York: Oxford University Press.
Nettel, A. L. (2009). Arguing for Principles in Different Legal Cultures. In J. Ritola (Ed.), Argument Cultures: Proceedings of OSSA 09, CD-ROM (pp. 1-9), Windsor, ON: OSSA.
O’Keefe, D. J. (2002). Persuasion: Theory and Research. Thousand Oaks, CA: Sage Publications.
Plantin, Ch. (1990). Essais sur l’argumentation: Introduction à l’étude linguistique de la parole argumentative. Paris: Éditions Kimé.
Plantin, Ch. (forthcoming). Dictionnaire critique de l’argumentation.
Siegel, H. and Biro, J. (2008). Rationality, Reasonableness, and Critical Rationalism: Problems with the Pragma-dialectical View. Argumentation 22, 191-203.
Rubinelli S. (2003). Topoi e Idia nella Retorica di Aristotele. Phronesis, XLVIII, 238-247.
Sperber, D. (1996). La contagion des idées. Paris: Editions Odile Jacob.
Walker, J. (1994) The Body of Persuasion: A Theory of Enthymeme. College English, 56-1, 46-65.
Wolff, F. (1995). Trois techniques de vérité dans la Grèce Classique. Hermès 15, 41-71.