ISSA Proceedings 2014 – The Agentive Approach To Argumentation: A Proposal
Abstract: This paper outlines an agent-centered theory of argumentation. Our working hypothesis is that the aim of argumentation depends upon the agenda agents are disposed to close or advance. The novelty of this idea is that our theory, unlike the main accounts of argumentation, does not establish a fixed function that agents have to achieve when arguing. Instead, we believe that the aims of argumentation depend upon the purposes agents are disposed to achieve (agendas).
Keywords: Agent, agent centered theory of argumentation, function of argumentation
The main goal of this paper is to outline an agent-centered theory of argumentation. Our working hypothesis is that the aim of argumentation depends upon the agenda agents are disposed to close or advance. The novelty of this idea is that our theory, unlike the main normative accounts of argumentation (i.e., rhetorical, dialogical and epistemological theories of argumentation), does not establish an a priori function that agents are expected to achieve when arguing. Instead, we believe that the aims of argumentation depend upon the purposes agents are disposed to achieve (i.e., their agendas). The problem with fixing an a priori function for argumentation is that some argumentative practices do not fit into the proposed end. Our concern is that when an agent does not aim for the fixed function of argumentation, his/her argumentative practice could be misunderstood or overlooked. That is why our agentive theory suggests that the agendas agents are disposed to close or advance by means of argumentation determine the goal of such communicative activity. If our intuitions are right, our account shows some promise understanding of a broader diversity of argumentative practices than each of the normative theories of argumentation individually considered.
Given the formal constraints of this presentation, we are not going to do a thorough reconstruction of each of the normative theories of argumentation. Instead, we are going to do cautious generalizations. First, we are going to make explicit the principle that normative theories of argumentation use to fix the putative goal for argumentation. Then, we will use a counter-example showing that the methodology of fixing an a priori function for argumentation is wrong. Finally, we will present the main concepts of our approach and show how it deals with the proposed counter-example.
2. The normative theories of argumentation
The normative theory of argumentation is an account providing responses to different issues concerning the analysis and evaluation of arguments. In dealing with the problem of the function of argumentation, normative theories fix an a priori goal that agents are suppose to satisfy. Three main claims are the object of our analysis.
(1) The goal of argumentation is to persuade (e.g., Perelman & Olbrechts-Tyteca, 1969; Tindale, 2004, Zarefsky, 2014).
(2) The goal of argumentation is to achieve a consensus resolving a difference of opinion (e.g., van Eemeren & Grootendorst, 1984, van Eemeren, 2010).
(3) The goal of argumentation is to establish truth and justified belief (e.g., Lumer, 2005a; 2005b)
The problem with fixing the aim of argumentation beforehand, is that some argumentative practices do not adjust to the fixed goal, and, consequently, the theory analyzing and evaluating argumentation tends to misunderstood or overcome such argumentative practices.
Let’s take a look at one fragment of the following counter-example proposed by Marianne Doury in the paper “Preaching to the Converted. Why Argue When Everyone Agrees?” For future reference, we will refer to Doury’s case as CAR RESTRICTION. In Doury’s words, this case is meant to show that “the goal of persuasion is but one goal among others that can be assigned to argumentation, and that, as a result, persuasion cannot be considered as the central element in the definition of argumentation” (2012, p. 100). To contextualize, CAR RESTRICTION is a transcription of a conversation between a vendor (hereafter V) and two clients (hereafter C1 and C2). All of them have seen each other before, but they know very little about each other.
V: Actually, what do you think of the law, er … we were actually talking about er… this law, there, that was just voted, that is in effect, you know, the law about traffic restriction for odd-even numbered license plates for the cars.
C1: Listen, I will tell you what I think, for Paris, we should be doing this all the time.
V: All the time.
C2: Exactly. We all agree then.
C1: I find this a great idea. First of all because at last, every day, there is already a maximum number of people who could find a way to organize their transportation… People do not need their cars all the days!
V: The opposition parties, actually, were against it at the beginning and we do not hear them speak anymore, now.
C2: They showed women who…who were actually commuting in the car of their friends, of a friend who came to pick them up; they can do this all the time.
C1: Of course! There are people…well, the problem is, that there need to be jobs or… or certain obligations that allow one to leave at a fixed time and to return at a fixed time. For example, in my case, this is not possible. But, ninety-nine percent of the time, I do not take the car!
V: Yes, you are all the time using public transportation.
C1: Exactly. … (Doury, 2012, p. 101).
According to Doury, CAR RESTRICTION is just an example of argumentative situations in which a controversy is proposed, and even though all the arguers agree on one same view, they provide arguments for their positions (p. 103). To be sure, the controversy is posed by the vendor when asking “what do you think of the law … about traffic restriction for odd-even numbered license plates for the cars?” The agreement between the arguer becomes explicit when C1 states “… we should [impose the restriction for odd-even numbered license plates for cars] all the time,” V assents saying “All the time,” and C2 responds claiming “Exactly. We all agree then.” Finally, without a detailed reconstruction, some of the arguments put forward are the following. C1 “finds [the idea of imposing restriction for odd-even numbered license plates for the cars all the time] great” because, in her words, “at last, every day, there is already a maximum number of people who could find a way to organize their transportation”. Additionally, from her perspective “People do not need their cars all the days.” C2 agrees with [the idea of imposing restriction for odd-even numbered license plates for cars all the time] because [with this restriction “[t]hey showed women who … were actually commuting in the car of their friends [that] they can do this all the time.”
For Doury, CAR RESTRICTION is a counter-example against the idea that the aim of argumentation is persuasion. Shortly, if “to persuade” is defined with the Merriam-Webster dictionary as “to move by argument, entreaty, or expostulation to a belief, position, or course of action,” then persuasion is not the goal of argumentation in CAR RESTRICTION. The reason for this is that one cannot “move” someone to believe something that he/she already believes. To clarify, the point is not that persuasion is never the end of argumentation, but to provide a negative instance for the claim that all argumentation aims to persuade.
We believe that CAR RESTRICTION also is a counter-example for the claims that all argumentation aims to resolve a difference of opinion, and that all argumentation aims to the establishment of justified true belief. To recall, from the pragma-dialectical approach, argumentation arises from a disagreement and ends with the dissolution of the different of opinions. Yet, in CAR RESTRICTION the argumentation does not finish with the agreement. Rather, that is trigger for the arguments put forward by the participants of the conversation. Similarly, CAR RESTRICTION presents a counter- example for the epistemological theories of argumentation because in it the arguers are not epistemically justified in believing that the restriction for odd-even numbered license plates for the cars should be imposed all the time. One of the features of knowledge is that it is factual, but the aforementioned proposition is not. Therefore, there is not knowledge to be established in CAR RESTRICTION.
3. The agentive proposal
Our proposal is that the problems posed by CAR RESTRICTION are explicated if we understand argumentation as a type of agenda an agent has. Briefly put, for our presentation purposes here, an arguer is an agent, and the purpose or objective he/she is trying to attain by arguing is his/her agenda (cf. Gabbay & Woods, 2003; 2005). The closure of each of these agendas is bound by a group of conditions of execution (CE). That is, requirements that, if satisfied by the agent, would count as an achievement of the agenda. These requirements include, in the case of epistemic agendas, things like time, information, computational capacity, and methodological strategies (Woods, 2013). Notice that CE are found in varying degrees. Broadly speaking, the most stringent extreme of the spectrum only authorizes belief formation when all possibilities of error are ruled out – including miscalculation – and/or complete information is achieved, while the other extreme allows for fallibilist belief formation with incomplete information. For instance, when argumentation takes place in scientific discovery, its aim can be taken to be the fixation of a justified (and, optimistically, true) belief. Yet such a demanding goal is not a requirement for argumentations that are directed towards practical purposes, such as putting a hypothesis under probation or justifying a practical decision against a background of incomplete information.
In contrast with other approaches mentioned above, we think the purposes of arguing vary accordingly with the agendas and sub-agendas advanced by the agents. This implies that arguing is an activity performed by agents embedded in other activities and as a part of the requirements of the fulfilment of other agendas. By the same token, arguing presupposes other agendas agents need to achieve if they want their argumentation to be successful. For instance, agents need to capture the attention of their addressees, as well as being warranted that these addressees do understand their arguments. For our present concerns, however, it suffices that we distinguish four kinds of agendas in which the act of arguing can intervene. These agendas are not presented in the spirit of showing an exhaustive list, but only as an example of the fruitfulness of our approach. The agendas in question are:
* Agendas of epistemic arrival (AEA), which aim at forming a particular belief. This is the case where people argue in order to create a belief (cf. Peirce, 1877). To be sure, a paradigmatic case of this kind of agenda is the verification of a scientific hypothesis, and in this sense, there must be some expectations about the grade of strictness of its justification and veracity. Of course, normative epistemological approaches provide an account of these kinds of examples. But not all AEA are so. If you have to engage in argument in the absence of complete information in order to take an immediate course of action, as e.g. in an emergency room, then to maintain the strictness of a scientific epistemic arrival would demand more time and, accordingly, the delay would turn out to be fatal – literally. There are times when, given the risks at hand, to aim at effecting an immediate educated guess is better than to wait for a warranted but temporally mediated truth. Still it is also true that sometimes you can try to close an AEA by simply asking somebody for information, as in the case of looking for an address in a new city (testimony references). As this last example shows, however, not all AEA are accomplished via argumentation.
* Agendas of epistemic defensibility (AED), which intend to present and defend (to other agents) a belief previously fixed by the arguer via the closing of an AEA. This includes the cases of political harangues, prosecutor accusations, and attorney’s allegations among others. Notice that these agendas do not seek to fixate the arguer’s own beliefs, but those of others. In this sense, AED are paradigmatic cases of persuasion. As such, they naturally fall under the jurisdiction of rhetorical theories of argumentation. Of course, an AED can be sincerely pursued or not. Thus, one can defend a belief, or defend a pretended belief, as in the case of the counsellor who does not believe in the innocence of his/her client.
* Agendas of epistemic maintenance (AEM), which aim at ratifying a belief previously fixed by the arguer via the closing of an AEA. This is clearly a case in CAR RESTRICTION. Yet it is important to stress that in this example, it is simply not part of the arguer’s agenda to review whether the belief is proper knowledge (the epistemological way), whether it ought to persuade others (the rhetorical way), or whether there is a difference of opinion to resolve (the pragma-dialectical way). On the contrary, the arguers advance their arguments in order to have a surplus of reasons for maintaining and preserving a particular epistemic position. And the peculiarity of this scenario is due to the fact that multiple agents carry out the agenda in a joint manner. But there are no obstacles for an AEM to be an individual agenda (as in Peirce’s ‘tenacity’ method for fixing belief) or a collective one (as many Colombians agreeing with the conclusion that James Rodriguez is the best player of the first round or stage of the 2014 World Cup – we imagine the Dutch people might feel the same about van Persie or Robben). In any event, the collective case can become a mechanism of ideology preservation.
* Agendas of epistemic obstruction (AEO), which aim at preventing the proper attainment of epistemic agendas by other agents. For instance, when you distract someone in order to avoid them from focusing on some problem (e.g. by arguing about some irrelevant topic), or when one prepares a diversion (e.g. by admitting herrings as premises in the argumentative scenario) you are preventing the proper attainment of epistemic agendas by other agents. In the first case, the obstruction consists in hindering or delaying a proper belief formation on the part of the other agent. In the second, it consists of facilitating the other agent in the formation of a false belief. However, in the last case, there is no pro or con persuasion as such, in the sense that it can be any of them. Notice that what is at stake is an epistemic agenda, yet not because the agent has an intrinsic epistemic agenda, but because he/she is interested in the epistemic agendas of his/her addressee. Of course, this instance of an AEO is a source of possible error in epistemic agendas and as such it does not need to always be achieved by means of arguing.
Let us observe that all these agendas (AEA, AED, AEM, and AEO) are actually sub-agendas, that is, agendas that are carried out as a means with respect to an ulterior end. In this sense, their role is primarily ‘methodological’ (in the etymological sense of the word). Indeed, in the examples discussed above, AEA serves as a means for determining truth, saving a life, or arriving at some place. In the AED examples, persuasion is pursued in order to obtain votes or to make a decision about the innocence or culpability of someone. In AEM, arguing serves the self-assertion of the arguer’s belief system. And in AEO, arguing functions as a strategy for weakening the potential course of action of other agents. In this sense, this approach explains why arguing is not an end in itself most of the time. Although it can be imagined of as an immediate agenda, as when agents argue as a way of training in argumentation; argumentation is an activity agents engage in order to obtain things different to more arguments.
Finally, our proposal is encompassing enough as to admit different types of epistemic agendas, but equally it is rigorous enough as to not admit relativism: insofar agendas are things that can be achieved totally, partially, presumptively, etc., their fulfilment can be evaluated as adequate or inadequate, better or worse, properly or improperly closed, etc.; and by keeping in mind the conditions of execution (CE) and the degrees of strictness with which an agenda has to be undertaken, our proposal helps to clarify, in an unified perspective, why there are different epistemic ‘contexts’, what they are and how to identify them (issues that Doury leaves underspecified), and why they bring varied —although, occasionally, mixed – results. All these topics, however, deserve another paper.
Doury. M. (2012). Preaching to the Converted. Why Argue When Everyone Agrees? Argumentation, 26(1), pp. 99-114.
Gabbay, D. & Woods, J. (2005). The reach of abduction: Insight and Trial. A practical logic of cognitive systems, Vol. 2. Elsevier, Amsterdam.
Gabbay, D. & Woods, J. (2003). Agenda relevance: A study of formal pragmatics. A practical logic of cognitive systems, Vol. 1. Elsevier, Amsterdam.
Lumer, C. (2005a). Introduction: The Epistemological Approach to Argumentation – A Map. Informal Logic, 25(3), pp. 189-212.
Lumer, C. (2005b). The Epistemological Theory of Argument – How and Why? Informal Logic 25(3): 214-232.
Peirce, Charles S. (1877). The Fixation of Belief. In: The Essential Peirce, Vol 1. Christian Kloesel and Nathan Houser (Eds.). Bloomington, Indiana University Press. pp. 109-123.
Tindale, C. (2004). Rhetorical Argumentation. Principles of Theory and Practice. Thousand Oaks, SAGE Publications. pp. 1-27.
van Eemeren, F. & Grootendorst, R. (1984). Speech Acts in Argumentative Discussions. Dordrecht: Foris Publications.
van Eemeren, F. (2010). Strategic Maneuvering in Argumentative Discourse : Extending the Pragma-dialectical Theory of Argumentation. Amsterdam: John Benjamins Publishing Company.
Woods, J. (2013) Errors of reasoning: Naturalizing the logic inference. College Publications, Milton Keynes.
Zarefsky, D. (2014). Rhetorical Perspective in Argumentation. Dordrecht, Springer.