ISSA Proceedings 2010 – Belief, Rationality And The “Jurisdiction Of Argumentation”

1. Introduction
A prevalent and sensible pre-theoretic intuition about the relationship between argumentation and belief is that argumentations are the sorts of things that ought to impact our beliefs about the issues over which we argue [i]. For example, we generally think that, if an agent concedes to a standpoint (or to a challenge to their standpoint) as a result of an argumentation, then ceteris paribus that agent should appropriately modify their mental attitude toward that standpoint. However, as David Godden (2010) shows, several influential “commitment-based” accounts of argumentation (in particular, Charles Hamblin’s (1970) dialectical theory, Douglas Walton and Erik Krabbe’s (1995) dialogue based theory and van Eemeren and Grootendorst’s (2004) pragma-dialectics) do not adequately attended to the pre-theoretic intuition that there is a normative relationship between what an agent ought to believe and the commitments the agent takes on in an argumentation. Commitment-based approaches to argumentation regard belief “to be too psychological a notion” (Godden 2010, p. 406) and instead of relying on the concept of belief such accounts focus on the commitments arguers publicly adopt during argumentations. Contrary to commitment-based theories Godden explicitly contends that, in typical cases, an agent should modify their mental attitude towards standpoints that the agent has conceded in an argumentation.

The purpose of this paper is to discuss some issues that arise from Godden’s criticism of commitment-based theories of argumentation. I start the paper with a description of a general and common method for distinguishing normative types. The discussion in this section is crucial since it is the foundation for drawing a distinction between the norms of argumentation, the norms of belief, and the norms of reasoning, a distinction that is central to the argument that I advance later in the paper. I will follow this section with a discussion of some other key concepts and definitions that will be used throughout the paper. Next, I explain Godden’s criticism of commitment-based theories of argumentation focusing in particular on his criticism of pragma-dialectics. In this section I develop and defend a formulation of a norm that links concessions made in argumentations to beliefs an arguer ought to possess. I call this norm the norm of argumentation compliance (or NAC for short). In the following section I proceed to examine the question of what sort of norm NAC is. While Godden is content to regard NAC as a norm of argumentation, I contend that there are advantages to regarding NAC as a non-argumentation norm such as a norm of belief or a norm of reasoning. I argue that there are theoretical advantages of two sorts to understanding NAC to be a non-argumentation norm. The first sort of theoretical advantage is simplicity. I contend that understanding NAC to be a non-argumentation norm simplifies the task of the argumentation theorist. The second theoretical advantage to understanding NAC as a non-argumentation norm is that so understanding NAC has to potential to help us address other difficult problems encountered in argumentation theory. In particular I think this understanding of NAC can help us understand the relationship between argumentation and other domains such as reasoning, critical thinking, problem solving and investigation. Finally, I will wrap up with a discussion of some unresolved issues that arise from understanding NAC as a non-argumentation norm.

2. Normative Types
There are different types of norms. Commonly we draw distinctions between epistemic norms, moral norms, social norms, political norms, legal norms, religious norms, and so on. However, it is unclear exactly what makes this colloquial talk of normative types a principled talk based on concrete differences among different types of norms. If this colloquial talk identifies a principled difference between different types of norms, then it should be possible to offer a description of what distinguishes one normative type from another.
Fortunately our colloquial talk of normative types does suggest a method for distinguishing normative types. The method suggested by our colloquial talk is to distinguish normative types according to the domain over which the norms of a normative type have authority. According to this method, a normative type N is a set of norms x1…xn where x1…xn have normative force over a common domain. This method of distinguishing normative types is analogous to the way we distinguish different types of governments based on the jurisdiction over which the authority of that government extends. For instance, in the Canadian system, there are civic governments, provincial and territorial governments, and governments for First Nation, Inuit and Métis peoples. We distinguish one type of government from another by identifying the unique geographical, constitutional, legal and political jurisdiction over which the government exercises its authority. Similarly, we can distinguish one normative type from another based on the common domain over which norms of that type share “normative jurisdiction.”

This characterization of normative types allows us to make important distinctions between normative types that are relevant to the study of argumentation. For instance, it shows us how we can legitimately distinguish between the norms of argumentation, the norms of belief, and the norms of reasoning. Each of these different types of norms possesses normative authority over divergent domains of jurisdiction. That is, the norms of argumentation have normative authority over a different domain than do the norms of belief and the norms of reasoning. The distinction between these different types of norms will be important to the argument I develop in section 5. I will further discuss these different types of norms in that section.

3. Other Definitions and Key Concepts
The general approach to argumentation from which I will work is the pragma-dialectic approach of Frans van Eemeren and Rob Grootendorst (2004). Van Eemeren and Grootendorst define argumentation as,
A verbal, social and rational activity aimed at convincing a reasonable critic of the acceptability of a standpoint by putting forward a constellation of propositions justifying or refuting the proposition expressed in the standpoint. (van Eemeren and Grootendorst 2004, p. 1)

This definition of argumentation focuses on the constellation of propositions that are marshalled to support the various positions put forward in an argumentative discussion. A reason for this focus is that the pragma-dialectic theory of argumentation describes how parties participating in an argumentative discussion use bits of language, in the form of speech acts, to achieve a rational resolution to a difference of opinion. Van Eemeren and Grootendorst develop a “model of critical discussion” that can be used as a standard for the evaluation of argumentations. Van Eemeren and Grootendorst claim that, “the model [of critical discussion] provides a series of norms by which it can be determined in what respects an argumentative exchange of ideas diverges from procedure that is the most conducive to the resolution of a difference of opinion.” (van Eemeren and Grootendorst 2004, p. 59) The norms are formulated in terms of a set of rules that argumentative discussions must satisfy in order to adequately resolve the disagreement that precipitated the argumentation. These rules are paradigmatic examples of norms of argumentation.

It is important for me to be more specific about the type of argumentation that I am concerned with in this paper. My focus is on argumentations in which the parties involved share epistemic goals such as arriving at the most rationally justified position, or arriving at the truth of the matter. This focus is meant to exclude argumentations in which arguer’s judge that the pursuit of epistemic objectives diverges from the pursuit of their own personal interests and aim to use the argumentation in order to advance their personal interests as opposed to pursing epistemic objectives. I want to be clear that I do not deny that investigating argumentations in which arguers are aiming to advance their own interests is a fruitful investigation. Rather, I focus on argumentations where all the parties share epistemic objectives for the reason that the arguments I develop in the paper most clearly apply to that form of argumentation. Thus, from here on, when I use the term ‘argumentation’ I am referring to those argumentations in which both participants share epistemic objectives.

Before I develop the substantive arguments of this paper I will explain some other important concepts that I will be using. The concepts are effective resolution and concession. In using these concepts I aim to follow Godden’s (2010) use of them as closely as possible. An effective resolution to a difference of opinion in an argumentation arises only when, “commitments undertaken in argumentation survive beyond its conclusion and go on to govern an arguer’s actions in everyday life, e.g. by serving as premises in her practical reasoning.” (Godden 2010, p. 397) The second important concept is concession. Godden uses the term concession to indicate “verbally accepting” some position. (Godden 2010, p. 400) A concession in this sense is a verbal speech act that can be made in an argumentation. This act involves overtly endorsing a position presented by another party in an argumentation. Important to Godden’s position is that psychological acts such as mental acceptance and believing “are logically and causally independent” from acts of verbal acceptance such as conceding. (Godden 2010, p. 400) The logical independence of mental and verbal acceptance implies that it is possible for an arguer to verbally accept some point without also mentally accepting that point. That is to say that it is possible for one to verbally accept some point made in an argumentation and yet not modify one’s reasoning in a way that is consistent with the point conceded. Similarly, the logical independence of verbal and mental acceptance also implies that it is possible for an arguer to modify their reasoning so that it is consistent with some standpoint even though the arguer has not conceded, or otherwise verbally accepted, that standpoint.

4. Godden’s Criticism of Pragma-Dialectics
One aspect of the pragma-dialectic account of argumentation that needs to be mentioned at this point is its commitment to externalization. The pragma-dialectic approach to argumentation is less concerned with the psychological states of arguers than it is with the externalized – that is to say, publicly available – commitments arguers express in argumentative discussions. Van Eemeren, Grootendorst and Snoeck Henkemans endorse this position when they assert, “the study of argumentation should not concentrate on the psychological dispositions of the people involved in an argumentation, but on their externalized – or externalizable – commitments.” (van Eemeren, Grootendorst, Snoeck Henkemans et all 1996, pp. 276-277)
However, contrary to the pragma-dialectic view, Godden argues that changes in commitment are typically inadequate for a successful resolution to a difference of opinion. Godden contends that what he calls “commitment-based” theories of argumentation, of which pragma-dialectics is an example, admit within the class of successfully resolved argumentations, argumentations that are not effectively resolved. So, some resolutions permitted by commitment-based theories of argument may not last beyond the conclusion of the argumentation and “go on to govern . . . [the] arguer’s actions in everyday life.” (Godden 2010, p. 397) Other versions of commitment based theories Godden thinks are subject to similar problems are Hamblin’s formal dialectics and Douglas Walton and Erik Krabbe’s contemporary dialectical theory. (Godden 2010, p. 406)
According to Godden (Godden 2010, p. 404) commitment-based models of argumentation jointly endorse the following three theses:
The goal of persuasive argumentation is to settle a difference of opinion by rational means.
Commitment and belief are logically and causally independent; a change in one does not always result in a corresponding change in the other.
A difference of opinion is resolved when the commitments of the disputants have reached a state of agreement with respect to the claims at issue.

Godden thinks this joint endorsement causes problems. The joint endorsement of these theses is problematic because together they admit, as viable resolutions to a difference opinion, resolutions that will not translate into changes in future action, reasoning, and argumentation. For instance, consider a scenario similar to one proposed by Godden (Godden 2010, p. 405) in which an agent S is involved in an argumentative discourse with an agent P. In the process of arguing S comes to think she is not skilful enough to rationally defend their standpoint against P’s reasoned criticism. Further suppose that S concedes to P’s challenge to S’s standpoint even while still holding the belief that her initial standpoint is correct. This scenario is surely recognizable to those who frequently engage in argumentation. In this circumstance S has agreed with P through externalized commitments even though she does not believe P’s challenge to her standpoint. This agreement constitutes a resolution to a difference of opinion that is sufficient to satisfy commitment based theorists whose only metric for measuring a resolution to a difference of opinion is verbal or written commitments. However, the resolution is not sufficient, Godden contends, to achieve an effective resolution to a difference of opinion. The resolution is not effective in this case because there was not a change in mental attitude sufficient to cause an appropriate change in future actions, reasoning, and argumentation by S[ii]. Part of what is valuable in the practice of argumentation is that arguers adjust their behaviour appropriately when a point has been rationally established as acceptable to all parties involved in the argumentation. However, Godden contends, without an appropriate change in mental attitude, the arguer who so concedes a point to a more skilful opponent (and is not held accountable through other social mechanisms such as the law) will not change their future actions, reasoning or argumentation relating to the point.

Godden says of situations similar to the one above that,
Having made a concession in the argumentative process . . . [the arguer] may be held accountable to it. That is, the dialectical rules constituting and governing the argumentative discussion in which she is participating place a set of social and normative responsibilities upon her. Thus, in whatever range of social activities the rules of argumentation are binding, and to whatever extent the results of the argumentation are enforceable, the arguer can be held responsible to the commitments she took on in the course of argumentation. But, not having accepted those commitments in her own mind, she does not hold herself accountable to them. So, as soon as she is no longer bound by the rules of the argumentative discussion she can act according to her own rational lights. (Godden 2010, p. 406)

A consequence of Godden’s argument is that a necessary condition on the effective resolution of a difference of opinion in typical argumentative discussions is that there is a change in mental attitude from one or other of the parties involved in the argumentative discussion[iii]. Godden asserts that “if argumentation is to effectively resolve differences of opinion then the jurisdiction of argumentation must include the arguer’s own belief system which forms the basis for her actions.” (Godden 2010, p. 413 italics added)
If we express Godden’s necessary condition for an effective resolution to a difference of opinion in conditional form we get the following statement,
(NC) If there is an effective resolution to the difference of opinion in an argumentative discussion D, then there is a change in mental attitude of one or other of the parties participating in D.

Agents engage in argumentation with the objective of effectively resolving a difference of opinion. The goal of arguers involved in an argumentation is hindered if NC is not satisfied. It would thus be reasonable to expect that there are normative rules binding on arguers in order to facilitate effective resolution. If there were none, then arguers who decided to engage in argumentation would not be accountable for failing to satisfy, or even frustrating satisfaction of, the necessary condition for achieving their own proclaimed (perhaps only implicitly proclaimed) objective. A lack of accountability in this respect, however, seems absurd.
Consider the following line of reasoning. Deciding to start an argumentation is similar to making a promise (at least implicitly) to rationally pursue an effective resolution to a disagreement. If the arguer hinders that objective, then the arguer is acting hypocritically since the arguer is acting in a way that is contrary to their implied promise when deciding to commence arguing. Therefore, arguers are normatively bound to act in manner that is compatible with genuinely trying to satisfy NC.

At this point, the question remains: “what exact norms are implied by NC?” While there may be several norms implied by NC, for the purposes of this paper, I can only focus on one. I take the norm I discuss here to be central to the achievement of an effective resolution. However, before I can formulate the specific norm, I argue for the norms existence by exploring implications of an arguer’s obligation to genuinely try to satisfy NC. The argument begins by considering the contrapositive of NC, (CNC) If there is not a change in mental attitude of one or other of the parties participating in an argumentative discussion D, then there is no effective resolution to the difference of opinion in D.

CNC brings to the fore the claim made in the previous paragraph that the achievement of effective resolution depends on an arguer changing their mental attitude. So given that an effective resolution depends on the satisfaction of NC, in trying to isolate the norms implied by NC, we should consider the different courses of action compatible with an arguer genuinely trying to satisfy NC. As far as I am aware this leaves three choices open to arguers. An arguer could either: (a) change their mind in regards to the issue over which there is disagreement so that an effective resolution is possible, (b) continue in the argumentative discussion until an effective resolution arises, or (c) decide that an effective resolution is not possible and withdraw from the argumentation recognizing that it cannot succeed[iv]. In a situation where an arguer has conceded the acceptability of the rational critic’s standpoint options (b) and (c) are not open to that arguer. Option (b) is not open because the dialogue cannot continue if the critic has, quite justifiably, come to the view that the arguer agrees with their position and an effective resolution has been achieved. Often when such a concession is made an effective resolution will result. However, as Godden’s case discussed earlier illustrates, there are occasions where the arguer will make a concession without changing their mental attitude and, consequently, an effective resolution to the difference of opinion will not result. Option (c) is not open in this situation since the arguer has accepted the position of their opponent and has not withdrawn from the argumentative dialogue but has conceded it. Therefore, in such a scenario, the only option consistent with the arguer’s obligation to act in a manner compatible with genuinely trying to satisfy NC is for the arguer to take option (a) and change their mind about their initial standpoint. We can express this norm with the following conditional sentence NAC:
(NAC) if an arguer concedes to a standpoint (or challenge to their standpoint), then ceteris paribus the arguer ought to modify their mental attitude appropriately.

There are a few things worth pointing out about the way I have phrased NAC. I understand the norm to make the arguer out to have an obligation to “modify their mental attitude appropriately” instead of something stronger like “changing their belief.” The reason for adopting the weaker formulation is that there are many possible changes in mental attitude different from a change in belief that may be sufficient to lead to an appropriate change in future action, reasoning, and argumentation.
NAC should also make accommodations for cases where a party has conceded a standpoint but that party later comes across strong reasons or evidence, of which they were previously unaware, that undermined the standpoint to which they previously conceded. I recognize that there is controversy over the meaning of ceteris paribus clauses. However, all I take that clause to mean in NAC is that the evidence and arguments available to the arguer remain the same. In other words, the arguer is not aware of any new evidence or arguments that she understands to override the rational grounds for which she conceded the standpoint.

5. What Type of Norm is NAC?
As explained above, NAC is a norm that must be followed in order to satisfy the necessary condition NC required for the effective resolution of a difference of opinion. It might seem natural, then, to think that NAC is a norm of argumentation. That is, to think of it might seem natural to think of NAC as a norm whose domain of jurisdiction is the argumentative dialogue. However, the normative force of NAC would extend beyond the domain of the argumentative discussion and cover some of an arguer’s psychological states. Godden is not troubled by this and suggests that there are a set of norms of argumentation that must go beyond the argumentative dialogue in the way that NAC does. Godden states that in order to be effective,

Argumentative commitments must be binding and enforceable, and typically must extend beyond the argumentative dialogue itself. To capture this idea it might be useful to speak of the jurisdiction of argumentation. Roughly by this I mean the domain over which the results of argumentation are binding; that is the domain over which argumentative rules have normative force or can act as norms. (Godden 2010, p. 412)

However, there are reasons to be sceptical about understanding NAC to be a norm of argumentation. Because arguers could always concede points or take on commitments that do not reflect their mental attitudes without anyone knowing, it is hard for an argumentation analyst to determine whether an arguer has satisfied NAC through the evaluation of argumentative discussions. The task of the argumentation analyst is more clearly delineated and simplified if she can focus on the externalized moves made during an argumentative discussion. If we expand the jurisdiction of the norms of argumentation beyond the argumentative discussion, it becomes very difficult for an analyst of argumentative text or speech to adequately determine whether an arguer is satisfying the norms of argumentation. This is the first theoretical upshot to understanding the jurisdiction of the argumentation norms to be restricted to the externalized commitments made in argumentation.

One consequence to restricting the jurisdiction of argumentative norms to the externalized commitments made in an argumentative discourse is worth flagging right off the bat. The consequence is that satisfying only argumentation norms would not be sufficient to determine if an effective resolution to the argumentative dialogue was achieved. The reason for this insufficiency is that argumentation norms would be restricted to the actual argumentative discussion when NAC demands the satisfaction of conditions beyond the argumentative discussion itself – i.e. the adoption of an appropriate mental attitude. Argumentation norms are still needed to determine whether all the externalized moves are in order in an argumentation, however, satisfaction of those norms does not determine whether NAC has been satisfied and, thus, satisfaction of those norms is not sufficient to determine whether or not an argumentation has been effectively resolved. I will have more to say about this consequence of regarding NAC as a non-argumentation norm in the next section.
My suggestion is that we understand NAC to be a non-argumentation norm of some other normative type. Good candidates for what normative type we could understand NAC to belong to are the norms of reasoning or the norms of belief (or more generally the norms of propositional attitudes). Following Finocchiaro (1984) and Johnson (2000) I understand the theory of reasoning “to formulate, . . . test, . . . clarify, and systematize concepts and principles for the interpretation, the evaluation, and the sound practice of reasoning.” (Finocchiaro 1984, p. 3) The norms of reasoning, then, as I understand them, are the norms the theory of reasoning establishes have jurisdiction over the practice and evaluation of reasoning. They are the norms that indicate which sort of reasoning practices are poor ones and which sort are exemplary ones. I understand the norms of belief – which could be more broadly formulated as the norms of propositional attitudes – roughly as the norms that have jurisdiction over the possession of propositional attitudes.

NAC’s jurisdiction covers what propositional attitudes an agent ought to posses. Therefore, NAC can naturally be understood as a norm of belief since it holds agents responsible for the propositional attitudes they posses and makes recommendations about what propositional attitudes they ought to possess. If an agent possess an inappropriate belief – say one that does not reflect the evidence available to them – then the agent would not be satisfying the norms of propositional attitudes. On this view, if an arguer made a concession and failed to adopt an appropriate mental attitude the arguer would be in violation of a norm for the proper formation of propositional attitudes instead of a norm of argumentation. By conceiving of NAC as a norm of belief, the argumentation analyst can focus their assessment on whether all the externalized moves of the argumentation are in order without concerning themselves about what beliefs an arguer has.

NAC could also naturally be construed as a norm of reasoning. If an arguer were to maintain a mental attitude the arguer understood to be incompatible with a concession made in an argumentation without having become aware of further reasons that would undermine the grounds for making the concession, then the arguer would be guilty of bad reasoning. Thus, a violation of NAC could be understood as a violation of proper reasoning and the arguer would be accountable for such a failure. On this view NAC has jurisdiction over the proper practice of reasoning. Also, in addition to allowing argumentation theorists to focus on overt commitments made during the argumentation, understanding NAC as a norm of reasoning could also have further theoretical upshots since it has potential to help us better understand the relationship between argumentation and other domains that rely on proper reasoning such as problem solving, critical thinking, knowledge, rationality and investigation. Ralph Johnson (Johnson 2000, pp. 21-23) describes the problem of specifying how all these domains interrelate as “the network problem.”

If we understand NAC as a norm of reasoning we may be able to make headway in understanding the relationship between norms of reasoning and argumentation(5). Recall that NAC must be satisfied for the necessary condition NC to be satisfied. In other words, if NAC is treated as a norm of reasoning, satisfaction of a norm of reasoning would be a necessary condition for effective resolution to a difference of opinion. So an agent adopting at least some proper reasoning practices is required if that agent is going to effectively resolve a difference of opinion through argumentation. I think a worthwhile project would be to determine what norms need to be satisfied in order to satisfy NC and what normative-types these norms fall under. This project could identify some of the relationships between the domain of reasoning and the domain of argumentation and thus add some clarity to the difficult network problem. This last suggestion is speculative at this point and needs much more development than I can offer here. Nevertheless, improvement on the network problem is a theoretical opportunity provided by understanding NAC as a norm of reasoning that I think is worth exploring.

6. Broader Implications
There are several interesting questions that emerge from the line of thought pursued in this paper that I cannot fully address here. It will, however, be worthwhile to mention some of these unresolved issues.
One of the consequences of regarding NAC as a non-argumentation norm is that a necessary condition of an argumentation being effectively resolved is that a non-argumentation norm be satisfied. Even if all the norms of argumentation are satisfied in some particular argumentation, it is possible that one of the parties in that argumentation could fail to appropriately modify their mental attitude and, thus, the argumentation would not be effectively resolved. Now if the normative study of argumentation studies anything it is the norms that govern an argumentative discussion. However, if my arguments are correct, these norms, on their own, are insufficient to determine whether an argumentation is effectively resolved. What does this consequence mean for argumentation theory as a whole? What does it tell us about the subject matter and scope of argumentation theory?

One option is to regard argumentation theory as having a scope that is restricted to the norms of argumentation. This view of argumentation theory accepts that argumentation theory cannot tell us whether or not an argumentation has been effectively resolved. Rather, on this view, it is the job of some other discipline – for example, psychology or critical thinking – to determine whether an argumentation has been effectively resolved and it is the job of argumentation theory (in so far as it is a normative theory at least) to determine whether the norms governing argumentative discussions have been satisfied. A different option would be to regard argumentation theory as the study of more than just the norms of argumentation, but as the study of the norms that, in some broad sense, bear on argumentation. On this view the normative study of argumentation is broader than the delineation of the norms of argumentation. The study of argumentation would include the study of all norms relevant to argumentation even if those norms are not norms of argumentation. One reason to adopt this view is that it would allow argumentation analysts to determine whether or not an argumentation has achieved its goal to effectively resolve a difference of opinion. However, which of these different ways of regarding the theory of argumentation is correct is beyond the scope of this paper and must be resolved on another occasion.

7. Summation
I began with a discussion of normative types in which I explained that we can understand what distinguishes one normative type from another based on the domain over which the norms of that normative type possesses normative jurisdiction. I then explained how Godden argues that a necessary condition on the effective resolution to an argumentation requires an arguer who has conceded a standpoint (or a challenge to a standpoint) to modify their mental attitude appropriately so that their future actions, reasoning, and argumentation reflect that modification. I formulated this necessary condition as NC. I argued that NC implies the norm of argumentation compliance NAC. I then examined Godden’s suggestion that because NAC has normative force over an arguer’s beliefs we may want to understand argumentative norms as extending beyond the argumentative discussion itself. I discussed some theoretical advantages of understanding NAC (and perhaps other similar norms) to be a non-argumentation norm such as a norm of belief or a norm of reasoning. The theoretical advantages were (a) such an understanding of NAC allows argumentation analysts to focus only on the externalized commitments of arguers instead of worrying about an arguer’s mental attitudes and (b) it may assist in better understanding the relationship between reasoning and argumentation. I concluded by considering some unresolved issues that are turned up by the discussion in this paper. In particular I considered what the arguments developed in this paper say about the scope of argumentation theory. Given that the norms of argumentation do not include NAC one might conclude that the scope of argumentation theory is limited in such a way that argumentation theory does not have the job of determining whether or not an argumentation is effectively resolved. However, as mentioned there are other options. One alternative discussed was that argumentation theory be understood as the study of norms that are broader than the norms of argumentation. Which of these different views of argumentation theory is correct is not possible to determine here, however, I think it is an important issue worth the thought of those interested in the theory of argumentation.

NOTES
[i] Godden adopts a broadly Davidsonian picture of the relationship between actions and beliefs under which belief together with a sufficiently strong pro-attitude (such as a desire) is a reason for an action. Godden also draws on the common intuitions that “normally and generally our beliefs about the world causally influence our behaviour” and that “our beliefs play a premissory role in our inferences and practical reasoning, they have a causal role in determining our actions” (Godden 2010, p. 400).
[ii] One important qualification on the scope of this necessary condition is required. A person could be held accountable in such a way that future actions, reasoning and argumentation were appropriately modified if social mechanisms for the enforcement of the results of argumentation were strong enough. Legal cases are examples in which the social mechanisms for enforcing the results of argumentation are sufficiently strong to cause a change in behaviour – and even at times a change in reasoning and argumentation – without requiring the presence of a change in mental attitude (Godden 2010, pp. 412-413). However, as Godden points out, typically these social enforcement mechanisms will not be present and it is up to the arguer to enforce the results of argumentation on themselves (Godden 2010, pp. 412-414).
[iii] I take the case of a partial-resolution to a disagreement to be embraced by option (a) since for a partial resolution to a disagreement to be effective there must also be a change in mental attitude. A partial-resolution may be achieved when one or both of the parties in an argumentation have moved closer – through making partial concessions – to the standpoint of the other without fully endorsing it. However, without an appropriate change in mental attitude the move closer to the alternative standpoint will not impact future reasoning, deliberation, and argumentation and, thus, the partial-resolution would not be an effective partial-resolution.
[iv] A reasonable response for someone who wanted to maintain that NAC is a norm of argumentation is that we can still make headway on the network problem and understand NAC to be a norm of argumentation. This is a real possibility which I do not want to deny. While I am unclear about what an account of the relationship between reasoning and argumentation would look like if we understand NAC as an argumentation norm, I do not want to deny the possibility that one could be develop such an account and that that account could be an appealing one.

REFERENCES
Eemeren, F.H. van, Grootendorst, R., Snoeck Henkemans, A.F., Blair, J.A., Johnson, R.H., Krabbe, E.C.W., et al. (1996). Fundamentals of Argumentation Theory. Handbook of Historical Backgrounds and Contemporary Developments. Mawhah: Lawrence Erlbaum.
Eemeren, F.H. van, Grootendorst, R., & Snoeck Henkemans, F. (2002). Argumentation. Mahwah: Lawrence Erlbaum.
Eemeren, F.H. van, & Grootendorst, R. (2004). A Systematic Theory of Argumentation: The pragma-dialectic approach. Cambridge: Cambridge University Press.
Finocchiaro, M.A. (1984). Informal logic and the theory of reasoning. Informal Logic 6, pp.3-8.
Godden, D.M. (2010). The importance of belief in argumentation: belief, commitment and the effective resolution of a difference of opinion. Synthese 172(3), pp. 397-414.
Hamblin, C.L. (1970). Fallacies. London: Methuen.
Johnson, R. H. (2000). Manifest Rationality: A pragmatic theory of argument. Mahwah: Lawrence Erlbaum Associates.
Walton, Douglas and Erik C.W. Krabbe. (1995). Commitment and Dialogue: Basic Concepts of Interpersonal Reasoning, Albany: State University of New York Press.