PDF31.4 05 van Laar 344-366 RTG © Jan Albert van Laar. Informal Logic, Vol. 31, No. 4 (2011), pp. 344-366. Argumentation schemes from Hamblin’s dialectical perspective JAN ALBERT VAN LAAR University of Groningen Oude Boteringestraat 52 9712 GL Groningen The Netherlands j.a.van.laar@rug.nl Abstract: This paper aims at a nor- mative account of non-deductive argumentation schemes in the spirit of Hamblin’s dialectical philosophy. First, three principles are presented that characterize Hamblin’s dialecti- cal stance. Second, argumentation schemes, which have hardly been examined in Hamblin’s book Falla- cies, shall be dealt with by applying these principles, taking an argumen- tation scheme from authority as the leading example. Third, a formal dialectical system, along the lines indicated by Hamblin, shall be de- veloped that includes norms for us- ing argumentation schemes and norms for responding to arguments that are presented as instantiating acceptable argumentation schemes. Résumé: Ce document vise à décrire des schèmes normatifs de l’argumentation non-déductive dans l'esprit de la philosophie dialectique de Hamblin. D'abord, je présente trois principes qui caractérisent la position dialectique de Hamblin. Deuxièmement, j’applique ces principes aux schèmes d'argumentation, qui n'ont guère été examinés dans Fallacies, le livre de Hamblin, en prenant un schème d'argumentation d'autorité comme le meilleur exemple. Troisièmement, j’élabore un système formel dialectique, dans le sens indiqué par Hamblin, qui comprend des normes pour utiliser des schèmes d'argumentation et des normes pour répondre aux arguments présentés comme l'instanciation des schèmes d'argumentation acceptables Keywords: argumentation scheme, connection premise,counterconsideration, criticism, dialectic, expert opinion, formal dialectic, Hamblin, premise, pre- sumption, scheme. 1. Introduction From a dialectical perspective, argumentation theory can be conceived of as the study of how participants in a dialogue do and should act with the aim of solving their self-defined prob- lems or differences by balancing considerations pro and con, so that the result is based upon their own conceptions of the merits of the case (cf. Van Eemeren and Grootendorst 2004; cf. Walton Argumentation Schemes from Hamblin's Perspective 345 1998). By expounding his philosophy of argument in his ground-breaking book Fallacies (Hamblin 1970), Hamblin has been one of the originators of the kind of argumentation theory according to which dialogue participants need not rely on exter- nal norms, but control each dialogue themselves. This paper deals with non-deductive argumentation schemes (Garssen 2001; Hitchcock 2010; Kienpointner 1992; Wagemans 2011; Walton, Reed and Macagno 2008), and it aims at a normative account of them in the spirit of Hamblin’s dialec- tical philosophy. First, three principles are presented that charac- terize Hamblin’s dialectical stance (Section 2). Next, I shall ex- amine how argumentation schemes, which have hardly been ex- amined in Hamblin’s book, can be dealt with from such a per- spective by applying these principles to arguments that are put forward with the pretence of instantiating an acceptable argu- mentation scheme, taking an argumentation scheme from authority as the leading example (Section 3). Finally, I shall in- dicate how a formal dialectical system, along the lines indicated by Hamblin, can be developed that includes norms for using ar- gumentation schemes and norms for responding to arguments that are presented as instantiating acceptable schemes (Section 4). 2. Hamblin’s dialectical perspective First, Hamblin’s dialectical philosophy can be characterized by, what he labeled, “the Dialectical Theory of Logical Form, or, perhaps, the Dialectical Theory of Meaning” (pp. 285-6). I shall refer to this theory as the Dialectical Meaning Principle: all properties of linguistic entities are … determinable from the broad pattern of their use (p. 285). Hamblin seems to apply this principle at three occasions at least. (a) When dealing with equivocation, the meaning of non-logical expressions is not to be seen as given in advance of the dialogue, but rather as determinable from the ways in which the expres- sions are used within the dialogue, among which the ways in which the expressions’ meanings are molded with clarifying re- marks. To elaborate on his example, the meaning of “locate” can be deciphered by examining what statements, containing that term, have been accepted or rejected, and, in special situations, by examining how the term at hand has been charged as equivo- cal—“your argument is based on a shift in the meaning of ‘lo- cate’” —, or how such a charge is responded to—“at all occur- rences, I have used the term as meaning the same as ‘decide Jan Albert van Laar 346 where to put x’” (Hamblin 1970, Chapter 9). (b) A different kind of application of this Dialectical Meaning Principle can be found in Hamblin’s discussion on the various kinds of locution (speech acts) that are needed within a dialogue. According to Hamblin, the so-called “syntactical rules” (p. 259, p. 269) of the dialogue provide the various types of locution with their meaning. To use one of his examples, a question “P or Q?” is the type of locution that is followed by the assertion “P”, or by the assertion “Q,” or by the expression of non-commitment to the disjunction of P and Q, or by the denial of the disjunction of P and Q, or by ex- pression of non-commitment to both P and Q (p. 266). These rules provide the meaning of this kind of question.1 A discussion about the meaning of the locution type of “assertion” prompts Hamblin to state his Dialectical Meaning Principle (p. 284-286). (c) According to Hamblin, a logically deductive scheme, such as modus ponens, is primarily a dialectical procedure (pp. 299- 300), which suggests that its meaning is exhausted by a set of procedural, dialogical rules. How could we apply the Dialectical Meaning Principle when developing a normative model of argumentation schemes? According to the principle, an argumentation scheme must be seen as provided with its meaning, either by the way participants actually use argumentation schemes (as in application (a) above), or by rules that govern the use of argumentation schemes within a dialogue (as in applications (b) and (c)). When aiming at a prescriptive formal system, argumentation schemes must be defined by formulating the general preconditions and postconditions that govern the speech acts that are employed when using the argumentation schemes (such as putting forward an argument that instantiates a particular argumentation scheme), or when responding to a speech act in which it has been used (such as critically reacting to the connection between the premises and the conclusion of an argument that instantiates an argumentation scheme). As far as a particular scheme is con- cerned, for example a particular argumentation scheme from authority, its meaning is also to be given by the specific proce- dural rules that make this argumentation scheme stand apart from other argumentation schemes. An example of such a rule would be one that enables the receiving party to critically react towards a kind of premise that is characteristic of the argumen- tation scheme at hand, such as the proposition that the source appealed to in the argument really is an expert in the field. Ar- gumentation schemes, then, are different in so far as different preconditions or postconditions obtain. 1 Hamblin contrasts syntactical rules with “discretionary rules,” which aim at improving the quality of the dialogue, and do not define speech acts. Argumentation Schemes from Hamblin's Perspective 347 Second, Hamblin’s dialectical philosophy can be charac- terized by his view on the evaluation of arguments. It is stressed by Hamblin at various occasions that the decision on whether an argument is “good” is in the end up to the participants them- selves. We can label this view as his Dialectical Evaluation Principle: The evaluation of arguments and of other dialogical contri- butions is to be left to the participants. Hamblin’s view on the role of the logician gives evidence of his acceptance of this principle. About this role, he writes: The logician does not stand above and outside practical argumen- tation or, necessarily, pass judgement on it. He is not a court of appeal, and there is no such judge or court: he is, at best, a trained advocate. It follows that it is not the logician’s particular job to declare the truth of any statement, or the validity of any argument (italics in original, p. 244). The task of a logician, or an argumentation theorist for that mat- ter, is to develop the devices with which the participants in an argumentative dialogue can themselves solve the differences and problems that they encounter, and to place these devices at their disposal. Accordingly, arguments should not be evaluated on the basis of norms that are external to the dialogue at hand. One consequence of the Dialectical Evaluation Principle is stated explicitly by Hamblin, to wit that the decision whether the connection between the argument’s premises and conclusion, or—as he names it—the passage from premises to conclusion, is “good” is, in the end, up to the participants themselves. When Hamblin discusses the dialectical criteria for good arguments, he states just one criterion that deals with this connection. I shall refer to this criterion as the Justificatory Force Principle, and consider this to be a consequence of the Dialectical Evaluation Principle: The passage from premisses to conclusion must be of an accepted kind (italics in original, p. 245). This Justificatory Force Principle becomes apparent at a number of places. Above, we have already seen that Hamblin conceives of a deductively valid argument form, such as modus ponens, as a dialectical procedure, which suggests that the procedure can be adopted or rejected, so that whether a particular argument is to be treated as a deductively valid argument is, in part, a conse- quence of a previous choice. The emphasis on acceptance fur- Jan Albert van Laar 348 ther becomes clear when discussing the situation where a recipi- ent happens to accept a particular form of reasoning as deduc- tively valid, and where he further acknowledges that a particular argument with acceptable premises instantiates that form of rea- soning, but where she nevertheless does not want to accept that argument’s conclusion. Even in such situations, there exists an “virtually ad infinitum” (p. 252) number of escape routes for the recipient of this argument. She may point out an equivocation, or take the argument to be question begging, or consider the ar- gument insufficiently translucent, or raise a counterargument (pp. 251-252), and so forth.2 Thus, even deductive arguments are only knock-down arguments, in Hamblin’s view, if they are accepted by the participants as such. How can we follow a similar course when dealing with arguments that aspire to instantiate an acceptable argumentation scheme? One consequence of Hamblin’s dialectical perspective seems to be that the dialogue participants themselves must be seen as being in the position to choose the argumentation schemes they consider appropriate, and that they have the pro- cedures at their disposal with which to determine whether a par- ticular argument satisfies their standards. In other words: in the end, it is up to the participants what argumentation schemes to adopt as prima facie acceptable, and how to determine whether a particular argument fulfills self-chosen requirements.3 In a dif- ferent, but related context of dealing with fallacy-mongers, Hamblin writes: “The control of each dialogue is in the hands of the participants themselves” (p. 283). 3. A dialectical theory of argumentation schemes In this section, I shall elaborate on how argumentation schemes can be incorporated in a dialogue theory, in line with Hamblin’s dialectical principles. The following questions shall be dealt with: How can we define “argumentation scheme”?; What are the options for the participants when choosing their preferred 2 Hamblin’s view seems to be that the passage of the argument is not lacking goodness merely as a result of the recipient’s refraining her consent to the passage from premises to conclusion. Instead, the idea seems to be that an argument is a failure to the extent that the recipient of the argument offers a relevant counterconsideration against the passage. There must be a ‘good reason’ for not accepting the passage, a requirement which, of course, can only be negotiated within the confines of the dialogue, given the dialectical perspective. 3 What remains for argumentation theorists? In line with Hamblin’s view, the argumentation theorist’s task would be to assist dialogue participants to make these choices, for example by developing useful concepts, and dialogical sys- tems or procedures. Argumentation Schemes from Hamblin's Perspective 349 argumentation schemes?; How can they use argumentation schemes in order to interpret and reconstruct arguments?; How can the proponent profit from employing an argumentation scheme that has been adopted?; How can the participants deter- mine whether an argumentation scheme has been correctly ap- plied?4 3.1 Defining “argumentation scheme” I conceive of an argumentation scheme as a scheme of reason- ing, containing a number of variables, that involves particular (conditional) obligations in a dialogue in which the scheme has been adopted by the participants. If an opponent has adopted a particular argumentation scheme as acceptable, and if she has been offered an argument that clearly instantiates that scheme, she must be considered as under the obligation to commit her- self to the standpoint (or conclusion) of that argument as soon as she has committed herself to the reasons (or premises) of that argument, be it that this commitment to the standpoint is can- celled if the opponent puts forward a counterconsideration that cannot be successfully refuted by the proponent. An adopted argumentation scheme is an argumentation scheme that has ac- tually been adopted by the dialogue participants. That an argu- mentation scheme has been adopted implies that, according to the discussants, the scheme is not a deductively valid scheme, but that it exhibits a justificatory force that is sufficiently strong, so that each instance has premises that suffice, in non-excepting circumstances, for the conclusion. A commonly adopted argu- mentation scheme is an argumentation scheme that is frequently adopted within a particular community, or within a particular class of dialogues. Though not always (Hitchcock 2010, p. 157), the notion of an argumentation scheme (or argument scheme) is often defined as a commonly adopted framework of argument (Van Eemeren and Grootendorst 1992, p. 96; Garssen 2001, p. 96; Walton, Reed and Macagno 2008, p. 1). In order to get a clear view on the nature and role of argumentation within a particular setting, such as for instance a political deliberation, it is useful to make use of a notion of commonly adopted argumentation schemes. However, we also need to be able to examine patterns of reason- ing that resemble common argumentation schemes in all re- spects except for being common, in order to be able to describe discussants as having adopted argumentation schemes that are not common but, in their view, worthy of acceptance. 4 An exposition of the ideas in Section 3 can also be found in Van Laar (2012) and in Van Laar and Krabbe (2012) . Jan Albert van Laar 350 3.2 Choosing argumentation schemes In order to control the dialogue, the participants themselves must choose the argumentation schemes they consider pertinent to resolving their differences, as well as specify the details of those schemes, in some situations regardless of whether the adopted schemes or the adopted specifications of schemes are frequently adopted ones. For example, they could choose to adopt an argumentation scheme From Expert Opinion (cf. Wal- ton, Reed and Macagno 2008, p. 310) as acceptable, and specify the preferred version as in Scheme 1: Person E is an expert in field F; Person E says that P; P is a proposition within field F; Therefore P. But then, they could specify more or less the same idea in a somewhat different manner, leading to Scheme 2, Person E is an expert in field F; Person E says that P; P is a proposition within field F; Person E is unbiased with re- spect to P; Therefore P. Or, they could even adopt Scheme 3: Person E is an expert in field F; Person E says that P; P is a proposition within field F; Person E is unbiased with re- spect to P; Person E has always shown to be reliable as far as propositions in field F are concerned; Therefore P. Argumentation schemes can be considered for adoption within a preliminary stage of the dialogue, such as the opening stage of a critical discussion (Van Eemeren and Grootendorst 2004, Chap- ter 3). Particular argumentation schemes can be considered as appropriate, or inappropriate for that matter, within a particular class of dialogues, such as those that are performed within more or less institutionalized “argumentative activity types” (van Eemeren and Houtlosser 2005). For example, schemes that un- derlie personal attacks are prima facie acceptable in many types of political debate, while hardly so within scholarly meetings between experts. The choice to enter a particular type of dia- logue can be considered to imply a choice to commit oneself to the argumentation schemes that go with that kind of dialogue. For the purpose of specifying dialogue rules for the use of argumentation schemes, I shall assume that choices about what argumentation schemes to adopt have been made in a prelimi- nary stage, and that it is clear to the participants what argumen- Argumentation Schemes from Hamblin's Perspective 351 tation schemes have been adopted. Different choices lead to dif- ferent rights and obligations. These will be discussed in the sub- sections below, in preparation of the system to be developed in Section 4. 3.3 Reconstructing arguments in the light of the adopted argu- mentation schemes An argument that is presented by the proponent as instantiating a particular adopted argumentation scheme will be interpreted and reconstructed in the light of that argumentation scheme. Thus, after having adopted Scheme 2 or Scheme 3, rather than Scheme 1, an argument “Erwin says that it is going to rain to- night, so, probably, it is going to rain tonight”, can be recon- structed as containing the unexpressed premise to the effect that Erwin is unbiased with respect to the proposition that it is going to rain tonight. In addition, the passage from premises to conclusion gets a particular content, as a result of the choice for a particular argu- mentation scheme. I assume that this passage can be considered as the substance of a separate premise that remains often, but not always, implicit: the so-called connection premise (Walton and Krabbe 1995, p. 128). The connection premise is a conditional statement, having the conjunction of the other premises as its antecedent and the standpoint as its consequent. It expresses that the passage between the argument’s premises and its conclusion is acceptable within the context at hand. The practical message of the connection premise is that the opponent’s commitment to the set of (explicit and implicit) regular premises entails, within the situation at hand, a commitment to the conclusion. After having adopted Scheme 1, the connection premise reads “If Er- win is a meteorologist and Erwin says that it is going to rain to- night, and if this is a statement within meteorology, then it is going to rain tonight.” However, if Scheme 2 had been adopted, the connection premise would read “If Erwin is a meteorologist, and Erwin says that it is going to rain tonight, and this is a statement within meteorology, and Erwin is unbiased in this re- spect, then it is going to rain tonight.” 3.4 Profiting from adopted argumentation schemes Krabbe makes a distinction between three types of concessions (2001), which can also be made to apply to the ways one may be committed to an argumentation scheme. A participant can accept a proposition as a free concession, meaning that she can retract her commitment to it without incurring any further obligations from retracting it. A plausible way in which a free concession Jan Albert van Laar 352 can be incurred is by refraining from challenging a statement by the other side. An argumentation scheme can also be the object of a free commitment. For example as a result of a participant’s choice to refrain from challenging the connection premise of an instance of the scheme at hand. Next, a participant can accept a proposition as a fixed concession within a particular dialogue, meaning that she cannot retract her commitment to it within the confines of this very dialogue. A fixed concession can be in- curred by explicitly tying oneself down to a proposition or to an argumentation scheme. Lastly, a participant can accept a propo- sition, or an argumentation scheme, as a presumption, meaning that she can retract her commitment to it, be it that she incurs special obligations by doing so. Entering a dialogue within a particular institutional context can be seen as entailing presump- tive commitments that characterize that institutional context. Even a philosophical skeptic is presumed to accept that he has a physical body at his athletics club. And by agreeing, as a politi- cian, to participate in a parliamentary inquiry into the causes of the financial crisis, one cannot, during that meeting, retract one’s commitment to arguments from authority,5 without chang- ing that kind of a meeting into quite a different kind of meeting. A commitment to a presumption involves special obligations when challenging, and thereby retracting commitment to it. When the opponent challenges a presumption, the propo- nent may always demand the opponent to explain her challenge, or demand her to defend the appropriateness of her challenge (see Van Laar and Krabbe 2012 on the opponent’s “burden of criticism”). An explanatory motivation of why a presumption has been withdrawn involves advancing counterconsiderations, by which the speaker makes it clear to the interlocutor why she is critical and how she considers her critical position to be a ten- able position. A defense of the appropriateness of challenging a presumption, which has been labeled a validation and an argued challenge (Van Laar and Krabbe 2012), involves giving argu- mentation, at a metalevel of dialogue, in favor of the appropri- ateness of the retraction of the presumption, for example by re- ferring to changed circumstances. However, in response to an opponent’s mere challenge of a presumption P, a request by the proponent for counterargumentation—that is, an argument by the opponent in favor of the denial of P—would amount to a Straw Man Fallacy. Only if the retraction of the presumption is conveyed through the rejection (strong denial) of it, it would be legitimate to request for genuine counterargumentation. 5 Of course, one must at least accept some version of an argument scheme with which an appeal to authority can be made, but not necessarily all ver- sions of such argument schemes. Argumentation Schemes from Hamblin's Perspective 353 There is an important difference between an explanatory motivation and a validation. In both cases, to meet the burden of criticism, the opponent puts forward reasoning. What makes these speech acts different is that the explanation may succeed even though the reasoning does not start from propositions that the proponent has conceded. In the case of validation, the oppo- nent is putting forward argumentation, and in order to succeed she must start from propositions conceded by the proponent. This is also the case in counterargumentation, but now she is defending a proposition at a meta-level. To be more specific, she is defending a proposition about the legitimacy of her previous move of challenging the connection premise. In order to elaborate on explanatory motivations, the no- tion of a counterconsideration must be dealt with in some more detail. A counterconsideration can be approached from the per- spective of giving an explanatory motivation, but it can also be approached from the perspective of giving strategic advice. First, a counterconsideration can be seen as revealing to the proponent how the opponent’s critical stance towards the main standpoint is a tenable position, in spite of her commitments. The opponent may explain her critical stance towards the con- nection between Erwin’s making his forecast and the correctness of the forecast by informing the proponent that she reckons, for instance, with the possibility that Erwin is making a mistake as a result of a lack of sleep. Second, a counterconsideration can be seen as revealing to the proponent how he should go about in his attempts to convince the opponent of his standpoint. Applied to our example: The proponent is advised to refute that Erwin might be making a mistake as a result of a lack of sleep, or to show that this possibility is far-fetched and need not be taken seriously. The opponent may stress either of these two perspec- tives when presenting her counterconsideration. She may stress the explanatory perspective, saying: “Why would we expect rain on account of Erwin’s say-so? As far as you’ve shown, Erwin might be making a mistake as a result of a lack of sleep.” Or, she could stress the perspective of strategic advice: “Why rain?; Can you exclude the possibility that Erwin was making a mis- take as a result of a lack of sleep?” In other words, a counter- consideration is a proposition that is put forward in order to both explain how the opponent takes her critical position to be a ten- able position and to provide the proponent with information about how to proceed in his attempts at defending his position. Offering a counterconsideration does not need to involve a (conditional) burden of proof. The opponent’s counterconsidera- tion can be put forward with the illocutionary force that Rescher called “cautious assertion.” When the opponent utters P as a cautious assertion, she expresses the message that “P is the case Jan Albert van Laar 354 for all that you (the adversary) have shown” or “P’s being the case is compatible with everything you’ve said (i.e., have main- tained or conceded)” (Rescher 1977, p. 6). Although a counter- consideration can also be asserted more strongly, such that the opponent must argue in favor of its correctness if asked to do so by the proponent, I shall assume in the remainder of this paper, that the opponent refrains from adopting a standpoint herself, and restricts herself to critically test whatever the proponent puts forward. In these situations, counterconsiderations are cau- tiously asserted and do not involve a burden of proof. Still, as we shall see, they can be successful in undermining the applica- tion of an acceptable argumentation scheme. Walton, Reed and Macagno (2008) discuss the problem of how an argumentation scheme can be binding, given that in- stances of argumentation schemes can be defeated. The twofold solution of the problem that I adopt here is the following (see also van Laar 2012). (a) If an argument instantiates an argumen- tation scheme that has been adopted by the opponent as a fixed commitment or a presumption, then the connection premise of that argument can always be treated as a presumption, that is, as a proposition that can only be challenged by incurring the obli- gation to provide an explanation or a validation, if so requested by the proponent. The reason that in these cases the connection premise is not a fixed concession is that argumentation schemes are by definition non-deductive schemes. The reason that the connection premise is not merely a free concession or not even a concession at all is that the proponent’s decision to use this spe- cific argument—that instantiates this very argumentation scheme—is plausibly based upon the presence of the opponent’s initial commitment to the argumentation scheme at hand. (b) If an argument instantiates an argumentation scheme that has not been adopted by the opponent, or merely as a free concession, then the connection premise must either be seen as not being a commitment of the opponent, or as merely a free concession. If, in the second case, the opponent challenges the connection premise, she simply retracts her free concession. In other words, if the argumentation scheme is not a commitment it is not bind- ing, and if it is a free concession, it is hardly binding. In order to clarify this position, I shall elaborate further on the Erwin case. Suppose the opponent is committed, either as a fixed concession or as a presumption, to the argumentation scheme from authority that was labeled Scheme 1: “Person E is an expert in field F; Person E says that P; P is a proposition within field F; Therefore P.” Suppose, further, that the propo- nent has put forward the following argument: “Erwin is a mete- orologist; Erwin says that it is going to rain tonight; That it’s going to rain tonight is a meteorological statement; Therefore it Argumentation Schemes from Hamblin's Perspective 355 is going to rain tonight.” Then, according to this view, the oppo- nent is committed to the following connection premise as a pre- sumption: “If Erwin is a meteorologist and Erwin says that it’s going to rain tonight, and that it’s going to rain tonight is a me- teorological statement, then it is going to rain tonight.” Given the status of this connection premise as a presumption, the op- ponent is allowed to challenge it, but she must then, upon the proponent’s request, provide a validation for this challenge, which is an argument for the thesis that challenging, and thereby retracting commitment to, this presumption is, within the con- text at hand, a legitimate move. For example, she might allege that Dutch meteorologists have a bad track record, which defeats the use of authority arguments in this special context, Erwin be- ing Dutch. In addition, after having challenged this presumptive connection premise, the opponent must, upon the proponent’s request, provide an explanation of her challenge of the connec- tion premise. For example, she might state that she reckons with the possibility that Erwin is making a mistake, possibly as a re- sult of lack of sleep, or of a neurological disorder, or she might, again, put the quality of Dutch meteorology into question. If the opponent provides such an explanation for her posi- tion of critical doubt towards the proponent’s connection prem- ise, she merely provides a counterconsideration that she does not need to defend on the proponent’s request, but that the propo- nent should refute in order to make the opponent withdraw her critical stance towards the connection premise. More in detail, if the opponent provides W as her explanation of her critical doubt towards the connection premise, she thereby provides the pro- ponent with the strategic advice to either (a) defend not-W, or (b) to defend that W does not refute the connection premise, or (c) to show that W points at a possible situation in which, admit- tedly the proponent’s premises are true and the conclusion false, but that is in the context at hand far-fetched and unrealistic. Therefore, if the opponent provides such an explanation, she is not defending a standpoint of her own. The opponent can defeat the proponent’s argument without advancing any genuine argu- ment (cf. Pollock 1995). To summarize this view, argumentation schemes are con- venient devices for proponents, for if the opponent is committed to a scheme as a fixed concession or as a presumption, this im- plies the further, presumptive commitment to the sufficiency of the justificatory force of any argument that instantiates that ar- gumentation scheme. The opponent’s criticism of the justifica- tory force of such an argument comes at the expense of having to provide explanatory counterconsiderations or even a valida- tion, if requested. If the proponent applies an argumentation scheme that has not been adopted by the opponent or merely Jan Albert van Laar 356 adopted as a free concession, or if he employs an argument without clearly instantiating any identifiable argumentation scheme, the opponent may choose to challenge the connection premise without incurring any such obligation to explain or validate her position.6 Non-deductive argumentation schemes can be incorporated in a dialogue theory in a way that does not mitigate the basic dialectical division of labor. 3.5 How to make the passage from premises to conclusion ac- ceptable? How can an opponent criticize an instance of an adopted argu- mentation scheme, and how can a proponent make the passage from premises to conclusion acceptable? In other words: how can the participants determine whether an adopted argumenta- tion scheme has been applied correctly? Instead of imposing on the participants a fixed list of critical questions, the contending parties themselves can fix the procedural arrangement for testing the quality of arguments. If the parties go about in a more or less systematic way, they can agree on testing both the acceptability of each of the regular premises of the argument, whether made explicit by the proponent or left implicit, and the acceptability of the connection premise (often left implicit).7 As we have seen in Section 3.3, what counts as a regular premise and what counts as the connection premise depends upon the choice of the specific argumentation schemes in a preliminary stage. The participants can choose to test the acceptability of each of these premises with various kinds of critical reactions. They can request for an argument of a statement with so-called “pure challenges” (Krabbe 2007, p. 56), “Why S?” – for exam- ple: “How are you going to convince me that Erwin is a meteor- ologist?” They can choose to add a counterconsideration to the challenge, such as “As far as I have heard, Erwin never has made a serious study of meteorology,” that explains why the op- ponent is not yet accepting the premise, thus turning the pure challenge into a “bound challenge” (Krabbe 2007, p. 56). Or, a 6 Nevertheless, it would be instrumental for the resolution process if she does convey her motives for challenging the connection premise, for the reason that it assists the proponent in developing a persuasive strategy. In other words, if the connection premise is not a presumptive commitment, it would be legitimate for the opponent to refuse to provide further explanation or validation, even though providing counterconsiderations can be expected to improve the quality of the exchange, thus going beyond what is strictly re- quired for conflict resolution. 7 Cf. Wagemans (2011) for a systematic approach to critical questions per- taining to arguments that appeal to expert opinions. Argumentation Schemes from Hamblin's Perspective 357 premise can be tested by strongly denying8 it, “Erwin certainly is not a meteorologist,” possibly accompanied by counterargu- mentation, “for Erwin makes elementary blunders in his fore- casts.” Each of these reactions imply, or at least prepare for, a negative evaluation of the proposition at issue, thereby making the reaction a “critical reaction” (see Krabbe and Van Laar 2011 for the various “ways of criticism”). Of course, the connection premise of the argument is no exception: it can be critically reacted to with a request for argu- ment, possibly accompanied by a counterconsideration, or by way of a strong denial, possibly accompanied by counterargu- mentation. As we have seen, even a pure challenge involves a burden of criticism, if the connection premise happens to be a presumption In the remainder of this paper, I restrict my attention to di- alogues in which the opponent advances a pure challenge and offers an explanatory counterconsideration upon the proponent’s request. Thus, I leave validations and counterargumentations aside. In the dialogue system of the next section, the fragment of dialogue that ensues from advancing an explanatory countercon- sideration against a connection premise is labeled a refutation dialogue. In a refutation dialogue, three issues can be expected to surface: can the proponent show to the opponent that the counterconsideration is false?; can he show that the explanatory counterconsideration is insufficient to defeat the connection premise?; and can he show that the counterconsideration is far- fetched and unrealistic and need not to be taken into account? The rules are such that the winner of the refutation dialogue wins the dialogue as a whole. When has the proponent defended his argument convinc- ingly, as far as the connection premise is concerned? The answer is that the opponent must have an unconditional right to put for- ward counterconsiderations, showing in some detail how she considers the conclusion worthy of critical doubt even if she were to accept the premises. The judgment whether or not the opponent succeeds in advancing a realistic counterconsideration should be “in the hand of the participants themselves.” The pro- ponent has defended his connection premise successfully, if for each counterconsideration put forward, he has been able (a) to refute it by defending its falsity or by defending the insuffi- ciency of the counterconsideration to undermine the connection 8 Krabbe and Van Laar (2011) make a distinction between strong denials, by which one incurs a conditional burden of proof, and weak denials, that mere- ly express that the proposition denied has not yet been convincingly argued for and moreover expresses the expectation that no convincing argument for it will be forthcoming. Jan Albert van Laar 358 premise, or (b) to show that the counterconsideration does not convey a sufficiently realistic counterexample.9 The following example illustrates the dialectic. 1. Proponent: It is going to rain tonight. [Standpoint] 2. Opponent: Why so? [Pure challenge] 3. Proponent: Erwin is a meteorologist, and Erwin says that it is going to rain tonight, and this is a statement within meteorol- ogy. Therefore, it is going to rain tonight. [Argument that, in this context, clearly instantiates the adopted Scheme 1] 4. Opponent: Why (If Erwin is a meteorologist, and Erwin says that it is going to rain tonight, and this is a statement within meteorology, then it is going to rain tonight)? [Pure chal- lenge of the connection premise] 5. Proponent: This is what you’re presumed to have accepted on account of your acceptance of Scheme 1. Explain! [Request for explanation] 6. Opponent: Erwin might be confused, as a result of his old age. [Counterconsideration 1, that starts a refutation dialogue] 7. Proponent: No, he is still widely esteemed by his colleagues. [Refutation by showing the falsity of counterconsideration 1] 8. Opponent: But, nobody likes rain! [Counterconsideration 2] 9. Proponent: It’s false that if nobody likes rain, Erwin’s predic- tion doesn’t hold water. [Refutation by showing the insuffi- ciency of counterconsideration 2] 10. Opponent: Okay, I stand corrected. But then, his employer might have put him under pressure to come up with this very prediction. [Counterconsideration 3] 11. Proponent: You know that’s nonsense. [Statement to the ef- fect that counterconsideration 3 does not yield a sufficiently realistic counterexample to the argument] 12. Opponent: Yeah, you’re right. I give up! [Acceptance of the connection premise, and also of the standpoint.] Thus, an argument can be accepted by the opponent as a proper instance of an adopted argumentation scheme when all of her counterconsiderations have been refuted or shown to be unreal- istic. This procedural arrangement enables the parties, in a situa- tion where the argument at hand instantiates an argumentation scheme that governs the dialogue, to come to an agreement as to 9 According to the view adopted in this paper, the opponent need not provide an argument in order to convince the proponent that the counterconsideration is realistic. Although this feature has not been implemented in the formal system presented in Section 4.2, the proponent must be the one who has the obligation to show that the counterconsideration is unrealistic. The reason, again, being that it is he who is convincing the opponent that her critical position is untenable. Argumentation Schemes from Hamblin's Perspective 359 whether the passage between premises and the conclusion is ac- ceptable. 4. Modeling argumentation schemes in dialogue systems In this section, I develop a formal dialectical system, called “Asymmetrical Why-because System for Argumentation schemes”, AWSA. This system resembles Hamblin’s formal system called “Why-Because system with questions” (1970, Chapter 8) at some points, but it also diverges from it in a num- ber of ways. AWSA is asymmetric. Dialectic is characterized by a divi- sion of labor, according to which there is a defending party and a critical party (cf. Rescher on probative asymmetries, 1977, p. 17-18). There is some reason to doubt that Hamblin’s system implements such a division of labor. First, in Hamblin’s dia- logues a party cannot win or lose the dialogue. Second, the rules of Hamblin’s system apply to both parties, who are referred to as “Black” and “White” rather than as, for example, “Opponent” and “Proponent.” Nevertheless, at a local level there is a clear division of labor, in so far as the argumentative exchanges are concerned. In the “Why-Because system with questions,” a re- quest for argumentation “Why S?” does not involve any com- mitment, while, instead arguments are made up from statements which do involve commitments and consequently must be de- fended against criticism. The idea of this paper is that this argu- mentative asymmetry can be upheld, also when presumptions and defeasible reasoning with argumentation schemes are con- cerned. I shall start by developing an “Asymmetrical Why- because System”, AWS, that forms a simplified version of Hamblin’s system in which the options of posing questions and of retracting commitments have been left out. The idea is that the participants can adopt deductive logical schemes, but not defeasible argumentation schemes. AWS has been made explic- itly asymmetrical by having a proponent and an opponent with distinct tasks. The proponent is defending a thesis, throughout the entire dialogue, by attempting to show that the opponent’s critical attitude towards his thesis makes the opponent’s position untenable. The opponent is critically challenging the thesis and the defense offered in favor of it, in an attempt to show that her critical position is tenable after all. What is also different from Hamblin’s system, is that in a dialogue according to the rules of AWS, a party can win the dialogue after having completed its task. Finally, the system is made less permissive by enforcing each move to be relevant to the issue with which the discussion Jan Albert van Laar 360 starts. Relevance is enforced by limiting challenges to be fo- cused only on the proponent’s thesis or on part of his argument, as well as by making all arguments respond to the challenges of the opponent. After having stated AWS, I shall continue by revising and adding rules that are needed to accommodate the use of argu- mentation schemes, instead of deductive schemes. This will be done by: (a) giving the proponent the option to provide an ar- gument that instantiates an adopted argumentation scheme; (b) by giving the opponent the option to challenge the premises, in- cluding the connection premise, of such an argument; and (c) by making the parties responsible for deciding the issue of the ac- ceptability of the connection premise by arguing over the ac- ceptability, relevance, and realism of the opponent’s counter- considerations. 4.1 Asymmetrical Why-Because System (AWS) Locution Rules: The following are permissible locutions, where S, T, … denote atomic or complex sentences of a language that at least contains the logical connective →. L1. Proponent may choose from the following moves: Thesis S; Argument {T, T→S};10 I give up. L2. Opponent may choose from the following moves: Why S?; I give up. Commitment Rules C1. At a preliminary stage it has been decided what proposi- tions, and what deductive logical schemes11 are part of Op- ponent’s commitment store.12 Structural Rules S1. The participants alternate and make one locution at a time. Proponent starts with “Thesis S.” 10 In this system the connection premise is modeled as a material implication, although the expressed message is, as we have seen, more involved. An ar- gument with several regular premises can be modeled by understanding these reasons as making up one conjunctive premise. 11 A similar feature of dialectical systems can be found in Mackenzie who enables the parties to choose “preferred valid argument schemata” (1990, p. 570). 12 Only the commitments of the Opponent are collected in a commitment store. Of course, the Proponent also is committed to the propositions he as- serts. However, the Proponent’s commitments at a certain stage can all be read from his previous dialogue moves. Argumentation Schemes from Hamblin's Perspective 361 S2. A move “Thesis S” is followed by “Why S?” or by “I give up.” S3. “Why S?” is followed by “Argument {T, T→S}” or “I give up.” S4. A move “Argument {T, T→S}” is followed by either “Why T?” or “Why T→S?” or “I give up.” S5. A move “Why S?” is not allowed if S is part of Opponent’s commitment store, or if S instantiates an adopted deductive logical scheme.13 Win-Loss rules W1. If a participant makes the move “I give up”, he loses the discussion and the other participant wins it.” 4.2 Asymmetrical Why-Because System for Argumentation Schemes (AWSA) Structural Rule S5 cannot be revised merely by replacing “de- ductive logical scheme” with “argumentation scheme.” For the parties can accept argumentation schemes as having justificatory force without accepting them as deductive, and allow instances of these schemes to get defeated with newly introduced informa- tion. Therefore, AWSA will be developed in line with the con- siderations of Section 3 by giving the proponent to right to re- quest for a counterconsideration when the opponent challenges the connection premise of an argument that instantiates an adopted argumentation scheme. Given that the emphasis in this paper is on the dialectic of non-deductive argumentation schemes, I refrain from dealing with presumptive commitments other than to the content of connection premises. In order to simplify matters further, attention shall be restricted to explana- tory counterconsiderations, while leaving out the rights and ob- ligations pertaining to validations and counterargumentations. Locution Rules: The following are permissible locutions, where S, T, … denote atomic or complex sentences of a language that at least contains the logical constants → and ¬. L1. Proponent may choose from the following moves: Thesis S; Argument {T, T→S}; Refutation S; Explain (Why S?); Un- realistic!; I give up. L2. Opponent may choose from the following moves: Why S?; Counter S; I give up. 13 Thus, Opponent is not allowed to utter “Why (If (if P then Q and not-Q), then not-P)?” if modus tollens has been accepted as a deductive logical scheme. Jan Albert van Laar 362 Definition: The move “Counter S” and all moves that follow upon it in the dialogue are said to be part of the refutation dialogue. Commitment Rules C1. At a preliminary stage it has been decided what propositions are part of Opponent’s commitment store. C2. At a preliminary stage it has been decided what argumenta- tion schemes are part of Opponent’s commitment store, such that both participants recognize for each connection premise whether or not it instantiates the associated conditional of one of these adopted argumentation schemes.14 C3. If Proponent uses an argument that instantiates an adopted argumentation scheme, then its connection premise becomes part of the Opponent’s commitment store, where it is marked as a presumption. C4. After a move “Why S?”, S is no longer an element of Oppo- nent’s commitments store (if it happened to be there). Structural Rules S1. The participants alternate and make one locution at a time. Proponent starts with “Thesis S.” S2. A move “Thesis S” is followed by “Why S?” or by “I give up.” S3. “Why S?” is followed by “Argument {T, T→S}”, or by “I give up”, or, in case both S is a commitment that is marked as a presumption (and consequently a connection premise of the form V→U), and “Why S?” is not part of a refutation dialogue,15 by “Explain (Why V→U?)”. S4. A move “Argument {T, T→ S}” is followed by either “Why T?” or “Why T→S?”, or by “I give up,” or, if “Argument {T, T→ S}” is part of a refutation dialogue, by “Counter U.” S5. “Explain (Why V→U?)” is followed by “Counter W” or by “I give up.” S6. “Counter W”, when offered in a refutation dialogue that was prompted by a move “Explain (Why V→U?),” is followed by “Refutation ¬W”, or by “Refutation ¬(W→¬(V→U)”, or by Unrealistic!, or by “I give up.” 14 For example, the associated conditional of Scheme 1 reads: “If Person E is an expert in field F and if Person E says that P and if P is a proposition within field F, then P”. 15 The model can be extended by enabling Proponent to request for an ex- planation of a challenge of a presumption within a refutation dialogue, which may lead to the embedding of refutation dialogues within refutation dia- logues. Argumentation Schemes from Hamblin's Perspective 363 S7. A move “Refutation ¬W” (or: “Refutation ¬(W→¬(V→U)”) by Proponent is followed by “I give up” or by “Counter X”, or by “Why ¬W?” (or: “Why ¬(W→¬(V→U)?” ). S8. Proponent’s move “Unrealistic” is followed by “Counter S,” or by “I give up.” S9. A move “Why S?” is not allowed if S is part of Opponent’s commitment store but not marked as a presumption. S10. It is not allowed to make the move “Counter S” if the move “Counter S” has been made before. Win-Loss rules W1. If a participants makes the move “I give up”, he loses the discussion and the other participant wins it. The example from Section 3.6 can now be formalized in the fol- lowing way: Move Commentary 1 Prop. Thesis S S = It is going to rain next week 2 Opp. Why S? 3 Prop. Argument {T, T→S} T = Erwin is a meteorologist, and Erwin says that it is going to rain tonight, which is a statement within meteorology 4 Opp. Why T→S? Assuming that Proponent’s argu- ment instantiated an adopted argu- mentation scheme, Opponent chal- lenges, and thereby retracts, a pre- sumption. 5 Prop. Explain (Why T→S?) 6 Opp. Counter W W = Erwin might be confused, due his old age 7a Prop. Refutation ¬W 7b Opp. Why ¬W? A move left implicit, but antici- pated by Proponent 7c Prop. Argument {X, X→¬W} X = He is still esteemed by his col- leagues 8 Opp. Counter Y Y = Nobody likes rain 9 Prop. Refutation ¬(Y→¬(T→S)) 10 Opp. Counter Z Z = His employer might have put him under pressure 11 Prop. Unrealistic! 12 Opp. I give up Jan Albert van Laar 364 Note that “Counter W” has no direct consequences for the commitment stores of any of the parties. More specifically, the proponent does not have the option to make the opponent defend W. This represents the idea that a counterconsideration merely conveys a possibility to be refuted by the proponent, rather than a statement to be defended by the opponent. Still, the counter- consideration goes much further than posing a mere critical question, such as “Why S?”, by introducing a piece of informa- tion W that, if not refuted by the proponent by getting ¬W or ¬(W → ¬(V→U) accepted (with V→U being the connection premise under attack), would defeat the proponent’s argument. According to these rules, the discussion about the accept- ability or refutatory sufficiency of these counterconsiderations decides the discussion as a whole. Given that also the partici- pants can themselves choose what argumentation schemes to adopt, within some preliminary stage of the dialogue, Hamblin’s Justificatory Force Principle has been satisfied. His Dialectical Meaning Principle can be seen as satisfied in so far as the mean- ing of the notion of argumentation scheme as well as the mean- ing of a particular argumentation scheme, such as a particular kind of argument from authority, is provided by the dialogical procedures specified in the rules. The model AWSA can be en- riched in various directions, as we have seen along the way. The system AWSA differs in an important respect from the Hamblin-style dialogue system ASD that has been proposed by Walton, Reed and Macagno (2008). In ASD, the opponent must provide an argument that the case is exceptional, if she challenges the connection premise on that account (pp. 388- 389). AWSA, however, accommodates defeaters in a way that does not interfere with the dialectical division of labor. 5. Conclusion In this paper, it has been shown how defeasible argumentation schemes, which were not explicitly dealt with by Hamblin, can be implemented in a dialogue system that closely resembles Hamblin’s own system “Why-Because system with questions”, and that, moreover, implements the principles that characterize his dialectical philosophy: the Dialectical Meaning Principle and the Dialectical Evaluation Principle, and more in particular his Justificatory Force Principle. The result can be seen as a way to specify, what van Eemeren and Grootendorst have called, the “intersubjective testing procedure” with which to determine whether an argumentation scheme has been applied correctly (2004, Chapter 6). Argumentation Schemes from Hamblin's Perspective 365 Acknowledgments: I thank Erik Krabbe and an anonymous ref- eree for very helpful comments. References Garssen, Bart (2001). Argument schemes. In Frans H. van Eemeren (Ed.), Crucial Concepts in Argumentation Theory, pp. 81-99. Amsterdam: SicSat. Eemeren, Frans H. van, and Grootendorst, Rob (2004). Argu- mentation, Communication, and Fallacies: A Pragma- Dialectical Perspective. Hillsdale, N.J.: Erlbaum. Eemeren, Frans H. van, and Grootendorst, Rob (2004). A Sys- tematic Theory of Argumentation. Cambridge: Cambridge University Press. Eemeren, Frans H. van & Houtlosser, Peter (2005). Theoretical construction and argumentative reality: An analytical model of critical discussion and conventionalised types of argumen- tative activity. In David Hitchcock (Ed.), The uses of argu- ment: Proceedings of a conference at McMaster University pp. 75-84. Hamilton, ON: Ontario Society for the Study of Argumentation. Hamblin, Charles L. (1970). Fallacies. Newport News, VA: Va- le Press. Hitchcock, David (2010). The generation of argumentation schemes. In Chris Reed & Christopher W. Tindale (Eds.), Dialectics, Dialogue and Argumentation: An Examination of Douglas Walton’s Theories of Reasoning and Argument, pp. 157-166. London: College Publications. Kienpointner, Manfred (1992). Alltagslogik. Struktur und Funk- tion von Argumentationsmustern. Stuttgart: Frommann- Holzboog. Krabbe, Erik C.W. (2001). The problem of retraction in critical discussion. Synthese, 127: 141-159. Krabbe, Erik C. W. (2007). Nothing but objections! In Hans V. Hansen and Robert C. Pinto (Eds.), Reason Reclaimed: Es- says in Honor of J. Anthony Blair and Ralph H. Johnson, pp. 51-63, (references pp. 283-299). Newport News, VA: Vale Press. Krabbe, Erik C.W. & van Laar, Jan Albert (2011). The ways of criticism. Argumentation 25: 199-227. DOI 10.1007/s10503- 011-9209-8. Laar, Jan Albert van. (2012). Logical criticism and argument schemes. In Henrique Jales Ribeiro (Ed.), Inside Arguments: Logic and the Study of Argumentation. Jan Albert van Laar 366 Laar, Jan Albert van and Krabbe, Erik (2012). The burden of criticism: Consequences of taking a critical stance. To be published. Mackenzie, Jim (1990). Four dialogue systems. Studia Logica 49: 567-583. Pollock, John (1995). Cognitive Carpentry: A Blueprint for how to Build a Person. Cambridge, MA: MIT Press. Rescher, Nicholas (1977). Dialectics: A Controversy-Oriented Approach to the Theory of Knowledge. Albany, NY: State University of New York Press. Wagemans, Jean H.M. (2011). The assessment of argumentation from expert opinion. Argumentation 25: 329-339. Walton, D.N. (1998), The New Dialectic: Conversational Con- texts of Argument. Toronto: University of Toronto Press. Walton, Douglas N., Chris Reed and Fabrizio Macagno (2008). Argumentation Schemes. Cambridge: Cambridge University Press. Walton, Douglas N., & Krabbe, Erik C. W. (1995). Commitment in Dialogue: Basic Concepts of Interpersonal Reasoning. Al- bany NY: State University of New York Press.