Argument and Its Uses! J. ANTHONY BLAIR University a/Windsor Abstract: Do not define argument by its use to persuade. for other uses of arguments exist. An argument is a proposition and a reason for it. and argumentation is an interchange involving two or more parties resulting in the assertion of one or more arguments coupled with anticipated or actual critical responses. A logically good argument has grounds adeq uate for the purposes at hand (true, probable, plausible, acceptable to the audience) and the grounds provide adequate support for the conclusion. The norms for good logic in arguments are different from the norms for the good use of arguments. Resume: Ce n'est pas l'usage persuasif d'un argument qui Ie definit, car un argument a d'autres usages. Un argument consiste en une proposition et une raison qui l'appuie. et une argumentation est un echcnge entre au moins deux personnes dans lequel s'exprime au moins un argument accompagne de l'anticipation ou la realisation de repliques critiques. Un argument logiquement bon a des raisons vraies, probables, plausibles ou acceptables pour son auditoire. et suffisantes pour appuyer la conclusion. Les criteres d'une bonne logique d'argument different des criteres du bon usage d'arguments. Keywords: argument. argumentation. persuasion, uses of arguments. informal logic. good argument, dialectic. rhetoric. Toulmin 1. Introduction The theme of the Ontario Society for the Study of Argumentation conference at which this paper was presented, "The Uses of Argument," (McMaster University, 18-21 May 2005) was inspired by the work of Stephen Toulmin (1958; Toulmin, Rieke, Janik, 1979). I hope that this paper reflects at least part of the spirit of Professor Toulmin's work-namely, a healthy irreverence toward received views. As will become evident, I borrow heavily from the substance of Professor Toulmin's work as well. The larger question motivating many of the ideas in this paper, and one that I expect is widely shared, is this: How does it all fit together? By "it all" I mean all the definitions of argument and argumentation, all the theories, all the perspectives, and all the norms that have been accumulating as the field of argumentation has developed since, let us say, 1958, the year The Uses of Argument (Toulmin 1958), and also another influential work in this field, La Nouvelle Rhetorique (Perelman and Olbrechts-Tyteca 1958), were first published. But what I will do here is much © In/ormal Logic Vol. 24, No. 2(2004): pp.137-151. 138 J Anthony Blair more modest, namely, suggest that the concept of argument needs to be slightly rethought. It is necessary to begin with a word about the historical development in philosophy that has come to be called "informal logic." It began partly as a reaction against a tradition of logic instruction. This was the tradition of teaching the skills of argument identification, analysis and assessment on the assumption that these tasks are sufficiently managed with the tools of formal logic. It was also partly a related reaction against the focus on decontextualized arguments and simplified, invented examples that was prominent in so much of the tradition of instruction centred on formal logic. Informal logic emerged in the process of developing a new kind of logic course offered by philosophy departments in the United States and Canada, one in which a key innovation was to examine texts that were chunks of real discourse, identify the arguments to be found in them, set them out perspicuously, and evaluate their logical cogency without relying on the traditional criterion of "soundness." Those active in this movement began to shape new conceptual tools to serve these tasks. But along the way, some unwarranted inferences were drawn. For starters a couple of babies were nearly thrown out with the bathwater. There was an initial perception that the turn towards informal logic implied an opposition to formal logic-hence the moniker: "informal logic. " But quite early on most recognized this mistake. The disagreement with the "formal logic is all you need for logic" camp, implies no disagreement with formal logic itself. (The name "informal logic" is therefore misleading, although it is too late to change it now.) So that baby was retrieved. Next, some time ago now many came to realize that they also disagree with those who think that all you need to understand arguments and argumentation is the logic of arguments-be it formal or informal. First came the realization that in order to understand argumentation, one needs to understand dialectic as well; and belatedly came the realization that it is necessary to understand rhetoric too. But now some who oppose "logic is all you need for argumentation" think that this position implies opposing logic, or at least minimizing its role in argumentation. This implication, just like the earlier one, is on the face of it suspect, and it is advisable to retrieve that baby, too. 2. Argument as persuasion But those are not the only bad inferences that have been made. By focussing almost exclusively on the persuasive function of arguments and on argumentation as a process of rational persuasion, many have tended to conceptualize argument as having an analytic connection with persuasion. There is a lot of evidence that this is how argument is conceived by informal logicians. Consider how many who are prominent in the field define the word Argument and Its Uses 139 'argument' or otherwise characterize what an argument is. Toulmin, Rieke and Janik (1979) say that an argument in one sense is a "chain of reasoning ... the sequence of interlinked claims and reasons that, between them, establish the content and force of the position for which the particular speaker is arguing" (1979, 14). Scriven says that "The function of an argument is to persuade you that since the premise is true, you must also accept the conclusion" (1976, 55-56). Pinto insists that "the word 'argument' ... is appropriately applied to sequences of propositions only when they serve as instruments ofpersuasion"(200 1,36). Freeman is interested in an argument as "a message which attempts to establish a statement as true or worthy of belief on the basis of other statements" (1988, 20). Hitchcock has called an argument "a set of claims, one of which is put forward on the basis of the rest" (1983, 31). Govier says that" An argument is a set of claims a person puts forward to show that some further claim is rationally acceptable" (2001, 3). Woods, Irvine and Walton say that in the broad sense of 'argument,' "an argument is a presentation of reasons or evidence in support of some claim. It is an attempt to build a case in favour of a conclusion" (2004, 2). [My emphasis in all the above quotations.] These definitions reveal how this prestigious group of authors are thinking about arguments. But if one makes reference to claims and messages and presentations, one is talking about assertions, which are communications, and in this case are messages aimed at affecting the beliefs, attitudes or conduct of others. So whether they are explicit about it or not, these accounts all make the concept of argument out to be analytically connected to the function of persuasion. They all conceive an argument as a particular instrument of persuasion. Now, persuasion is a kind of activity. Resolving a difference of opinion is a joint activity. These are things that people do. When someone persuades by arguing, what he or she is doing is using arguments to persuade. But is persuasion the only possible use of arguments? Does using arguments entail trying to persuade? Not at all. People use arguments in all sorts of different ways, and using them to try to influence an interlocutor or audience to accept some proposition, or to try to resolve a difference of opinion by getting the other party to accept one's position- that is, using them to try to persuade-is just one of many uses. Here is a list of seven other uses or types of uses of arguments, and there are probably more. By the way, because the names used below are in many cases the same as those that Walton and Krabbe (1995, 66) gave to their well-known list of dialogue-types, I need to make it clear that this is not a list of different kinds of dialogues. Perhaps they can all be analysed as if they occurred in dialogues, but that is another question--one that [ will not address in this paper. (1) Quasi-persuasion. Perelman and Olbrechts-Tyteca (1958) reminded us that people use arguments to strengthen adherence to an already-held point of view, and presumably, conversely, they can be used to weaken the grip of an already- doubted point of view. Goodwin 2 has noticed that people sometimes use arguments with the intent merely of opening their interlocutor's mind to a 140 J Anthony Blair possibility, of getting a foot in the door. These need not be cases of trying to get someone to abandon a belief, attitude or course of action, or to adopt a completely newone. (2) Inquiry/investigation and deliberation. People use arguments to try to think their way through to a considered opinion on an issue. A person will mull over the arguments on all sides, and try to see what they point to, all things considered. The person wants to determine for himself or herself what position on an issue seems justified. Doing that is different from self-persuasion, since a person don't necessarily begin such an inquiry with a standpoint that he or she is committed to. This is the use of arguments to inquire or to investigate if it is using them to decide what to believe, and it is their use to deliberate ifit is using them to decide what to do. (3) Justification. There is a use of arguments that is a lot like persuasion in that its goal is to gain the adherence of others to a thesis or proposal, but also a lot like inquiry or deliberation in that the arguer is presenting to the others the considerations that he or she finds compelling-that he or she thinks show that the thesis is true (or the most reasonable or most plausible) or show that the proposal is right (the best alternative, etc.). The arguer takes herself or himself to be explaining why the thesis is true or right. Perhaps this use of arguments is really nothing other than persuasion, but there are plenty of cases of persuasion in which the arguer's goal is to get the others to agree, whether or not that involves getting them to appreciate the truth of the matter. So I will list justification as a distinct use. (4) Collaboration. Some have recently urged that people ought to stop using arguments adversarially, and start to use them collaboratively (Gilbert 1997, Esp. Ch. 8; Tannen 1998, 284-290; Tindale 2002, esp. Ch. 4). Instead of identifying what is wrong with a view one disagrees with and trying to refute it, or instead trying to get another to come around to one's own view, one should look for what both parties think is right about one another's standpoints, and try to build on common ground. The use of arguments in collaboration is like inquiry in that its participants are trying to get at the truth of the matter: they are trying to get it right. But it is also like justification in that usually they are also writing up their findings in a way that, they hope, will gain the assent of others or at least be clearly understood by others. Perhaps collaboration should be classified as a species of inquiry, and not as a separate use of arguments, but I list it because it does emphasize the possibility of people using arguments constructively, trying to get at what is true in the other's position, a win-win activity, rather than adversarially, trying to defeat the other and win the argument, a zero-sum activity. (5) Rationale-giving. In some types of situation, someone's decision or judgement is expected to be accompanied by a rationale, no matter how the decision or judgement was arrived at. Administrators,judges and grant application assessment panels are often in the position of being required to accompany their judgements Argument and Its Uses 141 with such rationales. Although such rationales are addressed to audiences, the object is not to persuade the audience so much as it is to show that the judgement can be supported in terms of the criteria on the basis of which it was expected to have been made, or in some cases, legally required to have been made. Call this argument used as rationale-giving. (I don't like to use the term rationalization, since it suggests bad faith and there need be no bad faith in rationale-giving arguments.) (6) Edification/instruction. By discovering the arguments that have convinced someone else of a thesis or theory, along with the objections to that view that have been formulated, it is possible to arrive at a better understanding of the position than could have been acquired from just a statement of it by itself. University and college instructors require their students to study and come to an understanding of such arguments, with a view to deepening their understanding of the theory or thesis under study. People thus use arguments both for their own edification and also for the instruction of others. This use of arguments often entails working backwards from conclusions to premises, not from premises to conclusions. (7) Evaluation. Arguments can similarly be used to assess peoples' understanding and their intelligence. Someone's criticisms of others' positions and arguments reveal how, and thus whether, they have understand what they are criticizing. This is the purpose of assigning argumentative essays and examination answers to students. Arguments are thus used for evaluation. If all of these uses of arguments are different from persuasion, then it is just implausible to think of persuasion as the only or even the paradigm use of arguments. And if at least one of them is different from persuasion, then it is a mistake to define argument in terms of persuasion. 3. A revised concept of argument If argument is not to be identified with the use of arguments in persuasion, then there is a need for a definition and conception that makes it possible to think and speak of arguments as something that can be and are used to do these other things besides trying to persuade. At the heart of things, I suggest, are reasons-reasons for beliefs or for believing, reasons for attitudes or for emotions, or reasons for decisions about what to do. An argument is a reason for some such proposition, using 'proposition' in a broad sense. Arguments are, to borrow Mill's apt phrase, "considerations ... capable of determining the intellect either to give or withhold its assent" (Mill 1961, Ch. I). So ~ propose that we conceive a set of one or more propositions to be an argument (understanding "proposition" in the broad sense) just when all but one of them constitute a reason for the remaining one. And a set of propositions are a reason for an belief, attitude or decision, just when the former support the latter to some degree. What constitutes support is an epistemological question, understanding 142 J Anthony Blair epistemology in a broad way, so as to be the theory of the justification of attitudes and various kinds of normative propositions as well as of beliefs. Argument is to be distinguished from argumentation if argumentation is understood to be an interchange involving two or more parties resulting in the assertion of one or more arguments coupled with anticipated or actual critical responses, and any consequent chain of responses, including the assertion of other arguments. Argumentation in this sense is dialectical, understanding dialectic as an ordered interchange between two or more parties that is motivated by the question whether a proposition asserted by one party should be accepted by the others, and it presupposes that reasons are available as a means of establishing that it should or that it should not be accepted. Dialectic thus presupposes reason- giving as a tool or move, and reason-giving presupposes the possibility ofreasons supporting propositions, namely arguments. To take something to be an argument is to take a consideration to supply some amount of support for a proposition. So the identification of a set of propositions as an argument isa judgement, and individual people make judgements. It follows that whether some set of propositions is an argument is a judgement that someone makes. The Toulmin model (Toulmin 1958; Toulmin, Rieke and Janik 1979), with a slight modification, works well as a model of argument. (Recall that according to this model an argument is to be conceived as data adduced in support for a claim, with the inference from the data to the claim appropriately qualified and allowing for the possibility of rebutting factors, and the inference being supported by a warrant or inference-licence that itself can be backed up if challenged.) It is useful to employ the Toulmin model because its concept of "warrant" makes explicit the inference rule that is functioning in any argument, and being able to refer to the inference rule at work provides a way of distinguishing kinds of logical criteria. But I define "argument" in terms of propositions or sentences rather than claims because claims are tokens of a type of speech act, namely, the act of assertion, or putting forward a proposition or sentence as true, which is a kind of communication with others that carries with it the obligation to defend the proposition or sentence claimed if challenged, and so connects argument analytically with persuasion. (Because nothing about the concept rides on the resolution of the dispute over the ontological status of propositions, for simplicity of exposition, I will use the term "proposition" alone hereafter. Those who object to "proposition" talk may substitute 'sentence' or 'possible sentence' for 'proposition' in what follows.) (J would like to add two parenthetical remarks about the Toulmin model. First, criticisms of the Toulmin model for being insufficiently dialectical or insufficiently rhetorical mistake it for it something it is not. It is a model of argument, not a model of a use of argument or a model of argumentation. Second, calling it the "Toulmin" model should not be taken to imply that it is something Toulmin dreamed up, some confection ofToulmin's that has caught on, like a fad, in some quarters, Argument and Its Uses 143 and so might as readily be dropped. Toulmin deserves credit for bringing it to our attention once again, but it is nothing else than the epichairema, described in Cicero's Rhetorica ad Herennium and De Inventione. It has been around a long time (see van Eemeren, Grootendorst and Snoeck Henkemans 1996, 47-49).) Other things being equal, reasons can be judged to make it necessary, or more likely, or more plausible than otherwise that the proposition they support is true or worthy of acceptance. In some kinds of arguments, the warrant makes rejecting the conclusion while granting the grounds inconsistent in some way (see Scriven 1976, 30-32). That is, when a set of propositions support another one in such cases, there is some kind of inconsistency in accepting the former and rejecting the latter. In other kinds of arguments, the warrant in effect makes a prediction thatthe conclusion will be borne out, given the grounds. In such cases, the warrants' backing is supported by the success rate of such predictions. In other kinds of arguments, the warrant in effect postulates the conclusion as the best explanation of the grounds. In yet other kinds of cases, the warrant conveys an entitlement to shift the burden of proof to anyone who would disagree with the conclusion, given the grounds. And so on: all of this and more needs to be worked out, to be sure. The essential idea is that an argument, or more precisely, a unit of argument, is a compound proposition consisting of a proposition together with a consideration that supports it, other things being equal. The supporting consideration can include more than one proposition, so it is not a premise, but a group of premises. The consideration tends to show that the proposition is true, or reasonable, or probable or plausible, other things being equal. We often harmlessly speak as if the consideration itself is the argument, but this is always to be understood as short for "the consideration that does the supporting in the argument." Whether what counts as support is always relative to persons and situations is an open question so far as the concept of argument goes. The ceteris paribus rider is necessary, because there can in many cases be arguments for and arguments against a proposition--or, more precisely, arguments with the affirmation of a proposition as their conclusion and arguments with the denial of that same proposition as their conclusion. Someone will notice that by this definition there cannot be an argument with no support. It is important to distinguish between an argument and some person 's (or persons') argument. Someone's argument consists of a proposition and the consideration which that person takes to support it or offers as support for it (taking and offering are importantly different). A person's argument can contain premises that are irrelevant as support for their conclusion. Indeed, all the "reasons" a person takes or offers as support for a proposition can be completely irrelevant to it. We still correctly speak of that person's "argument." But if the offered support is completely irrelevant to the conclusion, then that person s argument-what he or she the "proposed" or "offered" or "understood" to be an argument-is in fact 144 J Anthony Blair not an argument. We are familiar with this ambiguity between attempt and success in other concepts. What someone offers as an excuse is "their excuse," but it can be judged to be no excuse at all; what someone presents as an explanation of some phenomenon or event is "their explanation," but upon examination it can be judged not to explain the phenomenon or event at all, so not to be an explanation. Thus it is quite appropriate to refer to someone S argument as their argument even when what the person adduces as support for a proposition does not support it at all. Such a person presumably thinks he has an argument, but he is mistaken. Although someone's argument with completely irrelevant "support" is not really an argument, someone's argument with extremely weak support is an argument, albeit an extremely weak one. Arguments come in degrees of strength, from extremely weak to extremely strong, from slightly suggestive support to completely compelling support. Thus I am suggesting that 'argument' is a normative concept. 4. Deflection of a possible criticism/misunderstanding One objection to the proposal to conceptualize argument independently of its uses runs as follows. Is doing so not falling back into the kind of abstraction from context that it was part of informal logic's founding spirit to avoid: does it not imply that arguments are in some way context-independent and can be understood and appreciated apart from the situations of their use? The answer is that the epistemic relation between supporting consideration and supported proposition is as context-independent as any epistemic relationship, but there is no implication that arguments in their uses can be recognized, reconstructed, or evaluated independently of the contexts of their uses. Discourse can be identified as argumentation or as containing arguments only in the I ight of a given particular interpretation of it. Arguments are embodiments of meaning, and meaning is generated by participants' understanding ofthe situation. The particular meanings of sentences, or how they are understood, are thus not accessible aside from their contexts, that is, the particular situations of their use. One's understanding of the meaning of a sentence will depend on what one understands the purpose of the communication to be, on what one takes the issue to be and on what one takes the communicative role of the utterance to be. So identifying the particular argument that someone has presented or that someone is thinking about relies on an understanding of the situation. This is part of the rhetorician's point. As well, identifying the ~rgument depends on an understanding ofthe role ofthe use of the sentences. This is part of the dialectician's point. But once there is a particular understanding of the discourse that makes it out to be argumentation, and there is a particular understanding of the argument in question, it can be then asked, from whatever perspective one occupies, whether the reasoning of that particular argument as it stands, so understood, and at that moment, is any good-that is, in the assessor's judgement to what extent do the considerations adduced support the proposition in question, or to what extent should they be taken to support it? Argument and Its Uses 145 5. Logical norms of arguments One reason not to lose sight of arguments as distinct from their uses is that it is important to keep in the forefront of theorists' attention the fact that there is not yet any settled theory of the logic of arguments. There is no normative logic for arguments that everyone agrees is right. I do not have a theory to propose, but I will sketch one possible approach that I think is promising. It reflects the views of many other people-Toulmin (1958), Perelman (1958, 1982), Hitchcock (2003), Walton (1996), Pinto (2001), van Eemeren & Grootendorst (1984, 2004), Johnson (2000), and no doubt many others-and there is really nothing new about it. 5.1 A suggested framework The question of whether an argument is logically any good is the question whether someone has any business drawing that conclusion from those premises, or accepting those considerations and taking them as warranting his or her assent to that proposition. In Toulmin's framework, to repeat, an argument consists of considerations from which the inference to a qualified proposition is licensed, other things being equal. The question can therefore be put this way: when is the logic of an argument, so understood, any good? When is one entitled to infer the conclusion of an argument? There are two parts to the answer. First, the grounds must be adequate for the purposes at hand. They must be true, or probably true, or plausible, or acceptable to the audience-whichever of these is required by the nature of the qualification attached to the conclusion. And, second, the support the grounds provide for the conclusion must be adequate to the strength of the confidence claimed for the conclusion. In assessing the adequacy of the support in any particular argument, one can ask, first, whether the grounds entail the proposition in question-whether the conclusion follows deductively from the grounds. If the answer is yes, well and good. What makes an argument deductively valid is that in the circumstances it has a defensible warrant that is a rule of inference with no qualifications: the warrant has the form, "given grounds of this sort, a proposition of that sort cannot possibly be false." If the argument is not deductively valid, then I don't think we should refashion it to make it deductively valid, unless there are unambiguous textual indicators that the arguer intended an entailment. Otherwise, doing so would produce a different argument. That is, although this is not the place to do it, I would argue against methodological deductivism-the doctrine that it is useful always to reconstruct arguments as if they were (intended to be) deductively valid. Instead, one should ask, second, whether the argument might be inductively strong. Does it have in the circumstances a warrant of the form, "given grounds 146 J Anthony Blair of this sort, a proposition of that sort is probably true" or something similar, perhaps with the probability quantified. If the argument is not deductively valid and its conclusion is not warranted by a probabilistic rule of inference, one can ask, third, whether in the circumstances the grounds support the proposition at issue on the basis of some other kind of rule of inference. Here is where the path forward is not so clear. One promising way to understand Walton's work on presumptive reasoning is to regard it as developing a conceptualization of a third kind of rule of inference: the presumptive warrant. And this is where argument schemes come into the picture. The various argument schemes-argument from analogy, argument from authority, argument from consequences, and so on-are to be understood as presumptive warrants. Attached to each particular kind of presumptive warrant is a set of types of critical questions. Tokens of those questions are to be asked about any particular argument that instantiates that scheme and they must be answered satisfactorily in order to justify the verdict that the argument in question does indeed have the presumptive force claimed for it. So, part A, check for premise adequacy. Part B, check for support adequacy. To do the latter, step one, check for deductive validity. If invalid, step two, check for inductive strength. If there is no probabilistic warrant, then, step three, see if the argument instantiates a presumptive argument scheme, and if so, run through the critical questions. It is not clear that such a procedure covers everything. What about evaluative arguments, for example? These are arguments that have a format something like this: "X is good, or a good of its kind, because the appropriate criteria for assessing Xs are A, Band C, and X satisfies A, Band C to a high degree." Is that one more argument scheme, and therefore covered by the above sketch, or is it a fourth kind of reasoning or argument? And are there, or can there be, presumptive arguments that do not have a scheme that has been described and named? Also, how do arguments from the best explanation fit into this picture? Are they a kind of inductive argument or a separate kind? There is clearly much work to be done here. 5.2 The Relevance, Acceptability, Sufficiency criteria Someone might infer that I have abandoned the criteria of argument cogency that Johnson and I introduced in 1977 in Logical Self-Defense: acceptability, relevance and sufficiency (Johnson and Blair 1977). A word about each of these is in order. (Hans Vilh. Hansen has pointed out to me that 'Perelman introduced a similar distinction in The Realm of Rhetoric. 3 Perelman's criteria seem descriptive; whereas Johnson's and mine are normative.) According to the concept of argument presented above, the judgement about whether grounds are relevant is one the arguer or the reasoner makes in deciding what belongs in the argument, one the interpreter makes in deciding what to attribute to the argument, and one the assessor makes in deciding whether they are both Argument and Its Uses 147 right. A consideration that is irrelevant to the proposition being argued for just does not belong to the argument, so an argument cannot have irrelevant premises, though of course someone's argument-what is presented as or taken to be an argument-assuredly can. So relevance is not a criterion of a logically good argument, but of argument itself. Acceptability is the generic name for the adequacy of an argument's grounds, and which criterion of adequacy is appropriate in any particular assessment will depend on the type of argument and the circumstances in which the person is appraising the argument. In some circumstances, it is important that the grounds offered in support of a proposition be true and known to be true before they are considered acceptable. But in other circumstances, it is quite legitimate to settle for what it is reasonable to believe. And in yet other circumstances, if the interlocutor accepts the grounds offered, that is all that is wanted. The criterion of sufficiency of the grounds as support for a proposition is more complicated. Distinguish sufficiency as a logical criterion from sufficiency as a mixed logical and dialectical criterion. An argument unit is logically sufficient if the strength of its support matches the qualification attached to the conclusion, other things being equal. But in many situations what is of interest is not just argument units, but a complex of many argument units that makes up a case for a proposition-the arguments for and against it, and for and against each other. So a sufficiency judgement about a case is a mixed judgement about both the logical adequacy ofthe argument units making up the case and the dialectical adequacy of the case itself. So the relevance, acceptability and sufficiency criteria are not precisely mistaken, but they require adjustment. 6. Logic and evaluation: Logic is only one of several perspectives I am arguing for distinguishing arguments from their uses and from argumentation, and for retaining a focus on the logic of arguments. The point is not that we can or should assess arguments outside of their situations of use, including their use in argumentation. The point, rather, is that we should not conflate the criteria for good logic in arguments with the criteria for the good use of arguments. These are not the same. The norms for good logic in arguments are different from the norms for the good use of arguments. As the strawman fall act illustrates, an argument can be logically sound but its use might be dialectical malfeasance in the sense that it changes the subject, violating a legitimate assumption of the discourse that the response is supposed to be relevant to the position being debated. Moreover, the use ofa straw man argument will in some cases be unethical as well as dialectically inappropriate, inviting unfair and even harmful reactions against the party to whom the view is falsely attributed. But using a straw man argument might be an effective debating tactic, and so deserve praise on that account. And it is even imaginable that a strawman attack focuses the public's attention on what the issue ought to 148 J. Anthony Blair be, and in that case is a politically praiseworthy move. So to suggest, as too many textbooks still do, that all we need to focus on is the logic of arguments is not just mistaken, but also an egregious oversimplification. What is the relation among the criteria of good logic, good dialectic and good rhetoric? It is misleading to present logic, dialectic and rhetoric as three parallel or three competing perspectives on argument or argumentation. Standard deductive logic, for instance, is not even about arguments. Deductive logic is the study of systems of patterns of necessary relations among propositions (or sentences). Some arguments exhibit such deductive patterns in their premise-conclusion relationship; many do not. Moreover, there are dialectical models of deductive logics, as Lorenzen (1982), Lorenz (1982), Krabbe (1982) and others have shown, so a dialectics/logic dichotomy seems out of place. In addition, it is difficult to imagine a use of argument that is not dialectical, and the question whether a particular argument is logically good in the sense that it asks whether one is justified in drawing a given conclusion from a given ground often cannot be answered without considering whether there are objections to the argument that have been adequately refuted, which seems to be a question about satisfying dialectical norms. Turning to rhetoric, for Aristotie (1984) it was the art of the proper modes of persuasion (Rhetoric 1354b21), specifically oratory, using arguments; but currently rhetoric is taken to have any form of symbolic representation as its subject matter, not just arguments. Argumentation theorists, taking themselves to be spelling out the implications of a rhetorical perspective, have made a valuable contribution by emphasizing the overlooked importance of paying attention to the situatedness of arguments-including reference to such elements as audience, occasion, venue and objective-when it comes to their interpretation and evaluation. But whether logical or dialectical norms are constructs of a collaboration between audience and arguer and hence in some sense basically rhetorical, as some seem to have suggested (e.g., Tindale 2004), is an epistemological question, and it is one I cannot take up here. So while it has been extremely valuable for Wenzel (1980) to have drawn attention to the importance of logic, dialectic and rhetoric for the study of argumentation, we do him no honour by sticking to the neat parallels that his formulation suggested. Moreover, there might be other perspectives on the uses of arguments and argumentation in addition to those of logic, dialectic and rhetoric. So even if logical criteria were worked out, that would not provide a complete normative theory for the evaluation of arguments or of argumentation. Nevertheless, losing sight of arguments as distinct from their uses risks neglecting the task of working out a satisfactory normative theory of their logic. > Argument and Its Uses 149 7. Concluding summary What I have been trying to do in this paper is sound an alarm about the direction that informal logic theorizing seems to be taking. It is right to attend to arguments on the hoof when considering their logic, and it is right to consider more than their logic when analysing and evaluating them. These legitimate moves seem to have led to a focus on the persuasive use of arguments to such a degree that many of now define argument as a tool of persuasion. But there are plenty of other uses of arguments and it is possible, and indeed desirable, to define 'argument' without reference to any particular use. It is important to focus on arguments so defined because we have not yet finished the job of providing a complete account of their logical norms. I sketched one way of framing their norms within the Toulmin model that assimilates much ofthe recent work of various theorists, and I indicated where I now stand on the relevance, acceptability, and sufficiency criteria that Johnson and I introduced many years ago. All that said, it is important nonetheless to insist that assessing the logic of an argument is not all there is to evaluating arguments. If this paper achieves nothing else, I hope it raises the issues of the need to rethink our conceptualization of argument, and of the unfinished task of working out a theory of the criteria for logically good arguments. Notes I This paper is a revision of a keynote address at the conference of the Ontario Society for the Study of Argumentation, "The Uses of Argument," at McMaster University, Hamilton, Ontario, Canada, 18-21 May 2005. Some of its revisions are due to conversations with Harvey Siegel and David Hitchcock, for which I am grateful. The keynote address was a much-revised version ofa paper presented at the Dry-Run series in the Department of Philosophy at the University of Windsor in March 2005. I wish to thank Patrick Bondy, Hans Vilh. Hansen, Ralph H. Johnson, Jeff Noonan, and Robert C. Pinto for insightful and constructive critical comments at that time. 1 Reported to the author by H.Y. Hansen as a point that has been made by Jean Goodwin. ) See Chapter 12, "The Fullness of Argumentation and the Strength of Arguments": "The strength of an argument depends upon the adherence ofthe listeners to the premises of the argumentation [acceptability]; upon the pertinence of the premises [relevance]; upon the close or distant relationship which they may have with the defended thesis; upon the objections that can be opposed to it; and upon the manner in which they can be refuted [sufficiency]" (p. 140). 4 By the "strawman fallacy" I mean an illegitimate argument against a distortion (that is, a misrepresentation) of the opponent's position as ifit were the opponent's position, in an adversary context. 150 J. Anthony Blair References Aristotle. (1984). Rhetoric. In l Barnes (Ed.), The Complete Works of Aristotle, Vol. 2. Princeton: Princeton University Press Barth, E.M. and J.L. Martens (Eds.). (1982). Argumentation: Approaches to Theory Formation. Amsterdam: John Benjamins B.Y. Eemeren, F.H. van and R. Grootendorst. (1984). Speech Acts in Argumentative Discllssions: A Theoretical Model for the Analysis of Discllssions Directed towards Solving Conflicts of Opinion. BerIin/Dordrecht: De Gruyter/Foris. Eemeren, F.H. van and R. Grootendorst. (2004). A Systematic Theory of Argumentation.' The Pragma Dialectical Approach. Cambridge: Cambridge University Press. Eemeren, F.H. van, R. Grootendorst, F. Snoeck Henkemans (Eds.). (1996). Fundamentals of Argumentation Theory. Mahwah, NJ: Lawrence ErIbaum Associates. Freeman, J.B. (1988). Thinking Logically: Basic Concepts for Reasoning. Englewood Cliffs, NJ: Prentice Hall. Gilbert, M.A. (1997). Coalescent Argumentation. Mahwah, NJ: Lawrence ErIbaum Associates. Govier, T. (200 I ). A Practical Study of Argument. 5th edition. Belmont, CA: Wadsworth I Thomson Learning. Hitchcock, D. (1983). Critical Thinking: A Guide to Evaluating Information. Toronto: Methuen. Hitchcock, D. (2003). Toulmin's warrants. In F.H. van Eemeren, J.A Blair, C.A. Willard and A.F. Snoeck-Henkemans (Eds.), Proceedings of the 'Fifth Conference of the International Society for the Study of Argumentation, pp. 485-490. Amsterdam: SicSat Johnson, R.H. (2000). Manifest Rationality: A Pragmatic Theory of Argument. Mahwah, NJ: Lawrence Erlbaum Associates. Johnson, R.H. and lA. Blair. (1977). Logical Self-Defense. Toronto: McGraw-HilI Ryerson. (3,d edition, 1993; I st U.S. edition, 1994, New York: McGraw Hill). Krabbe, E.C.W. (1982). Theory of argumentation and the dialectical garb offormal logic. In Barth and Martins, (Eds.), (1982), above, pp.123-132. Lorenz, K. (1982). On the criteria for the choice of rules of dialectical logic. In Barth and Martins, (Eds.), (1982), above, pp. 145-157. Lorenzen, P. (1982). Die dialogische BegrUndung van Logik-KalkUlen. In Barth and Martins, (Eds.), (1982), above, pp. 23-54. Mill, J .S. (186 I). Utilitarianism. Any edition. Perelman, Ch. (1982.). The Realm of Rhetoric. Translation by William Kluback of L 'Empire rherorique: rhetorique et argumentation (1977). Notre Dame, IN: University of Notre Dame Press. Argument and Its Uses 151 Perelman, Ch. and L. Olbrechts-Tyteca. (1958). La Nouvelle Rhetorique. Paris: Presses Universitaires de France. Pinto, R.C. (200 I). Argument, Inference and Dialectic: Collected Papers on Informal Logic. Dordrecht: Kluwer Academic Publishers. Scriven, M. (1976). Reasoning. New York: McGraw-Hill book Company. Tannen, D. (1998). The Argument Culture: Moving From Debate to Dialogue. New York: Random House. Tindale, C.W. (2004). Rhetorical Argumentation. Thousand Oaks, CA: Sage Publications. Toulmin, S.E. (1958). The Uses of Argument. Cambridge: Cambridge University Press. Toulmin, S., R. Rieke and A. Janik. (1979). An Introduction to Reasoning. New York: Macmillan Publishing Co., Inc. Walton, D.N. (1996). Argumentation Schemes for Presumptive Reasoning. Mahwah, NJ: Lawrence Erlbaum Associates. Walton, D.N. and E.C.W. Krabbe. (1995). Commitment in Dialogue: Basic Concepts of Interpersonal Reasoning. Albany: State University of New York Press. Wenzel, 1. W. (1980). Perspectives on argument. In 1. Rhodes & S.E. Newell (Eds), Dimensions of argument: Proceedings of the Summer Conference on Argumentation, pp. 112- 133. Annnandale, VA: Speech Communication Association. Woods, 1., A. Irvine and D. Walton. (2004). Argument: Critical Thinking. Logic and the Fallacies. 2nd edition. Toronto: Pearson/Prentice Hall. J. Anthony Blair Department of Philosophy University of Windsor Windsor, ON, N9B 3P4 tblair@ulVindsor.ca