The Principle of Vulnerability RALPH H. JOHNSON University of Windsor Key words: vulnerability, criticism, dialectical, manifest rationality, conclusive argument. Abstract: This paper seeks to articulate and defend the principle that every argument is susceptible to criticism and hence the arguer must not seek to immunize the argument from criticism. Considerations both in support of, and opposed to, the principle are considered. 1. Introduction In my view, an argument is best understood as the product of the practice of argumentation. To better understand what an argument is, we need to understand the practice within which it arises which is characterized by the following three properties. I First of all, like any practice, argumentation is teleological. It serves a number of purposes, but primary among them is rational persuasion. In the typical argumentative situation, the Arguer aims to persuade the other(s) of the truth of some thesis (the conclusion) and to do so by rational (as opposed to other, say, psychological) means. The practice of argumentation is thus tightly connected with rationality; and hence the theory of argumentation and the theory of rationality have a common interest. 2 Second, the practice of argumentation is dialectical. By that I mean that it typically and paradigmatically takes place between two (or more) minds. However, there is more to the notion of dialectical than just otherness. In this context, it means as well that the Arguer has agreed to let the feedback affect the argument. As a result of this criticism, the argument may well have to undergo some changes or modifications. (This agreement, this contract, is crucial in the justification of The Principle of Vulnerability, as we shall see later.) Third, the practice is characterized by a trait I call manifest rationality. In the practice of argumentation, rationality is not merely the inner reality but also the outward appearance of the practice. The practice must not just be rational; it must also appear rational. 3 This is why the Arguer is expected to respond to objections and criticisms from others, and not ignore them or sweep them under the carpet. It's not just that sweeping them aside would not be rational and hence not be in keeping with the spirit of the practice. It:S- that it would be such an obvious violation of it-and it would be seen to be such. An argument is a product that typically emerges from this practice and may be characterized as discourse aimed at rational persuasion. Its structure will include what I call, following Blair, an illative core which consists of the conclusion and the premises (reasons, evidence) that support it. 4 However, an illative core is not sufficient for an argument. An argument will also need what I call a dialectical tier in which the Arguer seeks to discharge hislher dialectical obligations. For the Arguer is obligated to deal with the objections and criticisms Informal Logic Vol. 17, No.2 (Spring 1995): 259-269 © Informal Logic 1995 260 Ralph H Johnson that have formed around the issue and the author's position. Why does this obligation exist? We have seen that the practice of argumentation presupposes a background of controversy. The first tier (the illative core) is meant to initiate the process of converting the Other(s), winning them over to the Arguer's position. But they will not easily be won over. nor should they be, if they are rational and wish to emphasize that rationality. For participants know that there will likely be objections to the Arguer's premises. Indeed, the Arguer must know this herself, and so it is not unusual that she attempts to defuse such objections within the course of her argument. If she does not deal with the objections and criticisms, then to that degree her argument is not going to satisfy the demands of rationality. Moreover, to that very degree her "argument" falls short of what is required in terms of structure--never mind the content; i.e., the adequacy of her response to those objections. Let me switch focus and discuss two types of response the Other may make to the argument. The Other may be content to make a judgement about the argument's overall value: viz., "What a lousy argument!" or: "Are you kidding?" or "That's a very interesting argument." This evaluation made by the Other may or may not be communicated to the Arguer. Nothing important is lost if it is not so communicated, since the primary function of the evaluation is to help the other to decide whether or not to be persuaded. A different type of response is criticism, the point of which is not to devastate the argument (much less the Arguer) but rather to give the Arguer guidance in rethinking the argument. The Arguer stands to profit from criticism, especially when the criticism is strong criticism by which I mean it is informed, reasoned, discriminating and challenging. 5 The requirement of manifest rationality explains why the Arguer must respond even to criticisms he believes (or knows) are misguided. If the Arguer were obligated only by the dictates of rationality (rather than those of manifest rationality), then one might ignore the criticism, as the rhetorician can. Having set forth my understanding of the practice of argumentation and roughly how I differentiate between evaluation and criticism,6 I tum now to a discussion of a fundamental principle of criticism. 2. The Principle of Vulnerability 2.1. Statement of the Principle The Principle of Vulnerability says that the Arguer may not represent his argument as immune to legitimate criticism. Observance of this principle is fundamental to the practice of argumentation, for it opens the door to criticism, while violation of it closes the door. This principle is to the realm of argumentation what The Principle of Falsifiability is to realm of scientific theory-a way of separating the genuine from the bogus. To be a legitimate scientific theory, according to Popper, a theory The Principle of Vulnerability 261 must be falsifiable. Likewise, to be a legitimate argument, it must be vulnerable to legitimate criticism. Although some such principle is implicit in many accounts of argumentation, it has rarely been made explicit. 7 2.2. Violations of the principle There are two different ways in which an Arguer might violate the principle. The first and most obvious form of insulation occurs when the Arguer seeks to undermine (or dismiss) in advance every possible criticism. George Will attributes a move something like this to Catherine MacKinnon: MacKinnon reasons serenely, as fanatics do, within a closed circle of logic: If you do not see our wicked society as she does, that just proves how wickedly society has "constructed" your faIse consciousness. Thus all critics are dismissable. 8 I do not here take a position on whether MacKinnon is gUilty as charged, but rather merely use this situation as a possible illustration of one way of violating the principle. Another example is furnished by an anecdote about a student who sought to challenge the viewpoint being presented by the professor in a University class. The professor told the student: "This is not subject to argument or disagreement; it has been conclusively established by consensus!" I take this response to be equivalent to the assertion that the argument under consideration was immune to criticism and so could not be criticized. 9 A second way the principle can be violated occurs when the Arguer refuses to accept any criticism as valid. The Arguer puts forth the argument; someone criticizes it; the Arguer rejects the criticism as not valid. Someone else puts forth another criticism; the Arguer rejects that criticism as not valid. A third person puts forth a criticism; the Arguer rejects it as well. A quotient of such rejection will naturally accompany the conviction of the Arguer who has done his or her homework, has anticipated objections and has been able, so he thinks, to defuse them. However, when this pattern continues, it begins to appear that the Arguer is not prepared to countenance any criticism as legitimate. We may then suspect that the Arguer has insulated his position against all possible criticism. Such insulation violates the principle and indicates that this "Arguer" is not a good- faith participant in the practice. In the case of empirical theories, the proponent of the theory must be prepared to indicate what would cause abandonment of the theory. Again there is a parallel in the case of arguments. It is perfectly legitimate to ask the Arguer: "What would or could upset your argument?"IO The Arguer may not know the answer to this question in some cases. All the same, the Arguer must be prepared to concede at least the possibility of such a challenge-else The Principle of Vulnerability has been violated. 262 Ralph H Johnson Before I leave this section and move to the question of how this principle is justified, it is important to distinguish violation of this principle from the legitimate and necessary step in which the Arguer seeks to discharge his dialectical duties by taking on and defusing objections and criticisms. Is not the Arguer here aiming at making the argument invulnerable? Is there anything wrong with that? Does that violate vulnerability? The difference between the two situations seems clear in principle yet hard to articulate clearly. In the former case, the Arguer attempts to determine in advance what is legitimate criticism and to disarm it. In the latter case, the Arguer acknowledges the legitimacy of certain types of criticism and seeks to show how his position handles them. But I am not confident that this characterization cAptures the difference. Having formulated the principle and discussed ways of violating it, I tum next to the task of justifying it. 3. Justification of the principle The basic line of justification stems from the very nature of argumentation, which I will supplement with some empirical considerations. 3.1. The pragmatic justification By its very nature, an argument is open to, indeed in some sense, expects criticism. The attempt to subvert all criticism subverts the practice. Therein lies the key to its justification, for I will argue that to violate vulnerability, to represent an argument as having been successfully insulated against all legitimate criticism, subverts the practice. Such an attempt does so because (1) it is not rational; (2) it violates the dialectical nature of argument; and (3) it violates manifest rationality. Take the latter point first-that such insulation violates manifest rationality. The Arguer who will brook no criticism or challenge appears as a fanatic or a zealot, and will certainly not appear to be rational and hence will not appear as a participant in the practice characterized by the requirement of manifest rationality. That the insulating move violates the requirement that the practice be dialectical is obvious, for in dismissing all possible criticism, the Arguer has turned hislher back on the Other. That the insulating move thwarts the telos of rational persuasion is also clear. To persuade the other rationally, the Arguer must be prepared to take the criticisms from the Other and show how hislher argument can meet those criticisms-not just dismiss them in advance as being ill-founded. The basic justification, then, is found within the nature of the practice of argumentation. I believe this line of justification is supported by empirical considerations. I want to tum to those now. The Principle of Vulnerability 263 3.2. The empirical dimension of justification Justification comes not only from the nature of the practice but also from reflection on its history. Arguments are human products; hence, liable one and all to be defective in some respect. We all know this. As Arguers, we are liable to certain recurrent flaws, some would call them "fallacies," such as failing to consider relevant evidence, or distorting the position we wish to criticize, or engaging in irrelevancies. In Logical Self-Defense, the authors write: Rarely is an argument so good that it cannot profit from criticism and seldom is an argument so bad that it cannot be improved by criticism. II (In effect, the authors have given a descriptive version of the Principle of Vulnerability. ) But, it might be objected, while it is generally true that arguments will have defects that criticism will reveal, there is nothing in principle that would rule out the possibility of a flawless argument which then would not be vulnerable to legitimate criticism. This happens in other areas of human endeavour. Occasionally, we hear talk of a flawless performance by a pianist. Or, someone may point to the 10's Nadia Comanici's racked up in the 1976 Olympics. Why is such achievement not possible in the realm of argumentation? What is different about argumentation? Here we can with profit consider the history of the practice of argumentation. Though some Arguers have sought to immunize their arguments from criticism, none has ever successfully done so. That is, to my knowledge there has never been a serious argument put forth by a serious Arguer which has not been thought criticizable and which has not been subjected to legitimate criticism. Not Plato, not Kant, not Russell, not Marx-no philosopher has been in this position with respect to any of his arguments. Not one of the greatest minds has been able to put forward an argument which has not, it seems, been subjected to legitimate criticism. If these great dialectical talents engaged in a discipline-philo sophy-that esp'ecially emphasizes the practice of argumentation have not been able to bring it off, then it seems highly likely that no one can. The empirical aspect of justification thus revolves around what I take to be a salient fact: there are not and have not been any conclusive arguments. If there were conclusive arguments, then there would be arguments which could rightly be said to be invulnerable; and vice-versa; if there were arguments that were invulnerable to criticism, they would be conclusive. '2 Before I develop my argument for this view, I need to pause to make four points. First, I may be wrong about this. I am open to counter-examples. My position is vulnerable. All anyone needs to do is point out an argument that has been conclusive. Second, suppose I am wrong about this salient fact, still I believe most would agree that such arguments are very rare. Even though very few arguments 264 Ralph H Johnson are conclusive, there are nevertheless many good arguments. Thus, it is not necessary that an argument be conclusive in order for it to be a good argument. lJ Third, I need to distinguish the claim I am making here from another to which it might be assimilated: i.e., that there are no conclusive statements. 14 I do not hold that view and certainly do not wish to be construed as arguing for it here. I believe that there are conclusive statements. Some examples would be statements that record perceptions, sense datum type statements, and analytic truths. (I suppose there is some tension between this view and fallibilism.) I also believe that some mundane statements are conclusive: i.e., the Green Bay Packers won the first Super Bowl, beating Kansas City 35-10 in 1967. Fourth, there is an objection, due to Jonathan Adler, that would undercut this empirical line of reflection. That objection goes like this: Even if an argument were conclusive, that would not mean that it would not be criticizable. For even if I have produced a conclusive argument (and so may be said to know that the conclusion is true, given the premises), still it does not follow that I know that I know; and this is what the criticism of the argument would show. Hence even conclusive arguments are criticizable. It would follow that my approach here is a red herring, because the question of whether the argument is conclusive is irrelevant to whether it is subject to legitimate criticism. The moral I wish to draw from this objection is that it allows us to see that the Arguer cannot be justified in claiming that the argument is conclusive until it has been tested against actual criticisms. Even if (per impossibile) an argument were conclusive, the Arguer would not be justified in claiming that until that argument has been tested. Hence even a conclusive argument could not present itself as invulnerable to criticism. On what grounds, then, do I claim that there are no conclusive arguments? Here is how I will argue the case. I begin by clarifying what is meant by a conclusive argument and what is required for an argument to be conclusive. Next, I discharge my burden of proof by arguing that there is good reason to think that no argument has ever satisfied these conditions. Third, I discuss some alleged counterexamples to my claim. 3.2.1. Requirements for a conclusive argument To begin the discussion of what a conclusive argument would look like, I take the following to be a nominal definition: DF A conclusive argument is one that conclusively establishes its conclusion on the basis of its premises. A conclusive argument is one that can successfully (and rationally) resist every attempt at legitimate criticism. (Overlook for the moment the fact that arguments do not and can not resist criticism; it is Arguers that so resist.) If there were such arguments, then there would be arguments that were invulnerable to criticism. If there are not such, or if there is good reason to believe that there are not such, then there is good reason The Principle o/Vulnerability 265 to believe that any argument will be criticizable, and hence attempt to immunize it against such criticism is misguided. Let's go further along the path of discussing what a conclusive argument would look like. Much will depend on how we define an argument. For that in turn has a bearing on how we understand the criteria for argument evaluation and criticism. I have already spelled out my own conception of argument in section 1 and that conception will be operative in what follows. Here I need to pause to consider yet another objection which goes this way: "There are conclusive arguments; proofs are conclusive and they are arguments." I do not find this argument compelling, largely because I disagree with the premise which says that proofs are arguments, but I cannot argue that case here. IS To be conclusive an argument would have to display four properties, three internal and one external property. I shall discuss each in turn. i. Internal properties. In order for an argument to be conclusive, the following would be necessary conditions: (C 1) Its premises would have to be unimpeachable\uncriticizable. Comment: It looks obvious enough that (C 1) is never going to be satisfied simpliciter. If one takes a fallibilist position on such matters, then it is evident straightaway that no argument can be conclusive because no premise can satisfy (C I ).16 One might attempt to get around this by adopting Hamblin's suggestion to relativize dialectical criteria to a particular discourse community. One might claim that while no premise can be unimpeachable simpliciter, a premise might have this status within a particular discourse community. However, then the argument in question becomes "conclusive-for" that community, not "conclusive" simpliciter. So for this discussion, that suggestion does not help. The second internal requirement: (C2) The connection between the premises and the conclusion would have to be unimpeachable-the strongest possible. Regarding (C2): I take it that no inductive argument, and no conductive argument (if there are such) could be conclusive, because they don't meet C2. Each of them is such that the premises could be true and the conclusion false. Hence no such argument could be conclusive. To be conclusive, the argument would have to be deductive in character. However, Deductivism as a theory of argument is every bit as questionable the view it is defending here. 11 I doubt that any argument can satisfy these internal requirement with respect to the illative core, and hence the conclusion follows. Still I want to continue my discussion to include the other criteria, for the sake of argument, as they say. Any conclusive argument would have to have what I call a dialectical tier in which the important objections to and criticisms of that illative core were dealt with in such fashion as to remove the ground from underneath them. For an argument to be conclusive, then, it would have to be the case that 266 Ralph H Johnson (e3) The argument meets all legitimate objections and criticisms. The problem is that no Arguer can undertake to head off all possible objections and criticisms; so some limitation would have to be placed on what is expected, something like "all significant (or well known) criticisms and objections." Here is one place an Arguer may be expected to run into difficulty. He may either underestimate or overestimate the force of some position or objection, or may fail to anticipate altogether an important objection, or may not know of some very important criticism. One of the basic reasons that we put our arguments out for criticism is that we recognize our own limitations in these matters. Indeed my experience has always been that the most important and useful criticisms of my argument are precisely those I did not and probably would not have thought of. ii. External property. To be conclusive, the argument would also have to satisfy one external property: (e4) The argument would be regarded as a conclusive argument. Part of an argument's being conclusive is that it is recognized as such. It would not only have succeeded in establishing its conclusion but it would be recognized as having achieved this. Now if they did know of such a specimen (and I take it for granted that we would know of it), then logicians and argumentation theorists would certainly advert to it in their teachings and writing. They would haul this magnificent piece out and say: "Ecce!" But no such thing happens, and for good reason; because no such specimen exists. IS Since, to my knowledge, no argument has satisfied all these conditions, I conclude that there are no conclusive arguments. 3.2.2. Another objection Now, I can imagine another objection that goes as follows: "Suppose someone violates this principle? Then what? Nothing serious happens, because really no one is going to be impressed by the Arguer's attempt to immunize the argument. They will just proceed to criticize the argument anyway. So it looks like your principle is gratuitous, is really not necessary." I agree that someone who violates this principle is not likely to "get away with it." So it is not a regulative principle but rather a constitutive one. If someone violates it, that violation suggests that this person is not really a participant in the practice of argumentation but rather that this person has some ulterior motive and is engaging the practice in bad faith. 3.2.3. Possible counterexamples It will be pointed out that some arguments satisfy (el) and (e2), so long as we deal with rather trivial issues. Thus the following argument, appropriately contextualized, is sound and to that degree invulnerable to criticism: If it rains the ground is wet. It rains. So, the ground is wet. The Principle of Vulnerability 267 I regard these specimens as limiting cases of argument. For one thing, the issue here is scarcely worth arguing over. Why would any rational person argue with someone whether the ground is wet? This sort of matter is typically settled by observation, not argumentation. Moreover the argument lacks a dialectical tier. Such limiting cases of argument are certainly not the sort that one should make policy over. 19 A different sort of counter-example is the famous Gettier counterexample to the classical view of knowledge as justified true belief. One wants to say: "Everyone agrees that this counter-example is indeed a counter-example; so it is conclusive." In response I want to say that a counter-example is not itself an argument, at least not in the paradigm sense discussed in Section I above. It has, for example, no dialectical tier. So it is not a clear case of an argument. More to the point, I can imagine two things which could impugn the Gettier counter- example. First, someone finds a significant (but as yet unknown) flaw in Gettier's reasoning. Second, someone shows how the justified true belief account can be maintained in the face of the Gettier counter-example. Unless both of these can be ruled out, it seems to me overly strong to claim that the counter-example is conclusive. However, this brings out an important point: that while arguments are not open to being conclusively established, they can, it seems, be conclusively criticized. Thus, I conclude that there are no conclusive arguments. Every argument should be regarded as subject to legitimate criticism; and hence the Arguer ought not represent the argument as invulnerable. 4. Conclusion In this paper, I have attempted to make explicit a principle that I believe is implicit in the practice of argumentation and one that most logicians and argumentation theorists would embrace. I call it The Principle of Vulnerability: arguments must be open to criticism; hence no Arguer should represent the argument as being invulnerable, or immune to criticism. I also sought to present justification for that principle and it come from two sources: from the nature of the practice, and from our experience with this practice. The force of my position means that I dare not claim that my own argument is conclusive. And I do not. For I expect that there will be some strong and legitimate criticisms of my argument. I have good reason to believe I shall not be disappointed. 20 1 See my "Agumentation: A Pragmatic Perspective" in Inquiry 13 (1994) pp. 23-34. 2 See JUrgen Habermas, The Theory of Communicative Action. Volume One: Reason and the Rationalization of Society, trans. Thomas McCarthy. Boston: The Beacon Press, 1981. p. 22. 268 Ralph H Johnson 3 In a way this parallels the requirements of the dispensation of justice, where justice must not only be done but it must be seen to be done. 4 See 1. A. Blair, "Premiss Adequacy," in F. H. van Eemeren, R. Grootendorst, 1. A. Blair and C. A. Willard (Eds.) Proceedings of the Third ISSA Conference. Amsterdam: Sic Sat, 1995. Vol.lI, pp. 191-202. 5 The arguer may be able to generate such criticism himself; but this appears to be the exception, not the rule. It has been said of both Plato and Wittgenstein that each was his own best critic, but how often does such a mind come along? 6 I realize that not everyone would agree with this distinction between criticism and evaluation. 1 take some consolation from the fact that my views are fairly close to those of Scriven as these are presented in various entries in his Evaluation Thesaurus, 4th edition, Newbury Park, CA: Sage Publications, 1991. 7 In Communication. Argumentation and Fallacies (Hillsdale, NJ: Erlbaum, 1992), van Eemeren and Grootendorst include something like the Principle of Vulnerability as one of their ten rules for rational discourse: "Parties must not prevent each other from advancing standpoints or casting doubts on standpoints." They defend it by citing freedom of speech (107), and also by arguing that it is not in keeping with the nature of a critical discussion. They describe the specific phenomenon 1 am talking about here: "This may also happen by declaring particular standpoints sacrosanct, so that the opponent is prohibited from casting doubt on them and they are rendered immune to criticism" (1992:108). They point out that doing this is "putting obstacles to the development of a critical discussion." In Understanding Arguments, 3rd ed. (New York: Harcourt, Brace and Jovanovich, 1987) Fogelin touches on something like this principle when he refers to what he calls self-sealing arguments (pp. 96-99): So far, we have seen two ways in which an argument can be self-sealing: (I) it can invent an ad hoc or arbitrary way of dismissing every possible criticism .... (2) A theory can counter criticism by attacking its critics. (98) (1) describes a way of violating the principle. And my point is that an argument which displays this feature is not a genuine argument. It may look like an argument, it may have all the elements of an argument, but if the arguer claims to have insulated the argument against all possible criticism, then this is no arguer and no argument. We have moved into a different mode of discourse which I am not sure how to characterize. S George Will, "MacKinnon leads assault on free speech," Detroit Free Press, February 1994. 9 I find it incredible that this should have been said in a University. And yet perhaps not so surprising if, as William James reminds us "there is a little dogmatist lurking inside all of us." The Will to Believe and Other Essays on Popular Philosophy. New York: Dover, 1956. pp.13-14. 10 In this connection, one thinks naturally of Flew's challenge in "Theology and Falsification," where he puts this question to the symposiasts: "What would have to occur or to have occurred to constitute for you a disproof of the love of, or the existence of, God?" In my terms, Flew is asking these symposiasts, particularly those who argue that God exists, or that God loves us, what might count against their argument. He is asking them to indicate their vulnerability. The same request could be directed to atheists, agnostics, and indeed to Flew himself. II R. H. Johnson and J. Anthony Blair, Logical Self-Defense, 3m ed. Toronta: McGraw-Hili Ryerson, 1993, p. 43. 12 But see the objection from Jonathan Adler dealt with below. 13 The implications of this point are of great importance to any theory of evaluation, but I do not have the space to develop them here. See my Manifest Rationality, forthcoming. The Principle of Vulnerability 269 14 Chris Tindale criticized an earlier draft of this on the ground that my position gives credibility to, e.g., the Holocaust deniers. I agree with those who believe that the evidence for the Holocaust is overwhelming. But that does not entail that each and every argument offered for it is conclusive. Nor do I agree with those who would suppress the views of the Holocaust deniers, no matter how much I disagree with them and dislike their views. 15 See my forthcoming Manifest Rationality, Chapter 9, for a fuller treatmeqt of this issue. 16 Harvey Siegel suggests that the position I am taking is really the fallibilist view as applied to arguments, and that fallibilism essentially is the view that with respect to any cognitive holding, the following proposition is always true and appropriate: "It is possible that 1 am wrong." In an argument, this translates into the caveat that "It is possible that there is something wrong with my argument." 17 For a critique of Deductivism, see Trudy Govier's Problems in Argument Analysis and Evaluation, Dordrecht: Foris, 1987, Chapters I, 3 and 4. 18 It is interesting to me that in all the history of writing about argumentation and all the texts that have been written, I am not aware of any which contains what the author regards as an exemplar of a conclusive argument, in the sense mentioned here. 19 This indicates how I would deal with the argument that "All men are mortal, Socrates is a man; therefore, Socrates is mortal." 20 I want to thank a number of colleagues for very helpful criticisms: Chris Tindale (who commented on the original version of this paper), Harvey Siegel, Maurice Finocchiaro, Gerald Nosich, Jerry Cederblom, Jill Binker, Dennis Matthies, and most especially Jonathan Adler. I wish also to thank two anonymous referees ofthe Proceedings for their comments. RALPH H. JOHNSON DEPARTMENT OF PHILOSOPHY UNIVERSITY OF WINDSOR WINDSOR, ONTARIO N9B 3P4 o