I nformal Logic VII.1, Winter 1985 The Logic of Deep Disagreements ROBERT J. FOCELIN The best evidence of the growing im- portance of the informal logic move- ment is the alarm it is causing in older and more conservative circles. Many think that something bad has happen- ned when truth tables, Centzen rules for natural deduction, and multiply general propositions in prenex normal form are replaced by a mixed and in- choate system of techniques which, among other things, do not easily ad- mit of machine grading. Now the informal logic movement is certainly diverse and loose jointed, but one thing that holds it together is a strong pedagogical commitment. The fundamental task of logic in a col- lege curriculum is to develop skills and techniques for the critical evaluation and coherent presentation of argu- ments. Negatively, it is held that formal logic, as it has commonly been taught, has not fulfilled this function. Of course, the disagreements here are not merely pedagogical. The philosophy of ordinary language still has its in- fluence and, as a result, many still hold that the techniques of modern symbolic logic are inadequate to the richness / subtlety / and power of ordi n- ary language. Its application distorts more than it illuminates. My own view, for what it is worth / is that the methods of formal logic can be profitably used to explicate such fundamental notions as validity, logical form, tautologies, con- tradictions/ etc. These techniques are of little use in the direct evaluation of actual arguments beyond the simplest level of complexity. But this is not what Dartmouth College I want to discuss. At its most timid (and worst) informal logic amounts to little more than in- formal formal logic. It takes over the prejudices-usually the deductivist prejudices-of traditional logic and re- peats them in a casual and inaccurate way. The idea persists, though people know better, that the only good argu- ment is a good deductive argument. This is revealed in practice when it is assumed that an argument is no good if it resists our charitable efforts at recasting it as a sound deductive argu- ment. Of course, given any argument that intuitively strikes us as sound, it's a snap to reconstruct it in a way that gives it the dignity of a deductively valid form. Simply introduce a condi- tional suppressed premise containing a conjunction of the stated premises as the antecedent and the conclusion as the consequent and voila you have a deductively valid argument that can be no more objectionable than the argu- ment you began with. r r s becomes s • t If r&s, then t t If the original argument strikes one as valid, there can, of course, be no ob- jection to the truth of the introduced suppressed premise. (And you do not even get Lewis Carroll's problem of Achilles and the Tortoise, for having produced a valid argument, you can stop.) 2 Robert 1. Fogelin Needless to say, a reconstruction of this kind is mere hand waving. Our goal in reconstruction is to ferret out those underlying principles that serve as the ground for our initial intuition that the argument is acceptable, and when the method of argumentative recon- struction is used in this way, there is much to be said in its favor. It often happens that when we surface those principles which seem to give an argu- ment its backing, they turn out to be false, and we are thus forced to reject, or at least modify, our original argu- ment. This is a useful procedure- the Socratic method applied to back- ground principles rather than to defini- tions - but it is also a method that can easily be distorted by deductive chauvinism. To insist that any accept- able argument must admit of a recon- struction (free of the hand waving de- vice noted above) revealing its deduct- ive validity, presupposes the non- existence of compelling inductive argu- ments. Again, to insist that our back- ground principles be exceptionless- immune to all counterexamples how- ever recherche-is to adopt a rationalist ideal and forget that many of our basic principles hold, not always, but, as Aristotle would say, only for the most part. But I think that the chief danger of adopting a deductive model for all reasoning-even as an ideal-is that it yields skeptical consequences. The demand that in an acceptable argument the conclusion must be entailed by ex- ceptionless premises yields the con- sequence that virtually all of those everyday arguments which seem per- fectly adequate are, in fact, no good. In the short run, students find this dis- covery of almost universal irrationality arresting. Debunking has its charms. The long run effect is less salutary. If students become convinced that they can always find something wrong with any (non-trivial) argument presented to them, then the distinction between good arguments and bad arguments is subverted, and the whole enterprise of arguing seems to lose its point. Indeed, a perSistent problem in teaching logic is that we will turn our students into radical, if short-termed, skeptics. For example, students will often treat the notion of vagueness in such a way that all concepts without sharp boundaries are considered vague. I suppose that the notion can be used in this way, though J do not like it, but the upshot is often that the stu- dent dismisses as invalid all argu- ments employing concepts that admit of borderline cases. A similar problem arises with ambiguity I appeals to authority, slanting, and so on. Applied in an external and mechanical way, the standard fallacies do not do what they are supposed to do-distinguish good arguments from bad arguments-since all arguments turn out to be bad. The answer to this, of course, is that the student must learn to analyze argu- ments in a way that is sensitive to the contexts in wh ich they occu r. A term in an argument may be vague I but the argument may not trade upon this vagueness. Or the vagueness may weaken the argument, though not in a serious way given its general intents and purposes. Finally, an argument may contain crucial terms that are hopelessly vague-a common and interesting expression which indicates that we really do not understand the range of application of such terms. The above may sound like a plati- tude. I don't think that I have ever heard anyone come right out and say: "Ignore context; it doesn't matter." But turning to context represents an important departure from the standard way of looking at arguments. An argu- ment is no longer thought of as merely a structure on propositions: one of which is designated as the conclusion, the others as premises presented in its behalf. An argument is produced by the activity of arguing and arguing is something people do. Furthermore, they do it for a wide variety of reasons, in an effort to achieve very different purposes. Toulmin was right in speak- ing about the uses of argument, not just the use of argument. Here I want to examine a very simple case of using an argument (or giving reasons) to explain a bit of conduct. A is asked why he is taking a particular road and he responds, "I want to pick up the fish last." We can imagine this being a conclusive reply. On the other hand, it might be met with the rejoinder, "No, go to the Grand Union last; I don't want the ice cream to melt." This too might be conclusive. But things could also become complic- ated. A might point out that the traffic that way is horrible this time of day, and it would be better to wait a bit to let it clear out. And he might be crushed by the reply "Today is Saturday." People being what they are, we can even imagine this discussion becoming quite heated. I wish to dignify this vignette drawn from suburban life with a Kantian question: how is such an argument pos- sible? The answer (or part of the an- swer) is that the parties to the conversa- tion share a great many beliefs and (if this is different) a great many preferences. They share a detailed knowledge of local geography, prefer- ences for frozen ice cream over melted ice cream and fresh fish over stinking fish, etc. An important feature of these shared beliefs and preferences is that they lie in the background, unmen- tioned. They guide the discussion, but they are not the subject of it. (Except occasionally: "You don't want rotten fish, do you?") They provide the frame- work or the structure within which reasons can be marshaled, where marshaling reasons is typically a matter of citing facts that others already know or of arranging facts in a way that their significance becomes clear. This is a high level enterprise, one that rests upon the thick sedimentary layer of the unchallenged. I shall say that an argument, or better, an argumentative exchange is normal when it takes place with ina context of broadly shared beliefs and preferences. I shall further insist that for an argumentative exchange to be Deep Disagreements 3 normal, there must exist shared pro- cedures for resolving disagreements. People often disagree over simple questions of fact, but, in general, they agree on the method for resolving their disagreement. If you think that Rod Carew hit more triples last year than George Brett, we can simply look it up. If the record book does not convince you, then it's a waste of time talking to you. This last point has sytematic import- ance. Record books have a privileged status in discussions of past athletic ach ievements. They can, of course, contain mistakes and it is possible to show this. But if someone raised general objections to relying on official record books (IIWhat makes you think you can trust them?"), the challenge would be so bizarre that we would dis- miss it rather than attempting to an- swer it. We can imagine a world where baseball record books are systematic- ally distorted without the players them- selves uttering a peep. There is, how- ever, no reason to suppose . that our world is anything like this imagined world. Indeed, the reliability of official record books is assumed as part of the framework in which discussions of this kind take place. These reflections bring me to the following thesis: the language of argu- ment, including the language of argu- mentative assessment, has its primary application in the context of a normal or near normal argumentative ex- changes. A claim that something shows (or proves) something else is much like a knowledge claim. In On Certainty, Wittgenstein remarks that "one says 'I know' when one is ready to give compelling ground. "I know' relates to a possibility of demonstrating the truth" (#243). Arguing is the process of producing these compelling grounds. But to be compelling, grounds must be true or at least thought to be true and I together with other accepted pro- positions, lend adequate support to the claim to be established. Thus arguing, Le., engaging in an argumentative exchange, presupposes a background of shared commitments. More Witt- 4 Robert J. Fogelin genstein: 341. That is to say, the questions that we raise and our doubts depend on the fact that some propositions are exempt from doubt, are as it were like hinges on which those turn. 342. That is to say, it belongs to the log ic of our scientific investigations that certain things are In deed not doubted. And more famously: 344. My life consists in my being content to accept many things. The possibility of arguments, the pos- sibilityof a genuine argumentative ex- change, depends, I am suggesting, on the fact that together we accept many things. But if arguments presuppose this rich background of agreement, how does disagreement even arise, and what job is left for argument to do? One obvious answer is that people in- volved in an argumentative exchange often have an interest in the way the argument is resolved. Arguing, like other human activities, is subject to abuse. Arguing is also a complex activity. It is a skill, something that people can do well or badly. Indeed, as Hume saw long ago (in his discussion of "Unphilosophical Probability" in the Treatise) and as cognitive psychol- ogists have recently rediscovered, human beings seem to be endowed with innate capacities for messing things up as soon as argumentative structures rise above the simplest level of com- plexity. (This is particularly true of inductive inferences where, for ex- ample, the tendency for hasty and un- founded generalization seems to be hardwired into the human brain.) Now, as Michael Scriven has insist- ed, our orindary non-technical lan- guage contains a rich system of critical tools for correcting and assessing argu~ ments. I think that he is right, but he tends to concentrate upon the more civilized part of this vocabulary. I think that some of the cruder argument- ative devices are also interesting. Sup- pose, for example, that I accuse some- one of being pig-headed. This is not a generous thing to say, but it is not a free floating insult either. To call some- one pig-headed is to make quite a specific charge: he continues to cling to a position despite the fact that compelling reasons have been brought against it. But compelling to whom? We are saying that they ought to be compelling for him, or else it wouldn't be right to call him pig-headed. He knows that they are true and in other, less interested, contexts would recog- nize their force. Again, consider the claim that some- one is biased. The person I accuse of bias has a right to ask "How am I biased?" Here the burden falls upon me to show that he has willfully sup- pressed certain facts that support the side of the issue he opposes or that he has suppressed facts that bear against his own position. Now the point that I am trying to make is that charges of bias and pig-headedness only make sense in a normal (or near normal) argumentative context, for in each case an appeal is made to common ground. And I want to make this claim general- ly: the significance of all of our argu- mentative devices is internal to normal (or near normal) argumentative con- texts. To use a Wittgensteinian notion, our argumentative words are odd- job words; they have very special func- tions that come into existence only when words with regular functions are already in place. II These ruminations bring me, at last, to the central question of this paper: what happens to arguments when the context is neither normal nor nearly normal? The asnwer that seems forced upon us is this: to the extent that the argumentative context becomes less normal, argument, to that extent, become impossible. This is not the weak claim that in such contexts argu- ments cannot be settled. It is the stronger claim that the conditions for argument do not exist. The language of argument may persist, but it becomes pointless since it makes an appeal to something that does not exist: a shared background of beliefs and preferences. Here I wish to speak about deep dis- agreements. My thesis, or rather Witt- genstein's thesis, is that deep disagree- ments cannot be resolved through the use of argument, for they undercut the conditions essential to arguing. What is a deep disagreement? First let me say what I don't mean by this notion. A disagreement can be intense without being deep. A disagreement can also be unresolvable without being deep. I can argue myself blue in the face trying to convince you of some- thing without succeeding. The explana- tion might be that one of us is dense or pig-headed. And this is a matter that could be established beyond doubt to, say, an impartial spectator. But we get a very different sort of disagreement when it proceeds from a clash in under- lying principles. Under these circum- stances, the parties may be unbiased, free of prejudice, consistent, coherent, precise and rigorous, yet still disagree. And disagree profoundly, not just marginally. Now when I speak about underlying principles, I am thinking about what others (Putnam) have called framework propositions or what Witt- genstein was inclined to call rules. We' get a deep disagreement when the argument is generated by a clash of framework propositions. An example may help. A distinctive feature of the informal logic movement is the replacement of artificial cooked- up arguments by genuine arguments that are complex and pressing. The morality of abortion has been a favorite topic. First of all, arguments on both sides of this issue can be subjected to normal criticism, i.e., criticisms that ought to be accepted unless the arguer is biased, pig-headed, etc. An argu- Deep Disagreements 5 ment on any subject can be question begging, biased, slanted, vague, and so on. It is characteristic of deep dis- agreements that they persist even when normal criticisms have been answered. Another feature of deep disagreements is that they are immune to appeals to facts. Parties on opposite sides of the abortion debate can agree on a wide range of biological facts-when the heartbeat begins in the fetus, when brain waves first appear, when viability occurs, etc.-yet continue to disagree on the moral issue. Their disagreement can even survive a general agreement on moral issues: for example, on the sanctity of human life, for the central issue of the abortion debate is the moral status of the fetus and that cannot be settled by an appeal to biological facts or by citing moral principles al- ready limited to moral agents or pa- tients. Here the following diagnosis of the situation is tempting. What I have called deep disagreements are gener- ated by conflicts between framework propositions. They remain recalcitrant to adjudication because the sources of the disagreement-the framework propositions-are allowed to lie in the background, working at a distance. The way to put the debate on a rational basis is surface these background propositions and then discuss them directly. This sounds attractive until we con- sider what such a discussion would look like in detail. On the one side someone will hold that at conception, or to be delicate, very shortly after conception, an immortal soul enters into the fertilized egg, and with this, personhood is attained. Why should one believe anything like this? Well, this is part of a wider tradition, ground- ed in revelation, and sustained and deepened by faith. I don't know how well I am doing in representing this position, my heart is not really in it, but the point that I am trying to make is that when we inquire into the source 6 Robert J. Fogelin of a deep disagreement, we do not simply find isolated propositions (liThe fetus is a person."), but instead a whole system of mutually supporting propositions (and paradigms, models, styles of acting and thinking) that constitute, if I may use the phrase, a form of life. I think that the notion of a form of life is dangerous, especially when used in the singular. We do better to say that a person participates in a variety of forms of life that overlap and criss- cross in a variety of ways. Some of these forms of I ife have I ittle to do with others. This explains why we can enter into discussions and reasonable argu- ments over a wide range of subjects with a person who bel ieves, as we think, things that are perfectly mad, e.g., that concern with conserving natural resources is impious, since it denies the imminence of the second coming. But if a person believes that, can we trust him on any subject? The answer might be yes. He might, for example, be an expert on short-term lease options. So I don't want to exaggerate. Claim- ing that deep disagreements exist does not mean that they are common. And again, a disagreement can be heated without being deep. But if deep dis- agreements can arise I what rational procedures can be used for their resolu- tion? The drift of this discussion leads to the answer NONE. Since these ideas come from Wittgenstein, let me cite his very words. Toward the end of On Certainty the following series of paragraphs occurs: 608. Is it wrong for me to be guided in my actions by the propOSitions of physics? Am I to say I have no good ground for doing so? Isn't precisely this what we call a 'good ground'? 609. Suppose we met people who did not regard that as a telling reason. Now, how do we imagine this? Instead of the physicist, they consult an oracle. (And for that we consider them primitive.) Is it wrong for them to consult an oracle and be guided by it?-If we call this "wrong" aren't we using our language- game as a base from which to combat theirs? 610. And are we right or wrong to combat it? Of course there are all sorts of slogans which weill be used to support our proceedings. 611. Where two principles really do meet which cannot be reconciled with one another, then each man declares the other a fool and heretic. 612. I said I would 'combat' the other man-but wouldn't I give him reasons? Certainly; but how far do they go? At the end of reasons comes persuasion. (Think what happens when missionaries convert natives.) If we turn back now to the informal logic movement, we will discover that it has not dodged deep disagreements. Indeed, it has seized upon them with zest. And sometimes, at least, works in informal logic give the impression that they possess the resources to resolve such disagreements. With Wittgen- stein I am skeptical of such claims. To illustrate thiS, I shall consider one case of a deep disagreement, the dispute over affirmative action quotas, so- called 'reverse discrimination' quotas. As informal logicians, we can say a great deal about the surface features of this debate. We can point out, for example, that the phrase 'reverse dis- crimination' is tendentious. Indeed, there may be any number of things that we might say about the coherence and forcefulness of arguments on each side of the issue. Down deep however those who are opposed to ~ffirmativ~ action quotas have a very strong argu- ment on their side. It is a fairness argu- ment. Affirmative action quotas are both overinclusive and underinclusive with respect to their distribution of benefits. Many who profit from such quotas have not themselves been sub- jected to prejudice, nor are they placed at any disadvantage because of pre- judice. At the same time, many who have suffered from prejudice and are disadvantaged because of this will not receive simi lar benefits simply because they are not in the class cov- ered by the affirmative action program. More troublesome still, affirmative action quotas are both underinclusive and overinclusive with respect to the burdens they impose. Many who have benefited from past discrimination are not burdened with the costs of rectifica- tion and, worse yet, many individuals who demonstrably have not profited from past discrimination are so bur- dened. Finally, these bad fits, as they are called, cannot be written off as minor and unavoidable administrative slippage. Not only are affirmative action quotas unfair, they are signific- antly and avoidably unfair. Of course, the above argument could be made stronger, for example by citing appropriate statistics, and, needless to say, certain of its factual premises can be challenged. Here I wish to con- sider quite a different response that does not, or at least need not, chal- lenge any of the factual claims in the above argument. It takes the following form. The prejudice we are consider- ing, for example, against minorities, was not directed against them as in- dividuals, but as members of a parti- cular group. It isn't this way: as it turns out, a great many people who have been subjected to racial prejudice hap- pen to be black; rather people were subjected to prejudice because they were black. Furthermore, they were subjected to prejudice by an identifi- able group: the ruling white majority. Prejudice, and its attendant disadvan- tages, has been a group phenomenon. The demand for rectification, then, is not the claim of one individual against another, but the claim of one group against another. That the black minor- ity has a just claim against the white majority is amply born out by an histor- ical record and by statistics that no one denies. I want to say that we are here dealing with a deep disagreement because the parties on both sides might agree on Deep Disagreements 7 all historical and statistical matters, but still disagree. The dispute is, in fact, one concerning moral standing. (In this way it is like disputes concerning abortion, and this may explain, at least in part, its intractability.) The anti- quota argument rests on the assump- tion that only individuals have moral claims. The pro-quota argument rests upon the assumption that social groups can have moral claims against other social groups. But the word 'assump- tion' is too weak. The arguments on each side are carried on within the framework of such commitments. Is there any way of adjudicating a clash of this kind? I confess that I do not see how. III When the Pythagoreans discovered the irrationality of the square root of two, they swore themselves to secrecy -or so the story goes. Perhaps we should adopt the same strategy, with respect to deep disagreements. Our enterprise depends, at least in part, on the assumption that earnest clear think- ing can resolve fundamental issues. But if in the end, and sometimes the end is very near, we have to fall back on persuasion, what's so bad about using these techniques right from the start? There is an answer to this. In the con- text of a normal argument, people claim to be invoking mutually accept- able grounds, and they can be held responsible for this claim. Still, what shall we say about deep disagree- ments? We can insist that not every disagreement is deep, that even with deep disagreements, people can argue well or badly. In the end, however, we should tell the truth: there are dis- agreements, sometimes on important issues, which by their nature, are not subject to rational resolution. 8 Robert j. Fogelin Note The passages from Wittgenstein's On Certainty, are from the edition edited by C.E.M. Anscombe and C.H. von Wright, translated by Denis Paul and C.E.M. Anscombe, J & J Harper, New York and Evanston, 1969. It has been my experience that every- one sees some force in those arguments against affirmative action quotas that appeal to the rights of individuals. Many, however, see little force in the counter-argument with its appeal to the rights of groups. For an articulate pre- sentation of this second standpoint, see Owen M. Fiss's "Croups and the Equal Protection Clause", in Equality and Preferential Treatment, edited by Marshal Cohen, Thomas Nagel, and Thomas Scanlon, Princeton University Press, Princeton, 1977. This reader presents a collection of forceful state- ments on both sides of this issue. In this essay I have concentrated on the deep disagreements that arise be- cause of conflicts between belief structures. Deep disagreements can also arise because of differences in preference scales. Needless to say, bel ief structures and preference scales are interrelated in important ways and for this reason deep disagreements are certainly more complex (and perhaps even more intractable) than here in- dicated.D Robert J. Fogelin, Department of Philosophy, Dartmouth College, Hanover, New Hampshire, 03755