How to Think Straight: An Introduction to Critical Reasoning

Home > Other > How to Think Straight: An Introduction to Critical Reasoning > Page 7
How to Think Straight: An Introduction to Critical Reasoning Page 7

by Antony Flew


  3.26 This objection is commonly launched against the appositeness of some stereotype, especially when the suggested explanation of its being apposite is something in the history of the human set to which it is applied which was beyond the control of members of that set. But, of course, any such explanation tacitly grants that the stereotype does indeed apply to that particular set, for thus to explain is precisely not to explain away.

  3.27 Paragraph 3.26 completes the main business of chapter 3. But the chapter can be usefully and even entertainingly rounded off with a consideration of a problem originally set by Aristotle in his treatise On Interpretation, now conventionally referred to by its Latin title de Interpretatione. Aristotle’s problem was to discover what, if anything, can be validly inferred from the logically necessary truth that Either there will be a seafight tomorrow or there will not be a seafight tomorrow. It has therefore come to be known as the Problem of the Seafight, even when the logically necessary truths being examined by Aristotle’s successors contain no reference to a sea battle.

  3.28 Thomas Hobbes (being English) preferred to work with a proposition about the weather. His treatment can be found in a pamphlet Of Liberty and Necessity. (The subtitle is too good to miss, even though it was apparently not provided by Hobbes himself but by a piratical publisher: A Treatise wherein All Controversy Concerning Predestination, Election, Freewill, Grace, Merits, Reprobation, etc. is Fully Decided and Cleared. Anyone seeking a more recent if less confidently definitive treatment of these questions may be referred to, for instance, Flew and Vesey 1987.) In this pamphlet Hobbes begins : “Let the case be put, for example, of the weather. It is necessary that tomorrow it shall rain or not rain. If therefore it be not necessary that it shall rain, it is necessary that it shall not rain; otherwise there is no necessity that the proposition, It shall rain or not rain, should be true” (emphasis original).

  3.29 It should by now be obvious to us that something is going badly wrong. For here, from an empty tautological truth, Hobbes is fallaciously deducing a substantial conclusion about the (causal) necessity of the occurrence of (all) future events. The crux lies in the misplacing and consequent misinterpretation of the term “necessary.” Recall that a logically necessary or necessarily true proposition is one the denial of which must involve asserting a self-contradiction. On the contrary, a logically contingent proposition is one that is logically possible, one that whether or not it happens to be true can always be denied without actual self-contradiction. The expressions “logically necessary” and “logically contingent” therefore do not belong to our ordinary, primary, first-order vocabulary for discourse about the nonlinguistic world. Instead they are elements in a vocabulary for second-order talk about talk; that is, for talk about propositions and the logical relations between propositions.

  3.30 Once the difference between these two orders of discourse is noticed and its importance is appreciated, we have an excellent reason for distinguishing the propositions about which we are talking by printing them in italics. If this is done, then it is easy to see how Hobbes went wrong. For his premise now becomes: “It is a logically necessary truth that Tomorrow it shall rain or not rain.” The logical necessity of this proposition is a consequence simply and solely of the fact that to deny it would be to contradict yourself. But the application of the same test reveals that neither the proposition It shall rain nor the proposition It shall not rain is itself a logically necessary truth. Hobbes himself was further confused by a failure to distinguish logical from physical or causal necessity. For although he certainly does not make that distinction here, he does nevertheless appear to believe that the conclusion he is entitled to draw is that whether it does or does not rain tomorrow, whatever actually does happen will have been causally necessitated.

  4.1 The No-true-Scotsman Move is one way in which some participants in a discussion may shift their ground, often without either those participants themselves or any of the others involved in the discussion clearly appreciating what is happening. There are many others. One that is very common is the But-those-people-will-never-agree Diversion. This is a move in which totally irrelevant bargaining considerations are intruded into factual discussion. If one is trying to thrash out some generally acceptable working compromise on how things are to be run, then one must consider the various sticking points of all concerned. But if instead you are inquiring into what is in fact the case and why, then that someone refuses to accept that this or that is true is neither here nor there. They may be right or in error, reasonable or unreasonable, in their refusal. But the questions which you were supposed to be discussing are not questions about some particular person and what that person does, or will or will not accept. Rather, they are a matter of what is actually true, regardless of what either that particular person or anyone else may either wish or think.

  4.2 This But-those-people-will-never-agree Diversion is often executed with the help of the wretched expression “prove to.” The point of this expression, and what makes it deplorable, is to confound producing a proof with persuading a person. Yet people may be persuaded by an abominable argument just as they may remain unconvinced by considerations which they certainly would accept if only they were more rational, or more honest, or both. It may very well be that no one can prove to—that is to say persuade—Biblical Fundamentalists that our species was not specially created but was instead produced by a process of evolution through natural selection. But that is simply irrelevant to the questions of whether our species was or was not specially created, and of whether that it was or that it was not can be known to be the truth. (For some discussion of the presuppositions and implications of Darwinian theory, see, for instance, Flew 1997.)

  4.3 The But-those-people-will-never-agree Diversion is one sort of move from whatever was the original subject of discussion to a different sort of question: a question about persons. Another move—let us call it the Subject/Motive Shift—is far more common. This move starts by discussing the truth or falsity of some proposition, and the grounds for holding that it has this one or that one of these alternative truth values. But it then goes on to discuss the different questions of what someone’s motives might be for asserting or denying the proposition and/or for wishing to believe or to reject it. Once this distinction is clearly made, it becomes obvious that these are indeed different kinds of questions. But we still need to find ways both to keep them distinct and to pick out some of the possible connections between the two.

  4.4 One reason why these two questions are so often confounded is that the word “reason” is relevantly ambiguous. When someone is said to have some reason for believing a certain proposition, we may need to ask whether this reason is a ground for holding that the proposition is actually true or whether it is a motive for persuading oneself of it, irrespective of whether it is true or not. In the former case we can speak either of a reason (ground) or of an evidencing reason, in the latter either of a reason (motive) or of a motivating reason. We might also distinguish a third sense of “reason.” For instance, in the phrase “one reason why” at the beginning of the present paragraph, what is being referred to is neither of the first two sorts of reason but rather a cause.

  4.5 The classic occasion for distinguishing the first two of these three senses of “reason” is provided by the famous argument known as Pascal’s Wager. Others have urged that the existence of a universe that exhibits certain special characteristics constitutes evidence for, or even proof of, the existence of some kind of God. There is, for instance, what is known as the Argument from, or, better, the Argument to, Design. But in his Pensées the philosopher-mathematician Blaise Pascal (1623–1662) begins by conceding that there are no grounds, no evidences, adequate to warrant the desired conclusion: “Reason can decide nothing here.” Pascal maintains that we are all confronted by a situation in which it is inescapably compulsory to make a bet. He sees only two possible ways to bet and insists that no one has any alternative but to stake his or her life in one or the ot
her of these two ways. The one bet is that the teachings of Roman Catholicism are true. If you bet this way you become or, as the case may be, remain a believing and practicing Roman Catholic. The other bet is that those teachings are false. If you bet that way you reject Roman Catholicism and you live your life on the assumption that your death will be your annihilation.

  4.6 In terms of the present distinction between senses of the word “reason,” what Pascal was saying was that, although we have no sufficient reasons (grounds) for believing that the teachings of Roman Catholicism are true, we do have the very best of reasons (motives) for trying to persuade ourselves that they are. For if they are true and we bet our lives on the truth, then we stand to make the infinite gain of Heaven. But if they are false and we had bet our lives on their being true, then our loss in leading Roman Catholic lives—even if we do persist in seeing that as a loss—will at worst have been strictly finite. Suppose, however, that we bet our lives on the secular alternative. If we were right to reject Roman Catholic teachings as false, then it is our gains which turn out to have been finite, which we always believed they would be. But if we were wrong, then it is our loss which will be infinite, an eternity of the most extreme tortures in Hell. It is therefore clear that, always given Pascal’s assumptions, he is right to conclude that anyone would be mad to make their bet in any sense other than that which he recommends. (For a critique of this Wager argument see, for instance, Flew 1984, chapter 5.)

  4.7 The further distinction between reasons (grounds) and reasons (causes) becomes essential if we want even to begin adequately to critique an argument such as the following: “In deriving mind and knowledge from nature, as science conceives it, ‘the naturalist’ must assume that his own account of nature is true. But on his premises, the truth of this account, like that of any other bit of knowledge, is merely the function of the adjustment of the organism to its environment. . . . This entire conception of knowledge refutes itself” (Urban 1949, p. 236).

  4.8 W. R. Urban, like so many other writers, is, in effect, arguing that if there are always physiological reasons why I utter the sounds which I do utter, then I cannot have, and know that I have, good reasons for believing the propositions which I assert by uttering those sounds. If it is the one, then it must be merely that and not the other. But now, in terms of our further distinction, the physiological reasons must be reasons (causes), whereas the good reasons which I may or may not have for believing can only be reasons (grounds). So no reason (ground) has been given why “reasons,” in these two different senses of “reason,” should be considered as necessarily rivals for the same space, with the presence of one precluding that of the other. (For a thorough examination of an argument of this kind, as C. S. Lewis presented it in the first edition of his Miracles, see Anscombe 1981, chapter 21.)

  4.9 A first general lesson of method for us to draw from the previous five paragraphs is that wherever we need to distinguish two or more senses of a word, there we also need to supply informative parentheses in order to maintain the distinctions made. The model to imitate, which I have been imitating, is that of funny (ha ha) as popularly and correctly distinguished from funny (peculiar). For even after an ambiguity has apparently been recognized, the temptation is to plunge on as if the crucial distinction had never been made. Thus, while claiming to have accepted the distinctions indicated earlier (see paragraph 4.4), someone may nevertheless continue to argue just as before that if the reason why I believe something is physiological, then I cannot have any sufficient reason for believing it. But it would scarcely be possible for anyone to do this if he or she had shown, by inserting the appropriate parentheses, that he or she had really recognized the ambiguity.

  4.10 A second lesson will perhaps only be completely mastered when we work—as here—with engagingly important and tricky examples. It can be hard to come fully to terms with the fact that a word may be entirely ambiguous, with senses as different and as unconnected logically as those of any two wholly different and etymologically unrelated words. For example, in chapter 1 we had occasion to distinguish two radically different senses of the word “democracy” (see paragraph 1.30).

  4.11 In one of these two senses—call it liberal—an institution is democratic to the extent that it is in due season possible for its members to vote the leadership out, if that is what those members want to do. In the other sense—call it paternalist—an institution is democratic to the extent that it serves the true needs and interests of its rank-and-file members, usually as determined by some particular elite individual or set of individuals. Given this distinction, it becomes tempting to speak of two varieties or species of democracy—especially if you yourself happen to want to appropriate for one of these supposed varieties or species favorable attitudes originally directed toward the other. But our word “democracy” derives from Greek words meaning people power, and it has retained this primary meaning in English. So democracy (paternalist) is the very opposite, rather than another variety, of democracy (liberal).

  4.12 The reason why mistakes of the kind discussed in the previous three paragraphs are so tempting—the cause of the trouble, that is—is that speech habits are just as much habits as any others and just as hard to form or to break. Maintaining a distinction between two senses of a word requires that some entrenched habits of association be overcome and that others be formed. The least one can do is to make suitable parenthetical insertions whenever we employ the ambiguous word in any possibly troublesome context. In this way the desired associations of what is inserted pull against the undesired associations of the ambiguous term. In the exceptionally intractable case, where this routine treatment proves ineffective, the ambiguous term should be jettisoned completely and replaced by two words or expressions that do not look alike and so do not have the same associations.

  4.13 The most common, indeed the most dully commonplace, case of the Subject/Motive Shift is that in which an assertion is dismissed as false or an argument is discredited as unsound for no other and better reason than that it is made or presented by an interested party. Certainly it is right always to be alert to the possibility that assertions and arguments are being corrupted by the self-interests of the asserters and the arguers. And there is little need to warn most of us to be alert to the possibility that the statements and arguments of businesspersons, politicians, labor-union spokespersons, and public-relations officers may be corrupted by various forms of self-interest. Everyone, too, knows the cynic’s definition of ambassadors as people sent abroad to lie for their countries.

  4.14 It may very well be that as an ambassador, labor-union official, or public-relations officer, he or she is paid to say a certain thing or argue in a certain way. But that material interest by itself does not constitute good or even any grounds for concluding that those representatives’ claims must be false and their arguments must be invalid. Those of us who do not belong to any of these three suspect classes often present sound arguments for conclusions that happen to be both true and to our advantage. The truth is no more necessarily disagreeable than it is necessarily agreeable.

  4.15 No one is likely to want to deny outright anything asserted in the two previous paragraphs. The problem is, as with many other important truisms, to discipline ourselves so that we are never carried away by the heat of the moment. We must always remember that even a truism may be true. One step in the right direction might be for everyone to collect his or her own sets of arresting examples. Such sets need to be individually collected and individually weighted to offset each individual’s particular assemblage of biases. For instance, if someone is generally inclined to suspect the truthfulness of businessmen or diplomats or labor spokespersons, then that person’s collection of examples has to contain statements from members of these particular, notoriously sinister, occupational groups, statements which, when made, have been contemptuously disbelieved by that person and by other opponents, but which nevertheless turned out to have contained the truth and nothing but the truth.


  4.16 Notwithstanding the recent development of the economics of public choice, however, many people apparently remain unaware of the very real and live possibility of corruptions resulting from the private interests of employees of public and semipublic organizations, corruptions resulting from their private interests precisely as employees of those particular organizations. Employees of agencies established to combat perceived evils, for instance, cannot but have strong job-preservation interests in the continuation of at least sufficient of those evils to justify the preservation of the agency which employs them. If and insofar as those evils are indeed diminished, either by the activities of the agencies themselves or by independent technological and social developments, it becomes necessary, just to maintain present levels of employment and funding, for those agencies somehow to identify further supposed examples of the evils in question. (For a brief account of the nature of the economics of public choice, consult the preface to Buchanan 1991. The classic contribution to that area of economics is Buchanan and Tullock 1962.)

  4.17 This is certainly no place for a general investigation of the ways in which such possibilities have in fact frequently been realized: for instance, in the Equal Employment Opportunities Commission (EEOC) and the Environmental Protection Agency (EPA) in the United States or in similar organizations in other countries. But it is for us very relevant to refer to a case in which the identification of an abundance of further supposed examples of the evils in question is achieved by tacitly transforming the meaning of the key word; the word, that is to say, used to refer to that evil.

 

‹ Prev