How to Think Straight: An Introduction to Critical Reasoning Read online

  Published 1998 by Prometheus Books

  How to Think Straight: An Introduction to Critical Reasoning. Copyright © 1998 by Antony Flew. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, digital, electronic, mechanical, photocopying, recording, or otherwise, or conveyed via the Internet or a Web site without prior written permission of the publisher, except in the case of brief quotations embodied in critical articles and reviews.

  Inquiries should be addressed to

  Prometheus Books

  59 John Glenn Drive

  Amherst, New York 14228–2197

  VOICE: 716–691–0133, ext. 207

  FAX: 716–564–2711

  WWW.PROMETHEUSBOOKS.COM

  06 05 04 03 02 8 7 6 5 4

  The Library of Congress has cataloged the printed edition as follows:

  Flew, Antony.

  [Thinking about thinking]

  How to think straight : an introduction to critical reasoning / by Antony Flew.

  p. cm.

  Originally published: Thinking about thinking.

  Includes bibliographical references and index.

  ISBN 1–57392–239–0 (alk. paper)

  ISBN 978-1-57392-239-5 (pbk.)

  ISBN 978-1-61592-215-4 (ebook)

  1. Thought and thinking. I. Title.

  BF455.F614 1998

  160—dc21

  98–8517

  CIP

  Printed in the United States of America on acid-free paper

  We must follow the argument wherever it leads.

  Socrates (Fifth Century B.C.E.)

  A man who will not reason about anything is no better than a vegetable.

  Aristotle (Fourth Century B.C.E.)

  A moment’s thought would have shown him. But a moment is a long time, and thought is a painful process.

  A. E. Housman (Twentieth Century C.E.)

  Many people would sooner die than think—in fact, they do so.

  Bertrand Russell (Twentieth Century C.E.)

  Foreword

  1 The Basic Equipment

  2 If/Then and All/None

  3 Evasion and Falsification

  4 Motives and Grounds

  5 Minding Our Language

  6 Figuring

  7 A Chapter of Errors

  8 The Final Foreword

  Selected Bibliography

  Index of Personal Names

  Index of Ideas

  The present work is a revised and greatly expanded version of a book originally published in the United Kingdom as Thinking about Thinking and later reissued in the United States of America as Thinking Straight. The revision is primarily an update of the original work. References to controversies since deceased have been replaced by others more topical. But there has also been some simplification, intended to make the contents easier for readers to master. The purpose of the expansion was to introduce additional useful material, especially material relevant to what have become major controversial issues since the appearance of the first edition. Among these are, for instance, controversies about pollution, conservation, and the proportionate representation of various perceived minorities in different areas of activity and achievement.

  Among members of the class of books aimed at the improvement of the quality of thinking, Thinking Straight was from the beginning distinctive in two ways. One was the frequency of references to the ideas both of various classical philosophers and of other major thinkers such as Karl Marx. I have been told by colleagues who have used Thinking Straight as the required text for critical thinking courses that these references have led some of their pupils to sign up for further courses in departments of philosophy. This has encouraged me to introduce into this second edition both paragraph-long quotations from René Descartes and David Hume and other equally relevant, shorter quotations from Thomas Aquinas, John Locke, Thomas Hobbes, and, again, David Hume. I have also introduced parenthetic references to various books with which those seriously engaging in these current controversies need somehow to come to terms. That, as was made very clear in the frequent use of that expression by Karl Marx, is by no means the same as coming to agree!

  The second distinctive feature of Thinking Straight is indicated by the subtitle given to the first edition. It was: “Or, Do I Sincerely Want to Be Right?” Other books about critical thinking at least tend to suggest to their readers that the authors’ chief concern is helping readers protect themselves against deception by unsound arguments urged, whether innocently or intentionally, by other people. Certainly this is necessary work. But it is not sufficient. A prime cause of our being deceived is, for all of us, always our own desire to be so deceived. How to Think Straight: An Introduction to Critical Reasoning therefore insists throughout that all of us constantly need to be asking ourselves what it is which we want to believe to be true, and whether our desires so to believe are stronger than our desires to know the truth, however uncongenial to us that truth may be. It is a truly existential challenge.

  Antony Flew

  26 Alexandra Road

  Reading RG1 5PD

  England

  1.1 The first thing to get straight in thinking about thinking is the difference between questions about validity and questions about truth. But in getting this straight we shall find that we are also sorting out every other really fundamental notion. For the indispensable notions are all connected. We cannot fully master any one without getting the same grasp upon the lot. Once the essential preparation is complete, we may proceed to the main business of the book. That business is to consider examples of thinking, usually of bad thinking, in order to learn how to do the job better. Here and now we have first to clean and tidy the tools.

  1.2 The reason to begin precisely where we are beginning is that thinking about thinking is concerned, at least in the first instance, with the validity or invalidity of arguments, rather than with the truth or falsity of propositions. What is true, or false, is propositions. What is valid, or invalid, is arguments. These notions and these distinctions are absolutely basic. To say that an argument is true or that a proposition is valid is as uncomprehending or as inept as to say that someone got to first base in basketball or that someone made a home run in tennis.

  1.3 Consider propositions. There are, of course, propositions and propositions. Both those mutually advantageous proposals which one businessman makes to another and those improper but delightful suggestions which playboys put to their intended playmates are called, quite properly, “propositions.” But in this book—perhaps regrettably—we shall engage with propositions only in a quite different sense of the word. In this, our relevant sense, the word “proposition” is defined as, “whatever may be asserted or denied.” So a proposition for us becomes whatever may be expressed by the that-clause in such sentences as, “She asserted that he had been there on Wednesday,” or “He denied that he had ever met her.”

  1.4 In the irrelevant, proposal sense a proposition may be said to be attractive or unattractive, profitable or unprofitable, and many other things besides. What it cannot be, or be said to be, is either valid or invalid. In our different sense there are again several things which a proposition may be: demonstrated, for instance, or probable, or refuted. Nevertheless, the primary characteristic is truth or falsity. For demonstration here is nothing else but proving that the proposition is true. The proposition which is probable just is probably true. Refutation, again, is not merely saying, but showing, that the proposition is false. It is because refutation involves more than denial that hard-pressed spoke
spersons so often assert that they have refuted charges when in fact all that they have done is deny them, perhaps dishonestly.

  1.5 Propositions in this understanding are not to be identified with arguments, although all arguments contain propositions. Piety demands that our first example be a dull hack hallowed by immemorial tradition. Its tedious, trite, and trivial character will ensure that no one is distracted from what is being illustrated by any interest in the illustration. Later I shall deploy interesting and important examples. I hope thus to escape the dangers of boring myself and everybody else, or suggesting that the subject itself is as trifling as this first illustration.

  1.6 Set out carefully and piously, the traditional example runs: if All men are mortal, and if Socrates was a man, then it follows necessarily that Socrates was mortal. This example includes three constituent propositions. The first two serve as premises, the last as a conclusion. In other contexts and in other arguments what is here conclusion might serve as premise, and what are here premises might be derived as conclusions from other premises.

  1.7 The italicization of the constituent propositions and the representation of the whole argument in a hypothetical (if this, then) form are both important. The first device brings out two things: first, in general, that arguments are concerned with the logical relations between propositions; and second, in particular, what proposition is being said to be necessarily connected with what two others. Later much more will be said both about logical relations and about logically necessary connections. For the moment it is sufficient, but necessary, to emphasize that these are always and only relations of, and connections between, propositions.

  1.8 The second of the two devices, that of representing a whole argument in hypothetical form, makes it clear why, in order to know whether the exemplary argument in which these three propositions are here embodied is valid, we do not need to know whether any of its constituent propositions are true. We do not for this purpose need to know because in offering the argument we are not actually saying anything about the truth or falsity of these constituent propositions. It is all hypothetical. Another argument of the same form would be no less valid even if all of its three constituent propositions happened in fact to be false. This would be true of the absurd argument: If All tigers are strictly vegetarian, and if Socrates the son of Sophroniscus was a tiger, then it follows necessarily that Socrates the son of Sophroniscus was strictly vegetarian. As we shall see in chapter 2, such hypothetical deductions, albeit from much more sensible premises, may serve as the initial steps in a more complex pattern of argument. Such deductions lead us from the actual falsity of the original conclusion in a valid argument to the further conclusion that at least one of the original premises must also be false.

  1.9 Although to say that the present argument is valid is thus not to say that any of the three constituent propositions are true, it does imply the truth of the complex hypothetical proposition: If All men are mortal, and if Socrates was a man, then it follows necessarily that Socrates was mortal. In asserting this truth, what is asserted is that the argument from the two constituent premise propositions to the constituent conclusion proposition is valid. The fact that you can say that the claim that this argument is valid (or invalid) is a true (or false) claim is, however, no more a justification for confounding validity with truth than the fact that you can say that the contention that a certain man is a homosexual is a true (or false) contention is a warrant for identifying homosexuality with truth.

  1.10 To say that an argument is deductively valid is, by definition, to say that it would be impossible to assert its premise or premises while denying its conclusion or conclusions without thereby contradicting oneself. That is what deduction is. We have just seen that an argument may be valid, notwithstanding that both its premise or premises and its conclusion or conclusions are false. Similarly, an argument may be invalid, notwithstanding that both its premise or premises and its conclusion or conclusions are true.

  1.11 Later we shall return to the relations and lack of relations between validity and truth, and I will provide mnemonic illustrations. But the first thing now is to underline the connection between the two concepts of deductive validity and of contradiction and to explain what is so wrong about contradiction. Suppose someone were to maintain that, although Socrates was a man, Socrates was not mortal. No doubt such apparent irrationality in so simple a case is somewhat hard to imagine. Yet that difficulty should, if anything, make it easier to appreciate that if ever people were to behave in this way, then we would have to choose between two alternative conjectures. Either they are being in some way disingenuous, or else they are not fully masters of the meanings of all the words which they have uttered.

  1.12 On the one hand, perhaps they have some sort of doctrinal commitment to affirm the two premises while nevertheless equally firmly denying the obvious conclusion. They may want to maintain that Socrates was a man and, as such, mortal, and yet that Socrates was a god and, as such, not mortal. Certainly there are those who hold that someone, though not Socrates (c. 470–399 B.C.E.), was at the same time both truly man and truly God. Or maybe our imaginary objectors have their reasons for wanting to say one thing in one context or to one group of people while saying something altogether inconsistent in another context or to another group of people. This temptation is familiar to us all. It is no prerogative of members of that scapegoat class, professional politicians.

  1.13 On the other hand, it is also possible that our imaginary objectors are careless or confused about the crucial difference between all and some. Some men are mortal is consistent, as All men are mortal is not, with Some men are not mortal. Again, there is no call for any far-fetched supposing. We meet all too many cases of people who, having noticed that something or other is true for a few instances of such and such a sort of thing, proceed forthwith to assume, or even to assert, that the same is true of all things of that sort. We have, surely, all done it ourselves? (Such generalizations about all and every something or other are, by the way, called universal propositions.)

  1.14 There may appear to be a third possibility, that an objector might be interpreting one of the key terms in one way in one of the premises and in another way in the other premise or in the conclusion. The word “Socrates,” for instance, might be employed to refer to one person on one occasion and to another on the other. Again, “mortal” might be construed as meaning “liable to death,” whereas “not mortal” was understood as metaphorically “immortal”—immortal, that is, in that wholly different sense in which a great person who indisputably has died, or will sometime die, may nevertheless truly and consistently be numbered among the immortals.

  1.15 This apparent third possibility is thus the possibility of equivocation. The word “equivocation” is here defined as “the employment of some word or expression in two or more different senses without distinction in the same context.” If equivocators realize that they are equivocating in their employment of one of the key terms in an argument, then their performances are certainly disingenuous. If they do not realize this, then, equally certainly, they are “not fully masters of the meanings of all the words they have uttered.” In the most literal sense they do not know what they are talking about.

  1.16 The basic point developed in the five previous paragraphs is extremely important. It is none the less so for having been made with a hackneyed, traditional example developed in a somewhat far-fetched way. This basic point is that the terms “valid” and “invalid,” as applied to deductive arguments, and the expression “deductive argument” itself have all to be defined in terms of self-contradiction and the avoidance of self-contradiction. It is because these are thus central notions that our concern with logic inextricably involves us also in concerns with both meaning and truth. The basis of the necessary and inescapable involvement with meaning will be immediately obvious. Given that a valid deductive argument is, by definition, one in which to assert the premises while denying the conclusion is to contradict y
ourself, then it becomes at once clear that no one can be in a position to know whether or not any argument is valid, except insofar as he or she has mastered the meanings of all its crucial terms.

  1.17 It may be more difficult to appreciate that there are necessary connections between logic and truth and why these connections make it so essential to argue validly and to avoid contradiction. For did not this chapter itself begin by insisting that “thinking about thinking is concerned, at least in the first instance, with the validity or invalidity of arguments rather than with the truth or falsity of propositions”? And have we not gone on to assert that arguments may be valid, though both their premises and their conclusions happen to be false, or invalid, though both their premises and their conclusions are in fact true?

  1.18 Yes, this was said. It is all true. But it is also true that, though sound argument and a reasonable appreciation of the available evidence may happen sometimes to lead to false conclusions, no man who is indifferent to argument and to evidence can claim to be concerned for truth. Abraham Lincoln was profoundly right when he wrote, chiding the editor of a Springfield, Illinois, newspaper: “It is an established maxim and moral that he who makes an assertion without knowing whether it is true or false is guilty of falsehood, and the accidental truth of the assertion does not justify or excuse him.” It is also true that to tolerate contradiction is similarly to be indifferent to truth. For people who, whether directly or by implication, knowingly both assert and deny one and the same proposition show by that behavior that they do not care whether they assert what is false and not true, or whether they deny what is true and not false.

  1.19 To grasp this point is to raise a perennial personal challenge. Like all such personal challenges, it should be seen as being at least as much a challenge to me and to us as it is to you and to them. For whenever and wherever I tolerate self-contradiction, then and there I make it evident, either that I do not care at all about truth, or that at any rate I do care about something else more. It was thus precisely because to affirm the premises of a valid deductive argument while denying the conclusion is, by the definition of “valid deductive argument,” to contradict yourself, that Socrates used to demand: “We must follow the argument wherever it leads.”