CLARENDON LIBRARY OF LOGIC AND PHILOSOPHY General Editor: L. Jonathan Cohen, The Queen's College, Oxford
RATIONALITY
...
72 downloads
996 Views
11MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
CLARENDON LIBRARY OF LOGIC AND PHILOSOPHY General Editor: L. Jonathan Cohen, The Queen's College, Oxford
RATIONALITY
The Clarendon Library of Logic and Philosophy brings together books, by new as well as by established authors, that combine originality of theme with rigour of statement. Its aim is to encourage new research of a professional standard into problems that are of current or perennial interest. General Editor: L. Jonathan Cohen, The Queen's College, Oxford. Also published in this series Quality and Concept by George Bealer Psychological Models and Neural Mechanisms by Austen ClaTk The Probable and the Provable by L. Jonathan Cohen The Diversity of Moral Thinking by Neil Cooper The Metaphysics of Modality by Graeme Forbes Interests and Rights: The Case Against Animals by R. G. Frey The Logic of Aspect: An Axiomatic Approach by Antony Galton Ontological Economy by Dale Gottlieb Equality, Liberty, and Perfectionism by Vinit Haksar Experiences: An Inquiry into some Ambiguities by J. M. Hinton The Fortunes of Inquiry by N. Jardine Metaphor: Its Cognitive Force and Linguistic Structure by Eva Feder Kittay Metaphysics and the Mind-Body Problem by Michael E. Levin The Cement of the Universe: A Study of Causation by J. L. Mackie Divine Commands and Moral Requirements by P. L. Quinn Simplicity by Elliott Sober The Logic of Natural Language by Fred Sommers Blindspots by Roy A. Sorensen The Coherence of Theism by Richard Swinburne Anti-Realism and Logic by Neil Tennant The Emergence of Norms by Edna Ullmann-Margalit Ignorance: A Case for Scepticism by Peter Unger The Scientific Image by Bas C. van Fraassen The Matter of Minds by Zeno Vendler Chance and Structure by John Vickers What is Existence? by C. J. F. Williams Works and Worlds of Art by Nicholas Wolterstorff
RATIONALITY A Philosophical Inquiry into the Nature and the Rationale of Reason Nicholas Rescher
C L A R E N D O N PRESS • OXFORD 1988
Oxford University Press, Walton Street, Oxford OX2 6DP Oxford New York Toronto Delhi Bombay Calcutta Madras Karachi Petaling Jaya Singapore Hong Kong Tokyo Nairobi Dar es Salaam Cape Town Melbourne Auckland and associated companies in Berlin Ibadan Oxford is a trade mark of Oxford University Press Published in the United States by Oxford University Press, New York © Nicholas Rescher 1988 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, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press British Library Cataloguing in Publication Data Rescher, Nicholas, 1928Rationality: a philosophical inquiry into the nature and the rationale of reason. (Clarendon library of logic and philosophy). 1. Rationality. I. Title 153.4'3 ISBN 0-19-824435-5 Library of Congress Cataloging in Publication Data Rescher, Nicholas. Rationality: a philosophical inquiry into the nature and the rationale of reason I Nicholas Rescher— (Clarendon library of logic and philosophy). Includes index. I. Reason. 2. Reasoning. 3. Rationalism. I. Title. II. Series. B833.R47 1988 128'.3—dcl9 88-15191 ISBN 0-19-824435-5 Set by Hope Services, Abingdon Printed in Great Britain at the University Printing House, Oxford by David Stanford Printer to the University
For Dorothy In gratitude
Preface This smallish book broaches a very large topic—the nature and status of rationality. It explains and defends the thesis that rationality consists in the intelligent pursuit of appropriate objectives. And it argues that cognitive, pragmatic, and evaluative rationality constitute a unified and indissoluble whole in which all three of these resources are inseparably co-present. Good reasons for believing, for evaluating, and for acting go together to make up a seamless and indivisible whole. More specifically, the book deals with the following issues: 1. the mechanics of reason (What is rationality and how does it work?) 2. the rationale of reason (Why be rational? What justifies a reliance on rationality? How is it validated?) 3. the rewards of reason (Does rationality pay? Are rational people happier?) The discussion seeks to clarify the workings of rationality and to provide a clear overview of this important resource, whose possession differentiates Homo sapiens from the remaining inhabitants of the planet and links our species with those other creatures in the cosmos (if such there be) who also possess the resource of reason. In writing this book I have one central objective—to protest against an overly narrow conception of what 'rationality' is all about. When cultivating the limited concerns of a particular discipline it is easy to lose sight of how complex and many-sided rationality is, and there has been a widespread tendency to take an over-narrow view. For the logician, the avoidance of inconsistency is seen as rationality's be-all and end-all. For the economist, it is efficiency in the pursuit of chosen objectives. For the decision theorist, it is correct cost-benefit calculation. Every specialty seems to opt for some narrow desideratum as the definitive feature of reason. In fact, however, rationality is something far-reaching and much-inclusive, and not merely a particular delimited good to
viii
Preface
be achieved by narrow technical means that conveniently lie within the scope limitations of a particular discipline. Properly construed, rationality is as wide-ranging and complex as the domain of intelligence at large. The central thesis of the book is that only a normative theory of rationality can be adequate to the complexities of the subject. In so far as this point is correct, it has an obvious and important consequence. For it means that the various social scientists— psychologists, economists, and decision theorists alike—who want on the one hand to present a theory of rationality while at the same time avoiding the vexing complexities of normative deliberations are engaged in a futile venture and condemned from the very outset to an inappropriate view of the rational enterprise. I began work on this project in Oxford in the summer of 1983, but then had to put it aside because of other commitments until the following summer there, when I managed to complete a first draft. This was reworked in odd moments in the course of the ensuing academic year, and subsequently completed in Oxford during the summers of 1985 and 1986. Like many other products of contemporary American philosophy, this book is indebted to influences astir on both sides of the Atlantic. I am grateful to Linda Butera and Christina Masucci for preparing a workable manuscript through many difficult revisions. And I am obliged to James Allis and to Martha Harty Scheines for reading a draft version of the book and helping me to improve its presentation. One final matter. In using pronouns of reference to generic, otherwise unidentified, agents or persons, I follow the established, yet eminently problematic, custom of employing the masculine form ('he', 'his', etc.) instead of the more appropriate but complex 'he or she', 'his or her'. This saves time for author and reader alike, and money for the purchaser. But let it be said once and for all that no comfort to sexism is intended, and that the reader is authorized and, indeed urged to make a mental substitution of the more complex, but more appropriate, locution. Moreover, by way of compensation, and as a token of bona fides, I have treated the protagonist of the discussion—namely reason—as feminine. In this book, as in most Indo-European languages, reason is a she. N.R. Pittsburgh, PA September, 1987
Contents
1. The Range of Rationality 1.1 Rationality as a Human Resource 1.2 Rationality and Intelligence 1.3 Ramifications of Rationality 1.4 The Systematic Nature of Reason
1 1 2 9 16
2. The Predicament of Reason 2.1 Stage-Setting for the Problem 2.2 Optimum-Instability 2.3 Implications of Optimum-Instability 2.4 Ideal vs. Practical Rationality: The Predicament of Reason
19 19 22 25
3. The Rationale of Rationality: Why Follow Reason? 3.1 The Problem of Validating Rationality 3.2 The Pragmatic Turn: Even Cognitive Rationality has a Pragmatic Basis 3.3 The Self-Reliance of Rationality is not Viciously Circular
33 33
4. Cognitive Rationality and Risk: A Critique of Scepticism 4.1 Is Cognitive Rationality Possible? 4.2 Scepticism and Risk 4.3 The Deficiency of Scepticism 4.4 Against Scepticism: The Pragmatic Dimension 4.5 Scepticism vs. Rationality 5. Cognitive Rationality and Consistency 5.1 Consistency: Absolute Requisite or Ultimate Ideal? 5.2 Preliminaries: Two very Different Sorts of Acceptability 5.3 Motivating the Acceptance of Inconsistent Theses 5.4 Linearly Inferential vs. Dialectically Cyclic Reasoning 5.5 Context and Dialectics 5.6 Consistency as a Cognitive Desideratum
27
39 42 48 48 54 61 64 69 73 73 76 79 83 87 90
x
Contents
6. The Rationality of Ends 6.1 A Critique of the Humean Conception of Reason: Rational Action requires Appropriate Ends 6.2 The Crucial Role of Interests: Wants and Preferences are not Enough for Rationality 7. Economic Rationality and Problems of Utility Maximization 7.1 The Problem of Utility 7.2 Preference vs. Preferability 7.3 Difficulties in Orthodox Decision Theory 8. The Systemic Unity of Reason 8.1 Practical Reason requires Cognitive and Evaluative Reason 8.2 Cognitive Reason has both Practical and Evaluative Dimensions 8.3 Evaluative Reason has Factual and Practical Dimensions 8.4 The Systemic Unity of Reason 8.5 The Parallelism between Rational Inquiry and Evaluation 9. Conceptual Egocentrism and the Limits Of Cognitive Relativism 9.1 Rationality and Cognitive Relativism 9.2 The Basis Problem 9.3 The Local Absoluteness of our own Standards of Rationality 9.4 More on Criteriological Egocentrism: The Primacy of our own Standards 9.5 Limits to Cognitive Relativism: Metacriterial Monism
92 92 97 107 107 111 115 119 119 122 125 126 128 133 133 136 139 144 149
10. The Universality of the Rational 10.1 The Universality of Reason 10.2 Cultivation Hierarchies 10.3 Objectivity and Consensus 10.4 Objectivity and Cognitive Consensuality Contrasted: The Import of the Ideal
157 157 163 169
11. The Rationality of the Real 11.1 The Intelligibility of Nature 11.2 'Our'Side 11.3 Nature's Side
176 176 180 182
172
Contents 11.4 A Recognition of Limits 11.5 A'Principle of Sufficient Reason'?
XI
188 188
12. Rationality and Humanity 12.1 People as Rational Agents: The Presumption of Rationality 12.2 On Failing to do the Rational Thing 12.3 Against the Greeks 12.4 Rationality as a Duty: Ontological Obligation and the Imperative to Reason
191
13. Rationality and Happiness 13.1 Are Rational People happier? 13.2 Two Modes of 'Happiness' 13.3 More on the Affective Rewards of Rationality 13.4 Is Rationality Inhumane? 13.5 Reason as a Basis for Reflective Happiness
210 210 211 216 218 223
14. Conclusion: The Grandeur and Misery of Reason 14.1 The Grandeur of Reason 14.2 The Autonomy of Reason 14.3 The Misery of Reason
224 224 227 229
Name Index Subject Index
231 233
191 194 200 204
1 The Range of Rationality
SYNOPSIS (1) Rationality consists in the intelligent pursuit of appropriate ends. It pivots on the use of intelligence or reason, the crucial survival instrument of the human race, in the management of our affairs. (2) The three main contexts of rationality are the cognitive, the practical, and the evaluative. All are united in the common task of implementing 'the best reasons'—reasons for belief, action, and evaluation, respectively. In each case, rationality requires the use of intelligence for optimizing—that is, for figuring out the best thing to do in the circumstances. Good reasons must be both cogent in themselves and, comparatively, the best available, referring to the real interests of the agent rather than mere wants. (3) Although the rational resolution of an issue depends on the contextual circumstances, nevertheless, rationality is universal in the sense that anyone in just the same circumstances would be rationally well advised to adopt the same resolution. (4) Above all, reason is systematic: it requires us to pursue intelligently adopted objectives in intelligent ways, acting on principles that make sense in a systematic way and whose appropriateness other intelligent agents can in principle also determine.
1.1
RATIONALITY AS A HUMAN R E S O U R C E
The ancients saw man as 'the rational animal' (£,£>ov koyov'txov), set apart from other creatures by capacities for speech and deliberation. Under the precedent of Greek philosophy, Western thinkers have generally deemed the use of thought for the guidance of our proceedings to be at once the glory and the duty of Homo sapiens. Rationality consists in the appropriate use of reason to resolve choices in the best possible way. To behave rationally is to make use of one's intelligence to figure out the best thing to do in the circumstances. Rationality is a matter of deliberately doing the best one can with the means at one's disposal—of striving for the best results that one can expect to achieve within the range of
2
The Range of Rationality
one's resources—specifically including one's intellectual resources. Optimization in what one thinks, does, and values is the crux of rationality. Rationality calls for the intelligent pursuit of appropriate ends. It is a matter of the recognizably effective pursuit of appropriately appreciated benefits. Rationality thus has a crucially economic dimension, seeing that the impetus to economy is an inherent part of intelligent comportment. Costs and benefits are the pivotal factors. Be it in matters of belief, action, or evaluation, rationality demands a deliberate endeavour to optimize benefits relative to the expenditure of available resources. Reason requires the cultivation of intelligently adopted objectives in intelligent ways. Rationality is not an inevitable feature of conscious organic life. Here on earth, at least, it is our specifically human instrumentality, a matter of our particular evolutionary heritage. Rational intelligence—the use of our brains to guide action by figuring out what is the apparent best—is the survival instrument of Homo sapiens, in much the same way that other creatures have managed to ensure their survival by being prolific, or tough, or well sheltered. It is a means to adaptive efficiency, enabling us—sometimes at least—to adjust our environment to our needs and wants rather than the reverse. But the fact that rationality forms part of our human condition does not automatically validate it. In particular, it does not establish that, while reason has served us well in the past, we might not now do even better by turning elsewhere. Against this background, the principal task of the present book is to clarify the rationale of rationality, and to answer that provocative and fundamental question: 'Why be rational? Are there good reasons for it?'
1.2
RATIONALITY AND I N T E L L I G E N C E
Reason can (and should) come into operation whenever we are in a position to decide what to do—whenever a choice or decision confronts us. Philosophical tradition since Kant sees three major contexts of choice, those of belief, of accepting or endorsing theses or claims, of action, of what overt acts to perform, and of evaluation, of what to value or disvalue. These represent the
The Range of Rationality
3
spheres of cognitive, practical, and evaluative reason, respectively (cf. Display 1.1). DISPLAY
1.1: Objects of Rational Deliberation
1.
C O G N I T I V E R A T I O N A L I T Y : What to believe or accept? What to maintain regarding states of affairs in both the formal and the empirical domains. Product: factual contentions (beliefs). 2. PRACTICAL R A T I O N A L I T Y : What to do or perform? What to decide regarding actions. Product: action recommendations (injunctions). 3. EVALUATIVE R A T I O N A L I T Y : What to prefer or prize? What to adopt regarding goals and ends. Product: evaluations (appraisals).
1 Rationality is not just a matter of thought, but of action as well.The person who proceeds unintelhgently in figuring out the proper thing to do thereby commits a rational fault. But the person who figures out correctly the rationally optimal thing to do in the circumstances and then fails to do it thereby also strays from the path of reason. Whatever sort of 'doing' may be at issue—be it in belief, action, or evaluation—it must be done appropriately if the demands of reason are to be met. One can characterize many different sorts of things as rational— not beliefs, actions, and evaluations alone, but also people, plans, arrangements, and so on. But these are all derivative uses. Rational people are people whose beliefs, evaluations, and actions are on the whole rational; rational plans are plans based upon rational beliefs, assessments, and actions; rational arrangements are those based on rational plans, and so on. The entire fabric of the conception of rationality is spun from the threads of rational belief, evaluation, and action. Someone acts rationally in matters of belief, action, and evaluation when his reasons are cogent reasons. Rationality thus involves the capacity 'to give an account' 1 —to use one's intelligence to provide a 'rationale' for what one does that establishes its appropriateness. It is a matter of conducting one's affairs responsibly—of being able to provide an account of (sufficiently) ' \6yov SiSdvat as Plato puts it in the Republic; cf. Koyov Hxoiv in the Aristotelian formula.
4
The Range of Rationality
telling reasons for what one does in a way that will enable people 'to see the point' and to accept that it makes good sense to proceed as one did. A belief or evaluation or action is accordingly rational if the agent can tell a story that succeeds in making sense of it by showing that—and how—it was a sensible thing to arrive at in the circumstances, due to its optimal conduciveness to appropriate ends. Where something is amiss in an agent's management of his beliefs, evaluations, or actions—where his information or his 'appraisals' or his decisions are inappropriate in the circumstances —there is a failure of rationality. Although rationality functions in all three of the very different domains of cognition, action, and evaluation, it is at bottom one and the same thing that is at issue throughout: effecting the resolution of choices in the best possible way—in line with the strongest reasons. Rational people are, ex officio, in a position to provide a rationale of good reasons for what they do—which is why the range of rationality has, since the days of the philosophers of ancient Greece, generally been deemed coextensive with the linguistically equipped intelligence needed for the articulation of warranting considerations—of reasons. Rationality accordingly pivots on the deployment of 'good reasons': I am being rational if my doings are governed by suitably good reasons—if I proceed in cognitive, in practical, and in evaluative contexts on the basis of cogent reasons for what I do. And the question of motivation is a crucial aspect of rationality; as with morality, it is a matter of doing the right things for the right reasons. (If someone does what is, in the circumstances, the intelligent thing to do, but does it simply by accident or by whim, he is not thereby comporting himself rationally.) To be sure, the reasons that support beliefs, actions, and evaluations may rest on rather general principles and lack any sort of fine-grained particularity. My 'good reason' for taking this medication is simply that the doctor prescribed it; my 'good reason' for believing that the population of Calcutta exceeds five million is simply that the encyclopaedia says so. And that is quite good enough to support the rationality of my beliefs in the circumstances; belief-formation at a remove from the sort of substantive reasons at issue in firsthand information still qualifies as rational.2 2 Note the difference between 'x knows that there is a good reason for his accepting p' and 'There is a consideration known to x which (to his knowledge)
The Range of Rationality
5
A rational agent's 'reason' for taking a certain step (adopting a belief or performing an action or making an evaluation) is a consideration or line of thought which provides this agent with a justifying ground for taking this step, and which can therefore— in the agent's own view—serve to explain or validate it. Of course, not all reasons for actions are good reasons in the sense of being cogent—that is, of such a sort that they would move someone who proceeded in an intelligent and sensible way. To do something rationally is to do it for good and cogent reasons. And this is not the same as just having a motive for doing it. All of us almost always act for motives, but valid reasons are (ex hypothesi) what motivate the rational agent, and most of us do not act rationally all of the time. All too often we are moved to what we do by desires or wants, and these may or may not be rationally well advised. The crux is that it may or may not be in one's best interests to get what one desires—that very much depends on exactly what it is that one happens to want. People automatically have a motive whenever there is a desire, but they only have a good reason for what they do when it is recognizably in their best interests.3 The gambler has reasons for persevering in squandering his resources ('it gives him a perverse sort of pleasure'), and the revenge-seeker also has reasons for stalking his victim. The mere fact that one wants something—that it accords with one's desires —is certainly some reason for opting for it, and provides a 'ground' of sorts. But such wilful agents are 'rational' only in potentiality and not in act. For mere unevaluated desires can provide us with 'reasons' for acting that are not necessarily anything like sufficiently good reasons. From the rational point of view, our mere wants have little significance. They can and should be outweighed by our interests and our needs. A 'minimally' rational agent does what he does for reasons, never mind for the moment whether they be good, bad, or indifferent. But a truly rational agent has good reasons for what he does and acts in such a way as to be well advised in so proceeding. provides a good reason for his accepting p.' The former answers to the pattern Kx(3q)[q & (qRp)] and the latter to the pattern (3q)Kx[q & (qRp)). 3 The idea that mere desires as such are not automatically reason-providing is stressed in Stephen Darwall's recent book, Impartial Reason (Ithaca, 1985). Strange to say, however, the conception of a real interest (which, after all, automatically does provide a good reason for doing something) does not figure much in Darwall's discussion. (The word 'interest' is in fact absent from his index.)
6
The Range of Rationality
Good reasons are those whose guidance optimally serves our real or best interests in the matters at issue. What makes a reason into a good reason is the fact that its implementation leads our efforts in appropriate directions, and the best reasons are those that achieve the most in this way. The rational thing to do is always the intelligent thing—that is, the adoption of the most cost-effective means towards the realization of one's appropriate goals. One can virtually always 'rationalize' something one wants to do, putting it in a favourable light with respect to rationality by finding some reasons for doing it.4 But that by itself does not of course suffice to render the act in question a rational one. For, despite its being supported by some reasons, further, still better, available reasons may well point another way. Rationality is not just a matter of having some reasons for what one does, but of aligning one's beliefs, actions, and evaluations effectively with the best or strongest available reasons. It pivots on doing that which, everything considered, one is 'well advised' to do. The matter of giving or following the course of intelligent and 'responsible' advice is the crux of rationality. To re-emphasize: rationality is a matter of seeking to do the very best we can (realistically) manage to do in the circumstances. To proceed contrariwise by way of 'rationalizing'—by putting reasons for what we want in the place of good reasons for what we ought to do—is an abuse of rationality. Rationality thus calls for appropriate resolutions intelligently arrived at and sensibly implemented. It is geared to the sensible pursuit of appropriate ends. (Here 'ends' is used as a generic term covering both concrete goals and more diffuse values.) Rationality is thus a two-sided, Janus-faced conception. On the side of means, it reflects a pragmatic concern for efficient process, while on the side of the 'appropriateness of ends' it reflects a value-geared concern for product. (Moreover, the acceptability of the means themselves also enters in.) Cognitive reason has historically been seen as the capacity to grasp facts through their underlying rationale by discerning 'the reason why' that lies behind them. It is thus reason that enables us to grasp the substantiating connections between facts (facultas nexum veritatum percipiendi, knowledge of the 'enchainement des 4 For a useful discussion see Robert Audi, 'Rationalization and Rationality', Synthise, 65 (1985), 159-84.
The Range of Rationality 5
7
verites' as Leibniz puts it. The use of reason involves maintaining the systemic linkages between diverse but connected items. 'Treat like case alike' is a salient principle of reason. The impetus to order, coherence, and system is inherent in its operations. The rationality of a particular belief—of accepting a particular contention—is of course a function of the evidence at one's disposal. (The Siamese king who refused to believe that rivers solidify in northern countries at a certain season of the year was perfectly rational, the freezing of water into ice lying wholly outside his experience.) Rationality is a matter of doing things in the best (most intelligent) way one can manage in the circumstances. Rationality is realistic; it does not require more than is possible. Just as a truthful person does not necessarily tell the truth as such, but tells the truth as he sees it—speaking on the basis of his information (which may be imperfect), so a rational person proceeds on the basis of the grounds that are available to him (which may well also be imperfect). The rational person is by definition someone who, in effect, uses his intelligence to maximize the probability—that is, the responsibly-formed, subjective probability—that matters will eventuate favourably for the realization of his real (or 'best') interests. Both cognition and evaluation are accordingly crucial. It is silly to dedicate shrewd means to inappropriate ends; it is foolish to pursue worthy ends by ineffective means. Throughout, rational appropriateness is a matter of determining in which direction the best (or strongest) reasons point. All domains of rationality are thus united by the common mission of finding 'the best reasons'. Rationality's cardinal injunctions are: 'Adopt what is, in the circumstances, the best available option! Pursue it in the way that is, in the circumstances, the best possible way!' The idea of optimization—of seeking for the best alternative —lies at the very core of rationality. The ancient Greek idea that vovs, or reason, is a force operating in the world to make things work out 'for the best', already codifies this critical linkage between rationality and optimality.6 While rationality calls for striving for 'the best' solution, this is frequently not a straightforward matter. Often several distinct 5 6
G. W. Leibniz, Nouveaux Essais, iv. 17, sect. 4. Cf. Plato, Timaeus, 41d4, 53b2-4.
8
The Range of Rationality
parameters of merit are operative—say, plausibility, probability, and naturalness (simplicity) in matters of explanation—and they may yield rather different modes of optimality. Generally, we should speak of a rational (and a best) solution of our problems rather than simply the rational (or the best) solution. There often is no uniquely best alternative and, when this is so, rationality leaves us with options. It must be stressed that while rationality requires us 'to do our very best' this is a matter of 'our best'. It is not 'the best that can be done' as distinct from 'the best that can be done by us, given the capacities and resources at our disposal'. Rationality does not make demands beyond the limits of what is genuinely possible for us—it does not require accomplishments beyond the limits of the possible. For rationality, no more is demanded of us than doing our realistic best to work efficiently and effectively towards the realization of our cognitive, practical, and evaluative goals. Rationality is, indeed, a matter of optimization—of doing the best we can by implementing the strongest reasons—but this must be understood as the best we can feasibly manage to do in the existing circumstances. Some problems are simply too complex to admit of a perfect resolution by the means actually at our disposal. Hence, rationality asks for no more than that we do the best we can manage with the means in hand. Being realistic is a part of being rational. To be sure, rationality is not just a passive matter of making good use of the materials one has on hand—in cognitive matters, say, the evidence in view. It is also a matter of actively seeking to enhance these materials: in the cognitive case, by developing new evidential resources that enable one to amplify and to test one's conclusions. The endeavour to make the most of one's opportunities is an aspect of intelligence that is crucial to rationality. (To rest content with unquestioned habit, for example, is a defect of intelligence that is not consonant with rationality.) The logician takes consistency to be the bulwark of rationality, the scientist evidential cogency, the economist efficiency. All are right, but each only partially so. Each focuses upon what is no more than a part of reason—that one particular aspect of intelligent procedure that is of primary importance for his own field. The crucial point is that rationality as such is something complex and many-sided—though all of its parts are embraced
The Range of Rationality
9
within the one, overall generic formula that rationality consists in the intelligent conduct of one's affairs. There is a difference between rationality and reasonableness— between being rational and being willing to 'listen to reason'. For it is not necessarily rational to 'be reasonable'—sometimes the best means to appropriate ends lie in terminating 'mere discussion' (in the training of children, for example, or in certain negotiations). Sometimes, moreover, it is clever—and indeed altogether rational—to act in an 'irrational' way. Think of the suitor who feigns indifference, or of the tactician who makes occasional 'foolish mistakes' to keep the opponent off balance with his unpredictability. This sort of thing may well prove rational after all, since there are perfectly good reasons for proceeding in such a manner as to fail, in certain circumstances, to do the 'rationally appropriate' thing. A deeper rationality may, on occasion, countermand the obvious-seeming requirements of reason. Sometimes 'rationality' is contrasted with 'feeling', and 'reason' with 'human sympathy'. But this overly cerebral conception represents a far too narrow and one-sided view of reason's domain. Rationality is broad and comprehensive. Feelings are generally not a matter of reasoning, but they are certainly not outside the province of reason. As Pascal saw, 'the heart too has its reasons, which are unknown to mind'. Feelings too can provide reasons and canalize the operations of intelligence. The human spirit extends beyond the human body that determines our 'material' interests and the human mind that determines our 'cognitive' interests. (Man lives not by bread or by knowledge alone—nor even by their combination.) Neither man's material interests nor yet his cognitive interests exhaust the realm of appropriate values. Reason herself recognizes the utility and appropriateness of our 'higher' (aesthetic and affectively social and even 'spiritual') values. The realm of rationality is as large and comprehensive as the domain of valid human concerns and interests.
1.3
RAMIFICATIONS OF RATIONALITY
Rationality makes demands upon us. It speaks in didactic tones: this or that is what you should do. Its declarations have a
10
The Range of Rationality
normative force, enjoining us as to how we ought to go about settling questions of what to believe, do, or value. 'These are the strongest reasons for belief, action, and evaluation—be governed by them!' Reason addresses us in the imperative mood in impelling us to meet her requirements. The rational man is committed to an optimal use of his resources—to doing as well as he possibly can with the inevitably limited means at his disposal. (He does not go out and buy the best available car, but settles for the best that fits into his overall budget.) In matters of belief, action, and evaluation we must proceed within the budget of our intellectual and other resources (of time, effort, finances, and ability), doing the best we can in the full recognition that this may not lead us to the ideally best. A rational person is someone who is impelled by reason in what he believes, does, and values—who endeavours to let all his proceedings be governed by, and shaped with a view to, the strongest reasons. Being rational consists in the disposition to make good reasons constitute the motives for what one does. Since this is something we can achieve only within limits, one must regard perfect rationality as an idealization and acknowledge that we humans are 'rational animals' because of our capacity for reason, and certainly not because of our achievement of perfected rationality. The fundamentals of rationality are the same on all sides, alike whether the 'doing' at issue is a matter of belief or evaluation or action: , ,r . A r 1. X does A 2. X has good, and indeed X does A rationally = \ overridingly good, reasons for doing A 3. X does A on the basis of these reasons That 'overridingly' of the preceding formula requires comment. We only have overridingly good reasons for doing something if our reasons for doing it are better (stronger) than our reasons for not doing it—and thus better than those for doing any one of its alternatives. This stress on overridingly good reasons reflects rationality's concern for optimization—for doing the very best we can in the prevailing conditions. What is crucial for rationality throughout is proceeding on the basis of preponderating reasons.
The Range of Rationality
11
Accordingly, rationality in all its forms calls for the comparative assessment of feasible alternatives, and so demands five faculties: 1. Imagination: the capacity to contemplate alternatives—to entertain alternative possibilities, project hypotheses, and perform 'if-then' thinking. 2. Information-processing: the capacity to determine what can and cannot be done—to map the boundaries of feasibility, so as to determine what abstract possibilities are actually realizable, what theoretical prospects are realistic alternatives. 3. Evaluation: the capacity to assess how desirable various sorts of situations are—to appraise alternatives. 4. Selection—Informed Choice: the capacity to effect a choice between alternatives, to adopt one and reject others—and to do this in the light of the considerations at issue in 1-3 above. 7 5. Agency: the capacity to implement choices. As the last two entries indicate, rationality is not purely intellectual, but also involves the will, the capacity of choice. If—as seems sensible—we define 'free-will' as the capacity for intelligent choice among alternatives, then it follows that only a creature endowed with free-will can be rational. Moreover, for rational agency—as opposed to rational judgement as such—the capacity to implement one's choices in action is also necessary. This ability separates rational agency from mere intelligence as such, thus setting persons apart from mere intelligences. (In consequence it makes for a gap between the 'artificial intelligence' of robots and fully endowed, rational creatures. For as long as an intelligence can only solve given problems, and cannot freely select its own problems and define the agenda of its own operations, it will not qualify as a rational agent.) These resources of imagination, cognition, evaluative appreciation, and judgementally guided choice are essential to any mode of rationality, while for practical rationality the capacity for action is also a requisite. A creature lacking any of the preceding capacities will not qualify as a fully rational being or person. Reason is something many-sided: only a complex creature can be rational. 7
Cf. Frederic Schick, Having Reasons (Princeton, 1984), esp. chap. 2.
12
The Range of Rationality
This complexity is mirrored in rationality itself. Overall, it has the structure set out in Display 1.2, and accordingly has two major divisions: the theoretical (which includes the cognitive and the evaluative sectors) and the practical or pragmatic. Being rational is a matter of doing the intelligent thing, where 'doing' is neutral between thought (mental doing, like accepting or prizing) and action (actual behavioural doing). DISPLAY
1.2: A Taxonomy of Reason
I. From the Angle of Inputs T H E O R E T I C A L : no action directives are (essential) inputs (purely) FACTUAL: only (purported) facts are inputs VALUE-INVOLVING: values are among the inputs PRACTICAL: action directives are among the inputs II. From the Angle of Outputs COGNITIVE: (purported) facts are outputs NORMATIVE: values or action directives are outputs EVALUATIVE: values are outputs PRAGMATIC: action directives are outputs Consider the following four theses: 1. A is the rationally appropriate thing for X to do (to perform or to believe or to value) in the circumstances. 2. X does ,4. 3. X does A out of a recognition that 1 is the case. 4. X acts rationally in doing A. Note the following points: A. We do not have it that: (1 & 2) -> (4). For if X fails to realize that A is the rational thing to do—indeed may even think that not-A is the rational thing to do—but nevertheless does A (say simply to please someone), then X is not acting rationally in doing A. B. We certainly do have it that: (1 & 3) -» (4).
The Range of Rationality
13
This inference is unproblematically valid. For here X {by hypothesis) does the rationally appropriate thing for a rationally cogent reason. C. However, it is not the case that 3 affords the only route from 1 to 4. For example, if X has been trained (habituated) to do A because of its appropriateness, and does it 'automatically' in the suitable circumstances, then we have a perfectly viable alternative route from 1 to 4. D. As we have seen, moreover, 4 is perfectly compatible with not-1. If in his prevailing epistemic circumstances the agent has every reason to believe that 1 is so, even though it in fact is not, for reasons he does not know and cannot reasonably be expected to know, then other things being equal, he will act altogether rationally in doing A. Point (C) deserves special emphasis. Rational procedure can be a matter simply of good insight rather than one of discursive inference—it can issue from recognizing a good reason rather than from arguing one's way to it by some inferential process. Rational agency requires intelligent behaviour. But it does not always require deliberation—Aristotle to the contrary notwithstanding. 8 A drowning man swirls in the rapids. He sees two branches coming within reach. Straight away, without any further thought, he clutches one of them. He certainly does something intelligently—and perhaps even, in a way, 'deliberately'—namely 'clutch a branch'. But he does not just 'clutch a branch'—he grasps this one rather than that one. But at that point further reasoning and deliberation do not (and ought not to) come into it. Deliberation is a luxury our drowning man just cannot afford. There must always be good reasons why a genuinely rational act was done as it was, and indeed the agent must have such reasons in an at least tacit or implicit way. But this recognition of reasons can be unconscious, as it were. I am tacitly aware of what I am doing when I 'instinctively' grasp that branch. I do not stop to think and deliberate. But all the same I acted rationally. In such a case it is in fact altogether rational not to deliberate. Agents can unquestionably act in an altogether rational way without any actual reasoning on their part on the spot, then and there. Sometimes rationality demands doing something straight off; it becomes necessary 'to get in there and do something', without any prior deliberation whatsoever. This, of course, is itself a 'lesson of reason'. In such cases we are not going against reason, 8
Aristotle, Nicomachean Ethics, III. ii. 15.
14
The Range of Rationality
we are being 'perfectly rational'. Again, there is sometimes simply no need to deliberate further. The sick man must take one of those pills from the bottle. Once he realizes that he must do so, there is no point in deliberating further about which one. It just does not matter. Then, too, the particular case at hand is sometimes covered by a general policy that is already in place. When I enter the car, I put the seat belt on automatically. It is doubtless rational to do so, but this is not something I need to deliberate about at this point, having (sensibly) schooled myself to do it in the past. In all such cases one acts—even without concurrent reasoning—out of a pre-established recognition that there is a good reason for what one does. While rationality is a matter of implementing the best of (available) reasons, this can be done, quite appropriately, through policy via habit rather than by overt intention entertained explicitly on the spot. Still, typically and quintessential^ rationality operates through explicit deliberation in the face of recognized alternatives. The cultivation of appropriate ends by suitable means is the life-blood of rationality. That is why foresight and the systematic coordination of actions into the framework of a coherent plan are crucially important instruments of rational action. Considering that reason-guided free agency is essential to being a person, the question seemingly arises: is one a better person in so far as one is more rational in one's actions? The answer is: 'Yes— in a way.' For the more rationally one acts, the more fully one realizes—other things being equal—the ideals and values at issue in being a person. For, it is this normative aspect that is crucial. To ask the question in view is not like asking: Given that having a backbone is essential to being a mammal, is one a better animal in so far as one has a larger backbone? For mammalhood—unlike personhood—is not a normative conception for which 'better mammalhood' makes sense. However, as an inherently normative conception, personhood admits of fuller or lesser degrees of realization. One can do better or worse at exercising and actualizing the sorts of capacities that constitute the characterizing requisites for being a person—rationality among them. The methods we use in cultivating rationality change in the light of the experience we have with them.9 New cognitive tools 9
See the author's Methodological Pragmatism (Oxford, 1977).
The Range of Rationality
15
(statistics, for example) can amplify the reach of cognitive reason, for example. But the basic commitment to the intelligent pursuit of intelligently adopted objectives remains the steadfast, guiding beacon for rational endeavour. Rationality as such is something fixed—its nature is constant. But while rationality itself is something stable, the course of action it requires of us changes with circumstances and conditions. Suppose, for example, that we are caught up in a particular interaction situation of the sort described in Display 1.3, fully recognized as such by both of us. In this situation, I do indeed have 'a (good) reason' for not doing A, since this might result in my most preferred alternative—and indeed represents the only route to this desideratum. But this is clearly not a wise step, since I would do well to reflect as follows: 'From your angle, doing B is clearly the best choice: it yields a preferred result for you regardless of how I make my choice. Hence, I should count on your doing B (once I credit you with enough intelligence to figure out that this is how the land lies). And therefore I should myself do A, which would in any case avert worse coming to worst in the realization of my least-preferred alternative.' The supposed 'good reason' was simply not good enough. Rational choice in a given situation generally requires a consideration of the wider context. DISPLAY
1.3: An Interaction Situation
I do A?
You do B?
My preference ranking
Your preference ranking
yes yes no no
yes no yes no
3 2 4 1
2 4 1 3
No particular concrete course of action is ever rational or irrational as such. For, it will always be possible that a situation may arise in which some seemingly irrational action is 'the lesser evil' because all the available alternatives make even less sense. Only if we characterize an action in a normatively 'loaded' way ('taking a needless risk', 'making an ill-considered choice') can we assure its irrationality. With neutrally described actions, however, a closer scrutiny of situational details will always be necessary to warrant a verdict of irrationality.
16
The Range of Rationality 1.4
THE SYSTEMATIC NATURE OF REASON
As already emphasized, rationality calls for proceeding on the basis of good reasons for whatever we do (be it in matters of belief, action, or evaluation). It consists in one's being in a position to provide, in principle, a cogent account of one's proceedings—an account that would convince other rational beings of the appropriateness of what one does. This demand for an account that is both cogent and coherent endows reason with a certain holistic and systemic character, seeing that local resolutions must be embedded in a global framework. Rationality is governed by certain pervasive 'desiderata of reason' that bind together the proceedings of different rational agents—and the proceedings on different occasions of a single agent—under the aegis of overarching principles. Above all, reason requires us to proceed in a way that 'makes sense'. Whether in matters of belief, action, or evaluation, rationality demands: consistency (avoid self-contradiction), uniformity (treat like cases alike), coherence (make sure that your commitments hang together), simplicity (avoid needless complications), economy (be efficient). These 'requisites of reason' require little commentary—they are familiar to the point of triteness. But, for the sake of a single example, let us contemplate coherence. For a rational being deliberately to (1) accept a contention, (2) adopt an end, or (3) make an evaluation, is also to commit himself to (1) accepting those things he deems to be consequences of this contention, (2) accept those things he deems to be means to this end, or (3) make like evaluations of those things he deems to be analogous with this valued item. All these ways of proceeding are simply instances of the demand for coherence in rational procedure. The 'virtues of rationality' are the characteristics of a system. They make for an organic (or systematic) unity of procedure, serving to make sure that everything fits together in an effective and mutually supportive way. Reason is the organizing force in the mental life of an intelligent creature—the orderer of chaotic events into coherent experience. It is the instrumentality by which
The Range of Rationality
17
mind secures its grasp upon a difficult and unstable world. The rules of reason are 'regulative principles'—instructions that keep the conduct of our various affairs on an efficient, effective, and thus intelligent basis. Reason is eminently practical—it wants what works (is efficient and effective). But it pre-eminently also wants what makes sense. The systematic aspect of reason inheres in its drive for intelligibility—in its demand for ways of proceeding whose appropriateness other intelligent agents can in principle perceive. It is in this sense alone that one can maintain the universality of the force of reason: Whatever considerations render it rational for someone to do A will ipso facto render it rational for anyone 'in his circumstances' to do A—anyone placed in conditions sufficiently like his. To act rationally is to do that which one is well advised to do in the circumstances, and good advice is a matter not of persons but of the situations in which they find themselves. This conditional character is a key aspect of the 'universality' of rationality. Sometimes, to be sure, it is very unlikely, or even impossible, that many or most others could find themselves in those particular circumstances. For example, it could happen that certain opportunities for advantageous action are opened up by the circumstance that most others are not doing something. It can be perfectly rational for people to act 'out of step' with the majority; for example, when a fashionable trend leaves a useful niche abandoned. But, clearly, contra-majority action is not something that everyone could do. The fact that something that is rational for someone is rational for anyone else who is 'in his shoes' thus does not mean that it is rational for everyone, full stop. It may (as in this case of productively eccentric action) even be logically impossible for everyone to act in this way at a given time. The question of what we would do if we were 'in someone else's shoes' of course raises the issue of just what are we to take along when we step into those shoes? If we take nothing of our own along, then we are bound to see the issue exactly as he does— there is no other possibility. If we take everything along, then of course he simply vanishes from the scene—his idiosyncratic wants and needs count for nothing. What we have to do to proceed realistically is assess what it would be appropriate to do in conditions 'sufficiently like' his, specifically what it would be intelligent to do in the light of his information and of his situation.
18
The Range of Rationality
It is not a matter of changing the rules of the game on our subject, but one of assessing how well he plays the game by his rules—the rules he does, or ought to, follow, given his valid needs and real interests. What we take along, then, is our native intelligence and our 'common sense'—our ability to think and to judge as sensible people. However, the various substantive commitments on matters of fact and of value are things that we must in large measure take over from the agent at issue. These various perspectives and theses regarding the nature and workings of rationality set the stage on which the deliberations of the book will unfold. Their elaboration and consolidation is the principal task before us.10 10 The article 'Reason' by G. J. Warnock in vol. 7 of The Encyclopedia of Philosophy, ed. P. Edwards (New York, 1967), also gives an overview of our terrain. However, the best encyclopaedia article on rationality in general is that in Jos6 Ferrater Mora, Diccionario de Filosofia, iv. 2774-9 (art. 'Raz6n'). Books well worth consulting include: KurtBaier, The Moral Point of View (Ithaca, 1958); S. T. Benn and G. W. Mortimore (eds.), Rationality and the Social Sciences (London, 1976); Jonathan Bennett, Rationality (London, 1964); Brand Blanshard, Reason and Analysis (London, 1962); George Boas, The Limits of Reason (New York, 1961); Morris R. Cohen, Reason and Nature (New York, 1931); Stephen Darwall, Impartial Reason (Ithaca, 1983); A. C. Ewing, Reason and Intuition (Oxford, 1942); Sigmund Freud, The Future of an Illusion, trans. W. D. Robson-Scott (London, 1928); Arthur E. Murphy, The Uses of Reason (New York, 1943); Ernest Nagel, Sovereign Reason (Glencoe, 1954); Stephen Nathanson, The Ideal of Rationality (Atlantic Highlands, 1985); George Santayana, The Life of Reason (rev. edn., New York, 1954); W. H. Walsh, Reason and Experience (Oxford, 1947); A. N. Whitehead, The Function of Reason (Princeton, 1929). An interesting anthology of recent essays is J. Margolis, M. Krausz, and R. M. Burian (eds.), Rationality, Relativism, and the Human Sciences, (Dordrecht, Boston, Lancaster, 1986). Other references are given in the encyclopaedia articles mentioned above.
2 The Predicament of Reason
(1) The rationally appropriate resolution of a problem arrived at on the basis of one body of information may well become unstuck when that body of information is enlarged. (2) Rationality is a matter of doing the best one can in the circumstances—of optimization relative to situational constraints—and an enlarged experience can always alter those constraints decisively. Rational solutions to problems are ampliatively context-sensitive: enhanced data can lead to altogether different results. (3) But we can never rest assured that the information we actually have is complete in all relevant regards—all we can ever do is to hope or presume that it is so. (4) Consequently, we face the predicament of reason inherent in the circumstance that rationality requires us to do what appears best in the full and clear recognition that this may well fail to be anything like the best thing to do. SYNOPSIS
2.1
S T A G E - S E T T I N G FOR THE PROBLEM
The road to rationality, like that of true love, does not always run smooth. For, his very rationality can lead a rational agent into difficulty. To see how this comes about, consider the following series of problem-solving situations: Case 1: Deductive information-extraction D A T A : A certain quantity x satisfies the following condition: x2 <9. Q U E S T I O N : Which of the following alternatives affords the best, most informative conclusion we can infer about the value of JC? 1. - 2 =£;c =£ 2 2. - 3 < x < 2 3. - 3 < x < 3 4. - 4 < x < 4
20
The Predicament of Reason Case 2: Informational gap-filling
A manuscript note contains the (partly illegible) passage: 'He sent her a 1-tter . . .'. Q U E S T I O N : HOW is that gap in '1-tter' to be filled in? DATA:
Case 3: Probabilistic reasoning D A T A : 1. X is a mechanical engineer. 2. 90 per cent of mechanical engineers are male. Q U E S T I O N : How probable is it that X is male? Case 4: Inductive inference D A T A : A sequence starts 1, 10, 100. Q U E S T I O N : What are we to expect at the 10th place? Case 5: Prudential decision D A T A : 1. It is starting to rain. 2. Yonder large tree affords the only shelter in the large, flat meadow we are crossing. P R O B L E M : T O decide where to go. Case 6: Practical decision D A T A : X i s involved in the following two-person game-situation, where a question mark in the tabulation indicates ignorance regarding the outcome: .u v, b ° Y s Alternatives No. 1
No. 2
„ „. No. 1 0/? -100/? Xs Alternatives _ _y+w NQ 2 y+w P R O B L E M : What course of action should we recommend to X? Case 7: Expert intervention 1. X suffers from asthma. 2. Antihistamines are the most effective available medicament for (most cases of) asthma. P R O B L E M : What to prescribe for XI
DATA:
Case 8: Moral reasoning D A T A : You promised Xto pay someone $10 today. You have the money. P R O B L E M : T O decide whether it is appropriate to repay that $10 or not.
The Predicament of Reason
21
Case 9: Criterion implementation D A T A : I placed my hand on John Jones's shoulder some hours ago and have kept it in place all the time, up to and including the present moment. PROBLEM: To determine who the person is on whose shoulder my hand now rests. In each case, we face a perfectly possible and clearly delineated situation of choice. And in each case the 'rationally appropriate resolution' seems rather obvious and straightforward. But, now consider what happens when some additional, supplementary information comes our way. Let us assume that in these nine cases we come by certain further data: Case 1: We are further given that x is an integer. Case!: The text continues: 'to transport her wounded brother'. Case 3: We are also informed that A'gave birth to a bouncing baby boy last week. Case 4: We are further told that the sequence continues 1,10,100, 1, 10, 100, for the next six entries. Case 5: We are given the supplemental datum that there is also much lightning and thunder. Case 6: Those question mark designated entries are filled in as follows: Y's Alternatives No. 1 No. 2 No. 1 0/+100 -100/-100 X's Alternatives No. 2 -1/+10 -1/+10 Case!: We are also informed that X is highly allergic to antihistamines. Case 8: Supplemental information: you had also promised Y earlier on to pay him $10 today and all of your assets beyond $10 have just been stolen. Case 9: Supplemental information: only an hour ago, they completed a brain-transplant operation on John Jones, installing Sam Smith's brain in his head, and all of the resultant individual's memories, cognitive capacities, psychological dispositions, and other mental traits are now those of Sam Smith.
22
The Predicament of Reason
Clearly, one and the same basic situation recurs throughout. In all sorts of cases, the rational resolution of problems is contextsensitive to the information in hand, in such a way that what is a patently sensible and appropriate resolution in a given datasituation can cease to be so in the light of additional information— information that does not abrogate or correct our prior data, but simply augments it. It is clear that this circumstance pervades all sectors of rational deliberation, regardless of whether we are dealing with deductive information-extraction, inductive inference, evaluative reasoning, practical reasoning, or any other type. The fact is that the rationally appropriate resolution of a problem on the basis of one body of evidence or experience can always become unstuck when that body of evidence or experience is not actually revised but merely enlarged. The ramifications of this fact are pervasive and their implications merit closer attention.
2.2
OPTIMUM-INSTABILITY
The rationally appropriate procedure in problem solving is to strive for the best resolution achievable in the light of the available data.1 Rationality enjoins us to adopt the best available option. Having surveyed the range of alternatives, the rational thing to do is to resolve the choice between them in what is, all considered, the most favourable way. What is 'favourable' will of course differ from context to context. With cognitive choices, 'favourable' primarily hinges on substantiation, with practical choices, on effectiveness, with evaluative choices, on preferability. The particular mode of merit at issue in 'optimality' will differ from domain to domain—and even, to some extent, from agent to agent. The salient fact is that rationality is always a matter of optimization relative to constraints, of doing the best one can in the prevailing circumstances. 1
In some situations (e.g. negotiations, games, and warfare) the intelligent thing at the level of policy or strategy may be occasionally to do a 'stupid' thing at the level of particular acts or tactics. It is sometimes advantageous 'to keep your opponent guessing' by not being too predictable and occasionally doing something unexpected—even though this is a 'stupid' thing to do in the circumstances—as part of a deeper cunning. Nothing said here about rational procedure is intended to contradict the prospect of such wise 'foolishness'.
The Predicament of Reason
23
On this basis, a course of action A affords the rationally appropriate resolution in the circumstances C if adopting A is the optimal, most promising thing to do given that C. Using some obvious symbolic abbreviations (with 'iff' = 'if and only if) we can write: (P) R\(A/C) iff A opt C This much is straightforward. Nevertheless, there is a serious problem here. The difficulty that revolves about principle (P) inheres in the fact that 'circumstances' must be construed as recognizable circumstances—circumstances as they can be recognized by the agent in terms of information at his disposal. The rationality of someone's resolution of an issue will depend on the circumstances as he faces them. It is a matter of the circumstances as determinable from where he stands, and not of what the real ones may actually happen to be. The rational thing to do in resolving an issue is to make the best use of all the (relevant) information at our disposal. Clearly, it is not—and cannot rationally be—required that we should use information that we do not have and cannot obtain. (Even if, unbeknownst to me, there will be a power failure in a few moments, it is nevertheless rational for me to take the elevator upstairs.) The rationality of our beliefs, actions, and evaluations is clearly a matter of the information that we can secure, not one of what might in theory be available to others who may be more favourably circumstanced than ourselves. All that rationality can ever manage—and all that can rationally be expected of us—is that at any given juncture we should make the best possible use of the available information. Twentiethcentury medicine is in no way deficient in point of rationality for failing to apply twenty-first-century remedies. Rationality, like politics, is an art of the possible—a matter of doing the best that is possible in the overall circumstances in which the agent functions —cognitive circumstances included. If we had 'complete information', and in particular if we knew how our decisions would eventuate—how matters will actually turn out when we decide one way or another—then rational decision-making and planning would of course be something very different from what they are. In this world we are constrained to decide, to operate, to plan, and to act in the light of incomplete
24
The Predicament of Reason
information. When we marry, take a job, invest our money, improve existing technology, and so on, we have no clear idea what eventual consequences will ensue. All we can ever do is to be rational in the circumstances as best we can determine them to be. If rationality were only possible in the light of complete information it would perforce become totally irrelevant for us. It lies in the inevitable nature of things that we must exercise our rationality amidst conditions of imperfect information. A mode of 'rationality' capable of implementation only in ideal circumstances is pointless; in this world, the real world, there is no work for it to do. We have to be realistic in our understanding of rationality—recognizing that we must practise this virtue in real rather than ideal circumstances. A conception of 'rationality' that asks no more of us than doing the best we possibly can is the only one that makes sense. (In fact, one that asks for more would not itself be rational.) Clearly, if rationality is to be something that we can actually implement then it has to be something whose demands we can meet in sub-ideal conditions— conditions of incomplete information as we (inevitably) confront them. The problem, then, is that rationality is 'information-sensitive': exactly what qualifies as the most rational resolution of a particular problem of belief, action, or evaluation depends on the precise content of our data about the situation at issue. And this dependency so functions that a 'mere addition' to our information can transform the optimality situation radically. For, as those preceding examples indicate, a mere amplification of the known circumstances may well indicate the appropriateness of doing something totally incompatible with that initial optimum. Merely enhanced data can, alas, induce radically different conclusions. It is a critically important fact that the optimality at issue in the aforementioned condition of 'A opt C is perfectly compatible with: (3 A')(3 C')[(C compat C ) & (A' opt (C + C')) & - (A' compat A)]. 2 2
Only in the purely deductive case are we secure against this sort of difficulty. When C and C" are compatible, then no information that hinges deductively on C + C" can be actually incompatible with something derivable from C. Additional (compatible) information can never yield contradictions deductively—it can only effect a narrower specification within a broader range, exactly as per Case 1 of Sect. 1.
The Predicament of Reason
25
Throughout the range of rational deliberation there is no 'optimuminvariance' under mere amplification of the prevailing conditions; constraint-relative optimality is generally not preserved when the operative constraints are enlarged. Be it in matters of belief, evaluation, or action, rationality is a matter of optimization subject to the existing information base. But, as we obtain more information we change the 'frame of reference' relative to which optimality is determined. And then what was once the best, most favourable resolution may well no longer continue to be so in the light of these further developments. Optimality, in sum, is 'context-sensitive' with respect to the informational context. The history of the empirical sciences affords a familiar illustration. Beliefs in the luminiferous aether, the conservation of matter, and the like, were all sensible and rational in their day. Achieving a substantial enlargement of the data base on which we erect the structures of our theorizing generally produces those changes of mind characterized as 'scientific revolutions'. As significantly enhanced experimental information comes to hand, people are led to resolve their problems of optimal question-resolution in radically different ways. It will not do to react to this state of affairs by saying: 'Delay decision until your experience is perfected, your information altogether complete.' To postpone a decision until then is tantamount to preventing its ever being made. A rationality we cannot deploy here and now, amidst the realities of an imperfect world, is altogether useless. We thus confront the 'predicament of reason' generated by the confluence of the two considerations, that the rationality of a problem-resolution is 'information-sensitive', and that in the real world our information is always incomplete. Mere additions to our information can always upset the applecart of rational decision.
2.3
IMPLICATIONS OF OPTIMUM-INSTABILITY
This failure of optimum-invariance in situations of rational problem-resolution has unwelcome consequences. In our reasoning regarding what it is rational to do, we seemingly require the following inferential rule of rational appropriateness:
26
The Predicament of Reason (R) A opt C C obtains
.-. R\A But optimum-instability shows that this rule is deeply problematic. For we can also have A' opt (C + C ) (C + C") obtains
for some A' that is flatly incompatible with A. The emergence of such inconsistencies shows that something has gone wrong. But what? The answer is not far to seek. What has gone wrong is the inferential principle at issue in the deconditionalization rule (R), which—given principle (P)—leads from the premisses lR\(A/Cy and 'C obtains' to the conclusion ' R W . For, the fact is that this rule is equivocal as between two crucially different interpretations. It emphatically does not obtain in the form: R\ (A/C) C obtains (inter alia) .-. R\A But it does indeed obtain in the amplified and qualified form: Rl(A/C) C obtains and furthermore C is A -complete, with no facts bearing relevantly on A omitted. ••• R\A The rationality of a given course in specified circumstances depends both on what is explicitly specified, and also, no less importantly, on what is omitted in the characterization of the circumstances. There is a tacit supposition that the specified circumstances include all relevant considerations. For the rule to work, a condition of relevancy-completeness must be supposed—an understanding that if any further relevant considerations had obtained, they would have been included. The 'Rule of Deconditionalization' at issue in rational deliber-
The Predicament of Reason
27
ation is, thus, not the simple two-premiss inference contemplated above, but rather the more complex three-premiss inference: R\(A/C) C obtains C is A -complete: no C-supplementary counter-indications to A actually obtain .'. R\A In this light, we now recognize that the simpler, two-premiss deconditionalization rule was actually enthymematic. In deliberations about rationality, deconditionalization always requires that (generally tacit) additional completeness premiss to the effect that the specified circumstances C represent an A -relevantly complete characterization of the existing circumstances—in omitting no actually obtaining circumstances that counter-indicate A. Actually applying this corrected inference-rule obviously involves us in the difficulty of establishing such a completeness premiss. And this difficulty is very real. For, the thesis 'There are no C-supplementing counter-considerations to A' has two parts: (i) there are none that are known to us, and (ii) there are none that are unknown to us. And it is of course this second clause that presents difficulties. In view of it, deconditionalization always requires 'a step beyond the evidence in hand' in its presupposition that the situation as we face it is not altered by information that we fail to have. This is something one can sometimes plausibly suppose, but not something one can ever establish. Categorical (unconditional) rational appropriateness always hinges on the total circumstances, involving the entirety of information present and absent. But obviously this second factor of 'absent information' poses difficulties. In this world, our circumstances are inevitably sub-ideal, our information unavoidably incomplete. The inconvenient fact is that we simply cannot determine that nothing outside our cognitive reach has a certain character or tendency. This is more than we can ever manage.
2.4
I D E A L V S . PRACTICAL R A T I O N A L I T Y : THE P R E D I C A M E N T OF REASON
The 'predicament of reason' at issue in the irresoluble tension
28
The Predicament of Reason
between the demands of rationality and its practical possibilities inheres in the following aporetic situation: 1. We ought to do that which, as best we can determine it, is the rationally apropriate thing to do. 2. We ought to act as perfectly rational agents do. 3. What is appropriate as best we can determine it (in our suboptimal circumstances) will generally differ from what a perfectly rational agent would (ipso facto) determine to be best. Since these theses are mutually incompatible, one of them must be sacrificed, and since 1 and 3 represent unavoidable 'facts of life', it is clearly 2 that must be abandoned in the interests of consistencyrestoration. Neither the rationality-abandonment of 1-rejection nor the unrealistic perfectionism of 3-abandonment are plausible options. This circumstance that thesis 2 must be abandoned sets the stage for what we may call 'the predicament of reason', implicit in the fact that rationality seems to demand something of us that is, in the final analysis, not actually realizable. Fortunately, the difficulty that arises here admits of a sensible resolution. The distinction between idealized and practicable rationality offers a way out. For we must distinguish between: ideal rationality, which is geared to resolutions that are rationally appropriate with (absolutely) everything relevant taken into account—that are optimal pure and simple, and practicable rationality, which is geared to resolutions that are rationally appropriate with everything relevant taken into account that we can effectively manage to take account of in the prevailing circumstances—that are optimal as best we can manage to tell. This distinction softens the impact of rejecting premiss (2) of the predicament. For, while we cannot indeed achieve the demands of ideal rationality, we clearly should do all we can in the direction of practicable rationality. After all, ideal rational optimality is something merely 'Utopian' and 'pie in the sky'. For us, the only practicable optimality is that which is realistic and achievable— optimality as best we can get hold of it, which accordingly remains merely apparent optimality (optimality as best we can determine it). All we can ever secure in real-life situations of rational deliberation are seeming optima arrived at in the light of
The Predicament of Reason
29
incomplete information. We can have no assurance that they will continue to be optimal in the light of a fuller appreciation of the circumstances. We can only do our best. This sort of situation obtains throughout all areas of rational deliberation: cognitive, prudential, evaluative—right across the board. Even our optimally evidentiated beliefs are not necessarily true; even our optimally well-advised actions are not necessarily successful; even our optimally crafted appraisals are not necessarily correct. The fact of limitation confronts us in every direction. The crux of the matter is that rationality is not a matter of absolute optimization but of circumstantial optimization, not of doing what is the best thing unqualifiedly but of doing the best that can be done in the circumstances—including the informational circumstances—that are at issue. Consider the following illustration: 1. One has a severe headache. 2. In actual fact, yonder tablet is a (perfectly harmless) aspirin pill that will cure one's headache. 3. One has every reason to believe that yonder tablet is a deadly poison. What is the rational thing to do? Clearly, it is to avoid taking the tablet. The irrationality of acting contrary to the indications of circumstantially manageable optimization is not redeemed by the unforeseeably favourable issue of events. Reason enjoins us to do what appears optimal—what is optimal relative to the circumstances as we discernably confront them, and so relative to the information as we have it. The most we can possibly do—and the most that can be asked of us in the name of rationality—is to do the best we can manage to do under the prevailing conditions. But, we have no assurance that this actually is optimal, no guarantee that what seems the best thing to do actually is so. The irony is that in opting for this apparent and local optimum there is no guarantee that we will achieve a real and global optimum. It is a trite fact which nevertheless has enormously far-reaching implications that the deployment of intelligence or incomplete information may well yield inappropriate solutions. When one combines this with the consideration that we humans virtually always labour in circumstances of incomplete information we see that rationality is a resource of inherently limited utility.
30
The Predicament of Reason
The inexorable circumstance represented by thKl^r&jijlpwing facts has to be faced: 1. The 'ideally rational' thing to do is to do what is optimal. 2. The 'rationally appropriate' or, simply, the 'rational' thing is for us to do our utmost towards the ideally rational thing—to do that which as best we can tell is the ideally rational thing to do. Accordingly, rationality calls for optimization in the light of the available information. 3. What actually is optimal, however, is what is determinable as such (as optimal) in the light of complete information. 4. What is optimal in the light of incomplete information may well fail to be actually optimal and indeed may be counterproductive. 5. The most we can ever do is to act in the light of the available information, which generally is incomplete. It follows inexorably from these premisses that, in general, the rational thing for us to do is something which, as we must recognize full well, may actually impede realization of our objectives. This constitutes what might be called the predicament of reason: the circumstance that reason constantly calls on us to do that which, for ought we know, may prove totally inappropriate. Rational action in this world has to proceed in the face of the sobering recognition that while we doubtless should do the best we can, it may nevertheless eventuate that the seeming best we ean do is quite the wrong thing. It is the course of reason (1) to aim at the absolutely best, but (2) to settle for the best that is realistically available. (After all, it would be unreasonable, nay irrational, to ask for more.) But the paradox lies in our clear recognition of the tension between the two. The 'predicament of reason' thus resides in the fact that while the ends of rationality are achieved only under the ideal conditions of global totality, nevertheless the actual practice of rationality must inevitably be conducted at the level of local and imperfect conditions. We can never rest complacently confident that in following reason's directions we are not frustrating the very purposes for whose sake we are calling upon the guidance of reason. We have to recognize the 'fact of life' that it is rationally advisable to do the best we can, while nevertheless realizing all the while that it may prove to be inappropriate.
The Predicament of Reason
31
'But the problem is created by mere ignorance.' True enough! But true in a way that provides no comfort. Imperfect information is an inevitable fact of life. A 'rationality' that could not be implemented in these circumstances would be totally pointless. Were rationality to hinge on complete information, it would thereby manifest its irrelevance for our concerns. There is nothing 'mere' about ignorance. We have to realize that, no matter how rationally we work things out, unforeseen and unforeseeable circumstances may intervene to blow our house of cards away. 'The best-laid schemes o' mice an' men . . .'. Genuine global optimality lies outside our range. There is no escaping the fact that the judgements of optimality we form on the basis of information on hand may simply not bear the weight we place on them. In this regard, a fundamental uniformity of condition prevails throughout our rational concerns—not just in the sphere of prudential praxis but in that of our cognitive and evaluative efforts as well. Throughout, we must be prepared to settle for the best we can manage— irrespective of any assurance that it is good enough. In situations of interpersonal communication, to be sure, we can and do operate on the presumption of completeness. We look upon our interlocutors as collaborators of sorts, assuming that the information someone provides in communicative interchange is all the information that counts; that we are provided with all the relevant information. But in our dealings with the real world itself that just is not so. It is wrong to see reality in the light of a cooperative partner who provides us with all the relevant information we need to draw rationally appropriate inferences. We cannot assume that what we do not know makes no difference. It is profoundly foolish to think 'What you don't know won't hurt you.' (Many is the person struck by an unseen bullet or bitten by an unseen germ!) Such deliberations lead into the terrain of philosophical anthropology. For, we here confront a fundamental aspect of the human condition. With creatures like ourselves that are of limited capacity, and whose cognitive range of reference is thus inevitably limited, no assurance can be attained that 'doing the best one can' to follow the dictates of rational appropriateness will produce a rationally optimal resolution—that those apparent optima we can attain will actually yield real optima. There is never any assurance
32
The Predicament of Reason
that we will actually succeed by following the best available advice. The cold and cruel reality is that (1) a decision based on incomplete information—no matter how intelligently made—can be totally counter-productive, and (2) in this life we always have to operate on the basis of incomplete information. There stands before us the profound lesson of the biblical story of the Fall of Man, that in this world there simply are no guarantees—not even for a life conducted on principles of reason. It is this sobering situation—doubtless unwelcome, but inevitable —that betokens the predicament of reason: the circumstance that rationality requires us to do 'what seems best' in the full and clear recognition that this may well fail to be, in actuality, anything like the best thing to do.
3 The Rationale of Rationality: Why Follow Reason?
(1) Why follow the guidance of reason? Simply because this affords, as best one can determine, the optimal chances of success in realizing one's appropriate ends. But how can one really be sure that what is optimal only 'as far as we can now tell' is actually optimal? One cannot! There is no way to prove that rationality pays. (2) We can only achieve the essentially pragmatic justification of showing that, as best we can judge the matter, the counsel of reason represents the most promising prospect of realizing our objectives. Even cognitive rationality is ultimately justifiable only in the practical order of reason. (3) In providing a rational justification of rationality—and what other kind would we want?—the most and best that one can do is to follow the essentially circular (but nonviciously circular!) line of establishing that reason herself endorses taking this course. SYNOPSIS
3.1
THE PROBLEM OF VALIDATING RATIONALITY
Why should one be rational? In a way, this is a silly question. For, the answer is only too obvious—given that the rational thing to do is (effectively, by definition) that for which the strongest reasons speak, we ipso facto have good reason to do it. Kurt Baier has put this point in a way difficult to improve upon: The question 'Why should I follow reason?' simply does not make sense. Asking it shows complete lack of understanding of the meaning of a 'why question'. 'Why should I do this?' is a request to be given the reason for saying that I should do this. It is normally asked when someone has already said, 'You should do this' and answered by giving the reason. But since 'Should I follow reason?' means 'Tell me whether doing what is supported by the best reasons is doing what is supported by the best reasons', there is simply no possibility of adding 'Why?' For the question now comes to this, 'Tell me the reason why doing what is supported by the
34
The Rationale of Rationality
best reasons is doing what is supported by the best reasons'. It is exactly like asking, 'Why is a circle a circle?'1 In virtue of its nature as such, the rational resolution to an issue is the best solution we can manage—the one that we should adopt and would adopt if we were to proceed intelligently. The impetus to rationality is grounded in our commitment to proceeding intelligently—to 'using our brains'. ('Why be rational?' 'It's the intelligent thing to do.' 'But why proceed intelligently?' 'Come now; surely you jest!') After all, once we admit that something is the best thing to do, what further reason could we possibly want for doing it? Once it is settled that A is the rational thing to do— which may itself take a lot of showing—then there is no more room for any further reason for doing A, no further point to asking 'Why do AT For at this stage, the best of reasons—by hypothesis —already speak for doing A. Once rationality is established, there are no further extra- (or supra-) rational reasons to which we could sensibly appeal for validation. In this sense, then, the question 'Why do the rational thing?' is simply foolish: it is a request for further reasons at a juncture at which, by hypothesis, all the needed reasons are already in. But this line of response to our question, though perfectly cogent, is a bit too facile. The job that needs to be accomplished is actually more complicated. Belief, action, and evaluation based on what really are—truly and actually—the 'best of reasons' must necessarily be successful. This contention is simply circular, since those theoretically 'best of reasons' are best exactly because it is they that assure realization of the best results. But, in this world, we are not in general in a position to proceed from the actual best as such, but only from the visible best that is at our disposal—'the best available (or discernible) reasons'. We have to content ourselves with doing 'the apparently best thing'—the best that is determinable in the prevailing circumstances. But, the fact remains that the alternatives whose adoption we ourselves sensibly and appropriately view as rational given the information at our disposal at the time are not necessarily actually optimal. The problem about doing the rational thing—doing that which we sensibly suppose to be supported by the best reasons—is that our information, being incomplete, may 1
Kurt Baier, The Moral Point of View (New York, 1965; abr. edn.), 160-1.
The Rationale of Rationality
35
well point us in the wrong direction. Facing this 'predicament of reason', we know the pitfalls, realizing full well the fragility of these 'best laid schemes'. So the problem remains: Why should we act on the most promising visible alternative, when visibility is restricted to the limited horizons of our own potentially inadequate vantage-point? In answering this question, let us begin by considering the situation in the light of the expected-value calculation set out in Display 3.1. It is clear in the context there postulated, that as long as d, the increment in the probability of success due to heeding rationality's advice, is greater than zero—that is, as long as 'doing the rationally advisable thing' will increase the probability of success somewhat, no matter how little—then this course is the decision-theoretically sensible one. As long as rationality improves the prospects of success, no matter how modestly, its call represents the best bet, the advisable course, the sensible thing to do. No guarantees are necessary.
DISPLAY
3.1: Rationality in Decision-Theoretic Perspective
Probability of achieving (optimal) success Probability of failing to achieve (optimal) success Value of achieving (optimal) success Value of failing to achieve (optimal) success
I do not do the rationally advisable thing
I do the rationally advisable thing
p
p+d
1- p
1 " (P + d)
x ~y
The following expected values emerge: EV (non-rat'l) = px + (1 - p)y = p(x - y) + y EV (rat'l) = (p + d)x + (l-p-d)y = p(x - y) + d(x - y) + y Note that: £V(rat'l) > £V(non-rat'l) iff d(x - y) > 0 iffd>0 (since obviously (x — y) > 0).
36
The Rationale of Rationality
After all, in this imperfect sublunary dispensation, probability is, as Bishop Butler said, 'the guide of life'. We must—rationally must—follow the guidance of the perceived probability that the general policy of doing 'the rationally indicated thing' will, on balance and over time, prove to be our best bet. Rationality is not a thing of the present moment and the case in hand, but of the whole picture and the long run. The rational person is, by definition, someone who uses intelligence to maximize the probability—that is, the responsibly formed subjective probability—that matters will eventuate favourably for the promotion of his real interests. It is just this that makes following the path of rationality the rational course. Rationality calls for adopting the overall best (visible) alternative—the best that is, in practice, available to us in the circumstances. And if A indeed is the rational thing to do in this sense, then we should expect to be worse off in doing something different from A. To be sure, things may not come to that; we could be lucky. But this is something we do not deserve and certainly have no grounds to expect. After all, why do we endorse the world-descriptions of the science of the day? Why do we follow the medical recommendations of the physicians of the day, or the policy recommendations of the economists of the day? Because we know them to be correct—or at any rate highly likely to be true? Not at all! We know or believe no such thing—historical experience is too strongly counterindicative. Rather, we accept them as guides only because we see them as more promising than any of the identifiable alternatives that we are in a position to envision. We accept them because they afford us the greatest available subjective probability of success— discernibly the best bet. We do not proceed with unalloyed confidence, but rather with the resigned recognition that we can do no better at the moment. Similarly, the recommendations of reason afford not assurance of success, but merely the best overall chances of reaching our goals. We act, in short, on the basis of faute de mieux considerations, of 'this or nothing better—as far as the eye reaches'. In real-world situations reason trades in courses of action whose efficacy is a matter of hope and whose rationalization is a matter of this-or-nothing-better argumentation. Like the drowning man, we clutch at the best available object. We recognize full well that even the most rationally laid scheme can
The Rationale of Rationality
37
misfire. Reality is not always and inevitably on the side of the strongest arguments. Reason affords no guarantee of success, but only the reassurance of having made the best rational bet—of having done as well as one could in the circumstances of the case. One cannot say, flatly and unqualifiedly. 'You"should be rational because rationality pays in rendering success if not certain then at any rate more probable.' Rather, we have to content ourselves with: 'You should be rational because this affords the best rationally foreseeable prospects of success—on the whole and in the long run.' And so, we follow reason because this makes good rational sense to do so, seeing that this affords us with the best visible prospect for realizing our objectives. One should be rational in general for just the same sort of reason as one should be rational in the specific case of the hungry man's choice between eating bread or sand—namely, that by all available indications this course represents the most promising prospect of attaining one's sensible goals.2 However, with all this said and done, the problem of real vs. apparent optimality that was our starting point still remains in place. For, this whole approach turns on the supposition that 'doing the rational thing' will indeed enhance our overall chances of bringing our affairs to a successful issue. Is this actually true? To all appearances, the answer is a resounding 'of course!' What is at issue here is seemingly a tautology. Clearly, if that object did not have a blade, we would not call it a 'knife'. Equally clearly, if we did not accept that doing A would enhance the chances of the relevant sort of success, then we would not characterize A as 'the rational thing to do in the circumstances'. We simply could not endorse that course of action as being 'rationally advisable in the circumstances' if we were not convinced that it enhanced the chances of a successful issue. For it is just this that makes something into 'the rational thing to do': its enhancing, as best we can tell, our chances of attaining success to a greater extent than any other available alternative. We may have trouble spotting 'the rational thing to do' in particular circumstances. But once our mind is made up about this, then the issue of rational advisability is closed. 2
Compare Rene Descartes, Discourse on Method, sect. iii. maxim 2.
38
The Rationale of Rationality
Unfortunately, however, this is still not quite the end of it. For the problem remains: Just exactly what are the probabilities with which we are operating? Of course, we intend them to be objective, real-world likelihoods; this is what we would ideally like to have. But, in fact, of course, they are no more than our considered estimates of such likelihoods as best we can shape them in the light of the available information. And this means that we are once again in the presence of rational resolutions effected on the basis of the available data. We are here confronted with an instant, local replay of the global problem that is being addressed. Striving to escape the predicament of reason, it mocks us by leaping ahead to bar our way. For we here confront once more the familiar and vexing issue of the actual optimality of apparent optima. And there is nothing we can do to escape this awkward circumstance—we simply have to take it in our stride. The fact of the matter is that we cannot prove that rationality pays—necessarily, or even only over the probabilistic long run. We do not know that acting rationally in the particular case at hand will actually pay off—nor can we even claim with unalloyed assurance that it will probably do so (with real likelihood rather than subjective probability). We can only say that, as best we can judge the matter, it represents the most promising course at our disposal. We have no guarantees—no means are at our disposal for pre-establishing that following rationality's counsel actually pays. Consider the sequence of theses that, by doing the rational thing in the present case we shall, as best we can tell: 1. assure ourselves of success; 2. assure ourselves of success if success is at all possible; 3. enhance the chances of success (its objective probability) in this present case; 4. enhance the chances of a good record of success overall in the whole series of similar cases. As we move down this list, we reach increasingly weaker and thus more plausible contentions. And to achieve something tenable we must go all the way down to the bottom of the list—we cannot stop short of item 4. And even here we must settle from the visible as opposed to real chances. The efficacy of rationality can only be maintained in the qualified way of a reasonable expectation.
The Rationale of Rationality
39
Throughout our cognitive and practical affairs we have to conduct our operations under conditions of risk. And so, when we do the rational thing but it just does not pan out, we simply have to 'grin and bear it'. The matter is one of calculated risks and plausibly expectable benefits. Rationality affords no guarantees. By the very nature of what is involved in rational procedure, the determinable odds are in its favour. But that may still be cold comfort when things go wrong. Then, all we have is the satisfaction of having done our best. The long and short of the matter is that nothing 'obliges' us to be rational except our rationality itself. Of course, one may somehow prefer not to be rational. With belief, I may prefer congeniality to truth. With action, I may prefer convenience to optimality. With value, I may prefer the pleasingly base to the more austere better. On all sides, I may wilfully opt for 'what I simply like', rather than for that which is normatively appropriate. But if I do this, I lose sight of the actual ends of the cognitive, practical, and evaluative enterprises, to the detriment of my real (as opposed to apparent) interests. It lies in the nature of things that reason is on the side of rationality. To be sure, she offers us no guarantees. Yet, if we abandon reason there is no place better that we can (rationally) go. 3 . 2 THE PRAGMATIC T U R N : EVEN COGNITIVE RATIONALITY HAS A PRAGMATIC BASIS
No considerations of theoretical, general principle can possibly establish that what is apparently the optimal course—what is so 'as best one can tell'—is actually optimal. In this matter we cannot proceed by way of cogent inference in the evidential/cognitive order of reason, but must turn in another direction altogether, to inference in the practical order of reason. The best available justification of rationality is a practical inference along the following lines: 1. We want and need rationally cogent answers to our questions —answers that optimally reflect the available information. 2. Following the path of cognitive rationality (as standardly construed) is the best available way to secure rationally cogent answers to our questions.
'40
The Rationale of Rationality Therefore: Following the path of standard cognitive rationality in matters of inquiry (that is, in answering our questions) is the rational thing to do: we are rationally well advised to answer our questions in line with the standard processes of cognitive rationality.
It must be stressed that this reasoning is of the following pattern: We have the inherently appropriate objective O; course of action A is the optimal available path to this objective; therefore, we are rationally well advised to follow this path. This, clearly, is a quintessentially pragmatic style of argumentation. It is appropriate to proceed rationally not because we know that by so doing we will (inevitably or probably) succeed, but because we realize that by doing so we will have done the very best we possibly can towards producing this outcome: we will have given the matter 'our best shot'.3 This practical turn is ultimately inevitable. We can do no more than to adopt an approach that represents the best and the most that we can do. For of course we cannot maintain: 'If you form a belief rationally, then it will turn out to be true.' This is simply not on the cards. The most we can do is to maintain: By all the relevant indications, there is good reason to think that a rationally formed belief is true. (That is exactly what we mean by 'a rationally formed belief.) The cogency of our practical argument rests on the fact that in real-life situations we simply have to do the best we can—that it would be senseless (and irrational!) to ask for more than this. To summarize: The answer to our pivotal question—'Why do the rational thing, given that we cannot guarantee its success?'— lies in a confluence of considerations: 3 In his interesting book on A Justification of Rationality (Albany, 1976), John Kekes argues that 'The justification of rationality is . . . [as] a device for problemsolving and it should be employed because everybody has problems, because it is in everybody's interest to solve his problems, and because rationality is the most promising way of doing so' (p. 168). This traditionally pragmatic view is very close to our own position except that it pivots rationality's justification on effectiveness in problem solving, while our own position is somewhat more cautious. It does not contend that the course of reason actually is our best recourse in problem solving, but only that it is so as best we can (rationally) judge. The present argumentation thus brings the aspect of reason's self-reliance to the fore as a critical aspect of the justification of rationality, and accordingly is not a pure pragmatism.
The Rationale of Rationality
41
1. As far as we can tell, it is the best thing to; do (it is the apparent optimum). 2. While apparent optima are not necessarily real optima—are not necessarily optimal as such—experience teaches that we cannot find a policy superior to that of doing what is apparently optimal. ('This or nothing better.') 3. By adopting this policy (on 'win or lose, it is the best we can do' grounds) we most effectively subserve the crucial desideratum of assuring ourselves of having done the very best that is possible for us in the circumstances. And the reason why we rest content with such a practical argument itself proceeds in the practical mode: 1. We have a certain (appropriate) objective—namely, to achieve the assurance of legitimacy and interpersonal efficacy of co-ordination essential to successful collaborative action. 2. We recognize that practical reason affords our only really practicable way to obtain such a rational co-ordination and validation. ('This or nothing.') Therefore: We are rationally well advised to adopt the way of practical reason in validating reason. The sort of argument for rationality that we have contemplated is thus a practical argument rather than one that proceeds in the strictly cognitive sector of reason. And this is the best that can be had. To decline it—to say that the best available is simply not good enough—is simply to be irrational. Not surprisingly, rationality here stands on the side of reason. Philosophers of pragmatic inclination have always stressed the ultimate inadequacy of any strictly theoretical defence of cognitive rationality. And their instincts in this regard are surely right. One cannot marshal an ultimately satisfactory defence of rational cognition by an appeal that proceeds wholly on its own grounds. In providing a viable justification the time must come for stepping outside the whole cognitive/theoretical sphere and seeking for some extra-cognitive support for our cognitive proceedings. It is at just this stage that a pragmatic appeal to the condition of effective action properly comes into operation. And this pragmatic aspect of the matter has yet another side. The pivotal role of rationality as a co-ordination principle must
42
The Rationale of Rationality
also be emphasized. The human condition is such that the adequate cultivation of our individual interest requires a co-ordination of effort with others and imposes the need for cooperation and collaboration. But this is achievable only if we 'understand' one another. And here rationality becomes critical. It is a crucial resource for mutual understanding, for rendering / people comprehensible to one another, so as to make effective communication and co-operation possible. The following three points are crucial in this regard. (1) It is a matter of life and death for us to live in a setting where we ourselves are in large measure predictable for others, because only on this basis of mutual predictability can we achieve conditions essential to our own welfare. (2) The easiest way to become predictable for others is to act in such a way that they can explain, understand, and anticipate my actions on the basis of the question 'What would I do if I were in his shoes?' (3) In this regard the 'apparent best' is the obvious choice, not only because of its (admittedly loose) linkage to optimality per se, but also because of its lsaliency\ The quest for 'the best available' leads one to fix on that alternative at which others too could be expected to arrive in the circumstances—so that they too can understand one's choices. The pursuit of optimality is accordingly a determinative factor for rationality not only through its direct benefits in yielding our best apparent chances of success, but also through its providing a principle for the guidance of action that achieves the crucial requisite of social co-ordination in the most efficient realizable way. (But why co-ordinate on what is rational, why not simply on habit or fashion or 'the done thing'? Partly, because these leave us in the lurch once we get off the beaten track of 'the usual course of things'. And partly because they are unstable and inherently unreliable.
3.3
THE S E L F - R E L I A N C E OF RATIONALITY IS NOT VICIOUSLY CIRCULAR
This practical line of argumentation may still seem to leave the situation in an unsatisfactory state. It says (roughly): 'You should be rational in resolving your choices because it is rational to believe that the best available prospects of optimality-attainment are
The Rationale of Rationality
43
effectively realized in this way.' To be sure, one might deem it preferable if that italicized clause were wholly suppressed. A sceptic is bound to press the following objection: The proposed practicalistic legitimation of reason conforms to the pattern: 'You should be rational just because that is the rational thing to do!' And this is clearly circular. It might seem questionable to establish the jurisdiction of reason by appeal to the judgement of reason itself. But, in fact, of course, this circularity is not really vicious at all. Vicious circularity stultifies by 'begging the question'; virtuous circularity merely coordinates related elements in their mutual interlinkage. The former presupposes what is to be proved, the latter simply shows how things are connected together in a well-co-ordinated and mutually supportive interrelationship. The self-reliance of rationality merely exemplifies this latter circumstance of an inherent co-ordination among its universe components. Admittedly, the reasoning at issue has an appearance of vitiating circularity because the force of the argument itself rests on an appeal to rationality: 'If you are going to be rational in your beliefs, then you must also act rationally, because it is rational to believe that rational action is optimal in point of goal attainment.' But this sort of question begging is simply unavoidable in the circumstances. It is exactly what we want and need. Where else should we look for a rational validation of rationality but to reason itself? The only reasons for being rational that it makes sense to ask for are rational reasons. In this epistemic dispensation, we have no way of getting at the facts directly, without the epistemic detour of securing grounds and reasons for them. And it is, of course, rationally cogent grounds and reasons that we want and need. The overall justification of rationality must be reflexive and self:referential. To provide a rationale of rationality is to show that rationality stands in appropriate alignment with the principles of rationality. From the angle of justification, rationality is a cyclic process that closes in on itself, not a linear process that ultimately rests on something outside itself. There is accordingly no basis for any rational discontent, no room for any dissatisfaction or complaint regarding a 'circular' justification of rationality. We would not (should not) want it otherwise. If we bother to want an answer to the question 'Why be
44
The Rationale of Rationality
rational?' at all, it is clearly a rational answer that we require. The only sort of justification of anything—rationality included—that is worth having at all is a rational one. That presupposition of rationality is not vitiating, not viciously circular, but essential—an unavoidable consequence of the self-sufficiency of cognitive reason. There is simply no satisfactory alternative to using reason in its own defence. Already embarked on the sea of rationality, we want such assurance as can now be made available that we have done the right thing. And such reassurance can indeed be given— exactly along the lines just indicated. Given the very nature of the justificatory enterprise at issue, one just cannot avoid letting rationality sit in judgement on itself. (What is being asked for, after all, is a rational argument for rational action, a basis for rational conviction, and not persuasion by something probatively irrelevant like threats of force majeure.) One would expect, nay demand, that rationality be self-substantiating in this way—that it must emerge as the best policy on its own telling. From the justificatory point of view, rationality is and must be autonomous. It can be subject to no external authority. Rationality in general is a matter of systematization, and the justification of rationality is correspondingly a matter of systemic self-sufficiency. Rather than indicating the defect of vicious circularity, the selfreferential character of a justification of rationality is a precondition of its adequacy! It is only a rational legitimation of rationality that we would want: any other sort would avail us nothing. And if such a rational validation were not forthcoming this would indicate a grave defect. To be sure, some theorists see rationality as heteronomous—as subject to some external sort of authority such as 'feeling' or 'the will'. Thus, one contemporary philosopher projects the idea that: [Underlying each . . . judgment there is a choice that the agent has made—a type of choice in which the individual is at the most fundamental level unconstrained by good reasons, precisely because his or her choice expresses a decision as to what is to count as a good reason for him or her.4 Such a view sees rational justification as linear and regressive— and thus as ultimately having to rest on an unrationalized rock4
Alasdair Maclntyre, in Maclntyre and Stanley Haverwas (eds.), Revisions (Notre Dame and London, 1983), 9.
The Rationale of Rationality
45
bottom that itself lies quite outside the domain of reason. But any such view is profoundly mistaken. For rational validation is not linear and regressive, but rather cyclical and systemically selfcontained. We need not—must not—subscribe to the RockBottom Fallacy. There is no way of grounding good reasons in arbitrary or otherwise unrationalizable decisions. ('Deciding as to what is to count as a good reason' forsooth! Not even God is in a position to do that!)5 No one decides what sorts of things are to count as good reasons. In general, we only learn in the school of bitter experience what qualifies as such. A certain irrationalism is astir in the world that rejects the quest for rationally validated reasons and advocates a free-wheeling 'anything goes'—even in the cognitive sphere of empirical inquiry.6 But, of course any sensible person not already committed to such a position would want to know if there is any good reason for taking it. And then we are at once back in the sphere of rationality and good reasons. Thus, the predicament of reason gives no comfort to scepticism or irrationalism and yields no grounds for abandoning reason or, worse yet, for turning against her. This self-supportive legitimation of rationality is the only cogent sort of validation that we are going to get. But, in the final analysis it is the only sort that it makes sense to ask for, seeing that rationality itself enjoins us to view the best we can possibly get as good enough.7 A desperate objection yet remains: 'So rationality speaks on its own behalf. Well and good. But why should I care for rationality? Why should I set myself to do the intelligent and appropriate thing?' At this point there is little more to be said. If I want a reason at all, I must want a rational reason. If I care about reasons at all, I am already within the project of rationality. But once I am within the project, there is nothing further external to reason that can or need be said to validate it. At that stage rationality is already at hand to provide its own support—it wears its justification on its sleeve. (The project of trying to reason with someone who stands 5 On this point, see Leibniz's correspondence with Arnauld regarding the Discourse on Metaphysics. 6 See e.g. Paul K. Feyerabend, Against Method (London and New York, 1978). 7 Est ridiculum quaerere quae habere non possumus, as Cicero wisely observed (Pro Archia, iv. 8). ? '••'•» :
46
The Rationale of Rationality
outside the range of rationality to convince them to come into its fold is clearly an exercise in pointlessness and futility.) One may of course quite appropriately ask questions like: 'Why should I cultivate the truth; why should I cultivate my best (or true) interests?' But in the very act of posing these questions I am asking for reasons—that is, I am evincing my commitment to the project of rationality. Caring for the truth and for one's best interests are simply part and parcel of this commitment. If I do not care for these things, then there is really no point in raising these questions. For in this event I have already taken my place outside the precincts of rationality, beyond the reach of reason. Irrationality—wishful thinking and self-deception—may be convenient and even, in some degree, psychologically comforting. But it is not cognitively satisfactory. If it is a viable defence of a position that we want, it is bound to be a rational one. The only validation of rationality that can reasonably be asked for—and the only one worth having—must lie in considerations of the systemic self-sufficiency of reason. In the final analysis, 'Why be rational?' must be answered with the only rationally appropriate response: 'Because rationality itself obliges us to be so.' In providing a rational justification of rationality—and what other kind would we want?—the best we can do is to follow the essentially circular (but non-viciously circular!) line of establishing that reason herself endorses taking this course. Reason's self-recommendation is an important and necessary aspect of the legitimation of this enterprise. Yet does reason's self-reliance not open the door to scepticism? For sceptics have always insisted on just this point that we cannot prove in advance of conceding reason's cogency that we will not go wrong by trusting our reason. And this—so we have granted—is quite correct. But, of course, what one can do is to establish that if we reject reason we cut ourselves off from any (rationally warranted) expectation of success. There are no guarantees that our ventures in trust are going to prove successful; whether our trust is actually warranted in any given circumstances (trust in ourselves, in our cognitive faculties, in other people, and the like) is something we cannot in the nature of things ascertain in advance of events. A conjunction of trust with hope and faith is germane alike to the cognitive project, the practical project, and the evaluative project. Throughout, we have to conduct our operations
The Rationale of Rationality
47
under conditions of risk, without assured confidence in outcomes and without advance guarantees of success. In all these matters efficacy is a matter of hope and confidence in the best available option whose rationalization is a matter of this-or-nothing-better argumentation. Such argumentation will not of course satisfy the sceptic. For him, the lack of guarantees undermines the whole project of rationality. The sceptic's objections to cognitive rationality deserve the fuller treatment of a separate chapter.
4 Cognitive Rationality and Risk: A Critique of Scepticism
(1) Recognition of the imperfection of our cognitive resources raises the spectre of scepticism. (2) Scepticism roots in risk aversion. The sceptic is exaggeratedly reluctant to run a risk ot error in the interest of acquiring information. (3) He fails to come to terms with the harsh fact that with cognition, as elsewhere, 'nothing ventured, nothing gained'. A sceptical stance blocks any prospect of success at cognitive enterprise from the very outset, without so much as a fair trial. (4) Moreover, there are also strong practical impediments to scepticism. (5) Theoretical deliberations cannot refute scepticism outright. But reflection on the nature of our purposes and aims—practical and theoretical alike—can develop a cogent argument against our assuming a sceptical stance in the first place. For, in rejecting the established standards of cognition the sceptic also cuts himself off from any prospect of effective communication. SYNOPSIS
4.1
IS C O G N I T I V E RATIONALITY P O S S I B L E ?
Reason's commitment to the cognitive enterprise of inquiry is absolute and establishes an insatiable demand for extending and deepening the range of our information. Reason cannot leave well enough alone, but insists upon a continual enhancement in the range and depth of our understanding of ourselves and of the world about us. By rejecting the very possibility of securing trustworthy information in factual matters, scepticism sets up a purportedly decisive obstacle to implementing these aims of reason. Cognitive rationality is a matter of using cogent reasons to govern one's acceptance of beliefs—of answering one's questions in the best feasible way. But is cognitive rationality realizable at all? We must consider the sceptic's long-standing challenge that it is not. For, the sceptic—in his more radical moments, at any
Cognitive Rationality and Risk
49
rate—insists that there is never a satisfactory justification for accepting anything whatsoever. The sceptical challenge to the project of empirical inquiry based on principles of cognitive rationality has a very plausible look about it. Our means for the acquisition of factual knowledge are unquestionably imperfect. Where, for example, are the 'scientific truths' of yester-year—those earth-shaking syntheses of, say, Newton or Maxwell? Virtually no part of them has survived unscathed. And given this past course of bitter experience, how can we possibly validate our present acceptance of factual contentions in a rationally convincing way? Accordingly, the principal aim of the present discussion is to indicate the problems and difficulties of a radical scepticism which maintains that rational cognition is unattainable, that the quest for knowledge is in principle a vain and vacuous pursuit. To be sure, if it is indeed the case that rationally justified belief must always be based upon rationally pre-justified inputs, then the sceptic is quite right. For, then the process of rationally validating our accepted beliefs can never get started. To all appearances, we here enter upon a regress that is either vitiatingly infinite or viciously circular. The rational justification of belief becomes in principle impossible—as sceptics have always insisted. But this sceptical challenge rests on a false supposition. For, the rational justification of a belief does not necessarily require prejustified inputs. The idea that even as human life can come only from prior human life so rational justification can come only from prior rational justification is deeply erroneous. For, the important distinction between discursive and presumptive justification becomes crucial here in a way that sceptics conveniently overlook. A belief is justified discursively when there is some other preestablished belief on whose basis this belief is evidentially grounded. The discursive justification of a belief lies in the fact that there is an already available, pre-justified belief which evidentiates it. In information-processing terms, this discursive sort of justification is homogeneous: there must be justified beliefs as inputs to arrive at justified beliefs as outputs. However, presumptive justification—unlike discursive justification—does not proceed through the evidential mediation of previously justified grounds, but directly and immediately through the force of a 'presumption'. A belief is justified in this way when
50
Cognitive Rationality and Risk
there is a standing presumption in its favour and no pre-established (rationally justified) reason that stands in the way of its acceptance. Presumption is the epistemic analogue of 'innocent until proven guilty'. The rational legitimation of a presumptively justified belief lies in the fact that some 'suitably favourable indication' speaks on its behalf, and no already justified counter-indication speaks against it. When, after a careful look, I am under the impression that there is a cat on the mat, I can (quite appropriately) base my acceptance of the contention 'There is a cat on the mat' not on certain preestablished premisses, but simply on my visual impression. The salient consideration is that there just is no good reason why (in this case) I should not indulge my inclinations to endorse a visually grounded impression of this kind as veridical. Presumptively justified beliefs are the raw materials of cognition. They represent to us contentions that—in the absence of preestablished counter-indications—are acceptable 'until further notice', thus permitting us to make a start in the venture of cognitive justification without the benefit of pre-justified materials. They are vulnerable to being overturned, but only by something else yet more secure, and thus remain in place until displaced by something superior. Accordingly, their impetus averts the dire consequences that would ensue if any and every cogent process of rational deliberation required inputs which themselves had to be authenticated by a prior process of rational deliberation—in which case the whole process could never get under way. This role of presumptions is absolutely crucial for cognitive rationality. For this mode rationality has two compartments, the discursive (or conditionalized) and the substantive (or categorical). The former is a matter of hypothetical reasoning—of adhering to the conditionalized principle that if you accept certain theses, then you should also accept their duly evidentiated consequences. But, of course this conditionalized principle cannot yield anything until one has already secured some acceptable theses from somewhere or other. And this is where substantive rationality comes in, by enabling us to make categorical moves. Presumptions determine our 'starter-set' of initial commitments, enabling us to make a start on whose basis further 'inferential' reasoning may proceed. The salient point is simply that there are some sorts of considerations which can cogently be taken as capable of establish-
Cognitive Rationality and Risk
51
ing at least a tentative or provisional presumption of rational acceptability. When the sceptic rejects any such presumptions, he automatically blocks any prospect of reasoning with him within the standard framework of discussion about the empirical facts of the world. The machinery of presumptions is part and parcel of the mechanisms of cognitive rationality; abandoning it aborts the entire project at the very outset. Presumptions accordingly play a crucial role. For, we cannot pursue the cognitive project—the quest for information about the world—without granting certain initial presumptions. They are reminiscent of Kantian 'conditions under which alone' the securing of answers to questions about the world is even possible. And prominent among these is that we can take our 'data' about the world as evidence, that a presumption of experiential veridicality is in order. In matters of sense perception, for example, we presume that mere appearances ('the data') provide an indication of how things actually stand (however imperfect an indication). That we can use the products of our experience of the world to form at least somewhat reliable views of it is the indispensable presupposition of our cognitive endeavours. If we systematically refuse, always and everywhere, to accept seeming evidence as real evidence (at any rate until the time comes when it is discredited as such), then we can get nowhere in the domain of practical cognition. We proceed in cognitive contexts in much the same manner in which banks proceed in financial contexts. We extend credit to others, doing so at first to a relatively modest extent. When and as they comport themselves in a way that indicates that this credit was warranted, then we extend more. By responding to trust in a 'responsible' way, one increases one's credit rating in cognitive much as in financial contexts. This is clearly so with our sources of information. The example of our senses is a particularly important case in point. Consider the contrast between sight and dreams. Dreams too are impressive and significant-seeming. Why then do we accept sight as a reliable cognitive source but not dreams? Surely not because of any 'inner' characteristics such as vividness, expressiveness, or memorability. The predisposition to an interest in dreams is clearly attested by their prominence in myth and literature. Our cognitive reliance on sight is not the consequence of its 'intrinsic' preferability, but is wholly the result of its success in having built up credit in the way
52
Cognitive Rationality and Risk
we have been considering. And much the same story holds for your instruments—for telescopes and microscopes, or computing machinery, and so on. We trust them because our experience with them has given us good reason to think them trustworthy. Our cognitive proceedings incorporate a host of fundamental presumptions of reliability, such as: Believe the evidence of your own senses. Accept at face value the declarations of other people (in the absence of any counter-indications and in the absence of any specific evidence undermining the generic trustworthiness of those others). Trust in the reliability of established cognitive aids and instruments (telescopes, calculating machines, reference works, logarithmic tables, etc.) in the absence of any specific indications to the contrary. Accept the declarations of established experts and authorities within the area of their expertise (again, in the absence of counter-indications). Principles of this sort are integral parts of the operational code of agents who transact their cognitive business rationally. Presumption is a matter of cognitive economy—of following 'the path of least resistance' to an acceptable conclusion. Its leading principle is: introduce complications only when you need to, always making do with the least complex resolution of an issue. There is, of course, nothing sacrosanct about the result of such a procedure. The choice of the easiest way out may fail us, that which serves adequately in the first analysis may well no longer do so in the end. But it is clearly the sensible way to begin. At this elemental level of presumption we proceed by 'doing what comes naturally'. What is fundamental here is the principle of letting appearance be our guide to reality—of accepting the evidence as evidence of actual fact, by taking its indications as decisive until such time as suitably weighty counter-indications come to countervail against them. But, just what sorts of claims are presumptively justified? The ordinary and standard probative practice of empirical inquiry stipulates a presumption in favour of such cognitive 'sources' of information as the senses and memory. And the literature further contemplates such alternatives as:
Cognitive Rationality and Risk
53
1. Natural inclination: a 'natural disposition' to accept (e.g. in the case of sense observation). 2. Epistemic utility in terms of the sorts of things that would, if accepted, explain things that need explanation. 3. Analogy with what has proved acceptable in other contexts. 4. Fit: coherence with other accepted theses. Even a weak reed like analogy—the assimilation of a present, problematic case to similar past ones—is rendered a useful and appropriate instrumentality of presumption through providing our readiest source of answers to questions.1 All in all, presumption favours the usual and the natural—its tendency is one of convenience and ease of operation in cognitive affairs. Of course, the question remains: What sorts of considerations validate our particular presumptions as such: how is it that they become entitled to this status? The basis of the answer has already been indicated. The validity of a presumption is not pre-established by some prior process of rational deliberation but emerges ex post facto through the utility (both cognitive and practical) of the results it yields. Legitimation is thus available but only through experiential retrovalidation, retrospective validation in the light of experience. For, the clear indications of experience are (1) that we do rather well in adhering to the established processes of cognitive rationality, and (2) that no alternative available to us is of better prospect and promise.2 In trusting the senses, in relying on other people, and even in being rational, we always run a risk. Whenever we place our faith in something, we risk being let down and disappointed in the end. Nevertheless, it seems perfectly reasonable to bet on the general trustworthiness of the senses, the general reliability of our fellow men, and the general utility of reason. In such matters, no absolute guarantees can be had. But, one may as well venture, for, if venturing fails, the cause is lost anyhow—we have no more promising alternative to turn to. There is little choice about the matter: it is a case of 'this or nothing'. If we want answers to factual questions, we have no real alternative but to trust in the ' This is why adherence to custom is a cardinal principle of cognitive as well as practical rationality. Cf. William James, 'The Sentiment of Rationality', in The Will to Believe and Other Essays in Popular Philosophy (New York, NY, 1897). 2 See the author's Methodological Pragmatism (Oxford, 1976) for a fuller development of this line of thought.
54
Cognitive Rationality and Risk
cognitively co-operative disposition of the natural order of things. We cannot pre-establish the appropriateness of this trust by somehow demonstrating, in advance of events, that it is actually warranted. Rather, its rationale is that without it we remove the basis on which alone creatures such as ourselves can confidently live a life of effective thought and action. In such cases, pragmatic rationality urges us to gamble on trust in reason, not because it cannot fail us, but because in so doing little is to be lost and much to be gained. A general policy of judicious trust is eminently cost-effective in the production of good results in matters of cognition. It is thus clear that all such cognition practices have a fundamentally economic rationale. They are all cost-effective within the setting of the project of inquiry to which we stand committed (by our place in the world's scheme of things). The concession of presumptive status to our pre-systematic indications of credibility is the most fundamental principle of cognitive rationality. The details of presumption management are clearly negotiable—and improvable over the course of time. But without a programmatic policy for presumption cognitive rationality cannot get under way at all. Yet, why grant those presumptions? What justifies taking that first step towards trust apart from its being convenient to do so.
4.2
SCEPTICISM AND RISK
To put scepticism into a sensible perspective, it is useful to consider the issue of cognitive rationality in the light of risk taking. There are three very different sorts of approaches to risk, and three very different sorts of 'personality' corresponding to these approaches. The general situation is summarized in Display 4.1. The Type 1 ('risk-avoidance') approach calls for risk aversion and evasion. Its adherents have little or no tolerance for risk and gambling. Their approach to risk is altogether negative. Their motto is: 'Take no chances', 'Play safe', 'Always expect the worst'. The Type 2 ('risk-calculating') approach is more 'realistic'. It is a guarded, middle-of-the-road approach to risk, based on due care and calculation. It comes in two varieties. The Type 2.1 ('cautiously calculating') approach sees risk taking
Cognitive Rationality and Risk DISPLAY
Type Type (1) (2) Type
55
4 . 1 : Approaches to Risk
1: Risk avoiders 2: Risk calculators cautious daring 3: Risk seekers
as subject to a negative, anti- presumption that can, however, be defeated by suitably large benefits. Its line is: 'Avoid risks unless it is relatively clear that a suitably large gain beckons at sufficiently auspicious odds.' It reflects the path of prudence and caution. The Type 2.2 ('daringly calculating') approach sees risk taking as subject to a positive, pro- presumption that can be defeated by suitably large negativities. Its line is: 'Take risks unless it is relatively clear that an unacceptably large loss threatens at sufficiently inauspicious odds.' It reflects the path of optimism and hopefulness. The Type 3 ('risk-seeking') approach calls for the courting of risks. Its adherents close their eyes to the dangers and take a rosy view of risk situations. The mind of the risk seeker is intent on the delightful consequences of a favourable issue of events: the sweet savour of success is already in his nostrils. Risk seekers are chance takers and go-for-broke gamblers. They react to risk the way an old war-horse responds to the sound of musketry: with eager anticipation and positive relish for the fray. Their motto is: 'Things will work out.' In the conduct of practical affairs the risk avoiders are hypercautious; with no stomach for uncertainty, they insist on playing it absolutely safe. In any potentially risky situation, the mind of the risk avoider is given to imagining the myriad things that could go wrong. The risk calculators proceed with care: they take due safeguards, but still run risks when the situation looks sufficiently favourable. The risk seekers, on the other hand, leap first and look later, apparently counting on a benign fate to ensure that all will be well; they dwell in the heady atmosphere of 'nothing can go wrong'. The three types of approach to risk correspond to three fundamentally diverse attitudes: pessimism, realism, and (over-)
56
Cognitive Rationality and Risk
optimism, respectively. Such tendencies characterize people's attitudes or 'personalities' in a way that is relatively stable over time. To be sure, armchair psychology teaches us that they are also predictably changeable within one's life span: as small children we tend to be risk avoiders; in youth we welcome risk; with maturity we become more calculating; in old age we incline to be risk aversive once more. (Actually, the situation is even more complex. For, one and the same person will at any stage of the game differ in his approach to different kinds of risk—for example, in being prepared to take chances in their investments but not in their interpersonal relationships.) The three approaches spread across the entire spectrum of human concern. This is illustrated in Display 4.2, about which some comments are in order. DISPLAY 4.2: The Three Basic Approaches to Risk in Different Contexts ETHICS/ PRAXIS
Type 1 (over-) cautiousness Type 2 prudence
Type 3
recklessness
COGNITION
MORALS
POLITICS
scepticism (demanding of certainty) evidentialism (cautious' moderation)
rigidity (strict rule-morality)
authoritarianism
pragmatism (loose-constructionist rule-morality) idealism (a morality of unrealizable idealizations)
realism
syncretism ('anything goes' laxity)
utopiamsm
In the ethical case, risk avoiders opt for a rigid rule-morality. They want things spelled out in black and white so as to know exactly what's what in terms of the expectations placed on them. They tend to insist that: 'People have to know just what is required of them.' Theirs is a legalistic approach that sticks to the rules. By contrast, risk seekers favour an ideal morality that sets few if any rules. They opt for the spirit rather than the letter of the law. Theirs is an easy-going approach that trusts to people's good sense; as they see it, 'Laying down rules discourages higher aspirations and puts a damper on works of supererogation.'
Cognitive Rationality and Risk
57
Finally, risk calculators favour the 'loose-constructionist' rulemorality that strives for a balance. Their line is: 'Rules, yes, but for general guidance and minimum adequacy only. Don't insist on rigid conformity; encourage effort above and beyond the call.' In the political case, risk avoiders desire a system that keeps people in line, a law-and-order mode of governance that 'takes no chances that things will get out of hand'. The maintenance of social regularity is their prime imperative and rigid rule-adherence their order of the day. Risk takers, on the other hand, incline to a confidence that chafes at sanctions and restraints. 'Just leave people to their own devices; don't direct, canalize, restrain; trust in their good nature and you'll release initiative and open wellsprings of good feeling that will make for a better social order.' They take a bright, hopeful, 'optimistic' view of human nature. Finally, risk calculators want to be 'realistic'—to try on the basis of experience to find a via media that combines restraint and encouragement, sanction and incentive, the carrot and the stick. The cognitive case is pivotal for our present concerns. Here, risk avoidance leads straight away to scepticism. The sceptic's line is: 'Run no risks of error; take no chances; accept nothing that does not come with iron-clad guarantees.' (And the proviso here is largely academic, seeing that little if anything in this world comes with iron-clad guarantees.) The daring syncretist stands at the other end of the spectrum. Like radical Popperians, such as P. K. Feyerabend, the syncretist is inclined to think that anything goes. He is 'gullible', as it were, and stands ready to endorse everything and see good on all sides. The evidentialist, by contrast, conducts his cognitive business with care and caution, regarding various sorts of claims as perfectly acceptable, provided that the evidential circumstances are duly favourable. The sceptic will accept nothing, the evidentialist only 'the chosen few', the syncretist inclines favourably towards virtually everything. So much, then, for the three basic approaches to risk in various areas of endeavour. It is important to recognize that two fundamentally different sorts of misfortunes are possible in situations where risks are run and chances taken: Misfortunes of the 1st kind: We reject something that, as it turns out, we should have accepted. We decline to 'take the chance'
58
Cognitive Rationality and Risk
and avoid running the risk at issue, but things turn out favourably after all, and we 'lose out on the gamble'. Misfortunes of the 2nd kind: We accept something that, as it turns out, we should have rejected. We do 'take the chance' and run the risk at issue, but things go wrong, and we 'lose the gamble'. If we are risk seekers, we will incur few misfortunes of the first kind, but—things being what they are—relatively many of the second kind will befall us. Conversely, if we are risk avoiders, we shall suffer few misfortunes of the second kind, but shall inevitably incur relatively many of the first. The overall situation is depicted in Display 4.3. D I S P L A Y 4 . 3 : Risk Acceptance and Misfortunes
Number of (significant) misfortunes
0
50 Type 1 (Risk avoiders)
Type 2.1 (Cautious calculators)
100 Type 3 (Risk seekers)
Type 2.2 (Daring calculators)
Increasing risk acceptance (in % of situations)
The rationally sensible thing to do is clearly to adopt a policy that minimizes misfortunes overall. It is thus evident that both Type 1 and Type 3 approaches will—in general—fail to be rationally optimal. Both approaches engender too many misfortunes for comfort. The sensible thing is to adopt the 'middle-of-the-road'
Cognitive Rationality and Risk
59
policy of risk calculation, acting as best we can to balance the positive risks of outright loss against the negative ones of lost opportunity. The path of reason calls for sensible calculation and prudent management: it standardly enjoins upon us the Aristotelian 'golden mean' between the extremes of risk avoidance and risk seeking. Rationality thus counter-indicates approaches of Type 1 and Type 3. Its line is, 'Neither avoid nor court risks, but manage them prudently in the search for an overall minimization of misfortunes', and it insists on proceeding by way of carefully calculated risks. In the cognitive case, in particular, the sceptic succeeds splendidly in averting misfortunes of the second kind. By accepting nothing, he accepts nothing false. But, of course, he loses out on the opportunity to obtain any sort of information. The sceptic thus errs on the side of safety, even as the syncretist errs on that of gullibility. In claiming that his position wins out because it makes the fewest mistakes, the sceptic uses a distorted system of scoring. For, while he indeed makes the fewest errors of one kind, he makes the most of another. Once we look on this matter of 'making mistakes' realistically, the sceptic's vaunted advantage vanishes. The sceptic is simply a risk avoider who is prepared 'to take no risks' and stubbornly insists on minimizing errors of the second kind alone. After all, what we want in inquiry—the object of the whole enterprise—is information. What we seek is the very best overall balance between answers to our questions and ignorance or misinformation. We face a trade off at this stage, however. Are we prepared to run a greater risk of error to secure the potential benefits of greater understanding? The judicious cognitivist is a risk calculator who recognizes the value of understanding and is prepared to gamble for its potential benefits. H. H. Price has put the salient point well: 'Safety first' is not a good motto, however tempting it may be to some philosophers. The end we seek to achieve is to acquire as many correct beliefs as possible on as many subjects as possible. No one of us is likely to achieve this end if he resolves to reject the evidence of testimony, and contents himself with what he can know, or have reason to believe, on the evidence of his own firsthand experience alone. It cannot be denied that if
60
Cognitive Rationality and Risk
someone follows the policy of accepting the testimony of others unless or until he has specific reason for doubting it, the results will not be all that he might wish. Some of the beliefs which he will thereby acquire will be totally incorrect, and others partly incorrect. In this sense, the policy is certainly a risky one . . . but it is reasonable to take this risk, and unreasonable not to take it. If we refuse to take it, we have no prospect of getting answers, not even the most tentative ones, for many of the questions which interest us.3 What Price says here about the testimony of others is even truer when applied to the testimony of our own senses. Again, in the ethical case, the rigid rule-moralist operates with a standard of approval that prevents him from making the mistakes of endorsing 'unworthy' actions. But he loses out at the opposite end by disapproving all sorts of things that intuitively seem altogether meritorious. The rigoristic rule-moralist errs on the side of censoriousness—even as the moral idealist will err on the side of slackness. The general situation is as follows. We have an intuitive, pre-systematic inclination to class certain acts as acceptable (+) or unacceptable ( - ) . In the context of its applications, our formal theory can run into 'problems' of two kinds. With problems of the first kind, theory condemns (classes as - ) something intuition approves (classes as + ) . With problems of the second kind, the situation is reversed: theory approves (classes as + ) something intuition condemns (classes as - ) . The most acceptable overall policy is presumably one that minimizes the overall volume of difficulties encountered. In general, then, the course of wisdom is to position one's standard of acceptance in the middle ground between the extremes—to try for the best prospect of minimizing errors overall. The details will of course hinge crucially on the exact form of those Display 4.3 curves, as shaped in the light of our standards for the appraisal of misfortunes. There is no single, uniquely rational policy of procedure in the face of risk—some scope must be allowed for the influence of one's subjective constitution—the sort of person one is. But the course of rationality clearly lies in a thoughtful endeavour to do the best one can, over all, within the constraints of one's situation.
3
H. H. Price, Belief (London, 1969), 128.
Cognitive Rationality and Risk 4.3
61
THE D E F I C I E N C Y OF S C E P T I C I S M
The sceptic too readily loses sight of the raison d'etre of our cognitive endeavours. The object of rational inquiry is not just to avoid error but to answer our questions, to secure information about the world. And here, as elsewhere, 'Nothing ventured, nothing gained' is the operative principle. Granted, a systematic abstention from cognitive involvement is a sure-fire safeguard against one kind of error. But, it affords this security at too steep a price. The shortcoming of that 'no risk' option is that it guarantees failure from the very outset. It is self-defeating to follow the radical sceptic into letting discretion be the whole of epistemic valour by systematically avoiding accepting anything whatsoever in the domain of empirical fact. To be sure, when we set out to acquire information we may well discover in the end that, try as we will, success in reaching our goal is beyond our means. But we shall certainly get nowhere at all if we do not even set out on the journey—which is exactly what the sceptic's blanket proscription of acceptance amounts to. In 'playing the game' of making assertions and laying claims to credence, we may well lose: our contentions may well turn out to be mistaken. But, in a refusal to play this game at all we face not just the possibility but the certainty of losing the prize—we abandon any chance to realize our cognitive objectives. A sceptical policy of systematic avoidance of acceptance is fundamentally irrational, because it blocks from the very outset any prospect of realizing the inherent goals of the enterprise of factual inquiry. In cognition, as in other sectors of life, there are no guarantees, no ways of averting risk altogether, no option that is totally safe and secure. The best and most we can do is to make optimal use of the resources at our disposal to 'manage' risks as best we can. To decline to do this by refusing to accept any sort of risk is to become immobilized. The sceptic thus pays a great price for the comfort of safety and security. If we want information—if we deem ignorance no less a negativity than error—then we must be prepared to 'take the gamble' of answering our questions in ways that risk some possibility of error. A middle-of-the-road evidentialism emerges as the most sensible approach. Perhaps, no other objection to radical scepticism in the factual
62
Cognitive Rationality and Risk
domain is as impressive as the fact that, for the all-out sceptic, any and all assertions about the world's objective facts must lie on the same cognitive plane. No contention—no matter how bizarre—is any better off than any other in point of its legitimative credentials. For the thoroughgoing sceptic there just is no rationality-relevant difference between 'More than three people are currently living in China' and 'There are at present fewer than three automobiles in North America'. As far as the cognitive venture goes, it stands committed to the view that there is 'nothing to choose' in point of warrant between one factual claim and another. Radical scepticism is an H-bomb that levels everything in the cognitive domain. The all-out sceptic writes off at the very outset a prospect whose abandonment would only be rationally defensible at the very end. As Charles Sanders Pierce never tired of maintaining, inquiry only has a point if we accept from the outset that there is some prospect that it may terminate in a satisfactory answer to our questions. He indicated the appropriate stance with trenchant cogency: 'The first question, then, which I have to ask is: Supposing such a thing to be true, what is the kind of proof which I ought to demand to satisfy me of its truth?'4 A general epistemic policy which would as a matter of principle make it impossible for us to discover something which is ex hypothesi the case is clearly irrational. And the sceptical proscription of all acceptance is obviously such a policy— one which abrogates the project of inquiry at the very outset, without according it the benefit of a fair trial. A presumption in favour of rationality—cognitive rationality included—is rationally inescapable. It could, to be sure, eventuate at the end of the day that satisfactory knowledge of physical reality is unachievable. But, until the end of the proverbial day arrives, we can and should proceed on the idea that this possibility is not in prospect. ('Never bar the path of inquiry', Peirce rightly insisted. The trouble with scepticism is that it aborts inquiry at the start.) The sceptic's favoured approach is geared to the level of individual contentions (propositions). He insists on addressing the issue of rational acceptance at the level of particular theses: are we really in a position to accept this or that p or not? And he then finds various seemingly plausible grounds for not doing so. But the 4
C. S. Peirce, Collected Papers, ii (Cambridge, Mass., 1931), sect. 2.112.
Cognitive Rationality and Risk
63
real issue is methodological, and relates to our general procedure with regard to acceptance. The sceptic's mistake is one of omission -he fails to see that the fundamental issue we face is not just that of particular acceptances, but that of a choice between policies of acceptance. The question 'Is the particular claim p justified in the prevailing circumstances?' is best approached indirectly. The sensible move is to step back and begin with the question: What are the appropriate methods (standards, criteria) for acceptability-determination? Once this issue is resolved, we then have in hand the instruments for deciding the justificatory status of p. To deal effectively with scepticism we had best begin by dealing with methods (standards, criteria) rather than particular theses.5 Our basic concern is—and should be—with the state of things at the level of cognitive policies. We clearly require some policy of acceptance, some general stance on the issue that views particular claims in the light of general principles. And whatever policy we adopt is one that we must be prepared to justify. But a non-sceptical policy is one that is validated vis-a-vis its rivals by the sort of systemic considerations that we have canvassed—that is, by minimizing, as much as possible, the overall extent to which one runs into problems and difficulties. And this is the policy of a cautiously managed evidentialism of just the sort at issue in scientific method. But, of course, once we have an acceptance policy in place, the issue of propositional acceptance—acceptance in those particular cases against which the sceptic rails—is now settled for us. Our hands are tied and no longer free: what it is appropriate to accept is now something that the policy we adopt prescribes, because that is exactly the sort of policy it is—an acceptable policy. At this point, having a general policy in place, the option of scepticism lies behind us—on the far side of a point of no return. Such a 'refutation' of sceptism does not proceed at the item level of showing that the sceptic's view of the matter is untenable in this or that particular case. Rather, it proceeds at the policy level, showing the pragmatic superiority of adopting a line of approach at variance with the sceptic's. Viewing the issue in this pragmatic light, we see that the sceptic's risk-avoidance policy is simply not 5 This position is developed at length in the author's Methodological Pragmatism (Oxford, 1977).
64
Cognitive Rationality and Risk
one that it is rational to adopt. It has to be recognized that any systematic cognitive policy that we could possibly implement in the real world is bound to allow some errors to arise. Given that errors can be of the two kinds we have been considering, there simply is no way of averting errors altogether. Now, the sceptic's problem is that his preferred stance represents a particular policy choice ('Accept nothing!') that is not only not mistake-proof, but rather a mistake-inviting, definite commitment of procedure—even though its only mistakes are those of the first kind. Viewed from the vantage point of representing a particular policy alternative, the sceptic's difficulty is that he avoids mistakes of only one kind, thereby maximizing those of another. 4.4
A G A I N S T S C E P T I C I S M : THE PRAGMATIC D I M E N S I O N
The radical sceptic's seemingly high-minded insistence on securing irrefrangible truth rather than settling for reasonable warrant for acceptance—duly followed by the mock-tragic recognition that this is actually unachievable—is totally counter-productive. It blocks from the outset any prospect of staking reasonable claims to information about the ways of the world. And this is a grave defect. The 'discomfort of unknowing' is a natural human sentiment. To be ignorant of what goes on about one is dangerous from an evolutionary point of view. As William James wisely observed: The utility of this emotional effect of expectation is perfectly obvious; 'natural selection', in fact, was bound to bring it about sooner or later. It is of the utmost practical importance to an animal that he should have prevision of the qualities of the objects that surround him.6 Man has evolved within nature to fill the ecological niche of an intelligent being. The demand for understanding, for a cognitive accommodation to one's environment, for 'knowing one's way 'about' is one of the most fundamental requirements of the human condition. Man is homo quaerens. We have questions and want (nay, need), answers. The need for information, for cognitive orientation in our environment, is as pressing a human need as that for food itself. We are rational animals and must feed our minds even as we must feed our bodies. In pursuing information, 6
William James, 'The Sentiment of Rationality', op. cit. pp. 78-9.
Cognitive Rationality and Risk
65
as in pursuing food, we have to settle for the best we can get at the time. We have questions and need answers—the best answers we can get here and now, regardless of their imperfections. The basic demand for information and understanding presses in upon us, and we are impelled towards (and are pragmatically justified in) bestirring ourselves towards getting it satisfied. The great Norwegian polar explorer F. Nansen put it well. What drives men to explore the uninviting polar regions, he said, is the power of the unknown over the human spirit. As ideas have cleared with the ages, so has this power extended its might, and driven Man willynilly onwards along the path of progress. It drives us in to Nature's hidden powers and secrets, down to the immeasurably little world of the microscopic, and out into the unprobed expanses of the Universe. . . . it gives us no peace until we know this planet on which we live, from the greatest depth of the ocean to the highest layers of the atmosphere. This Power runs like a strand through the whole history of polar exploration. In spite of all declarations of possible profit in one way or another, it was that which, in our hearts, has always driven us back there again, despite all setbacks and suffering.7 With us, the imperative to understanding is something altogether basic: things being as they are, we cannot function, let alone thrive, without knowledge of what goes on about us. The knowledge that orients our activities in this world is itself the most practical of things—a rational animal cannot feel at ease in situations of which it can make no cognitive sense. We have questions and want (nay, need), to have answers to them. And not just answers, but answers that cohere and fit together in an orderly way can alone satisfy a rational creature. This basic practical impetus to (coherent) information provides a fundamental imperative to cognitive intelligence. We humans want and need our cognitive commitments to comprise an intelligible story, to give a comprehensive and coherent account of things. For us, cognitive satisfaction is unattainable on any other basis—the need for information, for knowledge to nourish the mind, is every bit as critical as the need for food to nourish the body. Cognitive vacuity or dissonance is as distressing to us as physical pain. Bafflement and ignorance—to give suspensions of judgement the somewhat harsher name that is 7
200.
Quoted in Roland Huntford, The Last Place on Earth (New York, NY, 1985), ».,«V 0 .^«, >«•-•:,
66
Cognitive Rationality and Risk
their due—exact a substantial price from us. The quest for cognitive orientation in a difficult world represents a deeply practical requisite for us. That basic demand for information and understanding presses in upon us and we must do (and are pragmatically justified in doing) what is needed for its satisfaction. For us, cognition is the most practical of matters: Knowledge itself fulfils an acute practical need. To be sure, even the most radical of sceptics have historically recognized that man must act to survive and thrive in the world. Like everyone else, the sceptics acknowledge that we humans find ourselves emplaced in media res in an environment which will not satisfy our needs, wants, and desires automatically, without intervention on our part. This concession opens scepticism to the charge of immobilizing action, invoking a 'refutation' of scepticism on the grounds that it makes the conduct of life itself impossible. David Hume put this point as follows: But a Pyrrhonian . . . must acknowledge, if he will acknowledge anything, that all human life must perish, were his principles universally and steadily to prevail. All discourse, all action would immediately cease; and men remain in a total lethargy, till the necessities of nature, unsatisfied, put an end to their miserable existence.8 To this sort of charge, the ancient sceptics always replied that while we must indeed act, this action need not be based on knowledge at all. They have insisted on the sufficiency of noncognitive guides for action—'appearances', 'custom', 'the general consensus', 'instinct', or the like. As Sextus Empiricus insisted time and ag ain, the springs of action are desire and aversion—seeking and avoiding—and these can operate without the intervention of any sort of credence— without our endorsing some actual thesis to the effect that this or that is really the case. Life without knowledge or rationally evidenced belief is certainly not in principle impossible: animals, for example, manage very well. Or again, a somewhat less radical strategy is available—one that countenances acceptance (and belief), but only on a wholly unreasoned basis (say, instinct, 8 David Hume, An Enquiry Concerning Human Understanding, sect, xii, pt. ii. Compare John Locke; 'He that will not eat till he has demonstration that it will nourish him; he that will not stir till he infallibly knows the business he goes about will succeed, will have little else to do but sit still and perish' (Essay Concerning Human Understanding, bk. IV, ch. xiv, sect. 1).
A '} i
Cognitive Rationality and Risk
67
constraint by appearances, or the like). The sceptic can accordingly hold—and act on—all those beliefs which people ordinarily adopt, with only this difference, that he regards them as reflecting mere appearances, and denies that the holding of these beliefs is rationally justified. Hume's objection hits wide of the mark: scepticism need not immobilize action. Still, this sort of defence misses the real point. The core of the objection is not simply that the sceptic fails to favour some assertions over others as a basis for action. It is rather that he insists that there just are no REASONS for doing so. And this is deeply problematic. For, while the sceptic may indeed have guides to action—namely the non-cognitive guides of instinct or custom, etc.—he cannot defend his actions. He cannot justify doing A rather than B. He can tell us that he eats food to assuage hunger (rather than merely rubbing his stomach) because that is the done thing, but this mere explanation of what he does does not constitute a ground for it. Scepticism thus destroys the prospect of any rational recourse to the processes of praxis. Given the sceptic's total suspension of judgement, our behaviour becomes not necessarily irrational but altogether irrationalizable. AH linkage between action and rationality is severed. The scepticism-rejecting policy of allowing inherently nonconclusive evidence to justify the acceptance of beliefs is ultimately a practical policy. This approach takes the practicalist stance that even as it is appropriate in other cases for 'the prudent person' to do what is advantageous, so it is also in this matter of acceptance. It treats prepositional acceptance as just another possible human act—albeit one of a somewhat special sort, namely, a cognitive one of thesis endorsement. It is a matter of adopting or accepting certain contentions because in the circumstances doing so demonstrably furthers the realization of one's cognitive ends—practical as well as theoretical. It is the basic practical impetus to coherent information that underlies the fundamental imperatives of cognitive intelligence: 1. Do the best you can to obtain adequate answers to your questions! 2. Feel free to adopt those answers, to act on the principle that we must make do with the best we can get as good enough for present purposes.
68
Cognitive Rationality and Risk
In pursuing information, as in pursuing food, we have little alternative but to settle for the best we can get at the time. (We may have to make do with conjecture and make guesswork serve for knowledge—but that is rationally acceptable if it indeed is the very best we can do.) It is not difficult to discern the unsatisfactory nature of the sceptic's non-cognitive strategies for the guidance of action. For, in being non-cognitive, these approaches are also non-rational (which is, to be sure, something quite different from the irrational). While they indeed tell us what to do, they are stonily silent on the crucial issue of why it is to be done. And this is eminently unsatisfying. Man is an ineradicably rational animal: a creature is geared to satisfy not only its physical but its intellectual hunger. In virtually every circumstance and situation one needs 'to know the reason why'. The insistence on non-cognitive guidance is a line the sceptic himself may willingly take, but there is no earthly reason why those of us who are not already precommitted to his views should join him in doing so. The argument against scepticism deployed here is thus an essentially practical one. It does not establish the internal inconsistency or theoretical untenability of a sceptical position. Rather, it shows that the price we would pay in taking such a position is so high as to outweigh any real benefit that could possibly accrue from it. That basic demand for information and understanding presses in upon us, and we must do—and are thus pragmatically justified in doing—whatever it takes to get it satisfied. Whatever the merits or demerits of scepticism as a theoretical position, we are entitled on practical grounds to dismiss it unceremoniously. Since the days of Greek antiquity, philosophers have often answered our present question 'Why accept anything at all?' by taking the following stance: 'Man is a rational animal. Qua animal he must act, since his very survival depends upon action. But qua rational being he cannot act availingly save in so far as his actions are guided by what he accepts.' This line of argumentation was voiced by the ancients and revived in modern times by a succession of thinkers from David Hume to William James.9 Note, however, 9
As William James said: '(Someone) who says "Better to go without belief forever than believe a lie!" merely shows his own preponderant private horror of
Cognitive Rationality and Risk
69
that the present position does not run: 'If you want to act effectively then you must accept something.' Rather, its line is: 'If you want to enter into "the cognitive enterprise"—that is, if you wish to be in a position to secure information about the world— then you must be prepared to accept something.' Both approaches take a stance that is not categorical and unconditional but rather hypothetical and conditional. But, in the classically pragmatic case the focus is upon the requisites for effective action, while our present, cognitively oriented, approach has focused upon the requisites for rational inquiry. In the present perspective, it is the frustration of our basic cognitive aims (no matter how much the sceptic himself may be willing to turn his back upon them) that provides the salient theoretical reason for the rejection of scepticism.
4.5
S C E P T I C I S M V S . RATIONALITY
The sceptic seemingly moves within the orbit of rationality, but only seemingly so. For, in fact scepticism runs afoul of the only promising epistemological instrumentalities that we have. To be sure, there will have to be a justification for our epistemological procedures. And, if it were simply this rationalization that the sceptic asked for his demands would not be unreasonable—and could in principle be met. But, the defence itself would, of course, have to be conducted within the framework of the standard modus operandi of rational argumentation. There is and can be no rational justification outside the domain of rationality itself. The radical sceptic's demand for a justification of rationality ex nihilo is inappropriate because it defines its problem in such a way that any resolution is in principle unfeasible. Philosophical sceptics generally set up some abstract standard of absolutistic certainty and then try to show that no knowledge claims in a certain area (sense, memory, scientific theory, and the like) can possibly meet the conditions of this standard. From this circumstance, the impossibility of such a category of 'knowledge' is accordingly inferred. But this inference is totally misguided. For, what follows is rather the inappropriateness or incorrectness of the becoming a dupe . . . but I can believe that worse things than being duped may happen to a man in this world' (op. cit., pp. 18-19).
70
Cognitive Rationality and Risk
standard at issue. If the vaunted standard is such that knowledge claims cannot possibly meet it, the moral is not 'too bad for knowledge claims', but 'too bad for the standard'. Any position that precludes in principle the possibility of valid knowledge claims thereby effectively manifests its own unacceptability. The sceptic's argument is a double-edged sword that cuts both ways and inflicts the more serious damage upon itself. It is senseless to impose on something qualification conditions which it cannot in the very nature of things meet. An analogue of the old Roman legal precept is operative here—one is never obliged beyond the limits of the possible (ultra posse nemo obligatur). It cannot rationally be required of us to do more than the best that is possible in any situation—cognition included. But rationality also conveys the comforting realization that more than this cannot be required of us: we are clearly entitled to see the best that can be done as good enough. In a last-ditch defence of his position, the radical sceptic may still take the following line: 'Departing from your view of 'rationality' and falling short by your standards is not something that I need regard as a genuine failing. Indeed, my very thesis is that your "rationality" has no suitable credentials.' To meet this desperate but profound tactic we must shift the ground of argumentation. It now becomes advantageous to approach the whole issue from a new point of departure, namely that of the prospects of communication. For, it turns out that not only does his refusal to accept claims preclude the radical sceptic from participating in the enterprise of inquiry, he is blocked from the enterprise of communication as well. All informative communication is predicated on the fundamental convention that normally and standardly what one declares to be so is something that (1) one accepts as true, and (2) one claims to be rationally warranted in accepting.10 In rejecting the ground rules of our reasoning as inappropriate, the sceptic also abandons the '" Peter Unger suggests a 'reformation' of language in the interests of scepticism: 'We want linguistic institutions and practices where our (universal) ignorance will not enjoin silence' {Ignorance (Oxford, 1975), 271). But, he offers no concrete proposals along these lines, and this is very understandable. It is difficult to see how a language could be formed, taught, and above all used in which assertion played no role and declaration carried with it no claims to veracity. And even if (per impossibile) one had such a 'language', what would be the point of using it in communicative contexts?
Cognitive Rationality and Risk
71
ground rules of communication. In denying the prospect of any sort of rational warrant—however tentative—the all-out sceptic embarks upon a self-imposed exile from the community of communicators, seeing that the communicative use of language is predicated on conceding the warranting presuppositions of language use. To enter into a discussion at all, one must acquiesce in the underlying rules that make discussion in general possible. But, if nothing can appropriately be accepted, then no rules can be established—and thus no statements made, since meaningful discourse requires rules. Communication becomes impossible. A radical scepticism, in the final analysis, engenders the collapse of communication and deliberation, thereby leading to a withdrawal from the human community. It is not possible both to engage in enterprises that in their very nature involve a subscription to standards—enterprises like linguistic communication—and at the same time to refuse to subscribe to standards. Where nothing can be said appropriately, nothing can be communicated. Perceptive sceptics have not been blind to this fact; it was already remarked by Pyrrho himself— witness his famous reply to a critic that it is not easy to divest oneself entirely of one's humanity." Historically, sceptics have, to be sure, occasionally been prepared to grant that they are destroyers of Aoyo? (reason and discourse), insisting that they use reason and discourse simply as instruments for their own destruction. Some have been prepared to grasp the nettle and admit—perhaps even welcome—the consequence that their position does not only gainsay knowledge, but actually rejects the whole project of cognitive rationality (at any rate at the level of our factual beliefs). But, whatever satisfaction this posture may afford the sceptic it is hardly likely to appeal to those who do not already share his position. The collapse of the prospect of rational inquiry and communication is the ultimate sanction barring the way to any rational espousal of radical scepticism. It is a price that a fanatically dedicated devotee of the sceptical position may be willing to meet, but it is clearly one that people who are not so pre-committed cannot possibly pay. Sensible people require cogent reasons for what they do, and only with the abandonment of a rigoristic all-out scepticism can such " Diogenes Laertius, Lives of the Philosophers, ix. 66.
72
Cognitive Rationality and Risk
reasons be obtained. We must be realistic about cognitive rationality, and must be sensible about what it is that we demand of knowledge, lest we espouse a perfectionism that leaves us empty-handed and deprives us of the cognitive sustenance that is a practical imperative of our condition. Despite its deficiencies, scepticism does serve one eminently useful function, it reminds us of the inherent riskiness of claims to certainty, knowledge, definitive truth. We cannot say that simply because a certain claim is cognitively warranted, plausible, justified (or the like) what it asserts is totally certain, absolutely correct, established beyond possibility of error. The sceptical tradition reminds us that all our claims to knowledge and truth carry some element of risk—that between the cognitive justification of a factual claim and its actual truth there always is a gap that needs to be filled, and that in the practical circumstances of our inquiries this gap can be narrowed to triviality but never completely closed. The sceptic is right to insist that some theoretical risk of error always characterizes our factual claims. Where he is wrong is in maintaining that this risk is never worth running. To be sure, merely theoretical argumentation cannot dislodge the sceptic from his stance of accepting no theses at all. Argumentation is unavailing because any probatively cogent argument must proceed from conceded premisses and the sceptic can always simply refuse to make concessions. All that argumentation can do is to forestall scepticism by showing the incompatibility of sceptic with positions acceptable to sensible people in general. Such argumentation may not dislodge someone from a sceptical position, but it should prevent somebody who has not (yet) taken this position from doing so. The only sort of critique of scepticism that it makes sense to ask for is a rational critique. And, viewed from this standpoint, the decisive flaw of scepticism is that it makes rationality itself impossible.12 12 This chapter draws upon the author's Scepticism (Oxford, 1980). For an interesting discussion of relevant issues see also John Kekes, A Justification of Rationality (Albany, 1976).
5 Cognitive Rationality and Consistency
SYNOPSIS (1) Is the maintenance of consistency—the avoidance of selfcontradiction, or at any rate recognized self-contradiction—an indispensable requisite and absolute sine qua non of cognitive rationality? Presumably not. Modern innovations in non-standard logic have rendered inconsistency-tolerance a real option. (2) The distinction between belief as certain and a more tentative sort or belief as provisionally acceptable is important for the epistemology of inconsistency-tolerance. (3) A total refusal to countenance inconsistency can exact a high cognitive cost in terms of ignorance and cognitive impoverishment. (4) In various inquiries it makes perfectly good sense to proceed dialectically by reasoning from seemingly acceptable yet nevertheless incompatible premisses. (5) Differentiations of 'context' can prevent the clashes that arise here from becoming altogether vitiating or vicious. In the human sciences in particular, inconsistency-tolerance can be strategically advantageous. (6) The maintenance of consistency—like the achievement of completeness, uniformity, or truth—must certainly count as an epistemic desideratum, something to be cultivated and prized. But it is not an absolutely indispensable requisite in whose absence the whole cognitive enterprise comes to grief.
5.1
C O N S I S T E N C Y : A B S O L U T E R E Q U I S I T E OR ULTIMATE
IDEAL?
Inconsistency has had a bad press for a long time. From classical antiquity onwards, theorists have insisted on its total exclusion from the precincts of rational thought. Aristotle was already emphatic on this head: For it is impossible for anyone to believe the same thing both to be and not to be, as some think Heraclitus says. For . . . it is not possible for contraries to belong at the same time to the same thing (and let us add the customary qualifications to this premiss as well), and the contrary of a belief is the belief in its contradictory, it is apparent that it is impossible
74
Cognitive Rationality and Consistency
for the same person to believe at the same time that the same thing is and is not; for someone who went wrong about this would have contrary beliefs. That is why everyone who proves [anything] comes back to this belief in the end, for it is by nature the principle for all other axioms also.! Most logicians have followed Aristotle's lead here, holding that anything whatsoever follows from inconsistent premisses, as per the argument: 1. 2. 3. 4.
P (premiss) not-P (premiss) P-ox-Q, from 1 Q, from 2 and 3
Accepting inconsistent premisses thus seemingly commits one to the absurdity of accepting anything whatsoever. Accordingly, it was long held that inconsistent beliefs are ipso facto totally absurd. Only logically incompetent people could possibly indulge in inconsistency. And so, anthropologists like Durkheim and LevyBruhl have often regarded inconsistency as the hallmark of 'primitive' and 'pre-logical' thought that 'does not bind itself down, as our thought does, to avoiding contradiction'. 2 Of late, however, logical theorists have developed various systems of non-standard logic ('relevance' logics, 'paraconsistent' logics, 'dialectical' logics, and the like), all of which block the preceding argumentation and avert the catastrophic consequence that inconsistent premisses entail anything. In one way or another, all of them so operate that inconsistency does not automatically lead to cognitive disaster. Such investigations have shown that one emphatically need not adopt the stance that once P and not-P are both admitted, then 'anything goes' and any arbitrary Q obtains. Inconsistency does not spread like a logical cancer into an all-pervasive logical chaos: it can represent a local and not necessarily global anomaly. It is not inevitably ubiquitous and destructive, but may prove to be sporadic and harmless, like a 'singularity' in mathematics. It is helpful to note that the idea of tolerating inconsistencies 1
Metaphysics, iii. 3, 1056b30-5. Lucien Levy-Bruhl, How Natives Think, trans. Lilian A. Clare (New York, NY, 1966), 63 (and cf. 263). See also £mile Durkheim and Marcel Mauss, Primitive Classification, trans. Rodney Needham (Chicago, 1963), 5-6, 21, 70-1, 73, 88-9. 2
Cognitive Rationality and Consistency
75
may be construed in the following four ways, arranged in order of increasing unpalatability. 1. Weak Inconsistency: To admit the prospect that for some thesis p we knowingly accept both it and its negation: A(p) and A(—p), for some p. 2. Strong Inconsistency: To admit the prospect that: A(p & —p), for some p. 3. Hyperinconsistency: To admit the prospect that: A(p & -p), for all p. 4. Logical Chaos: To admit the prospect that: A(p), for all p (and accordingly A(p) and A(—p), for all p). There are inconsistencies and inconsistencies. And the presently envisioned tolerance of inconsistency emphatically does not extend beyond case 1. It is appropriate to join with Aristotle in the general (and doubtless right-minded) abhorrence for proposals 24. Consider the supposition that a world is inconsistent in mode 2 (which 3 and 4 also entail). Only one rationally viable perspective now seems possible. While cogent considerations might move us to accept the truth of p and (concurrently) that of not-p—in that 'there is much to be said on both sides'—surely nothing could (rationally) move us to accept an outright SELF-contradiction of the form: p & not-/?. The point is at bottom an epistemic one—no cogent line of consideration could ever move us to accept such a contention, and a world-picture in which this circumstance is projected can thus serve no workable ontological purpose. We have little choice but to regard the very hypothesis we are making as self-destructive—as simply annihilating itself. In uttering an outright self-contradiction we literally 'cannot say' what we have in mind to assert. On this basis, weak consistency should be viewed as a condition of rational intelligibility. The size of a manifold of inconsistency clearly matters. Any single statement of ours must not be self-inconsistent. Our smallscale theories too should be consistent. Local consistency is crucial, but globally, within entire disciplines—let alone the body of our 'knowledge' as a whole—inconsistency can be tolerated with greater equanimity. Such a toleration of weak inconsistency is not all that bizarre or far-fetched. After all, no criterion of acceptability that is capable of affording us a 'realistic' standard of acceptability can prevent
76
Cognitive Rationality and Consistency
occasional contradictions from slipping through. No screening mechanism is perfect. Realistically speaking, we have to be prepared to tolerate occasional inconsistencies within the overall structure of our knowledge—not locally by accepting p-and-not-p in some particular setting, but globally, through/? being realized in one setting and not-p in another. To be sure, such a perspective only indicates the logical viability of inconsistency-tolerance. The question remains: Is it something that we would ever actually want to avail ourselves of? It is one thing to concede the technical feasibility of something (say, squeezing twelve college students into a telephone booth). And it is something very different to concede that it makes good sense to realize this prospect. Could an adequate theory of cognition tolerate inconsistencies? At this stage, a crowd of orthodox epistemologists rise up in opposition. Holding aloft the proud banner of cognitive rationality, they insist that it is utterly irrational ever to accept inconsistency—that no sensible basis can be found, ever or anywhere, to justify the acceptance of inconsistent contentions. 3 However, it will be maintained here that, on the contrary, consistency, while unquestionably representing an important cognitive desideratum (as do certainty and completeness), is nevertheless not an indispensably necessary precondition for sensible belief and deliberation. Desirable though it is, consistency is not something on which we must insist unconditionally from the very outset, before embarking on the enterprise at all.
5.2
P R E L I M I N A R I E S : TWO VERY D I F F E R E N T SORTS OF ACCEPTABILITY
In contemplating the prospect of 'accepting' incompatible claims, very different sorts of things can be envisioned. In particular, 3 A small literature on the topic has sprung up in recent years. Some useful contributions include: Robert Ackerman, Belief and Knowledge (Garden City, NJ, 1972); Richard Foley, 'Justified Inconsistent Beliefs,' American Philosophical Quarterly, 16 (1979), 247-57; idem, 'Is it Possible to Have Contradicting Beliefs?, Midwest Studies in Philosophy, 10 (Minneapolis, Minn., 1986), 327-55; Jaakko Hintikka, Knowledge and Belief (Ithaca, NY, 1962); Keith Lehrer, Knowledge (Oxford, 1974); Nicholas Rescher and Robert Brandom, The Logic of Inconsistency (Oxford, 1979); Marshall Swain, 'The Consistency of Rational Belief, in his Induction, Acceptance, and Rational Belief (Dordrecht, 1970), 25-54.
Cognitive Rationality and Consistency
77
acceptability as a matter of certain fact must be contrasted with acceptability as provisionally credible. The former is a matter of unqualified endorsement, the latter one of merely tentative or provisional acceptance. Taking a contention as an established fact is something very different from merely accepting it qualifiedly as the most promising prospect available at this stage. We do not regard the great bulk of what we accept in theoretical and scientific matters as graven in granite for all ages to come. To motivate these ideas, consider the situation depicted in Display 5.1. Clearly, this state of affairs confronts us with an outright contradiction. Something presumably must give way. But what? DISPLAY
5 . 1 : A Paradox of Rational Belief
1. 2. 3. 4.
One believes it to be the case that p. One is a rational agent. Rational agents always act on their beliefs. One will act (in any and all circumstances) on one's belief thatp (from 1-3). 5. One recognizes (concedes, allows) that there is some small (remote) chance that p might be false. 6. One is offered a bet that will pay one cent if p is true and will engender an awful catastrophe (say the end of all organic life in the universe) if not-p is true. 7. One would bet on p in this case (from 4 and 6). 8. Rational agents do not allow minute inducements to impel them into running even very small chances of (sufficiently great) disasters. 9. One would bet on not-/? in this case (from 5, 6, and 8). Note: 1 contradicts 9. Since 1 and 2 and 6 are merely hypotheses here, they stand secure. Moreover, theses 4 and 7 both follow logically from others. Three alternatives are thus open as exits from contradiction: 1. One can abandon 3 and dissociate rational belief from action. 2. One can abandon 5 as untenable in the presence of 1 and 2, refusing to view one's beliefs in a fallibilistic perspective.
78
Cognitive Rationality and Consistency
3. One can abandon 8 and regard rational agents as all-out Bayesians who invariably follow the guidance of expectedvalue calculations. None of these three options is particularly attractive. Abandoning 3 is hardly an appealing prospect, since rationality clearly requires us to gear action to belief. Nor is it all that easy to abandon 5. For it is not an appealing option to insist that rational believers cannot concede the prospect or possibility that their beliefs might be false, so that they are constrained to see all their beliefs as absolutely and definitively certain. Finally, we do not want to jettison 8, since we are well-advisedly reluctant to forego the expected-value approach to rational decision-making. Given our attachment to all these incompatible commitments, how are we to exit from our difficulty? As is often the case with such theoretical difficulties, the best way out lies through the door of a distinction. Some of our beliefs we are prepared to chisel in stone. Others we readily admit to be written in sand. Some things we accept and assert in a dogmatic frame of mind, others only guardedly. Our beliefs are not all of a piece. There are two possibilities as regards their epistemic status. There are these things we 'believe to be absolutely certain' (C-beliefs). We view these as totally secure and utterly safe. We would bet literally everything on them. Note now that the inference from 1-3 to 4 holds good with respect to these beliefs— but only these! Still, such wholly unconditional beliefs are relatively rare. The rational man uses due epistemic caution. Most of what we believe we 'believe to be plausible' (P-beliefs). We view these as adequately secure and relatively safe—but not as altogether certain. For these P-beliefs we would certainly not put 'everything' at risk: our attachment to them, though real, is not that firm. With them, the inference from 1-3 to 4 does not go through. On the basis of this distinction, then, we are able to avert the paradox of Display 5.1. The reading of 'belief in 1 that authorizes the move to 4 via 2 and 3 presumes that we are dealing with Cbeliefs. But, the reading of 'belief in 1 that allows us to invoke 5 is predicated on its being P-beliefs that are at issue. Accordingly, the distinction of these two types of belief enables us to escape from difficulty.
Cognitive Rationality and Consistency
79
What is at issue with the C/P distinction is not of course a difference in content of beliefs. Rather, the difference is one of their status. Some contentions (namely, those represented by Cbeliefs) I accept absolutely. Other contentions I accept all right (and, of course, accept as true—because that is what 'acceptance' is), but in a less committed, more tentative frame of mind. When I set down my contentions I write the former in red and the latter in black. But, regardless of what colour ink I use when I make the statement, I use it to say the same thing. It makes exactly the same claim either way; the difference is only in respect to my confidence in this claim.4 Now the important fact for present purposes is that, with respect to qualified beliefs (P-beliefs), inconsistencies can quite plausibly arise. The distinction between belief as certain (C-belief) and provisional belief (P-belief) is thus crucial for the epistemology of inconsistency-tolerance. For once our approval and belief is recognized as being in some degree tentative and fallibilistic, it makes perfectly good sense to contemplate the acceptance of incompatibilities—to take inconsistencies in our stride.
5 . 3 MOTIVATING THE ACCEPTANCE OF INCONSISTENT THESES
In general, a variety of sources of information (and misinformation) will be available to us. These may be human (different witnesses), documentary or sensory (sight and touch, for example), or disciplinary (different branches of inquiry). Each source is presumably internally self-consistent in its claims, but the data of one may in some instances contradict those of another. When this happens, what are we to do by way of an epistemic exploitation of their data? We can, of course, choose either to suspend judgement altogether (accept nothing), or to favour one source over its competitors, by accepting its claims and rejecting those of the 4 The well-known 'preface paradox' should be considered on this basis. The author is not really contradicting himself in acknowledging in the preface the falsity of some of the statements of his main text. He is just drawing to our notice that the text is printed in black ink (so to speak). The author believes those statements of the text all right, and invites us to believe them as well. But, only in the qualified and somewhat tentative way that characterizes the rational man's approach to the great bulk of factual issues.
80
Cognitive Rationality and Consistency
others (which is capricious, save in the presence of a sufficiently strong concrete justification). Or, rejecting these more drastic options, we may instead prefer to grant provisional acceptance (at any rate) to the data of all the sources, hoping to 'straighten things out' as we go along. This latter course is in various ways quite attractive; but it requires us to take inconsistencies in our stride. Someone is bound to object to this by insisting that we aim at truth in inquiry. And so, since we know a priori that inconsistent sets of statements cannot be true (in toto), why ever accept them? The answer lies in the fact that the idea of 'the truth, the whole truth, and nothing but the truth' is an idealization. We cannot— should not—expect actually to realize it in the realistic setting of our actual epistemic proceedings. The potential presence of inconsistency is no more than yet another token of the imperfection which we already know, on general principles, to characterize our epistemic situation in 'the real world'. Yet, what sort of considerations can reconcile the rational inquirer to accepting inconsistency? The reply must run roughly as follows. The adequacy with which we pursue the cognitive enterprise hinges on a scoring system geared to: 1. Questions answered satisfactorily 2. Questions unanswered because of informational underdetermination (ignorance) 3. Questions answered unsatisfactorily (1) because of incorrect informative determination (error) (2) because of informational over-determination (inconsistency) The epistemic policy that best meets our needs is clearly one that gives the best overall results by maximizing type 1 instances in relation to those of types 2 and 3. To proceed satisfactorily here we may just have to take a certain amount of cognitive dissonance in our stride. And there is no decisive reason why we should not accept a certain sort of inconsistency in the interests of broadening substantially the range of questions that we can resolve satisfactorily. As long as contradictions can be confined to particular local areas, there is no reason why they should be regarded as something totally distinct from other sorts of negativity. The crucial fact is that people have questions and require answers to them, and that these answers are only available through
Cognitive Rationality and Consistency
81
implementing certain general methods or procedures for warranting the acceptance of theses en masse. In the real world, our confirmation methods are always bound to be imperfect. No workable acceptance criterion that is available to and implementable by us can avoid errors and mistakes altogether. Any such screening procedure for acceptability is going to commit 'errors'—it is not only going to let falsehoods slip through along with truths, but is also going to allow contradictions to slip through on occasion. There is always some 'noise' in the system—some degree of entropy and imperfection: it is inevitable that any workable epistemic policy will allow some errors to slip in amongst the truths. And this allows for the presence of chaff amongst the wheat —a certain amount of misinformation along with the information. We are thus put into the position of having to pay a price for the relief of ignorance—by coming to terms with the circumstance that the avoidance of 'errors of omission' unavoidably carries the prospect of 'errors of commission' in its wake. In cognitive as in physical engineering there are no altogether fool-proof processes or procedures. We face the harsh fact of life that any cognitive process or policy we can actually get hold of is subject to the general condition that error increases with informativeness—that an enlargement of range can be achieved only at the cost of an increased frequency of malfunction. This cognitive recalcitrance of the real world is something that cannot be altered in this regard. And it has the consequence that the price in terms of information loss of averting inconsistency altogether is too large for comfort. The trade off between sporadic (localizable) inconsistencies and large-scale information forfeits is so balanced that we are 'rationally well advised' to take (occasional and localizable) inconsistencies in our stride.5 The salient point is this: if we are going to use the standard machinery of decision-theoretic choice resolution then we must recognize that there are cases where accepting an inconsistent family of theses yields greater informative utility overall than abiding by the available consistent alternatives. Thus, if we let informative utility be our guide, we should not automatically 5
This of course presupposes a body of logical machinery that does not allow one to deduce anything and everything from an inconsistency, so that even a single inconsistency renders a body of assertions cognitively incoherent. But such inconsistency-tolerant (or 'paraconsistent') logics are abundant nowadays.
82
Cognitive Rationality and Consistency
dismiss inconsistent data sets as useless. We should face bodies of information sets in a thoroughly pragmatic spirit, with a view first and foremost to the overall range of the issues they enable us to resolve satisfactorily. Inconsistency as such is not the sole consideration, but simply one element in a complex story. Its avoidance is not the be-all and end-all, but one consideration to be weighed and balanced in combination with others. Yet, seeing that the acceptance of (occasional) inconsistencies guarantees the acceptance of (some) falsehoods, is this not automatically self-defeating for the cognitive project? Not necessarily.The aim of the cognitive project is to secure the best achievable overall balance between information and misinformation. If error avoidance were all, we could simply accept nothing and become totally sceptical. If truth acquisition were all, we could accept everything and become totally gullible. But the best epistemic policy is clearly one that optimizes the overall balance of information, minimizing the sum total of errors of the two kinds. And an epistemic policy that does best on this standard will have to tolerate errors and inconsistencies, being such that an inconsistent family of contentions will occasionally (though no doubt rarely) manage to slip through the net. Of course, when we accept recognized inconsistencies we know that we cannot be entirely right in what we accept. But what is so terrible about that? It is, after all, something which, if we are at all realistic—or even moderately informed about the history of science—we must acknowledge anyway. We are well advised to look back to fundamentals at this stage. The crux of rationality—even cognitive rationality—is costeffectiveness; accomplishing a worthwhile job by optimally efficient means. And consistency is therefore no indispensable requisite. It is perfectly possible that the toleration of occasional inconsistency should more than pay for itself in terms of augmented knowledge overall. To say this is clearly not to say that inconsistencyacceptance is something positive and to be welcomed, but rather that it is a negativity of manageable scope—one that can be offset in the wider scheme of things by other positivities. In the final analysis, inconsistency is no worse than error; its presence represents no more than just another mistake somewhere along the line, unidentifiable in the epistemic fog. It is not an utter disaster, but merely a sign that our cognitive procedures are
Cognitive Rationality and Consistency
83
imperfect. Realizing that no inquiry procedure is 100 per cent foolproof, we may tolerate inconsistencies in the same spirit in which other negativities are taken in our stride in order to arrive at a greater good. (We realize full well that one cannot operate a transportation system without accidents.) To say this is not, of course, to say that we should welcome inconsistency. The point is that we can—and should—regard it as the price we pay for achieving a greater (informative) good.
5.4
L I N E A R L Y I N F E R E N T I A L V S . D I A L E C T I C A L L Y CYCLIC REASONING
Much if not most of our thinking is bound to be carried on under conditions in which we cannot justifiably regard the data that provide the premisses of our reasoning as absolutely certain (C-beliefs), but merely as provisionally acceptable plausibilities (P-beliefs). This situation has far-reaching implications for the appropriate character and structure of our reasoning, implications which generally go unheeded and unrecognized. Whenever we reason in a deductively valid way from assured premisses—from C-beliefs—we automatically know (1) that our conclusions are themselves certain and thus one and all unproblematically acceptable, and, moreover, we also know (2) that they are mutually consistent not only with the totality of our initial premisses but also with each other. Accordingly, our reasoning can be linear and progressive. We can march straight ahead, never having to look back over our shoulders at earlier findings to assure ourselves that the new findings have not rendered the old unacceptable or that prior results may fail to be reconcilable with the new ones. We can constantly forge ahead into new territory, confident that there is no point in stopping to re-examine old results in the light of new ones. By contrast, 'dialectical' reasoning is a matter of the repeated reconsideration of old issues from newly attained points of view. The root idea of such reasoning is that of a multi-stage process whereby we repeatedly re-examine one selfsame issue from different, and mutually inconsistent, points of view. It is a matter of developing a course of reasoning in several phases or 'moments'. We proceed in circles or cycles, returning repeatedly to a certain
84
Cognitive Rationality and Consistency
issue, viewing it now in this light and now in that. We do not constantly press onwards into new ground—continually 'changing the subject', as it were—but criss-cross the same terrain, approaching the old issues from different and often discordant angles. We have a plurality of witnesses, for example, who assert conflicting claims with respect to an historical episode. In such cases, we must re-examine the entire situation repeatedly, reconsidering it in the light of different 'versions' to distil out a plausible overall account that may—even in the end—leave various issues unresolved and unreconciled. Reviewers, critics, and commentators often reproach an author with inconsistency for reasoning on page ten from some contention P and on page eighty from some other contention incompatible with P. But such a condemnation can, in some cases, be superficial and unthinking. The iterative reconsideration of the same issues from mutually inconsistent cognitive perspectives is the feature which, above all, sets 'dialectical' thinking apart from standard, linearly inferential reasoning. Such a cyclical procedure yields a sequential deepening of the case for the conclusions one is endeavouring to substantiate. One 'tightens the net', so to speak, through continuous construction of an increasingly adequate case, consolidating the issue now in this aspect and now in that, returning to the same issue from different angles, making use of varying, and sometimes even mutually inconsistent, premisses for its substantiation. This sort of thing is dispensable and pointless in the case of strictly deductive reasoning. For here anything that follows from contradictory premisses can also be established in other ways. Whenever one can deductively establish that q given p and also establish that q given not-/?, then one can establish q unconditionally via the constructive dilemma: 1. p^q 2- -p -» q 3. p v -p
While this course of reasoning exemplifies the quintessentially 'dialectical' move to a conclusion (q) from contradictory startingpoints (p and not-p, respectively), still, in the specifically
\ Cognitive Rationality and Consistency
85
deductive case, this process is unnecessary and altogether avoidable. Whenever we can demonstrate q via this route, we can always in principle establish it without any reference top at all. We accordingly have no need for 'dialectical reasoning' in those rigoristic contexts, where—as (ideally) in mathematics—deductively cogent reasoning from categorically accepted premisses is the order of the day. However, with probable and plausible reasoning of the sort constantly encountered in non-deductive arguments, the situation is very different. Let us first consider the situation of plausible reasoning. Here, we begin with a collection of premisses or data that afford promising prospects—not validated certainties but tempting prospects that have some substantial claim on our acceptance, but are by no means certain. In such circumstances, we not only have no pre-assured guarantee that the data from which we reason are actually true, we even lack the assurance that they are mutually compatible. (Certified truths, of course, must be compatible, but merely plausible ones need not be.) The theses we are dealing with are not assured truths but only contentions that we are provisionally prepared to let into the range of what we accept. And this more relaxed and lenient stance towards cognitive 'acceptability' inevitably carries a very different attitude towards consistency in its wake. If one insists upon an unshakeably solid and secure foundation of absolute certainty for an ampliative process of inference, then of course consistency is everything. But, if one is more realistic about it then inconsistencies will not occasion all that much chagrin. For, our concern is now not with final certainty but with provisional credibility. Suppose, for example, that three (otherwise reliable) witnesses provide us with the following inconsistent body of information: 1. There is one and only one X, which is located in a fixed position somewhere in a given three-by-three (tick-tack-toelike) matrix. 2. Yes, and it is in the upper-left-hand square. 3. Yes, and it is in the lower-right-hand square. If one does not simply throw up one's hands in the face of this contradiction, then one can exploit the given data to draw such conclusions as
86
Cognitive Rationality and Consistency
(1) 1 is unproblematically acceptable as true (since everyone agrees on it) (2) the X is located in a corner square (since no matter where we turn, this conclusion emerges). From such a perspective, even an inconsistent set of premisses manages to divide the overall 'spectrum of conceivability' into a tractable partitioning of possible and impossible cases. We emphatically do not confront the chaos of 'anything goes'. But note here that we do not simply establish conclusion (2) deductively by proceeding on a one-shot basis to reason from any or all of the data. We establish it dialectically by showing that, no matter how we proceed from the different consistency-restoring curtailments of our inconsistent data, this conclusion is still something that we can secure. Those different arguments for our conclusion are only seemingly redundant; in the overall probative situation, they are all needed. As this illustration shows, there is a vast difference between the case of reasoning from premisses pre-established as certain (as certainly true), and that of reasoning from premisses whose acceptability is based on a footing of mere plausibility (as plausibly or presumably true). In this sort of circumstance, it can make eminent sense to proceed cyclically, returning to the same contention from different angles of consideration. Let us now turn from plausibilities to probabilities. It deserves note that a perfectly valid course of reasoning is at issue in a 'probability dilemma' of the following form: 1. q is (conditionally) probable given p 2. q is (conditionally) probable given not-p .". q is probable (simpliciter) For, premiss 1 is tantamount to:
Pr (qlp) =-E~^y-
> 1/2 o r 2M<7 & P) > PKP).
And premiss (2) is tantamount to: pr
(q/-p) = ^ ^ p f
> l/2 or 2pr(q & - p ) > pr(-p).
Combining these we have: 2[pr(q & p) + pr(q & -p)] > pr(p) + pr(-p).
Cognitive Rationality and Consistency
87
Whence we obtain: pr(q) > V2. And observe that it would have made no difference if we had set the operative probability threshold for qualifying as 'probably true' at 3/4 or at 99/100 instead of V2. The upshot that lq is probably true' would have ensued all the same. Now here too we have a 'dialectical' mode of probative argumentation, where we establish a thesis q by showing it to be forthcoming from distinct, and, indeed, incompatible, perspectives (p and not-/? respectively). Note, however, that in this probabilistic situation, contrary to the deductive case, it makes a crucial difference which particular 'inconsistency' we exploit—the particular, concrete thesis p that we make use of is something that does real work for us. (The mere fact that q is probable per se does not mean that it is probable relative to any and every sort of condition p whatsoever.) The probabilistically mediating thesis p here plays a probative role akin to that of the middle term in Aristotelian syllogistic logic. The task of finding a suitable link is a matter of cleverness and ingenuity which, if unresolved, prevents this whole line of reasoning from getting off to a start. There is an important lesson here. In situations where the cognitive ground beneath our feet is not totally firm—where our deliberations proceed on the basis of probabilistic and plausibilistic approaches—it may make perfectly good sense to operate dialectically and consider an issue prismatically in the variable light of inconsistent perspectives. It is possible to envision situations in which as good a case can be made for inconsistent claims as for almost anything else that one accepts. We are well advised to steel ourselves to tolerate occasional inconsistencies. There is nothing irrational about this. 5 . 5 CONTEXT AND DIALECTICS
The so-called doctrine of Averroism, the theory of 'double truth' —of incompatible 'truths' in theology and secular learning—was in all likelihood never actually held by any historical figure.6 In 6 See 'La Doctrine de la double \6rit6' in fitienne Gilson, Etudes de philosophic midie'vale (Strasbourg, 1921), 51-75; and see also Stuart MacClintock, Perversity and Error: Studies on the 'Averroist' John of Jandun (Bloomington, Ind., 1956).
88
Cognitive Rationality and Consistency
admitting both that 'The world is uncreated, according to the teachings of philosophy' and 'The world is created, according to the teachings of religion', the medieval 'Averroists' did not proceed to the outright absurdity of holding both of these teachings to be correct, but always gave one priority over the other! (The aversion they created among their contemporaries lay in their toughminded insistence that it was necessary to make a choice and indeed to make it in favour of secular knowledge via the principle that fides non habet meritum ubi ratio humana praebet experimentum.) In historical accuracy, an outright double-truth theory seems to be one of those artificial doctrines that, like solipsism and the radical version of scepticism maintaining that any one belief or opinion is every bit as good and as tenable as any other, has never actually been held by any flesh-and-blood exponent, but was simply invented by some theorists as a rod with which to beat others whom they opposed. But, be this as it may, if the present perspective is acceptable, a neo-Averroist theory of actually 'discordant truth' is indeed viable in principle, given two locally consistent but globally incompatible information contexts. Consider an example to illustrate this idea. The realm of our factual knowledge includes both the substantive inventory of what we know about the world, and also 'metascience'—the environing domain of what we know about our knowledge. Now, if the history of science teaches us any one thing it is that not all of our 'scientific knowledge' is correct—that our scientific successors of the year 3,000 will deem our science to be every bit as laden with errors as we ourselves deem that of our predecessors of 100 years ago. Any realistic appreciation of the status of our present-day science must acknowledge its inclusion of falsehoods—though we cannot, of course, identify just where they lie. The overall body of 'our knowledge' is thus collectively inconsistent. It includes both those various assertions in matters of scientific detail and the metaassertion that some of our scientific claims are false. But, of course, context separation prevents difficulties here. We never mix science with metascience in one seamless discussion, and never resort to metascientific deliberations in transacting our scientific business. As this example indicates, an epistemology of contextual variation can, clearly, cushion the impact of inconsistency.7 7
Regarding the implementation of this line of thought see Nicholas Rescher and Robert Brandom, The Logic of Inconsistency (Oxford, 1979).
Cognitive Rationality and Consistency
89
When inconsistency arises, the 'context' of the deliberation can thus come to our rescue by providing a saving limitation. For, we may well be able to achieve intracontextual consistency in various situations, even though the claims of one context may fail to be consistent with those of another. Global consistency all across the board may well elude us. The 'resolution criteria' we use to extract a consistent conclusion from an inconsistent set of initial data may well operate in a context-sensitive way, yielding results that are acceptable here but unacceptable there. Accordingly, considerations of context can and generally do mitigate the impact of inconsistency-tolerance. Reasoning can proceed not just inferentially (linearly) but also dialectically (cyclically). What is at issue here is not a difference in the rigour but rather a difference in the style of the reasoning. In the abstract disciplines—mathematics, for example—where the formal demands of logic are paramount, inconsistency is a disaster, a fatal flaw. And in the natural sciences, too, cognitive systematization almost always conforms to the linear paradigm through the mathematico-deductive mode of disciplinary articulation. We begin with a relatively modest starter-set of well-established 'axioms' and then proceed by the deductively inferential means of mathematical derivation. In the human sciences, however, the situation is generally very different. Paradigmatically (in historiography, for example) one proceeds not inferentially but dialectically. One gathers in all the promising 'data'—inconsistent though they be—and shapes them as best one can into a coherent structure. Outside the domain of mathematical reasoning, a 'prismatic' approach of looking at an issue from different angles—proceeding dialectically to consider it now in the light of P and again in the light of not-P—can be a perfectly sensible and altogether useful way of proceeding. Human thought is inherently complex and many-sided, a matter of inner tensions, of competing pushes and pulls. In the biographical interpretation of individual actions, in the description and explanation of historical transactions, in the interpretation of literary or artistic creations—throughout the humanities, in short—this prismatic complexity comes to the fore. The processes of inferential and dialectical reasoning, however different in other ways, are not different in that one is a matter of rigorous thinking and the other of irresponsible sloppiness—that
90
Cognitive Rationality and Consistency
one is 'genuinely scientific' and the other 'mere basket-weaving'. The widespread tendency to think of the 'human sciences' (the historical and humanistic disciplines) as non-scientific is based on an overly narrow and narrow-minded conception of rigorous reasoning. (It is not that there are no differences between the formal and the natural sciences on the one hand and the human sciences on the others, but merely that these differences are not such as to render the former acceptably 'scientific' and the latter not.)
5.6
C O N S I S T E N C Y AS A C O G N I T I V E D E S I D E R A T U M
Aristotle's influence has much to answer for—his own explicit disavowal of the quest for unwarranted rigour notwithstanding. By his insistence that deductively articulated science alone is genuinely scientific, and that dialectics is a matter of 'mere rhetoric', he set back for some 2,000 years the development of rational dialectics (and the dialectical disciplines themselves). Even now, some two millennia later, we have not freed ourselves from his (typically Greek) prejudice that only ampliatively deductive reasoning of the sort typified by mathematics is truly cogent, and that reductively dialectical reasoning is something inferior and second rate. We are still caught up in the backwash of the ancient Greek conviction that only those disciplines whose rational systematization proceeds by way of mathematico-linear development are rigorous and solid, and that those disciplines that proceed by way of dialectically cyclic argumentation are somehow inferior and unsatisfactory. Surely a more liberal and open-minded attitude is in order. In those cases where the linear approach can be implemented effectively, well and good. But, there is nothing to apologize for in those cases where a more complex and 'messy' dialectical approach to cognitive systematization is needed to accomplish the work in hand. Contrary to widespread opinion, one need not take the stance that consistency is an indispensably inevitable precondition for cogent thought and rationally acceptable reflection. Consistency may reasonably be regarded as a feature that is essential only in limited areas of formalizable deliberation (mathematics, pure and applied), but need not characterize rational cognition in general.
Cognitive Rationality and Consistency
91
The maintenance of consistency may be regarded as essentially akin to the achievement of completeness, uniformity, or simplicity —as an important cognitive value that should be cultivated and prized. Consistency must obviously and unquestionably count as an epistemic desideratum—a regulative idea or re'Aos for dealing with the murky complexity of the difficult situations in which we generally have to work. It is something towards which we should strive in the end, not something we should require from the outset. It need certainly not be seen as an absolute requisite for rationality —something in whose absence the whole enterprise comes to utter grief. The prospect of inconsistency can be entertained without propelling us across the boundaries of intelligibility. The root idea of rationality is that of cost-benefit co-ordination —of seeking for the best means to valid ends. The cognitive enterprise falls within this framework just as much as any other. Consistency is no be-all and end-all, and its maintenance is not indispensably necessary. Since the principal aim is knowledge as such, it is only realistic to reconcile ourselves to the fact that we may be called upon to tolerate occasional singularities of inconsistency in the interests of achieving the best overall balance of information over ignorance and misinformation. Cognitive rationality as such does not involve an absolute, unswerving commitment to consistency at any cost. Consistency is a prime desideratum of rationality, but not an absolutely indispensable requisite for it. It should be viewed not as a present demand, but as an ultimate ideal. A preparedness to tolerate conflict and dissonance—nay, even inconsistency—is often in the interests of doing as well as we can overall at this stage of the game. There is little to be said for an impatient insistence on securing here and now and all across the board that finally configured order and orderliness which is available only in the domain of the dead, the cut-and-dry region of museums, encyclopaedias, and graveyards.
6 The Rationality of Ends
(1) The rationality of ends—evaluative rationality—is an indispensable component of rationality at large. (2) If we adopt inappropriate ends and goals then we are not being rational, no matter how efficiently and effectively we pursue them. SYNOPSIS
6.1
A C R I T I Q U E OF THE H U M E A N C O N C E P T I O N OF R E A S O N : R A T I O N A L ACTION R E Q U I R E S A P P R O P R I A T E E N D S
In an oft-cited passage in Book III of the Nicomachean Aristotle wrote:
Ethics,
We deliberate not about ends but about means. For a doctor does not deliberate whether he shall heal, nor an orator whether he shall persuade, nor a statesman whether he shall produce law and order, nor does any one else deliberate about his end. They assume the end and consider how and by what means it is to be attained; and if it seems to be produced by several means they consider by which it is most easily and best produced, while if it is achieved by one only they consider how it will be achieved by this and by what means this will be achieved, till they come to the first cause, which in the order of discovery is last. (1112bl2—20) The sort of thinking that Aristotle has in view here—deliberation about efficient means to pre-established ends—is unquestionably important in human affairs. 'I need a bed; to make a bed I need a hammer and saw; I can borrow a hammer; so I shall go and buy a saw.' Aristotle's own examples of practical reasoning are exactly of this common and familiar sort—and are plausible enough in their way. 1 But not all deliberative reasoning is means-end reasoning. Admittedly, the doctor does not deliberate about 1
Metaphysics, 1032617-22; De Motu Animalium, 701al8-20. A helpful guide to Aristotle's theory of practical reasoning is Norman O. Dahl, Practical Reason, Aristotle, and Weakness of the Will (Minneapolis, Minn., 1984).
The Rationality of Ends
93
treating illness—that choice is already settled, included as part of his decision to become a doctor. But, a young person may well deliberate about whether to become a doctor in the first place— reflecting on whether this would be something good for him, given his abilities, skills, interests, and so on. And this sort of deliberation is not a question of means to pre-established ends at all. The long and short of it is that there are two very different sorts of deliberations: cognitive deliberations regarding matters of information and evaluative deliberations regarding matters of value. Whether certain means are appropriate to given ends is a question whose resolution must be addressed in the former, informational order of deliberation. But, whether the ends we have are appropriate as such, whether they merit adoption, is an issue which can and must be addressed in the latter, evaluative order of deliberation. A rational agent certainly cannot say 'I adopt G as a goal of mine, but am indifferent regarding the efficiency and effectiveness of means towards this goal.' But no more can a rational person say: 'I adopt G as a goal of mine, but am indifferent regarding its validity; I just don't care about the larger issue of its appropriateness as such.' Both matters—the efficacy of means and the validity of goals—are essential aspects of practical rationality.2 David Hume drew a sharp contrast between a narrowly construed 'reason' that is concerned only with means and a reasondetached faculty of motivation that concerns itself with ends— namely the passions. And he considered these motivating passions as autonomous forces operating outside the rule of reason proper. He regarded an impetus toward or away from some object—a desire or aversion—as simply the wrong sort of thing to be rational or sensible; as lying outside the rational domain altogether. As Hume saw it, apart from the formal issues of logic and mathematics reason merely deals in descriptive information about the world's 2 On the rationality of ends see Stephen Nathanson, The Ideal of Rationality (Atlantic Highlands, NJ, 1985). Social-science aspects of the issue are treated in S. I. Benn and G. W. Mortimore, Rationality and the Social Sciences (London, Henley, and Boston Mass., 1976), esp. pt. ii, 'Rationality in Action'. As these authors point out, social scientists are caught in a dilemma between the impetus of Max Weber's influential contention that the social science must be value free and the idea that the social scientist should be able to proceed prescriptively and render policy advice. It is clearly one thing to inform a client about how to get what he wants and another to counsel him about where his real interests lie.
94
The Rationality of Ends
states of affairs and relationships of cause and effect. Accordingly, reason is strictly instrumental: it can inform me about what I must do if I wish to arrive at a certain destination, but only 'passion'— desire or aversion—can make something into a destination for me. When one asks what is to be done, reason as such has no instructions—it is wholly a matter of what one happens to want. Reason is thus a 'slave of the passions'. Its modus operandi is strictly conditional: it dictates hypothetically that if you accept this, then you cannot (in all consistency) fail to accept that. But, all this is a matter of the hypothetical if-then. The categorical 'accept this!' is never a mandate of reason, but of that extrarational faculty of 'the passions', which dictates the bestowal of one's unconditional allegiances. Reason herself is inherently conditionalized: she says not what one must (or must not) opt for, but only what one is consequentially committed to if one already stands committed to something else. And so, Hume insisted: It is not contrary to reason to prefer the destruction of the whole world to the scratching of myfinger.It is not contrary to reason for me to choose my total ruin. . . . It is as little contrary to reason to refer even my own acknowledged lesser good to my greater, and to have a more ardent affection for the former than the latter.3 But this is clearly strange stuff. Reason cannot simply beg off from considering the validity of ends. Our motivating 'passions' can surely themselves be rational or otherwise: those that impel us towards things that are bad for us or away from things that are good for us go against reason, those that impel us away from things that are bad for us and towards things that are good for us are altogether rational. Reason can and should deliberate not only about what it is ill advised to believe (because it is probably at odds with the truth), but also about what it is ill advised to esteem (because it is probably at odds with our interests). Like various beliefs, various evaluations are palpably crazy.4 Reason, after all, is not just a matter of the comparability or consistency of pre-given commitments, but of the warrant that there is for undertaking 3
David Hume, A Treatise on Human Nature, bk. II. pt. iii. sect. 3. For strict consistency, a rigorous Humean should, by analogy, hold that cognitive reason too is only hypothetical—that it only tells us that certain beliefs must be abandoned if we hold certain others, and that no beliefs are contrary to reason as such, so that 'it is not contrary to reason to think one's finger larger than the entire earth'. 4
The Rationality of Ends
95
certain commitments in the first place. An evaluative rationality which informs us that certain preferences are absurd—preferences which wantonly violate our nature, impair our being, or diminish our opportunities—fortunately lies within the human repertoire. To be sure, philosophers even now often follow Hume in saying things like: Reason is wholly instrumental. It cannot tell us where to go; at best it can tell us how to get there. It is a gun for hire that can be employed in the service of any goals we have, good or bad.5 On such a view, reason has no concern with goals as such—all it can do is to inform us about the efficiency of means to ends. It can neither guide us in setting ends nor advise us about priorities, about how conflicts among divergent ends are to be settled. Ends, priorities, and values all lie outside the range of reason. They are no more than our value allegiances, the product of a rationally blind attachment to some fundamentally extrarational commitment. (In this context, oddly enough, Hume and Nietzsche are birds of the same feather.) But this is plainly too narrow a view of what reason is all about. To re-emphasize: There is not only an informative rationality that relates to means, but also an evaluative rationality that relates to ends. We can reason not only about matters of efficiency of goal attainment but about the appropriateness of our goals as well. One can reason not only about matters of fact, but also about matters of value. It is surely a dictum of reason not only to accept that which (in the light of the available evidence) is acceptance-worthy, but also to prefer that which (in the light of the available indications) is preference-worthy. It is contrary to reason (albeit to evaluative rather than cognitive reason) to prefer the lesser good to the greater or the greater evil to the lesser or to subordinate real needs to feckless wants. Only by doing violence to the nature of reason—only by ignoring or dismissing the evaluative side of reason with its concern for what is worthy of preference—can Hume maintain the sort of position he does. Even Hume himself stands committed (both in the History of England and An Inquiry Concerning the Principles of Morals) to the idea that passion as such is not really quite the end of the matter, because some passions (the 'good' ones associated with the 5
Herbert A. Simon, Reason in Human Affairs (Stanford, Calif. 1983), 7-8.
96
The Rationality of Ends
objects of the 'Glorious Revolution' of 1688) are worthy, and others (especially those of 'enthusiasm') are not. But what instrument have we for this crucial work of appraisal, save reason? Hume thought that 'rational sympathy' would do—but it too is in the end deeply problematic, since instinct or natural inclination is a doubtful justificatory resource, given its apparent tendency to pull different people in different directions. Unevaluated 'natural sympathy' is just as problematic as unevaluated passions. Rational evaluation, however difficult, is in the end the only way to go. Hume's profound error lay in his taking a part of reason to be the whole of it. For, reason at large must care for ends as well as means. If our ends (our goals and values) are themselves inappropriate—if they run counter to our real and legitimate interests—then no matter how sagaciously we cultivate them, we are not being fully rational. (A voyage to a foolish destination—no matter how efficiently conducted—is a foolish enterprise.) Hume mistakenly effected a total divorce between reason and choice: 'I have prov'd that reason is perfectly inert, and can never either prevent or produce any action or affection,'6 But, while reason indeed cannot of itself 'prevent or produce' action, the fact remains that it can motivate action by providing good reasons for it. When rational inquiry indicates to me that doing A is beneficial, then—in so far as I am rational—it impels me towards this action. Alternatively, if it indicates that the action is detrimental, it impels me away from it. Reason's task in relation to action is to provide grounds for or against. And this means that any disconnection of reason from action is quite mistaken. To see reason as irrelevant to action is to misrepresent it to the point of caricature. An interesting and somewhat desperate move to transcend the gulf between wants and interests—between 'what one wants' and 'what is good for one'—is represented by Henry Sidgwick's influential proposal to equate the latter with what one would want if—if one were fully informed, undisturbed by passion, painstaking in visualizing consequences, etc.7 But such a stance is predicated on the highly questionable idea that lack of information is the only impediment to appropriate evaluation. Clearly, it is not lack of information alone that prevents the monomaniac or the 6 7
A Treatise on Human Nature, ed. L. A. Selby Bigge (Oxford, 1964), 458. Henry Sidgwick, A Method of Ethics, 7th edn. (London, 1928), 111-12.
The Rationality of Ends
97
masochist from evaluating matters aright. Failures to assess means to ends are one thing, failures to think sensibly about values and priorities another. The crucial fact is that there is not only inferential ('logical') reason but also evaluative ('axiological') reason. Just as the rational man only believes what is belief-worthy for someone in his circumstances, so he only values what is value-worthy—deserving of being valued. And the determination of value-worthiness requires the sensible application of appropriate standards—in short, reasoning. For, it is quintessentially the work of reasoning to determine what sorts of commitments are rational (reasonconforming) and what sorts are not. (And this is so whether the 'commitments' at issue be beliefs or evaluations.)8 Concern for the rationality of ends is important precisely because cognitive rationality is not all, because information is not the only thing that counts in life. Knowledge of matters of descriptively non-evaluative fact is only one good among others. Reason has other matters to attend to as well.9 The instrumental rationality at issue in finding the effective means to chosen ends is only a part of rationality. For, means may well be directed towards inappropriate ends. An embezzler, say, or a self-destructively neurotic person can be quite efficient in figuring out how to attain objectives. But this partial sort of rationality does not render such activities rational tout court. The Humean dogma that the nature of our ends is immaterial to rationality must accordingly be rejected. Being intelligent about some things does not make one unqualifiedly intelligent. Evaluative rationality is an indispensable component of rationality overall. 6 . 2 THE C R U C I A L ROLE OF I N T E R E S T S : WANTS AND P R E F E R E N C E S ARE NOT E N O U G H FOR RATIONALITY
As this perspective indicates, rationality involves two sorts of issue—means and ends. The rationality of means is a matter of factual information alone—of what sorts of moves and measures lead efficiently to objectives. But the rationality of ends is a matter 8
On the matter of rational vs. irrational ends see Kurt Baier, The Moral Point of View (Ithaca, 1958), and Bernard Gent, The Moral Rules (New York, NY, 1973). 9 On this theme see the final chapter of the author's The Limits of Science (Berkeley and Los Angeles, Calif., 1984).
98
The Rationality of Ends
not of information but of legitimation. It is not settled just by factual inquiry, but involves evaluative judgement. And both aspects are crucial: ends without requisite means are frustrating, means without suitable ends are unproductive and pointless. Accordingly, rationality has two sides: an axiological (evaluative) concern for the appropriateness of ends and an instrumental (cognitive) concern for effectiveness and efficiency in their cultivation. The conception of rationality fuses these two elements into one integral and well-co-ordinated whole. Contentions like 'Smith is selfish, inconsiderate, and boorish' do not lie outside the sphere of rational inquiry—nor for that matter do contentions like 'Behaviour that is selfish/inconsiderate/boorish is against the best interests of people'. The issue of appropriate action in the circumstances in which we find ourselves is pivotal for rationality. Be it in matters of belief, action, or evaluation, we want—that is to say, often do and always should want—to do the best we can. For, one cannot be rational without due care for the desirability of what one desires—the issue of its alignment with our real, as distinguished from merely putative, interests. The sensible attunement of means to ends that is characteristic of rationality calls for an appropriate balancing of costs and benefits in our choice among alternative ways of resolving our cognitive, practical, and evaluative problems. Reason accordingly demands determination of the true value of things. Even as cognitive reason requires us to assess the evidence for theses at its true worth in determining what we are to accept, so evaluative reason requires us to appraise the values of options at their true worth in determining what we are to choose or prefer. And this calls for an appropriate cost-benefit analysis. Values must be managed as an overall 'economy' in a rational way to achieve overall harmonization and optimization. (Economic rationality is not the only sort of rationality there is, but it is an important aspect of overall rationality.) Someone who rejects economic considerations—who, in the absence of any envisioned compensating advantages, deliberately purchases benefits he deems to be worth a few pennies at the expense of millions—is simply not rational. It is just as irrational to let one's efforts in the pursuit of chosen objectives incur costs that outrun their true worth as it is to let one's beliefs run afoul of the evidence. It is a grave mistake to think that one cannot reason about
The Rationality of Ends
99
values—that values are simply a matter of taste and thus beyond the role of reason, because 'there's no reasoning about tastes'. The fact that valid values implement and pivot upon our needs and our appropriate interests means that a rational critique of values is not only possible but necessary. For, values that impede the realization of a person's best interests are clearly inappropriate. A priority scheme which sets mere wants above real needs or sets important objectives aside to avert trivial inconveniences is thereby deeply flawed. Even great values will have to yield to the yet greater. (Some things are rightly dearer to us than life itself.) Economists, decision theorists, and utilitarian philosophers to the contrary notwithstanding, the genuinely rational person is the one who proceeds in situations of choice by asking himself not the introspective question 'What do I prefer?' but the objective question 'What is to be deemed preferable? What ought I to prefer?'10 Rational comportment does not just call for desire satisfaction, it demands desire management as well. The question of appropriateness is crucial. And this is an issue about which people can be—and often are—irrational; not just careless but even perverse or crazy. There is nothing automatically appropriate—let alone sacred— about our own ends, objectives, and preferences. For rationality, the crucial question is that of the true value of the item at issue. What counts is not preference but preferability—not what people do want, but what they ought to want; not what people want, but what sensible or right-thinking people want. The normative aspect is ineliminable. There is an indissoluble connection between the true value of something (its being good or right or useful) and its being rational to choose or prefer this thing. And so, the crucial question for rationality is not that of what we prefer, but that of what is in our best interests—not simply what we may happen to desire, but what is good for us in the sense of contributing to the realization of our real interests. The pursuit of what we want is rational only in so far as we have sound reasons for deeming this to be want-deserving. The question whether what we prefer is preferable, in the sense of deserving this preference, is always relevant. Ends can and (in the context of rationality) must be evaluated. It is not just beliefs that can be '" A good exposition of the opposing position may be found in Frederick Schick, Having Reasons: An Essay in Rationality and Sociality (Princeton, NJ, 1984).
100
The Rationality of Ends
stupid, ill-advised, and inappropriate—that is to say, irrational— but ends as well. Evaluation is at the very heart and core of rationality. For, rationality is a matter of balancing costs and benefits—of best serving our overall interests. The question of worth is thus never far removed from the thoughts of a rational mind. The rationality of ends is an indispensable component of rationality at large. The rationality of our actions hinges critically both on the appropriateness of our ends and on the suitability of the means by which we pursue their cultivation. Both of these components—the cogently cognitive ('intelligent pursuit') and the normatively purposive ('appropriate ends')—are alike essential to full-fledged rationality. Rationality consists in the intelligent pursuit of appropriate ends. The standard line among economists is that rationality turns on the intelligent cultivation of one's preferences. But this is problematic in the extreme (as the preceding chapter argued). For, the two formulas are equivalent only if one happens to have sufficiently enlightened preferences. And this of course is not necessarily so. What I want or merely may think to be good for me is one thing; what I need and what actually is good for me is another. To move from preferences and perceived interests to genuine benefits and real interests I must be prepared to get involved in a rational critique of ends—to examine in the light of objective standards whether what I desire is desirable, whether my actual ends are rational ends, whether my putative interests are real interests. But just what is it that is in a person's real (best) interests? Partly, this is indeed a matter of meeting the needs that people universally have in common—health, normal functioning of body and mind, adequate resources, human companionship and affection, and so on.11 Partly, it is a matter of the particular role one plays: co-operative children are in the interests of a parent, customer loyalty in those of a shopkeeper. Partly, it is a matter of what one simply happens to want. (If John loves Mary, then engaging Mary's attention and affections are in John's interests— some things are in a person's interests simply because he takes 1
' The issue goes back to the idea of the 'basics' (principia) of the human good in the Middle Academy (Carneades)—things like the soundness and maintenance of the parts of the body, health, sound senses, freedom from pain, physical vigour, and physical attractiveness. Compare Cicero, Definibus, V. vii. 19.
The Rationality of Ends
101
an interest in them.) But these want-related interests are valid only by virtue of their relation to universal interests. Mary's approbation is in John's interest only because 'having the approbation of someone we love' is in anyone's interest. Any valid specific interest must fall within the scope of an appropriate, universal covering principle of interest legitimation. (The development of my stamp collection is in my interest only because it is part of a hobby that constitutes an avocation for me and 'securing adequate relaxation and diversion from the stress of one's daily cares' is something that is in anyone's interests.) A specific (concrete, particular) interest of a person is valid as such only if it can be subordinated to a universal interest. Some writers (J. P. Sartre, for example) see reason-providing considerations as locked into a potentially infinite regress that can only be broken by an ultimate appeal to unreasoned reasons. But this is just not how things go in the explanation and the validation of actions. Here, the regress of reasons (A because B because C) will and must terminate automatically and naturally with any normatively valid universal reason—an interest which it is only proper and appropriate for anyone to have when other things are anything like equal. I want this sandwich because I am hungry, and I want to stop feeling hungry (i.e., relieve those hunger pangs) because it is painful. But there is just no point in going further— and no need for it. When such a universal is reached, no further elaboration is called for. (It is this circumstance that endows the matter of the rational validation of ends with its importance.) And so, in assessing the rationality of actions we cannot look just to personal motives, but must invoke universally appropriate values as well. The fact that X wants A remains a mere motive for his action in pursuing A (in contradistinction to a reason) until such time as it is rationalized through the fact that X recognizes A to have the desirable feature F, which is not just something that X wants, but is something that any and every (reasonable) person would want. (Note, when X wants 'to marry Mary' this remains unrationalized until such time as it is 'covered' by the universal desideratum of 'marrying a person one loves deeply'.) Only such a legitimation sub ratione boni, as part of a universally cogent desideratum, can rationalize a valuation (or a choice or preference that flows from it). Ultimately, universal considerations can provide an adequate rationale. 'X wants A.' Why? 'He wants B and sees A
102
The Rationality of Ends
as leading to Z?.' But why does he want B? With a rational want we can extend this regress until we reach something that is generally (universally) desirable—-something the wanting of which we, the questioners, see to make sense, in that we value it, and think that ' everyone should do so, and think it pointless and needless to raise further questions. Only when X does that which we ourselves see as being 'normal and natural' for people in general do we stop asking for further special explanations. At that point, the factor of rationality accomplishes its characteristic work. Consider the contrast between: professed wants: what I say or declare that I want or prefer; felt wants: what I (actually) do want or prefer; real (or appropriate) wants: what the reasonable (impartial, well-informed, well-intentioned, understanding) bystander would think that I ought to want on the basis of what is 'in my best interests'. It is this last item that is decisive for rationality—namely what is in my 'real' or 'best' interests. Rationality is not just a matter of doing what we want (if this were so, it would be far simpler to attain!), it is a matter of doing what we (rationally) ought, given the situation in which we find ourselves. Relative value (utility, means-to-ends serviceability) is no doubt important. But, without due heed to the categorically normative status of those ends themselves, relative evaluation is a futile exercise from the rational point of view. To proceed rationally we must care not just for the efficacy of means but for the worth of ends. Man is not only homo sapiens but homo aestimans. The most fundamental judgement we make regarding even merely hypothetical developments is whether they are or are not 'a good thing'. Being rational involves endeavouring to do well (intelligently) what we must by nature do—and evaluation is, emphatically, a part of this. Action in pursuit of what we desire is not rendered rational by this fact. The crucial issue is one of evaluating that desire itself—of determining whether the desired object is actually desirable, something worthy of desire. (Desire may be enough to explain an action, but it is not thereby enough to qualify it as rational.) Other things being equal, it is rational to pursue one's wants. But generally other things are not equal. In the main, the point is not
The Rationality of Ends
103
what we do want but what we ought to want, not what we desire but 'what's good for us'. And when these differ rationality and desire part ways. (From the rational point of view it is counterproductive to pursue wants at the expense of needs and real interests.) Being desired does not automatically make something desirable, nor being valued valuable. The pivot is how matters ought to be. People's ends and purposes are certainly not automatically valid. We can be every bit as irrational in the adoption of ends as in any other choice. Apparent interests are not automatically real, getting what one wants is not necessarily to one's benefit, goals are not rendered valid by their mere adoption. People's ends can be self-destructive, self-defeating impediments to the realization of their true needs. Rationality calls for objective judgement—for an assessment of preferability, rather than for a mere expression of preference. The rationality of ends, their rational appropriateness and legitimacy, is accordingly a crucial aspect of rationality. More is at issue with rationality than a matter of strict instrumentality— mere effectiveness in the pursuit of ends no matter how inappropriate they may be. When we impute to our ends a weight and value they do not in fact have, we pursue mere will-o'-the-wisps. The rationality of ends is essential to rationality as such; there is no point in running—however swiftly—to a destination whose attainment conveys no benefit. It is useless to maintain 'rational consonance' with what we believe or do or value if those items with respect to which we relativize are not rational in the first place. Principles of relative rationality are pointless in the absence of principles of categorical rationality. Wants per se (wants unexamined and unevaluated) may well provide impelling motives for action, but will not thereby constitute good reasons for action. To be sure, it is among our needs to have some of our wants satisfied. But, it is needs that are determinative for interests, and not wants as such. A person's true interests are not those he does have but those he would have if he conducted his investigative business and his evaluative business properly. A person's welfare is often ill served by his wishes—which may be altogether irrational, perverse, or pathological.12 This distinction of appropriateness— 12 See the author's Welfare (Pittsburgh, Pa., 1972). Cf. John Rawls, A Theory of Justice (Cambridge, Mass., 1971), 421. Rawls traces this line of thought back to Henry Sidgwick.
104
The Rationality of Ends
of real, as opposed to merely seeming, wants and interests—is crucial for rationality. The latter turns on what we merely happen to want at the time, the former on what we should want, and thus on 'what we would want if—if we were all those things that 'being intelligent' about the conduct of one's life requires: prudent, sensible, conscientious, well-considered, and the like.13 To be sure, a person's 'appropriate interests' will have a substantial element of personal relativity. One person's self-ideal, shaped in the light of his own value structure, will—quite appropriately—be different from that of another. And, moreover, what sorts of interests a person has will hinge in significant measure on the particular circumstances and conditions in which he finds himself—including his wishes and desires. (In the absence of any countervailing considerations, getting what I want is in my best interests.) All the same, there is also a large body of real interests that people share in common—for example, as regards standard of living (health and resources) and quality of life (opportunities and conditions)—and it is these that are ultimately determinative of the validity of individualized interests. Both sorts of interests—the idiosyncratic and the general—play a determinative role in the operations of rationality. The rationality of ends inheres in the simple fact that we humans have various needs—that we require not only nourishment and protection for the maintenance of health, but also information ('cognitive orientation'), affection, freedom of action, and much else besides. Without such varied goods we cannot flourish as human beings—we cannot achieve the condition of human wellbeing that Aristotle called 'flourishing' {ebdaiiJiovia). The person who does not give these manifold desiderata their due—who may even set out to frustrate their realization—is clearly not being rational. These various 'goods' are not simply instrumental means to other goods, but aspects or components of a good end—human flourishing. But is that 'good end' really a good end? It obviously is—for us. Flourishing as humans, as the sorts of creatures we are, patently is for us an intrinsic good (though not necessarily the supreme good). We are so situated that from our vantage point 13
The contrast goes back to Aristotle's distinction between desire (opc&s) as such and rational preference (/3ovA7j
The Rationality of Ends
105
(and who else's is relevant for us?) it is clearly something that must be seen as good. We need not deliberate about it—need not endeavour to excogitate it from other premisses; for us, it comes direct—as an inevitable 'given'. No doubt Xenophanes was right. Even as different creatures may well have different gods so they might well have different goods. But no matter. For us the perfectly appropriate sort of good is our sort of good—the human good. In this regard, Aristotle did indeed get to the heart of the matter. For us, the human good (eiSai^ovia) is indeed an adequate foundation for substantive, practical rationality. We have to go on from where we are. It is in this sense alone that there is no deliberation about ends. The universally appropriate ends at issue in our human condition are not somehow freely chosen by us; they are fixed by the (for us) inescapable ontological circumstance that—like it or not—we find ourselves to exist as human beings, and thus as free rational agents. Their ultimate inherence in (generic) human needs determines the appropriateness of our particular, individual ends. The springs of human agency are diverse. Our actions can be engendered in different ways. We act not for reasons alone, but frequently from 'mere motives'—out of fear, desire, habit, impulse. In such cases we also have ends and purposes in view— ' but generally not appropriate ones. If rationality were merely a matter of unevaluated goals and purposes as such—if it were merely the 'technical rationality' of goal-efficient action—then the established line between the rational and the irrational would have to be redrawn in a very different place, and its linkage with what is intelligent and well advised would be severed. But where there is no appropriate and thus no meaningful end, rational agency ceases. (There will, of course, still be room for goal-directed action, but without appropriate goals it will be problematic from the rational point of view.) The rationality of ends is an indispensable component of rationality at large for two reasons. (1) Rationally valued ends must be evaluatively appropriate ones. If we adopt inappropriate ends we are not being rational, no matter how efficiently and effectively we pursue them. (2) We cannot proceed rationally without considering the ends-relative value of our means, inquiring whether the cost of those means (the resources we are expending
106
The Rationality of Ends
through them) is consonant with the values supposedly being realized through the ends, asking: 'If those costs are involved in the means, then are the ends really worth it?' Without rational evaluation, practical rationality becomes unfeasible—with fatal consequences for rationality as a whole, given the systematic unity of reason.
7 Economic Rationality and Problems of Utility Maximization
SYNOPSIS (1) To join with economists and others in locating rationality in 'utility maximization' presupposes that there is a meaningful (and measurable) 'utility' that people are rationally committed to maximize. And this is a very dubious proposition. (2) An 'economic rationality' based on bare, unevaluated preference as such is rationality in name only; it can be altogether irrational. True rationality calls for the pursuit of appropriate ends based on valid human interests, rather than following the siren call of unexamined wants or preferences. (3) The orthodox approach of decision theory in terms of 'maximizing expected utility' encounters particular problems. It is not the arbiter or master of rationality, but only the servant of a more fundamental and deep-rooted conception of rational appraisal.
7.1
THE PROBLEM OF UTILITY
Does the economists' favoured process of utility maximization provide an adequate basis for an account of practical rationality? Many theorists nowadays deem it so. Practitioners of economics, social-choice theory, game theory, and management science, and other exponents of 'the theory of rational decision-making' incline to see the crux of rationality in the maximization of something called 'utility'. They commonly hold that the rational man always aims at a measurable utility correlative with 'satisfaction' or 'well-being' or some such, and always chooses among alternatives on this basis. But, the pursuit of utility has its difficulties when cast in the role of a governing desideratum. For, the idea of maximizing the good rests on the presupposition that all of the different items of value at issue can be evaluated by a common, uniform measure. Measurability and
108
Economic Rationality
comparability are basic to any meaningful value maximization. And hereby hangs a problem. No doubt, the balancing of costs and benefits is a crucial aspect of rationality in general. But, these costs and benefits can be of an extremely variegated and diversified nature. Their assessment in terms of a common measure of 'utility' is something altogether visionary and unrealistic—an over-simplification that makes for convenience in economic analyses but is altogether unrealistic as a description of the real world. It is important for the issue being deliberated here to distinguish between indexing and measurement. It is one thing to assign numbers through some rule or procedure (for example, by giving identification numbers to museum artifacts in serial order of acquisition), and something else to use numbers to reflect a quantifiable characteristic of an object—a measurable property like length or weight. Science has succeeded in mathematicizing the realm of our knowledge to such an extent that we incline to lose sight of the fact that the realm of our experience is not all that congenial to measurement. It is full of colours, odours, and tastes, of likes and dislikes, apprehensions and expectations, loves and hates, pains and joys, that allow precious little room for measurement. We readily forget how very special a situation measurability is—even in contexts of seeming precision. Goods are commensurable only if, despite their evident seeming to be of different kinds, they nevertheless can all be assessed in terms of a 'common denominator', differing in value only in point of varying quantities of this factor in the same way in which baskets of apples and of oranges differ in (monetary) value only in terms of their prices. Commensurability presupposes a common unit—a neutral and pervasive parameter within which all value comparisons can be conducted, in the same way in which all money-value comparisons can be effected in terms of a market price. For the economist, to be sure, there is nothing all that problematic here. He is comfortably settled into the assumption of a market that establishes exchangeability and underwrites a price mechanism to provide a general standard of comparison. For him, it is normal and natural to suppose that different goods can be evaluated in terms of a common unit—money. He can equate the
Economic Rationality
109
value of x apples with that of v oranges because the price mechanism yields a 'rate of exchange' to establish convertibility between them. For him accordingly exchangeability in a market can be invoked to provide for commensurability. But, this envisions a very special set of conditions—conditions that are often not met at all. People constantly make—or endeavour to make—rational decisions: in driving their cars, investing their assets, choosing their careers, tidying up their closets, purchasing their food, and so on. Throughout, they are doing all sorts of well-advised and intelligent things—saving money, prolonging longevity, enhancing comfort, enlarging their friendships, increasing their knowledge, and the like. But, to say that they are throughout doing the same sort of thing—promoting 'utility'—is an eminently problematic contention. And so, at this point economists incline to proceed strictly in terms of preferences, since these seem to be a common denominator in people's choices. Surely (so the economists insist), if you have a preference for A over B we can assess its strength quantitatively by asking: 'How much will you pay to ensure having A instead of BT And they may well be right in holding that this question can always be asked and often answered. But it does not follow that in securing an answer here one is measuring anything objective—that some independently pre-existing, quantifiable attributes of things are being assessed when we spend hypothetical money to indulge our hypothetical 'preferences'. A hypothetical market is no more a market than a wooden owl is an owl. Moreover, preference quantification is not all that simple. When I contemplate painting the room, one colour sample looks more attractive than another, but I cannot quantify by how much. One landscape is prettier than another, but I can put no number to it. But, even if we were to concede the prospect of setting up a strength-of-preference measure, the use to be made of this sort of 'utility' remains problematic. For, as long as one is not in a position to regard utility in the light of a measure of something like goodness or acceptability or desirability or such, any linkage between rationality and maximization is broken. Exactly here is where the approach of most current economists and decision-theoretic utilitarians runs into trouble. They maintain, in effect, something like the following:
110
Economic
Rationality
The idea that utility literally measures something—against which you have tried to argue—is a hopelessly old-fashioned one. Utility does not reflect a measurement at all. It is simply and solely an index of preference. It does not measure intrinsic preferability but merely expresses the strength of a like or dislike. But, in taking this line these 'mere-preference utilitarians' (as we may call them) abandon all claims upon rationality. One person prefers Shakespeare, another the Birdman comics; one gets his pleasure from gardening, another from incinerating moths. Preference as such is simply inclination, with no pretence at appraisal and evaluation within the framework of a rational life-plan. Any pretence of a linkage between utility and rationality is thus abandoned. Clearly, if the use that we propose to make of preference evaluations is actually to have a bearing on judgements of rationality, then preference must reflect preferability. There is no basis and indeed no justification for seeing wants and preferences as something final—something outside the pale of appropriate examination and evaluation. For Jeremy Bentham and the early utilitarians utility was a measurement of value. And so, for them the injunction 'maximize utility!' was a perfectly sensible answer to the question 'What would it be rational to do?' But, once we reconstrue utility in terms of mere preference cultivation and abandon any recourse to value the bearing on rationality is also severed. For, the rationality of someone's striving to implement his preferences crucially depends on what those preferences are—whether, for example, he prefers self-inflicted pain or the suffering of others. (The man who prefers that people should think of him as a flowerpot is scarcely rational, no matter how cogently he may labour in this direction.) A utility theory that abandons the link of utility to value—that takes utility as a mere index of preference rather than a measure of preferability—thereby also severs any connection between maximization and rationality. When utility no longer reflects what is in someone's real interest, there is no longer any good reason to maintain that maximizing it is 'the rational thing to do'. Unless utility can be portrayed in the light of measure of value there is no earthly reason to question the rationality of someone who does not bother all that much about utility. But, if the quantity at issue is something adventitious—something as inherently
Economic Rationality
111
insignificant and potentially unreasonable as mere preference or desire as such—then it is no longer critical for matters of rationality. To re-emphasize: Once the link between utility and value is broken, the link between utility maximization and rational choice is also severed. The utility-maximization approach to rationality standard among economists and decision theorists is thus caught in a dilemma: 1. When utility is approached in the way of the old-fashioned value-utilitarians, then rationality could indeed be construed as calling for utility maximization—if only this utility were a well-defined quantity (which it certainly is not). 2. When utility is approached in the way of the latter-day preference-utilitarians, then utility is indeed a feasibly maximizable quantity, but the bearing of utility maximization on rationality is abrogated. Either way, the view of rationality as utility maximization comes to grief. To join with economists and others in locating rationality in 'utility maximization' presupposes that there is a meaningful (and measurable) 'utility' that people are rationally committed to maximize. And this is a very dubious proposition.
7.2
PREFERENCE VS. PREFERABILITY
The validation of someone's ends of course calls for taking his wants and preferences into account, but it certainly does not call for taking them at face value. If my objectives are ill-conceived or inappropriate, then there is no justification—prudential or otherwise—for their pursuit. If they go against prudence, the general good, or morality—against true self-interest in its widest and most inclusive sense—then rationality is no longer on their side. Whatever urges their pursuit is no longer an imperative of reason. Taken by themselves, 'Pursue your goals!' and 'Strive for your ends' are problematic injunctions from the rational point of view. For, the questions immediately arise: 'Are these goals and ends benign or malign?' Are they in your real interests?' 'Are they as such appropriate?' To ignore these questions is not intelligent but the very reverse. The ruthlessly efficient pursuit of desire is no
112
Economic Rationality
doubt a human trait of sorts—but one that deserves the name of selfishness rather than rationality. The pivot of rationality is not wants but true interests. Wants and preferences are no be-all and end-all. Clearly, not every sort of want satisfaction bears equal weight on the scale of rationality: a person's satisfactions may consist in masochism, Schadenfreude, utter perversity. It is evident that people's wants and preferences can go very wrong indeed—they may be selfdestructive, antisocial, totally inappropriate. A person's legitimate interests (health, knowledge, and so on) can thus stand in sharp contrast with his actual objectives, which, after all, may well emanate from greed, power-madness, or other such motives. Rational choice is a matter of opting not for what is preferred, but for what is preferable. For rationality, what matters is my appropriate wants, and appropriate wants are not just wants—they must also be validatable. Some ends are clearly appropriate qua ends—namely, those that benefit oneself alone (self-preservation or self-development, for example) or in combination with others (communal preservation, and communal development). Someone whose ends run counter to these—who is prepared to act against the welfare and the best interests of himself and his community in the absence of preponderant counter-indications—is simply not being intelligent about the conduct of his affairs. By definition, the rational act or choice is that for which the strongest case can be made out, everything considered. In ignoring the issue of the appropriateness of the operative ends, one weakens to the point of futility the case that is being made for an action or choice that fulfils such an end. The adoption of an end always has consequences and ramifications that cry out for deliberative appraisal. The issue of the validity (legitimacy, overall appropriateness) of ends is indispensible to rationality as such. Without a concern for the rationality of ends there is no true rationality at all. Rationality hinges not on what we do want, but on what we ought to want—on what ends we are well advised to have in the actually prevailing circumstances. Preferability, and with it the issue of the legitimacy of our ends, pivots on the issue of our needs and our appropriate wants. People's needs are complex, and the register of valid human aims and enterprises is correspondingly complicated. There is no one single highest good, but a variety of diversified ones which must be
Economic Rationality
113
addressed by way of achieving a harmonious mixture. Our complexity and many-sidedness must be recognized—man does not live by bread alone, nor sleep, nor learning, nor diversion. But, we need each and all of these diverse goods in mutual supplementation. An intelligently planned day or month or lifetime requires the co-ordination of diverse goods in appropriate conjunction. Economists and decision theorists often talk as though all of a person's wants and preferences are equally rational—as though any end whatsoever is automatically valid (appropriate, legitimate) simply by virtue of its mere adoption. But, this rides roughshod over the crucial differences between real and apparent interests. Consider the sequence: 1. what I want (here and now) 2. what I will want when the time comes 3. what I would want if I knew more (had ampler information) I deliberated more carefully (on the basis of my existing information base) I managed to make certain (recognizedly desirable) changes in myself (i.e., made myself over more fully in the direction of my own ideals). The further we work our way along this sequence of additional qualifications, the more fully do we effect the transition from seeming (or apparent) to real (or legitimate) interests. The various distinctions operative here (short-run vs. long run, well-informed vs. ill-informed, naive vs. reflective, actual vs. ideal) are all crucial for the determination of real as opposed to merely apparent interests. And it is real interests rather than mere wants as such that are central to rationality—not what is desired but what is desirable. The matter is less one of what people do want or prefer than one of what they should want or prefer given an appropriate appreciation of their needs. Free agents, alas, do not always do what they think it best to do. Only rational free agents do this. Many or most of us do what we please without paying more than 'lip service' to what, all things considered, we ought to do to most effectively serve our best interests. Rationality does not (and cannot) turn on preferences pure and simple. This is because rationality pivots on reasons, and
114
Economic Rationality
the question of whether what is preferred actually deserves preference cannot be omitted. With rationality, the issue of there being good reasons for our preferring what we do is ever present. The issue of whether what is valued is indeed valuable can never be brushed aside. For rationality, what matters is not preference itself, but preferability. When utility no longer reflects what is in one's real interest, there is no longer any good reason to maintain that maximizing it is 'the rational thing to do'. There is nothing rational about a wholly preference-geared utility as such—for rationality the question is not what is preferred but what is to be (i.e., deserves being) preferred. Only in so far as our preferences reflect authentic needs and appropriate wants can we view them as rational. Just here lies a salient defect of classical utilitarianism—it sees utility as an ultimate standard in rational choice, whereas the issue of the rational status of utility itself cannot itself be avoided permanently. As even John Stuart Mill was driven to admit rather reluctantly, utility and preference as such just do not allow us to latch on to what people rationally ought to want.1 It lies in the nature of rationality to place a constraint on a person's choices through the distinction that some alternatives are rational and others are not. To be sure, it is not that certain particular options or actions are as such closed off to the rational man—anything can be the lesser of two evils. What is unconditionally mandatory for rationality is not the performing of certain actions but the cultivation of certain values—values which in their turn constrain and canalize actions. The person whose objectives are inappropriate—who heedlessly acts against his own best interest or against those of his fellows—can certainly be efficient in his conduct, but nevertheless fails to qualify as rational. Behaviour aimed at inappropriate ends is thereby irrational. The satisfaction of our preferences is of course an important de facto end of ours. But, as with all other ends, the question of its rational legitimation remains a pivotal issue. And this is something that will depend not just on the fact of preference but on its nature—on the matter of what sort of preference is at issue. It is precisely this fact that makes authentic rationality a much trickier issue than that of the mere 'rationality' of means efficiency alone. 1 See John Stuart Mill, A System of Logic (London, 1843), bk. vi, ch. 12, 'On the Logic of the Moral Sciences'.
Economic Rationality
115
At this point, one comes up against the shortcomings of the concept of economic man and the economists' traditional conception of rationality in terms of the efficient pursuit of personal preferences. What moral philosophers find particularly objectionable in the proceedings of their colleagues in economics and decision theory is the way they reinterpret the honorific rubric rationality to suit their own ends. For, when economists, decision theorists, and social-choice theorists speak so casually of what the rational man does, they manage to conceal under the sheep's clothing of a seemingly descriptive rubric the wolf of a highly dubious and debatable normative commitment to the authority of unevaluated wants and preferences. The issue of evaluation of ends is absolutely crucial for rationality. Mere preference as such is just not good enough. We cannot address the rationality of actions without concerning ourselves with the prior issue of the rationality of the ends towards which they are oriented. We cannot sensibly endorse preference satisfaction as the quintessence of rationality until after we have made an independent assessment of the rationality of the preferences themselves. A narrowly construed 'economic rationality' based on unevaluated desires and mere preferences as such is rationality in name only; it can be altogether irrational. Rationality is a matter of appropriate alignment all along the line—not just of choices with preferences but of preferences with evaluation and of evaluations with values. True rationality demands the pursuit of appropriate ends based on valid human interests, rather than that of unevaluated wants or preferences.
7.3
DIFFICULTIES IN ORTHODOX DECISION THEORY
Among writers in economics and decision theory, it is virtually a dogma that the correct approach to rational choice is represented by the orthodox programme of expected-value calculations which weight alternatives through their respective utilities and likelihoods. To be rational, on such an approach, is to make one's choices via a balance of probabilities and utilities. However, this standard recourse to expected-value comparisons itself encounters substantial obstacles. For one thing, probabilities are not secure givens delivered into
116
Economic Rationality
our hands with accuracy guaranteed by the Recording Angel on high. They are vulnerable, fragile constructions based on limited and often problematic mechanisms. They are not arbiters of rationality, but themselves objects of rational criticism—representing factual judgements that themselves need to be rationally substantiated. Moreover, one must never forget that expectations are averages. Choices by means of expected-valued calculations are always based on the idea that the situation in hand can be resolved with a view to an average result. And in particular cases this can be far off the mark. For example, on a short-term basis, 'risk aversion' can very properly come into play. That is, it would not be at all irrational to prefer the top alternative in Figure 7.1, even though its expected value is the lesser. As this example shows, an expected value may fail to give a realistic measure of certain equivalency in the presence of an aversion (or propensity) towards putting a 'bird at hand' at risk. FIGURE
7.1: Gambling Aversion and 'Bird in Hand' Preference
Note: The expectation for A is +50; that for not-A is + 50.5.
Since they are averages of sorts, expected values lose sight of details and distinctions that may prove important for particular cases. They equate small probabilities of large gains or losses with large probabilities of small gains or losses. For example, the expected values for a one-in-a-thousand chance of losing $10,000 and for a fifty-fifty chance of losing $20 are equal. But, whether
Economic Rationality
111
mishaps occur frequently but with minor effect or rarely but with enormous impact is clearly going to affect the judgements of rational agents. The rational man will not follow the guidance of expected-value comparisons blindly and mechanically, without care for the details of the situation. Another problem with the expected-value approach is reflected in the fact that there are often good reasons for dismissing sufficiently improbable possibilities. For one thing, there are just too many of them: to be asked to reckon with such remote possibilities as unforeseeable natural disasters and the like is to baffle our thought by sending it on a futile chase after endless alternatives. A further reason lies in our need and desire to avoid stultifying action. It is simply 'human nature' to dismiss sufficiently remote eventualities in one's personal calculations. The 'Vacationer's Dilemma' of Figure 7.2 illustrates this. A rational vacationer takes the plausible line of viewing the chance of disaster as effectively zero, thereby eliminating that unacceptable possible outcome from playing a role by way of intimidation. People generally (and justifiedly) proceed on the assumption that the probability of sufficiently unlikely disasters can be set at zero; that unpleasant eventuations of 'substantial improbability' can be dismissed and taken to lie outside the realm of 'real' possibilities. FIGURE
7.2: The Vacationer's Dilemma
f S\
Stay at y_ home J
—CO—
CHOICE*^ /9999999V-
Have a safe but dull time of it. Value: small. Have a good time Value: positive, but modest.
•'1 + 100
^ y ' G o off on Y \. vacation J\
Meet with some 'perfectly awful' \ ^10,000,000)l 1 \ disaster or outright catastrophe. Value: negatively enormous-
Note: expectation for top alternative= + l; expectation for bottom alternative=99.99999-100= -.000001
-lO- 9
118
Economic Rationality
Only by artificially dismissing certain sufficiently remote, catastrophic possibilities as outside the range of real possibilities can we avoid the stultification of action engendered by anything like the standard decision-making approach via expected-value comparisons. The orthodox decision-theoretic programme of 'rational decisionmaking' through the comparison of mathematical expectations that mechanically combine probabilities and utilities is a useful but limited tool. Its advantages are obvious—clarity, precision, versatility. But its limitations are not insubstantial. Only under special, restrictive assumptions are probabilities and utilities defined, and only over limited ranges of conditions does their combination by way of expected values provide a sensible guide to decision-making. Expectations provide a useful decision-making tool in situations of rational choice, but one that cannot be used in a mindless and automatic way, since it is appropriate only where special conditions hold that often fail us. The mechanism of expected-value comparisons is an instrumentality of great power. But, like most other useful tools, there are limits to its utility. And herein lies the defect of a view that sees orthodox decision theory as the arbiter of rationality. The very fact that expected-value comparisons can be implemented in a rational (intelligent) and an irrational (unintelligent) way—depending on how meaningful those utilities are and how well-based those probability values are—shows that rationality cannot consist in expected-value appraisals alone. And so, while the maximization of actual or expected utilities is an important instrumentality of rational procedure, it is not itself the determinative arbiter of rationality. Probabilities and utilities do not determine rationality: they themselves are subject to rational constraints. Rationality is something too fundamental to be embraced within the boundaries of standard economic practice and orthodox decision theory. Expected-utility calculation is no more than a serviceable tool whose proper use must itself be conditioned and canalized by considerations of rational appropriateness.
8 The Systemic Unity of Reason
SYNOPSIS (1) Practical reasoning about particular actions requires both cognitive and evaluative considerations. (2) Both practical and evaluative rationality are inherent in cognitive rationality. (3) Evaluative reason has both cognitive and practical dimensions. (4) Reason is an organic unity, an indivisible whole. The fabric of rationality is seamless: all of its several departments are inseparably intertwined. (5) Rational inquiry and rational evaluation operate in a wholly parallel way. Both consist in the rational systematization of experience—our informative and our evaluative experience, respectively.
8.1
PRACTICAL REASON REQUIRES COGNITIVE AND EVALUATIVE REASON
As Chapter 1 has already noted, there are three major contexts for rationality, namely, belief, action, and evaluation. Correspondingly, there are three types of reasoning, the theoretical or cognitive (reasoning about matters of information), the practical (reasoning about actions), and the evaluative (reasoning about values, ends, properties, and preferences). It is the burden of this chapter to show that these all overlap in so intertwined a manner that they just cannot be strictly separated. This is a conclusion which must be argued in stages, and so we shall proceed step by step, beginning with the involvements of practical reason. Practical rationality requires that our actions and their objectives be duly validated and co-ordinated under the aegis of intelligence. In its generic structure, practical reasoning about particular actions takes the following form: 1. All things considered, my attaining the result X would be to the good. 2. Of the various alternative actions (or courses of action)
120
The Systemic Unity of Reason
available to me, it is A which, on the available information, is optimally conducive (overall) to my attaining the result X. 3. Doing A meets the condition of everything—considered optimally in relation to the whole spectrum of my appropriately operative ends. It involves no Jt-irrelevant side-effects that would offset its X-connected benefits. Therefore: A is the rational thing to do in the circumstances at hand: I am rationally well advised to do A. That first premiss, patently, involves a reference to values, while the second premiss makes a strictly factual claim about how things work in the world. (The third premiss is of mixed character.) Cogent practical reasoning in the context of particular decisions accordingly involves reference both to matters of fact and to matters of value. We cannot even begin to effect a rational choice among alternative courses of action without determinate beliefs about the consequences of these actions. Even if performing an action is in fact conducive to realizing someone's appropriate ends (if, say, ingesting yonder chemical substance will actually cure his illness), it is nevertheless not rational so to act if the agent has no knowledge of this circumstance (and all the more so if such information as he has points the other way). Even when we happen by luck or chance to do that which is, in the circumstances, the best thing to do, we have not acted rationally if we have proceeded without having any good reason to think that our actions would prove appropriate—let alone if we had good reason to think that they would be inappropriate. The agent who has no good reason to think that what he does conduces to his appropriate ends is not acting rationally. And his deficiency is not redeemed by unmerited good fortune—by luck's having it that things turn out all right. Rationality in action is not simply a matter of acting effectively towards our ends, but one of acting intelligently, and, given the role of chance in the world's events, these are not necessarily the same. Practical rationality must accordingly pivot on proper cognitive backing of duly exploited information. If someone's action is to qualify as fully rational, then (1) the agent must be in a position to recognize his actual reasons for doing it (there must be no selfdeception here), and (2) the agent must be able to evaluate his
The Systemic Unity of Reason
121
reasons as appropriate ones. A misunderstanding either about the character or about the merit of one's reasons for action is a defect of rationality. Rationally warranted action must be intelligently guided by rationally warranted information. Practical rationality accordingly requires cognitive rationality, seeing that rational action is a matter of doing what we have good reason to think to be effective on the basis of the information that is available to us. For practical rationality it is not belief as such but sensible (that is, rational) belief that counts. The issue of rational cognition is thus critical for practical rationality as well. If we have no idea how things work in the world, we remain impotent to act intelligently in the effective pursuit of our goals. Lapses from cognitive rationality would also involve us in lapses from pragmatic rationality, seeing that they involve an avoidable risk of frustrating our own purposes and interests. Intelligent action cannot proceed without knowledge, in whose absence we would no longer be able to perform those acts that afford our best-warranted hopes for optimal results. (We are in sad straits when, for instance, we cannot tell foods from poisons.) Rational action within the world must be predicated upon rationally warranted beliefs about it: we cannot expect to succeed with a praxis that is not underwritten by knowledge. And a similar story holds on the side of values. If our practical decisions are directed at 'unsuitable' ends—ends at odds with the indications of appropriate evaluative reasoning—then we are not conducting our business of practical reasoning aright. Actions aimed at inappropriate objectives or geared to mistaken priorities are thereby automatically flawed from the rational point of view. For, practical activity in the pursuit of an objective is rational only in so far as we have sound reason for deeming the realization of this objective to be a good. Accordingly, evaluation too is an essential resource for practice. Practical action only makes sound rational sense when it proceeds with a view to an end of some sort, and ends are appropriate only in so far as they can secure the legitimatizing authorization of values. When the adoption of one or another of a set of alternatives is under consideration, rationality requires that two sorts of issue must be addressed; their relative feasibility (costs), and their relative desirability (benefits). Without evaluative as well as
122
The Systemic Unity of Reason
cognitive reason, the conditions for practical, action-oriented rationality cannot be realized.
8.2
C O G N I T I V E REASON HAS BOTH PRACTICAL AND EVALUATIVE D I M E N S I O N S
Not only does practical rationality require cognitive rationality (since we cannot sensibly identify practically effective steps in the absence of factual information), but the converse holds as well: cognitive rationality requires practical rationality. Even in the sphere of our specifically cognitive concerns, praxis and action are not something external—a supplemental afterthought. For, inquiry and the acquisition of information is itself a practical activity on the same footing with any other—a process that must be governed by the standard justificatory ground rules of practical reason. In falling afoul of practical reason we compromise our claims to cognitive rationality as well, because rational belief about this world can only emerge from a rationally effective methodology of inquiry—from an appropriate process of information acquisition. Inquiry too is an activity which, properly conducted, must be governed by the ground rules of practical rationality. And practical reason plays yet another crucial role in our cognitive affairs. If experience cannot validate our factual information about the world, then nothing can. Yet, how is experience to be exploited to secure information about objective facts? Experience is inevitably episodic, personalized, subjective. It operates at the level of appearances. I am under the impression that yonder object looks rose-like; I judge that it emits a rosy fragrance; it feels like a rose to my touch. But, this sort of thing is all I get at the level of actual experience. In describing my experiences, I must use such autobiographical-sounding reports about how things strike me. Experience as such goes no further than to place me in a situation where / take something to be a rose. And the move from this merely subjective circumstance to the objective fact that it actually is a rose before me traverses a substantial distance. (As sceptics have always been ready to point out, there is—in theory—a lot that can conceivably go wrong.) However, the following complex piece of practical reasoning
The Systemic Unity of Reason
123
provides our only practicable way of validating the inference from subjectively experiential indications to objective fact: 1. On present indications, P. 2. In the prevailing circumstances, we can do no better (epistemically) than to go by present indications. 3. Accepting P is our best available option (in the prevailing circumstances) (from 1 and 2). 4. Let us adopt the best available option. (A matter of 'practical policy'.) 5. Let us accept P (from 3 and 4). 6. P (from 5). As this perspective indicates, what is at issue when we ground objective judgements on an experiential basis is not a matter of reasoning in the purely theoretical order of deliberation at all, but one of reasoning in the practical order of deliberation. We do not—nay, cannot—move from 1 to 6 directly, but only via the practical policy represented by premiss 4. What is at issue when we move from 'on present indications, P' to P itself is not a matter of any duly consolidated inference at all; it is a matter oi practical procedure. As such this policy does indeed have a backing of rational warrant. But this policy-supportive warrant itself lies in the practical rather than purely cognitive sector. We have no alternative but to proceed to cross the evidential gap from experience to objective factual claims (from phenomenal evidence to actual fact, from putative to real truth, from subjective data to objective claims) by what is, to all intents and purposes, a strictly practical inference whose ultimate foundation lies not in the sphere of information but in that of practical policy. Our rational procedure in the accession of substantive information about the world is part and parcel of our cognitive praxis under the aegis of interests. That basic demand for information and understanding presses in upon us and we must do (and are pragmatically justified in doing) what is needed for its satisfaction. (On one reading, this is the lesson Hume himself drew from his critique of 'induction'.) Cognitive reason as such has two sides, the hypothetical and the categorical. The hypothetical proceeds by way of: 'If you accept P, then you must (or, alternatively, cannot) in due consistency also accept Q.' Its conditionalized, if-then aspect does not involve us in
124
The Systemic Unity of Reason
any substantive commitments whatever. But of course it also provides no substantive information. Only in the categorically committed use of cognitive reason do we obtain outright information—effective answers to our questions. And this categorical dimension of cognitive reason has an ineliminably practical basis. For, we cannot get answers to our questions about the world without taking action. To accept a thesis is to do something—even mental action is itself a kind of action, and belief-formation a kind of praxis. The following objection arises: You maintain that cognitive reason involves practical reason through construing acceptance as an act—an 'act of acceptance'. But you also hold that practical reasonings always require premisses that can only be secured through cognitive reason. Do these together not make for a sceptical regress that blocks all prospect of cognition via the vitiating consequence that every valid practical argument presupposes others? The answer to this question is negative. For one must observe: 'Requires, yes, but not pre-requires—demands, yes, but not presupposes in the literal sense of something that must go before.' Requirements and requisites are not necessarily preliminaries. If I endorse 'We must leave the house' I am also required to endorse the claims: 'We must not remain in the living room', 'We must not remain in the dining room', etc. But I need not pre-establish these other requisites en route to validating that initial claim. The later claims stand co-ordinate with the earlier: they form part of the overall fabric of concomitant claims, but we do not have to validate them in advance as prerequisites. It is quite enough that they can eventually be validated—though perhaps only 'with the wisdom of hindsight'. And that holds for the present case as well. Those concomitant contentions must hold good and be available in the overall scheme of things. But, we need not secure them in advance. The structure of rationality is a matter of system, not of sequence. The mutual involvement at issue is reciprocal and harmless rather than presuppositional and vitiating. It does no more than illustrate the systemic unity of reason—the cohesive conjunction of its several components. Not only are cognitive reason and practical reason intertwined, but cognitive reason has evaluative involvements as well. Once one has
The Systemic Unity of Reason
125
certain facts at one's disposal, one can, of course, proceed to derive others from them by logical inference. But, one must make a start at the inferential venture by way of accepting some facts not on the basis of other facts, but on the basis of mere indications. We cannot proceed in factual matters without making some nondiscursive judgements of fact, such 'acts of acceptance' must be guided by evaluative processes. The stance that 'The indications are indeed strong enough to warrant acceptance in the particular case at hand' presupposes judgements along the lines of 'This issue is important enough for us to chance a resolution on present evidence notwithstanding the inherent risk of error'. Without these judgements regarding issues of cognitive value, there is no sensible basis for holding that the prevailing conditions justify running the risks involved in the acceptance of something that could, in the end, turn out to be incorrect. Thus, evaluation is also bound to enter into our cognitive deliberations. But the salient fact in this regard is the particularly close interconnection between cognitive and practical reasoning. For, on the one hand we need cognitive impact to operate practical reasoning, and on the other we must make greater considerations to validate the processes of our cognitive operations (as we saw in Chapter 3). 8.3
EVALUATIVE REASON HAS FACTUAL AND PRACTICAL DIMENSIONS
Even as cognitive rationality demands both practical and evaluative rationality, so evaluative reason has both cognitive and practical aspects. Evaluative judgements often take the generalized form: 'Items of a certain kind (i.e., of type T) have positive (or negative) value of such-and-such a sort, and to such-and-such an extent.' ('Acts of stealing are wrong.') But, of course such a general rule remains unproductively ineffectual, spinning inoperatively like a disengaged gear, until it is connected up with concrete situations. If we are ever to make actual use of such a generalization we must bring it to bear on a particular case: The existing situation 5 actually represents an instance of an item of the type at issue (type T).
126
The Systemic Unity of Reason
We can only move from the premiss 'Type T things have value V to the conclusion 'X has value V via the linking premiss lX is of Type 7"—moving from 'Theft is wrong' to .'AT has done something wrong' via the premiss lX has played the thief (has taken something that actually did not belong to him).' And this last item is clearly something that involves factual information. Only on the basis of such factual inputs are we able to apply or implement our values; in the absence of factual information, our value generalizations are no more than bloodless abstractions. Moreover, evaluative rationality demands practical rationality as well. This follows from its cognitive involvement through the consideration, already examined above, that evaluation's demand for factual information invariably requires a practically grounded leap across the gap from subjective evidence to objective conclusions.
8.4
THE SYSTEMIC UNITY OF REASON
Rationality is a matter of the intelligent pursuit of appropriate ends. Here, 'intelligence' bespeaks knowledge, 'pursuit' indicates action, and 'appropriate ends' call for evaluation. All sectors of reason must be invoked and co-ordinated in any formula that adequately characterizes the overall nature of rationality. To serve its function as a guide to human actions and interactions, reason must eliminate disorder and dissonance and similar impediments to co-ordinated thought, and accordingly she strives—always and everywhere—for consistency, uniformity, generality, and orderly harmony of all sorts. Rationality demands that our beliefs, evaluations, and actions should 'make sense'. And, as has been shown step by step, this means that the whole fabric of rationality must be seamless—cognition, evaluation, and action must form a cohesive unit. People have questions and demand answers. And we require that these answers fit, in mutual coherence, attuned to one another in systemic unity, forming a body that we are willing to implement in practice and use as a guide to action. Accordingly, we must be (cognitively) confident that—as best we can tell—such (practical) action will eventuate in (evaluatively) good effects. Where an agent's beliefs, values, and actions are not in themselves appropriate, and not mutually
The Systemic Unity of Reason
127
concordant, the preconditions for rationality remain unsatisfied. Reason is an organic unity, an indivisible whole. To be sure, philosophers sometimes envision a bifurcation of reason into two separate domains with altogether different ground rules, distinguishing sharply between theoretical and practical matters, between deupia and -n-pa^i?. In theoretical matters, they maintain, there is no urgency to make up our minds—we can always postpone a decision until further evidence comes to view, and simply suspend judgement until all requisite returns are in. In practical matters, on the other hand, there is an urgency to make up our minds—we must act in one way or another (since even inaction is a mode of action), and so must reach a decision of some sort. But, this seemingly clear contrast turns murky on closer scrutiny. Theoretical and practical reason are not disparate domains but interconnected sectors of one comprehensively unified whole. A single, uniform governing principle obtains throughout: 'Proceed everywhere on the basis of the best available reasons to do what is appropriate in the situation at hand.' The division between the practical and the theoretical does not bifurcate reason, it simply illustrates the fundamental uniformity of the rational through a diversity of situations. Our factual judgements (beliefs) and our evaluative judgements (appraisals) are both validated by rational processes whose fundamental structure is one and the same—and on both sides ultimately rest on considerations of praxis. Rationality is a matter of establishing harmony within (and between) the spheres of action and belief in the interests of the efficient pursuit of legitimate ends. Its characteristic demand is for good reasons—good reasons for what we believe, for what we do, and for what we value. Under the aegis of rationality, these three domains form part of a single, uniform, and co-ordinated whole. If our acts are based on inappropriate beliefs, they lack rational justification; if our beliefs do not admit of implementation in practice, they too suffer a defect of rationality; if our values are inappropriate they go against reason. In no such case would a rational agent be able to muster the confidence necessary for effectual thought and action. As these considerations indicate, reason is an organic unity, an indivisible whole. Each of its sectors interpenetrates the rest, seeing that the rational person must, in a duly co-ordinated way:
128
The Systemic Unity of Reason
1. strive to secure information adequate for the effective management of his affairs; 2. endeavour to assure the appropriateness of his values and goals; 3. act (in the absence of putative impediments) in ways which, given his beliefs, are serviceable for the implementation of his values and goals. The fabric of rationality is seamless: its several departments are inseparably interconnected.
8.5
THE PARALLELISM BETWEEN RATIONAL INQUIRY A N D EVALUATION
Contentions are normative when they go beyond mere description either in expressing a value-judgement or in having a valuejudgement as an unavoidable presupposition or consequence. The distinction between description and normative evaluation is real and important. To say that someone's face is asymmetrical is only to describe it, but to say that it is ugly is to make an evaluative judgement upon it. To say that John's bone was broken is merely to depict the situation descriptively; to say that it was injured or harmed is also to evaluate by indicating that the transformation at issue was a bad one. 'Killing' is merely descriptive, 'murdering' is also normative—seeing that murder is an unjustified killing. Sometimes the distinction between the descriptive and the normatively evaluative is expounded through a distinction between 'facts' and 'values'. But, this is very misleading. Factuality as such is something that is neutral as between strictly informative facts and additionally evaluative ones. Descriptive facts and normative ones are more closely interrelated than is generally recognized. They share a common epistemology. Rational cognition and rational evaluation run wholly parallel in point of validation because cognition too is an ultimately evaluative enterprise. Values and descriptive facts are both governed by norms. Our knowledge of both sorts of facts—the descriptive/ informative and the normative/evaluative—hinges on the criteriological bearing of the question: What merits approbation? To be sure, this overarching question bears a very different construction
The Systemic Unity of Reason
129
on each side of the issue, with approbation = acceptance on the side of descriptive information, and approbation = preference on the side of evaluative judgement. But acceptance too is a preference of sorts: an epistemic preference. Rational appropriateness in the criteriology of cognition is _, determined as follows: Those descriptively informative theses (descriptive judgements) qualify as rationally acceptable (= cognitively valid) which optimally systematize our cognitive data (where systematization proceeds under the aegis of descriptively factual generalizations). In an entirely parallel way, we have the following situation on the side of the criteriology of evaluation: Those normatively evaluative theses (evaluative judgements) qualify as rationally acceptable (= normatively valid) which optimally systematize our evaluative data (where systematization proceeds under the aegis of normative rules). Both the descriptive and the normative sides begin with the 'data of experience'. In the alethic (descriptively truth oriented) case, these are the data of sensation (sense perception) and their systemic extensions in factual theories. In the evaluative (normatively value oriented) case, these are the data of evaluation—of pro-orcon appraisal and their systemic extensions in normative rules. In both cases alike we proceed criteriologically in terms of the optimal systematization of experience—that is, by just the same device of seeking the best available extrapolation of the data—the interpretation that best coheres with the rest of our experience. In the cognitive case, validity (here amounting to presumptive truth or factual correctness) calls for the optimal inductive systematization of our informative experiences under the aegis of principles of explanation. In the normative case, validity (here amounting to presumptive appropriateness or evaluative correctness) calls for the optimal normative systematization of our evaluative experiences under the aegis of principles of justification. The parallelism between the two cases is depicted in Display 8.1, which portrays a value-cognitivism that sees the processes of rational inquiry and of rational evaluation as proceeding in a strictly parallel way. On both sides, system building provides us with a screening process that includes the fitting and excludes the
The Systemic Unity of Reason
130 DISPLAY
8.1: The Cognitive Parallelism of Inquiry and Evalu-
ation I.
ALETHIC
EPISTEMOLOGY
(Truth Criteriology) Inductive Systematization DATA
INFORMATION
The 'Raw' Data of ,' Informative the Senses / Beliefs Experiential Data
(Presumptively) -=• Correct Statements
Judgemental Extrapolations 2.
AXIOLOGICAL
EPISTEMOLOGY
Experiential Data
(Value Criteriology) Judgemental Extrapolations
The 'Raw' Data of Value-Sensitivity
Evaluative Prescriptions
DATA
(Prescriptively) -> Appropriate Evaluations EVALUATION
Evaluative Systematization
unassimilable, discriminating between what is tenable and what is not. The fundamental idea is that of controlling validity through the optimal systematization of the relevant data that runs uniformly across both the cognitive and normative domains. In either case, this is a matter of systematizaton of experience: cognition: rational systematization of informative experience through principles of explanation. evaluation: rational systematization of affective experience through principles of justification Essentially, the same standard applies throughout: a judgement is valid if it belongs to the most cogent systematization of the whole range of our relevant, alethically fact-oriented experience on the one side and that of our relevant, axiologically value-oriented
The Systemic Unity of Reason
131
experience on the other. (Note that this position that the best available systematization will do—that it is good enough as a basis for commitment—carries us back to the pragmatic line of justification sketched near the outset of Section 2 above.) The coherence approach to value criteriology in terms of judgemental systematization accordingly runs wholly parallel to the coherence approach to acceptance criteriology.1 The parallelism of the systemic process operative in both the cognitive and evaluative sectors engenders a symmetry of validation on the sides of cognitive and evaluative reason that once again exhibits the fundamental unity of reason. No wonder, then, that altogether analogous issues arise on both the cognitive and the evaluative sides. In the alethic case, we face the problem of bridging a (seemingly insuperable) gap between appearance and reality, between phenomenally subjective claims at the level of appearances and impressions, and ontologically objective claims at the level of being and actuality. The coherence criteriology of factual truth is of good avail here. For it authorizes the move from claims on the order of 'There looks to be a cat on the mat' to claims on the order of 'There actually is a cat on the mat' through optimal systematization. In making this move, we exploit the circumstance that this particular ruling regarding the nature of the real best systematizes our cognitive commitments overall. And an entirely parallel situation prevails on the evaluative, axiological side, where we face the problem of bridging a (seemingly insuperable) gap between subjective claims at the level of evaluative feelings and objective claims at the level of actual evaluation—between what strikes us as wicked {seems wicked) and what is wicked. And of course what the coherence criteriology of norms enables us to do in this case is to leap across just exactly this gap. It puts us in a position to move from claims on the order of 'Stealing seems wicked to me' to claims on the order of 'Stealing is wrong' through the mediation of the principle of best systematization. On both sides of experience—both with the sensory 'observation' and evaluative 'assessment'—we thus leap across the gap separating subjective seeming ('appearance') from objective being ('reality') 1 On coherence criteriology see the author's The Coherence Theory of Truth (Oxford, 1973) as well as ch. 2, 'Truth as Ideal Coherence', of his Forbidden Knowledge (Dordrecht, 1987), 17-27.
132
The Systemic Unity of Reason
by one selfsame device—systematization of the data. In each case we enter into a realm of objective claims through triangulation from the data of experience. (Of course, the greater extent of the interpersonal uniformity of sensory as compared with evaluative experience makes the case of sensation simpler than that of evaluation. But the difference is one of degree rather than of kind.) This fundamental parallelism means that value issues should also be seen in a 'realistic' light. Matters of value too can and should be regarded as objectively factual—the difference is just that we are dealing with evaluative rather than simply informative facts. The possibility of rational agreement/disagreement, criticism, correction, and the like, arises on the evaluative side also. One must avoid the confusion of values with tastes. 'There's no disputing about tastes' may be true, but 'There's no disputing about values' certainly is not. Values too can be altogether objective, in that value claims admit of rational support through impersonally cogent considerations. To re-emphasize: the rational validation of descriptively factual claims in empirical inquiry and of evaluative claims in normative evaluation proceed in closely analogous ways. Both consist in the rational systematization of experience—informative and evaluative experience, respectively. The parallelism of alethic and axiological criteriology indicates that what is sauce for the informative/ inductive goose is also sauce for the evaluative/normative gander. And this circumstance is highly important from the angle of our present concerns. For one thing, it illustrates from yet another direction of approach the holistic unity and integrity of reason. For another, it indicates that the very existence of an evaluative sector of reason hinges on the prospects of an objective rational inquiry into the nature and bearing of evaluative considerations. And this is all to the good. Given the systemic unity of reason, the whole of rationality would collapse into uselessness if rational deliberation about matters of value were in principle impossible.
9 Conceptual Egocentrism and the Limits of Cognitive Relativism
SYNOPSIS (1) Are the rules of reason uniform for all rational beings,or are they variable and relative? A specifically cognitive relativism is in fact unavoidable; it is rationally quite appropriate for different people, eras, and cultures to have not only different bodies of accepted beliefs, but also different standards and criteria of rational acceptability—different bases for the rational conduct of affairs. (2) There are different forms of relativism, and they have very different degrees of merit. The version endorsed here is an epistemic relativism with respect to different cognitive bases that nevertheless calls for an emphatic rejection of basis-egalitarianism. (3) Relativism as such does not underwrite indifferentism. Our commitment to our own position is (or should be) unalloyed. (4) Although one must recognize the reality of alternative cognitive frameworks, one certainly need not see them as valid equally with one's own. (5) The implications of our own conception of rationality are absolutely decisive for our deliberations. We ourselves are bound to see our own (rationally adopted) standards as superior to the available alternatives— and are, presumably, rationally entitled to do so in the light of the cognitive values we ourselves endorse.
9.1
RATIONALITY AND C O G N I T I V E RELATIVISM
Can a good reason for one person to believe (or evaluate or act) in a certain way fail to constitute a good reason for another to do so? Rationality is supposedly homogeneous—uniform for all rational agents alike. But is this indeed so? Are the rulings of reason always universally binding on all rational beings alike? It certainly seems that a good reason for one person is not necessarily a good reason for another—that, say, if something renders a certain action well advised for one person to perform then this is not necessarily the case with someone else. (What is a
134
Conceptual Egocentrism
good medication for my cold will not necessarily be a good one for yours.) Different people (let alone different cultures!) occupy different 'situations' that make the things that are rational for some to believe or to do quite different from those that are rational for others. To all appearances, rational validation is something that is variable with individual and group circumstances, and thus ceases to have any claim to universality and objectivity. And so, the problem of relativism arises—one person's or group's rationality may conceivably be another's foolishness.1 But would not this sort of pluralistic relativism destroy rationality as such? Does it not make every man his own arbiter, so that reason collapses into chaotic fragments? It is clearly necessary for any adequate treatment of rationality to address this vexing issue of interpersonal diversity and its ramifications. Let us begin by focusing on cognitive relativism. It is only rational to conform one's beliefs to the course of one's experiences. And this makes a relativistic plurality of beliefs inevitable. For, rationality itself requires differently circumstanced people to have different beliefs. The possession of the information that provides good reasons does not automatically transfer from one rational believer (or group) to another. Seeing that they have different experiences on the basis of which to form their judgements, it would not be rational of them not to differ in what they (quite rationally) believe. (Recall John Godfrey Saxe's poem of the blind men and the elephant.2) On the basis of his data, Newton was every bit as justified in his conclusions as Einstein was on the basis of his. The truth estimates of each were plausibly warranted 1 For a conspectus of recent discussions of relativism, as well as references to the literature, see: Michael Krausz and Jack Meiland (eds.), Relativism: Cognitive and Moral (Notre Dame, Ind. 1982); Michael Hollis and Steven Lukes (eds.), Rationality and Relativism (Oxford, 1982); Joseph Margolis (ed.), Is Relativism Defensible?, special issue of The Monist, 67, no. 3 (July, 1984). 2 The poem tells of the blind sages, those
. . . six men of Indostan To learning much inclined. Who went to see the elephant, (Though all of them were blind). One sage touched the elephant's 'broad and sturdy side' and declared the beast to be 'very like a wall'. The second, who had felt its tusk, announced the elephant to resemble a spear. The third, who took the elephant's squirming trunk in his hands, compared it to a snake; while the fourth, who put his arms around the elephant's knee, was sure that the animal resembled a tree. A flapping ear convinced another
Conceptual Egocentrism
135
in the prevailing circumstances by the available information. Not only can and do different people make different judgements about what is true, but they can do this with complete cogency and rational appropriateness. In principle, at least, it is thus entirely possible that the 'primitive' tribal people who accept the pronouncements of the witch-doctor and swear by his incantations are acting no less rationally than we 'civilized' moderns who accept the declarations of the physician with his medicines. Very possibly, that is, each is making equally adequate use (by prevailing standards) of the best evidence available to them. And this conformation of acceptance to potentially variable evidence is what cognitive rationality is all ' about. The judgements people can rationally make about matters of factual truth are inevitably basis relative—linked to variable evidence, rules, and methods. In the cognitive sphere, pluralism is inescapable. Here, we simply have to proceed from where we are, making our determinations as best we can on whatever basis the circumstances of era and culture place at our disposal. My evidence may not point the same way as yours, nor need my criteria be your criteria, given our differences in education and experience. And so, different cultures, different eras, different people can—quite appropriately—play the truth-determination game by different rules. No world spirit legislates that one selfsame inquiry-process must be used by all men in all times and places and circumstances. (Weather forecasting does not operate on the same principles here today as in ancient Rome.) No matter where we stand regarding 'relativism' as a philosophical doctrine, we have to accept the fact of relativity—the fact that different people can, quite appropriately and rationally, proceed differently in the conduct of their intellectual business. A specifically cognitive relativism is in fact unavoidable; it is rationally appropriate for different people, eras, and cultures to have not only different that the elephant had the form of a fan; while the sixth blind man thought that it had the form of a rope, since he had taken hold of the tail. And so these men of Indostan, Disputed loud and long; Each in his own opinion Exceeding stiff and strong: Though each was partly in the right. And all were in the wrong.
136
Conceptual Egocentrism
bodies of accepted beliefs, but also different standards and criteria of rational acceptability—different bases for the rational conduct of affairs. (If the science of statistics has not yet been introduced, they cannot make use of statistical data and methods.) Whatever may be the way in which rationality binds all rational beings alike, it has to come to terms with this sort of relativity at least. 9.2
THE BASIS PROBLEM
What does this basis relativity of cognitive rationality portend for the validity of our own beliefs? The following plausible-sounding propositions clearly form an inconsistent triad: 1. Different cultures have different views of the truth. 2. If different cultures do indeed have generally different views of the truth, then the different truth views of different cultures are all equally valid. 3. If different cultures do indeed have generally different views of the truth, then one culture's view of the truth—our own— is rationally superior to others. Given that these three contentions are collectively incompatible, at least one of these three propositions must be abandoned. So we face a choice among: 1-rejection. Monism: All cultures 'at bottom' think alike. Intercultural disagreement is an illusion that inheres in a superficial analysis of the facts; 2-rejection. Preferentialism: There is disagreement all right, but one culture's position—presumably our own—can validly be seen as rationally superior to others; 3-rejection. Egalitarian Relativism: Our own position is only one among others that are all equal to it in merit. Since the first, monistic line is scarcely plausible in view of the evident facts of the matter, it is worthwhile to look more closely at the residual choice between the other two approaches. Epistemic relativism maintains that the truth status of propositions differs relative to different bases and can, therefore, quite appropriately be adjudged differently by different people. But, what stance is one to take towards those different bases themselves?
Conceptual Egocentrism DISPLAY
137
9 . 1 : Versions of Basis Relativism
1. Basis egalitarianism Several bases are possible, and all of them are of an equivalent validity-status. The choice between them is rationally indifferent. All alike are equally: (1) invalid S C E P T I C A L RELATIVISM
(2) valid I N D I F F E R E N T I S T RELATIVISM (Protagorean Relativism) 2. Basis preferentialism Several bases are possible, but one particular basis has a privileged status, in that: (1) one alone is objectively (impersonally) correct OBJECTIVIST RELATIVISM
(2) one alone is subjectively (person- or group-relatively) correct (a) simply through custom or some other rationally arbitrary commitment C O N V E N T I O N A L I S T I C RELATIVISM
(b) on the basis of principles of some sort which themselves are seen as adopted (i) a-rationally (through taste, congeniality, custom, or the like) A - R A T I O N A L I S T I C RELATIVISM
(ii) rationally (on the basis of certain appropriate cognitive values) A X I O L O G I C A L RELATIVISM
The range of possibilities here is canvassed in Display 9.1, which effectively exhausts the alternatives for basis-relativism. The two egalitarian modes of basis relativism can be given relatively short shrift. 1. Sceptical relativism. The sceptical relativist maintains that rational validation is simply unattainable. People can maintain various things to be true and factual. But this is mere opinion— merely a reflection of how things seem to them. All judgement is merely so much opinion, all knowledge is simply purported (or imagined) knowledge. No rationally secure results are achievable
138
Conceptual Egocentrism
at all. Falsity prevails on all sides. Truth and fact are illusions that lie entirely in the eyes of the beholder. We arrive at a posture of an irrationalism that negates any prospect of a cogent, rational warrant for the acceptance of theses. Realistically speaking, there are no facts of the matter. There is just your purported truth and my purported truth, but no such thing as the real truth—or at any rate none that is determinable by us. (The sceptic could in principle concede that ontologically there might be such a thing as 'the truth', while yet insisting that epistemically we simply have no way of getting there from here.) 2. Indifferentist relativism (Protagorean relativism). Thissyncretist position sees contentions as tenable only relative to a basis whose adoption is entirely arbitrary. Validation is omnipresent just because it is whatever the assentor happens to take it to be. No viable distinction between what is right and what merely seems right can be maintained. The distinction between truth and error, sense and nonsense, becomes inoperative not because (as with the sceptic) error prevails over truth everywhere, but because everything is true on its own ground. In so far as a radical relativist maintains indifferentism ('It just doesn't matter . . .') or scepticism ('Any position is just as justified as any other—namely, not at all'), his position clearly cannot be secured by his relativism alone. Relativism as such is merely committed to holding that positions vary and are justifiable only from the vantage-point of a basis of some sort. Indifferentism and scepticism would follow from this contention only if one could somehow show that there is nothing to choose between different bases—that any one basis is as good (or bad) as any other. But (1) there is not good objective reason to think that this is so, and (2) if there were good objective reason to think this, then there would indeed be such a thing as a good objective reason to think something, and indifferentism and scepticism would become untenable. Relativism—that is to say, a relativism of a sensible sort—certainly need not engender indifferentism. (Indeed, it is hard to see how it could do so, seeing that we cannot at one and the same time coherently both be convinced basis-egalitarians and maintain the correctness of our own particular convictions.) Even a superficial scrutiny is thus enough to indicate that neither of these egalitarian approaches has much appeal. Any view that precludes us from differentiating between superstition and science,
Conceptual Egocentrism
139
or between uninformed guesswork and expertise, is not particularly attractive. As one contemplates these positions, the prospect of a basis preferentialism becomes increasingly appealing. This unacceptability of egalitarian relativism means that a sensible relativism must take the preferentialist form. In particular, we cannot but maintain the superiority of our own rationally held beliefs over their alternative, since in failing to do so we would cease to maintain their rational appropriateness. But can such a preferentialist relativism be defended?
9.3
THE LOCAL A B S O L U T E N E S S OF OUR OWN S T A N D A R D S OF RATIONALITY
The crucial question is not 'Are there indeed different norms and standards of rationality?', for the answer here is an immediate and emphatic 'yes'. Even as autres temps, autres moeurs, so also, other cultures, other standards. Rather, the salient question is: 'Are we well advised—perhaps even rationally obligated—to see all those various alternative norms and standards as equally appropriate, equally correct?' Must we adopt an egalitarian Principle of Basis Equality: 'AH of the various standards of judgement are equivalent in justification. Ours is on an equal footing with theirs in point of acceptability. It is a matter of indifference which basis we adopt! Any one is every bit as good (or poor) as the next?' 'Must we see all those other alternatives as on a par with our own?' The answer is: 'By no means!' The indifferentist thesis is daring and exciting—but also absurd. For, at this point we must turn relativism against itself by asking: 'Indifferent to whom?' Certainly not to us! For, we have in place our own basis of rational judgement, and it speaks loud and clear on its own behalf. Nor yet by parity of reasoning is ours equally acceptable to them. For whom, then, does that equivalency hold, dear indifferentist? For the world spirit—the disembodied arbiter of impersonal fact? But how can we say so? (And what price indifferentism then?) It lies in the very nature of what rationality is that it is rationally impossible to view two discordant standards of rationality as equal in merit. To be sure, people often say the following sort of thing:
140
Conceptual Egocentrism
[All] those concepts that philosophers have taken to be the most fundamental—whether it be the concept of rationality, truth, reality, sight, the good, or norms—all such concepts must be understood as relative to a specific conceptual scheme, theoretical framework, paradigm, form of life, society, or culture, [since] there is no one single framework or single metalanguage by which we can rationally adjudicate or universally evaluate the competing claims of alternative paradigms.3 But this exciting contention is deeply problematic. For, there indeed is, of course, a single and unique scheme that we can use in making such appraisals and adjudications, namely, our own—the one we actually accept. (Were we to trade it in for anofrier, then of course that other one would automatically become ours.) Only for someone who has no scheme or framework at all—who is located wholly outside the realm of linguistic and epistemic and cognitive commitments—is there an open and uncommitted choice among alternatives. But of course none of us do or can find ourselves in that position. Consider the following set of individually plausible but collectively inconsistent contentions: 1. We are rationally justified in thinking what we think. 2. They (those people of another era or culture or tradition) think differently from ourselves. 3. They are rationally justified in so doing. 4. If somebody else is rationally justified in disagreeing with us, then we cannot be rationally justified in thinking what we think. Here 2-4, taken together, entail not-1. Thus, 1-4 are inconsistent. Something within the group must be abandoned. Given that 2 is, or must be assumed to be, a 'fact of life', three options are open: Abandon 1. Lose all sense of cognitive legitimacy. Become a sceptic. Abandon 3. Become a dogmatist. Deny that, at bottom, those who disagree with us really 'know what they are doing'. Abandon 4. Adopt the stance of a qualified relativism. Accept the idea that while they may well be fully justified in their beliefs, evaluations, and actions relative to their basis of 3
Richard J. Bernstein, Beyond Objectivism and Relativism (Philadelphia, Pa., 1983), 8.
Conceptual Egocentrism
141
judgement, we ourselves are fully (and presumably rightly) justified relative to ours. The best way out of this perplexity is clearly to bite on the bullet by abandoning 4 and taking the stance that while there indeed are different viable positions, our own is—from where we ourselves stand, at any rate—the really appropriate one. For us, our position is decisive and we are rationally constrained—and thus entitled—to deem it so. Not only is this position rationally available to us, it is, in the final analysis, the only one that we can sensibly take. If we did not see our position in this light, then it would not really be our own rationally maintained position, contrary to hypothesis. The basis-egalitarian thunders: 'You have no right to your own particular basis! How can you possibly justify taking it to be superior to others?' But of course we can justify it—many philosophical discussions endeavour to do so, and some do quite well. 'But this justification is something you can only do by reasoning, by reflective and judicious appraisal. And on what basis can this reasoning proceed?' 'Why . . . by our own, of course!' (Who else's would we use?) 'Is this approach not circularly vitiating, self-undermining?' 'By no means! It is inevitable—and desirable.' As our earlier discussion of the self-justification of rationality indicates (see Chapter 3), a rational basis that would not thus support itself would ipso facto stand condemned as inadequate. 'But what makes you think your basis is any better than all those others?' That is a perfectly fair question. We had better have a good story to tell to justify our greater satisfaction with our own epistemic basis. (And of course we do—though this is hardly the place to unfold it.)4 But this validation of cognitive rationality will clearly have to proceed in its own terms. Consider the thesis: Our justification for thinking what we think is exactly on a level with their justification for thinking what they think. And so we are not entitled to deem our position superior to theirs. Despite its plausible appearance, this contention encounters deep difficulties. For, one must press the question: From just what 4 The author has told this story in his own way as regards 'the scientific method' in Methodological Pragmatism (Oxford, 1977).
142
Conceptual Egocentrism
'angle of consideration' is it that characterizations of 'is exactly on a level with' are going to be made? Not from ours surely—for this, after all, is ours precisely because we deem it superior. And by parity of reasoning not from theirs either. (From God's? Well . . . perhaps. But he of course is not a party to the discussion.) The salient point is that we are entitled—indeed, rationally constrained—to see our own criteriological basis of rational judgement as rationally superior to the available alternatives. If we did not take this stance—if we did not deem our cognitive posture effectively optimal—then we could not sensibly see ourselves as rationally justified in adopting it. It would, ipso facto, fail to be our real position, contrary to hypothesis. To be sure, this criteriological egocentrism can and should be tempered by criteriological humility. The wisdom of hindsight and the school of bitter experience teaches us the chastening lesson that our cognitive standards—and the judgements we base on them—are by no means necessarily perfect. All the same, we have no real alternative to using our standards—to doing the very best we can with the means at our disposal. While we have to bear in mind the sobering thought that our best just may not be good enough, we are nevertheless bound to see the standards we have adopted in the pursuit of rationality as superior to the available alternatives and to regard ourselves as rationally entitled to do so here and now. (Future improvement 'from within' can of course be envisioned.) To refrain from making this commitment is simply to opt out of the project of rationality altogether. In the pursuit of rationality we must, as with any other pursuit, begin from where we are. Anti-relativists often feel obliged to embark on a quixotic quest for 'cognitive universals' at the level of substantive beliefs or cognitive procedures that all rational beings share in common. One can in principle contemplate two very different lines of approach here: (1) We find that certain types of creatures have somehow been predesignated as rational (by the world spirit?), and we then inquire empirically (synthetically) into what it is that all of these predeterminedly rational beings have in common. Or again, (2) We make use of our own conception of what rationality is to characterize certain types of creatures as rational, and then ask (analytically) what it is that all of them must have in common simply in virtue of qualifying under this conception of ours.
Conceptual Egocentrism
143
Clearly, the second approach is the only practicable one; it makes no sense to try to implement the first, seeing that we simply have no way to get in touch with the world spirit. However, when we proceed in this second way, the only commonalities we can get out are the ones that we put in. We must ask what features beings must possess in virtue of qualifying as the sort of creature that we ourselves are prepared to accept as answering to our conception of rationality. Clearly, this approach puts that conception of ours at the forefront as the determinative pivot-point. After all, it is our conception of rationality that fixes the 'rules of the game' at issue when we pursue our deliberations about these matters. We have to play the rationality game by our ground rules because it is exactly those ground rules that define and determine what 'the rationality game' is that is at issue in our deliberations. If we were not playing the game on this basis, it would not be the rationality game that we were engaged in—it would not be rationality that is the subject of our concern. It is the determinative role of our own rationality standards that makes them absolute for us. To be sure, what makes our own conceptions authoritatively determinative in such matters is nothing special about us. Clearly, the Aristotelian cosmos is no longer with us—we are not the centre of the world. But we certainly and inevitably are at the centre of our thought world. Our inquiries have to be conducted within our frame of reference. We have to pose our questions in line with our ideas, to frame our perplexities by means of our concepts, to consider our issues in our terminology. If we ourselves are to classify someone as rational at all (and who else's attribution is now at issue?) then we must deem him qualified under the aegis of rationality as we understand it. If we ask about someone 'Is she tall?', we are clearly asking about her height as we conceive it. What she herself thinks about height is beside the point. And exactly the same situation holds with respect to rationality. A condition of 'questioner's prerogative' prevails—it is the person who puts the issue on the table who sets the frame of reference for determining what that issue involves—it is, after all, his question. With all these questions about rationality, it is of course 'rationality' as we ourselves conceive of it that is operative. The topic being ours, it is we who set the terms of reference for
144
Conceptual Egocentrism
what is at issue. At this point 'epistemic relativism' comes to a stop. Paradoxically, it is precisely the inevitable relativization of our questions and concerns and puzzlements to our terms of reference that makes those particular terms of reference absolute in our own discussions. Being framed in our terminology, it is our terminology, that is decisive for the questions that we raise and the inquiries that we conduct. If we ask if X is being rational in believing (or doing or evaluating) a certain thing, then the issue is clearly one of its being rational on the basis of the conception of rationality as we understand it. The governing absoluteness of our conception inheres in 'questioner's prerogative'—in the fact that the questions and issues we address in our deliberations about rationality are in fact our own and that, since the questions are ours, it is our conceptions that are determinative for what is at issue.
9.4
MORE ON C R I T E R I O L O G I C A L E G O C E N T R I S M : THE PRIMACY OF OUR OWN S T A N D A R D S
One recent writer maintains that 'it is necessary to have standards of rational acceptability in order to have a world at all'.5 But that is taking things too far. Cats and dogs certainly have a world, and possibly even a world-picture, while yet lacking standards of rational acceptability. What they do not have is a rationally formed world-picture—but that is something else again. (For this, rational standards are undoubtedly needed—but that, of course, is tautologous and trivial.) The salient point is that one cannot consistently both stake a claim to the rational validity of one's views and at the same time reject all commerce with rational standards and criteria. In this regard, our commitment to our own cognitive position is (or should be) unalloyed. We ourselves are bound to see our own (rationally adopted) standards as superior to the available alternatives—and are, presumably, rationally entitled to do so on the basis of the cognitive values we ourselves endorse. To be sure, there are cognitive bases different from ours— different sorts of standards altogether. But what does that mean 5
Hilary Putnam, Reason, Truth, and History (Camb., 1981), 147.
Conceptual Egocentrism
145
for usl What are we to do about it? Several stances towards those various bases are in theory open to us: 1. 2. 3. 4.
accept none: reject ours accept one: retain ours accept several: conjoin others with ours 'rise above the conflict': say 'a plague on all your houses' and take the path of idealization invoking the 'ideal observer', the 'wise man' of the Stoics, the 'ideally rational agent' of the economists, or the like
The first option is mere petulance—a matter of stalking off in 'foxand-grapes' fashion because we cannot have it all our own preferred way. The third option is unfeasible: different bases do not combine, they make mutually incompatible demands, and in conjoining them we will not get something more comprehensive and complete—we will get a mess. The fourth alternative is Utopian and unrealistic. We have no way to get there from here. Only the second step makes sense: to have the courage of our convictions and stand by our own guns. £mile Durkheim was no doubt right in insisting that 'all that constitutes reason, its principles and categories, is made [by particular societies operating] in the course of history'.6 But the fact that everyone's rational knowledge, standards, and processes are historically and culturally conditioned—our own of course included—does not preclude their binding stringency for those to whom they appertain (ourselves pre-eminently included). In conducting our cognitive, practical, and evaluative affairs in this world—as in conducting our movements within it—we have no choice but to go on from where we are. It is the very fact that we take it to be binding on us that makes an accepted standard our standard. If we are rational, then our standards and criteria of rationality are ours precisely because we deem them to be (not necessarily the best possible, but at any rate) the best available to us. Of course, to see our own standards as the appropriate ones for us to use here and now is not to deny the prospect of a change of standards. But when this happens our then stance toward the new standards is identical with our present stance towards those existing ones. A commitment to the appropriateness of his present 6
This view has become axiomatic for the entire 'sociology of knowledge'.
146
Conceptual Egocentrism
standards follows the rational man about like his own shadow. But are the beliefs of primitive, pre-scientific cultures indeed less rational than ours? A resounding negative is maintained in Peter Winch's widely cited article on 'Understanding a Primitive Society',7 which maintains vividly that Azande beliefs about witchcraft and oracles cannot be rejected as rationally inappropriate despite their clear violation of the evidential canons of modern Western scientific culture. Winch maintains that the Azande can quite 'rationally' see those occult beliefs to be justifiable in their own (deviant) way. But just here lies the problem. The answer you get depends on the question you ask. If we ask 'Do they hold their beliefs rationally?' we, of course, mean 'rationally on our understanding of the matter'. And the answer here is clearly lNo\ because in fact this sort of rationality does not figure in their thinking at all. The fact that they deem their beliefs somehow 'justified' by some considerations or other (which in fact provide no rationally adequate justification at all) is going to cut no ice in our deliberations. The issues that arise at this juncture go back to the quarrel between Evans-Pritchard and Levy-Bruhl. In his book on Primitive Mentality,8 Lucien L6vy-Bruhl maintained that primitive people have a 'pre-logical mentality'. Against this view, E. E. EvansPritchard9 argued that primitive people were perfectly 'logical' all right, but simply used a logic different from ours. When, for example, the Nuer maintain that swamp light is identical with spirit, but deny that spirit is identical with swamp light, they are not being illogical, but simply have in view a logic of 'identity' different from that of the identity claims in vogue in Western cultures. The obvious trouble with this sort of thing is that nothing apart from bafflement and confusion can be achieved by translating Nuer talk by use of our identity language, if what is at issue in their thought and discourse is simply not reflected by our identity conception. Instead of translating the claim at issue as 'Swamp light is identical with Spirit' and then going on to explain that 'is identical with' does not really mean what it says because the 7 Peter Winch, 'Understanding a Primitive Society', American Philosophical Quarterly, 1 (1964), 307-24, repr. in HoIIis and Lukes, op. cit. 8 L6vy-Bruhl, Primitive Mentality (London, 1923; 1st pub. in Fr., Paris, 1921). v Witchcraft, Oracles and Magic Among the Azandi (Oxford, 1937); Nuer Religion (Oxford, 1956).
Conceptual Egocentrism
147
ground rules that govern this idea are not applicable, an anthropologist would do well to paraphrase (if need be) the claim at issue in such a way as to explain what is actually going on. The fact that the Nuer have different (and to us strange-seeming) beliefs about 'spirits' no more means that they have a logic different from ours than the fact that they eat different (and to us strange-seeming) foods means that they have a digestive chemistry different from ours. 10 And so, in discussions about 'alternative modes of rationality' we do indeed have a 'higher standpoint' available to us—namely our own. And this is rationally justified by the consideration that no alternative is open to us—we have to go on from where we are. Accordingly, while one must recognize the reality of alternative cognitive methodologies, one certainly need not see them as equally valid with one's own. 'You have your standards and I have mine. There are alternatives.' But this fact leaves me unaffected. For, I myself have no real choice about it: I must judge matters by my own lights. (Even if I turn to you as a consultant, I must ultimately appraise the acceptability of your recommendations on my own terms.) Still, absolutism can go too far. One recent writer finds a powerful argument against relativism in the circumstance that [w]e must assume that other cultures share some of our standards of rationality if we are to understand their words and works; so we cannot consistently claim that no standards of rationality are universal, even when we find that some cultures do not respect all the standards of rationality that we do." But there is a big problem here. Even if one concedes (as presumably one must) the evidential point that we will not be able to recognize as rational those who do not share many (perhaps, even, most) of our own standards of rationality, this fact does not assure any actual overlap of standards among these cultures themselves. Suppose we ourselves hold standards A, B, and C, while culture no. 1 holds A, B, and D, no. 2 holds A, C, and F, and no. 3 holds B, C, and G. Then, clearly, each culture shares with 10
The relevant issues are interestingly treated in John Kekes's book, A Justification of Rationality (Albany, NY, 1976), 137-49. 11 C. Behan McCullagh, 'The Intelligibility of Cognitive Relativism', The Monist, 67 (July, 1984), 327-40 (see p. 332).
148
Conceptual Egocentrism
each other a majority of its standards, but without there being any one single universal standard that all cultures share alike. The 'necessary overlap' argument simply collapses as a case for the universality of standards. A chastened relativism—namely an axiological one—can accordingly take the stance that, while other positions are 'available', abstractly speaking, one's own responsibly determined position is, for oneself, altogether appropriate and valid. It is a relativism which pays tribute to dogmatism. Our own position is justified (as we see it) in terms of our own standards—which, again, we hold for good reasons. But, of course those standards are 'internal' to our position itself—and are not necessarily validatable from a different vantage-point. Still, this is immaterial for us, since, by hypothesis, we do not have this different vantage-point. Such a relativism leaves no room for indifferentism. It pivots on the idea of contextual appropriateness—appropriateness in the context that is delimited and defined by the specific circumstances of one's situation. We realize (relativistically) that pluralism prevails—that other standards are used by others. But we can (and must) nevertheless accept (absolutistically) our own standards as appropriate for ourselves. To recognize a standard as rationally valid is—where rational agents are at issue—already to have adopted it as one's own. In taking a cognitive position in adopting a certain set of standards and criteria of truth and validity, we assume an evaluative position, and such a position is by its very nature incompatible with the prospect of accepting alternatives, because the holding of an evaluative position consists in a rejection of the rest. Even when conceding the prospect of someone's having another position, we cannot see it as available to ourselves. This chastened version of pluralism is, in effect, a relativistic monism. Relativity obtains, in that different cognitive agents use different standards on a basis that one must regard as appropriate for them. But we ourselves have no alternative but to treat our standards of truth and rationality as rationally decisive because this view of the matter is implicit in their serving as standards for self-proclaimed rational agents. The only standard it makes any sense for us to use is the one we endorse. There is no point in my applying someone else's standard of value or worth or interest or appropriateness or whatever. God's standard I cannot apply, since I do not have it. Your
Conceptual Egocentrism
149
standard I will not apply, if I do not share it. The standard we have got to use is just exactly the standard that we have got. And we have no way to evaluate standards save in our terms. Whose superstandard should we use? God's? Since our expulsion from the Garden of Eden, he is no longer at hand to be asked. Somebody else's? But if we do not subscribe to it, then how can we rationally use it; and if we do, then we have thereby already made it ours! Thus, even while acknowledging that other judgements regarding matters of rationality may exist, we can, do, and must persist in deeming ours to be the best; their superiority according to our own standards is a foregone conclusion. We have no choice but to see our standards as appropriate for us. (In using someone else's with 'no questions asked' we would, ipso facto, be making them ours!) To use another standard (categorically, not hypothetically) is to make it ours—to make it no longer another standard. Like 'now' and 'here', our standards follow us about no matter where we go. These deliberations accordingly lead to a result that might be characterized as criteriological egocentrism. We can and indeed must see our own standards as optimal with respect to the available alternatives. If we did not so see them, they would thereby cease to be our standards. 'But cannot one go out and get another normative standard?' One certainly can. But on what basis would one do this? You might force me to change standards. Or you can, perhaps, brainwash me. But you cannot rationally persuade me. For, rational persuasion at the normative level has to proceed in terms of norms that I accept and, by the norms I actually have, my present standards are bound to prevail, if I am rational in the first place.12 9.5
LIMITS TO COGNITIVE RELATIVISM: METACRITERIAL MONISM
To discern more clearly the limits of relativism in the field of cognition it is useful to consider the following sequence of questions: 12
To be sure, someone could convince me that my understanding of the implications of my standards is incomplete and lead me to an internally motivated revision of my rational proceedings.
ISO
Conceptual Egocentrism
1. What do they, those people of whatever group we are considering, appropriately (or warrantedly) take to be true? 2. What do they appropriately (or sensibly) take to be a good reason for claiming something to be true? 3. What do they appropriately (or justifiedly) take as a standard or criterion for being a good reason for taking something to be true. 4. What do they appropriately (or correctly) take to be the functional metastandard or aim or purpose for appraising standards as criteria of good reasons? With this question sequence, a counter-question recurrently arises: 'Please remove ambiguity—does that 'appropriately' mean by their light or by oursT This issue arises at every step except for the last. Here, at this final point of 'what rationality is all about', we simply have to take our own position as decisive; no alternative is available to us. To qualify those of an alien culture as fully rational we must maintain both that they are playing their game intelligently by their own rules and also that in our sight these rules make good rational sense given their situation. It is ultimately intelligent comportment' and 'making sense' according to our standards of appraisal that makes what is at issue rationally invariant. The fact that we do (and must) apply our own idea of the matter is what makes for the universal element of rationality. What is universal about rationality is not something profound about sociology, but something rather trivial about language use; namely, that to accredit another culture as rational at all is to accept it as being 'rational' in our sense of the term—which may, to be sure, involve deciding whether their actions measure up to their standards. The absoluteness of (ideal) rationality is inherent in the very concept at issue. But is this really so? Could they not have a conception of rationality different from ours? Anthropologists do sometimes say that a certain society has a conception of rationality that is different from ours. But that is literally nonsense. Those others can no more have a conception of rationality that addresses an object different from ours, than they can have a conception of iron that addresses an object different from ours, or a conception of elephants that addresses an object
Conceptual Egocentrism
151
different from ours. If they are to conceive of those particular things at all, then their conception must substantially accord with ours. Iron objects are by definition what we take them to be: 'elephant' is our word and elephant our conception. If you are not talking about that, then you are not talking about elephants at all. You have simply 'changed the subject', and exited from the domain of the discussion. Similarly, if their conception is not close to a conception of what we call rationality, then it just is not a conception of rationality—it does not address the topic that we are discussing when we put the theme of rationality on the agenda. Of course, they may think that what we call pencils are chopsticks and use them as such. Or they may think books to be doorstops and use them as such. But, that does not mean that they conceive pencils or books differently from us, or that they have a different conception of pencils and books. 'They take pencils to be something we do not (namely chopsticks)', is fine as a way of talking. But 'They believe pencils to be chopsticks' is nonsense unless it is glossed as: 'They believe these sorts of things called "pencils" to be chopsticks.' And when this happens, then they do not conceive of pencils differently from ourselves, they just do not conceive of pencils at all. They simply do not have the (one and only) conception of pencils at all—namely, ours. If they do not have our concept, then they do not have the concept. Because that is what 'having the concept is' when it is we who carry on the discussion. What is at issue is our concept and to have it at all is to have it in the way we do. There is no difficulty with the idea that 'They implement and apply the conception of rationality differently from ourselves'. After all, we implement and apply the idea of a medication very differently from the ancient Greeks, using medications they never dreamt of. But the matter stands differently with the conception of a medication. This remains what it always has been: 'a substance used as a remedy for an ailment'. When one ceases to operate with that conception, then {ex hypothesi) one is no longer dealing with medications at all. The discussion has moved on to other topics. To have 'a different conception' of pencils or elephants or rational actions is simply not to have that conception at all. If they do not have our conception of scissors, they do not have a conception of scissors, full stop. For, when we ask about their dealings with scissors it is our own conception that defines the terms of
152
Conceptual Egocentrism
reference. If we recognize agents as rational at all, then we can make sense of what they do! This is not because we are so talented and versatile, nor yet because the slogan 'rationality is universal' gets it right. It is just because we could not say that they are rational (would not characterize the phenomena in this particular way) if we could not make rational sense of what they do. So, it is literally nonsense to say 'The X's have a different conception of rationality from the one we have.' For, if they do not have ours, they do not have any. Whatever analogue or functional equivalent there may be with which they are working, it just is not something that we, in our language, can call 'a conception of rationality'. The pivotal fact is that of 'questioner's prerogative'. Since the question 'What is their mode of rationality?' is ours, so is the 'rationality' that is at issue. Thus, on the crucial issues—'What is rationality all about?', 'What sorts of considerations characterize the rationality at issue?', 'What is appropriately at stake?'—it is our own position that is determinative. When the questions are ours, the concepts that figure in them are ours as well. At this stage of establishing the ultimate ground rules of appropriateness for rationality, it is our own position that is decisive. Consider the contention: Surely there are no historically and culturally invariant principles of rationality. People's (altogether plausible) views about what is rational change with changes in place and time. Yes and no. Of course, different people in different places and times conduct their 'rational' affairs quite differently. But, at the level of basics, of first fundamentals, there is bound to be a uniformity. For, what all modes of 'rationality' have in common is precisely this—that they all qualify as 'modes of rationality' under our conception of the matter. At this level of deliberation, 'questioner's prerogative' prevails, and our own conception of the matter becomes determinative. It is helpful to contemplate some analogies. There are many sorts of blades for knives. But, the fundamental principle that knives have blades at all does not depend on how people choose to make knives but on our conception of what a 'knife' is. If the given objects, whatever they might be, do not have blades then they are not knives. It would, clearly, be the height of folly to go about in
Conceptual Egocentrism
153
another culture asking people 'Must knives have blades?' The answer is a foregone conclusion. A negative response would not counter-indicate the thesis at issue, but would simply betoken a failure to comprehend. Again, there are various quite different sorts of information that people take as evidence to substantiate a claim. But 'deeper' principles like 'Give more credence to that for which the evidence is stronger' or 'Endeavour to expand and extend the evidence for your claims' do not depend on the evidential practices of people, but on our conception of what evidence is all about. (If people do not proceed in ways that conform to these principles of ours, then their practices—whatever they might be—are not evidential practices.) And so, while rationally appropriate knowledge claims and rationally appropriate actions and even criteria of appropriateness vary across times and cultures, the determinative principles of rationality do not. But, this interesting circumstance does not so much reflect a fact about different times and cultures as the fact that what counts as a 'standard of rationality' at all is something that rests with us, because we are the arbiters of the conceptual make-up of an issue within the framework of our discussions of the matter. What sort of thing we ourselves understand by 'rationality', becomes determinative for our own discussions of the matter. People can (appropriately) differ about what is rational—about what the appropriate resolution to a question of rationality is. But, the ultimate principles of rationality remain uniform and invariant throughout. And this uniform conception of 'what rationality is' suffices to establish and render uniform those top-level, metacriteriological standards by whose means each of us can judge the rationality of another's resolutions relative to that other's own basis of appraisal. For, those 'deeper principles' of reason are inherent in the very conception of what is at issue. If you 'violate' certain sorts of rules then—for merely conceptual reasons—you simply are not engaged in the evidential enterprise at all. The most basic principles of knifehood or evidence or rationality are 'culture dependent' only in the sense that some cultures may not pursue a particular project (the cutting project, the evidence project, the rationality project) at all. It is not that they can pursue it in a different way—that they have learned how to make knives without blades, to evidentiate without grounds, or make rational deliber-
154
Conceptual Egocentrism
ations without subscription to the fundamental principles we take as definitive of what rationality is all about. This ultimate conceptual uniformity makes it possible to criticize the proceedings of others from a rational point of view. And, no less importantly, this uniformity alone makes it possible to render advice—not in the mode of 'Here is the rational thing for anybody to do', but in the mode of 'Here is the rational thing for someone in your shoes to do'. For, rational advice must operate not from the advisor's own personal standpoint, but from the client's point of view (not necessarily, to be sure, with reference to what he wants, but with reference to his best interests). Relativity ends where charity begins—at home. For, our discourse is governed by our conceptions which are absolute at any rate for us. It is care for the concepts involved which chastens the impetus to relativism with absolutistic constraints. When social scientists say that alien cultures have a different 'rationality' from ourselves what they generally mean (strictly speaking) is (1) that they have different objectives (for example, that we seek to control and change our environment to suit our purposes, while they tend to reconstitute their purposes to suit their environment—to endeavour to come into 'harmony' with nature), and/or (2) that they use problem-solving techniques which are different from ours (for example, that we employ empirical investigation, evidence, science, while they use divination, omens, or oracles). But, if they pursue different sorts of ends by different sorts of means they, perhaps, have a different thought style and a different intellectual ethos, but not a different rationality. The anthropologists' talk of different rationality is simply an overly dramatic (and also misleading) way of making a valid point— namely, that they do their intellectual problem-solving business in a way different from ours. But, those different processes of theirs do not mean that they have a different rationality any more than those blow-guns of theirs mean that they have a different rifle. But how, short of megalomania, can one take the stance that one's own view of what is rational is right—that it ought to be binding on everyone? How can I maintain this agreement between my position and that of 'all sensible people'? Not, surely, because I seek to impose my standard on them, but because I do—or should!—endeavour to take account of their standards in the course of shaping my own. Co-ordination is achieved not because I
Conceptual Egocentrism
155
insist on their conforming to me, but because I have made every reasonable effort to make mine only that which (as best I can tell) ought to be everyone's. The issue is one not of domineering but of submissive conformity. In the end, I can thus insist that they should use the same standard that I do because it is on this very basis of a commitment to commonality that I have made that standard my own in the first place. The conformity of rational standards is—or ought to be—produced not by megalomania but by humility. The normatively guided absoluteness of rationality itself also accounts for the absoluteness of truth and of good reasons. The inference at issue runs: Ideal rationality is monolithically absolute. The real truth is what ideal rationality endorses. Therefore: The real truth is monolithically absolute. Truth and good reasons per se are idealizations—rendered such by the very nature of the concepts at issue. And this fact provides for a certain absolutism. What people do in these regards is unquestionably relativistic—being relativized to the bases on which they operate. But what they ought ideally to think is something absolute in virtue of the very nature of the idealization at issue. The normative absolutism does not, of course, extend to the practical politics of our actual doings. The real truth may be something absolute, but, obviously enough, our putative truths are emphatically not so. Still, our (defining) conception of rationality is geared to these idealizations. A sensible cognitive relativism, which is to say a duly designed evidential relativism, accordingly does not imply that one must: 1. abandon the truth (scepticism); 2. fragment truth by adopting a doctrine of truth pluralism (Averroism); 3. remain indifferent as between different truth criteria (indifferentism); 4. abandon rationality altogether (irrationalism). A sensible relativism will recognize that we can (quite appropriately) disagree about what it is that is true and what good reasons are at hand, while yet maintaining an (appropriately) absolutistic view of what truth and good reasons are. The ideal nature of actual truth and of actual good reasons that inhere in our (defining) conception
156
Conceptual Egocentrism
of inquiry establishes a clear limit to the implications of cognitive relativism.13 To re-emphasize: A cognitive relativism of potential basis-diversity is altogether compatible with an absolutistic commitment to our own basis. One can combine a relativistic pluralism of possible alternatives with an absolutistic position regarding ideal rationality and a firm and reasoned commitment to the standards intrinsic to one's own position. We ourselves are bound to see our own (rationally adopted) standards as superior to the available alternatives—and are, presumably, rationally entitled to do so in the light of the cognitive values we ourselves endorse. Relativism thus has its limits. The implications of our own conception of rationality are absolutely decisive for our deliberations. We ourselves must be the arbiters of tenability when the discussion at issue is one that we are conducting. And so, we cannot at once maintain our own rational commitments as such while yet ceasing to regard them as results at which all rational inquirers who proceed appropriately ought also to arrive if the circumstances were the same. In this sort of way, the claims of rationality are inherently universal. 13 Compare Hilary Putnam, 'Why Can't Reason be Naturalized?' Synthase, 52 (1982), 3-23 (see pp. 12-13).
10 The Universality of the Rational
(1) What is rational for one person to do, to believe, or to value is also of necessity equally rational for anyone else in the same circumstances. Rationality is universal, but circumstantially universal. This circumstantiality of reason means that people can legitimately differ in their verdicts of rationality, and even to some extent in their views about rationality itself. All the same, the fundamental principles of reason are universal—inherent in the characteristic aims that define the rational enterprise as such. (2) There is no real paradox here. The relativism of good reasons can be reconciled with the universality of rationality itself by taking a hierarchical view of the process through which the absolutistic conception of ideal rationality is brought to bear on the resolution of concrete cases and particular situations. (3) Some theorists equate rationality with consensuality. But, it is highly problematic to argue: 'Consensus has not been achieved, therefore consensus cannot be achieved.' We must distinguish between an epistemically grounded, cognitive dissensus, based on a relativism of evidence, which is altogether harmless, and an objectivity-destroying, substantive dissensus, that removes any possibility of rational deliberation. (4) Rationality, of whatever sort, pivots on the availability of good reasons. And this matter of good reasons is not something subjective or idiosyncratic; it is objective and lies in the public domain—although not, to be sure, in a way that constrains consensus. Objectivity is thus essential to rationality, but consensuality is not. Consensus hinges on what people think, while objectivity hinges on what they ought to think, and the two coincide only in the ideal case. SYNOPSIS
1 0 . 1 THE U N I V E R S A L I T Y OF REASON
Be it in cognitive, practical, or evaluative matters, rationality has two distinguishable but inseparable aspects, the one personal, private, and particular, the other impersonal, public, and universal. The private (particularized) aspect turns on what is advisable for the agent, duly considering his own personal situation in point of
158
The Universality of the Rational
his circumstances—his opportunities, capabilities, talents, objectives, aspirations, values, needs, and wants. (Note that we here construe 'circumstances' very broadly, including not only the outer and situational, but also the inner conditions that relate to a person's physical and psychological make-up.) The universalized aspect of rationality turns on its being advisable by personindifferent and objectively cogent standards for anyone in those circumstances to do the 'rationally appropriate' things at issue. The standards of rational cogency are general in the sense that what is rational for one person is also rational for anyone else in his shoes. Both aspects, the situational and the universal, are inseparable facets of rationality as we standardly conceive it. For a belief, action, or evaluation to qualify as rational, the agent must (in theory at least) be in a position to 'give an account' of it on whose basis others can see that 'it is only right and proper' for him to resolve the issue in that way. An intelligent, detached observer, apprised of the facts of the case, must be in a position to say: 'While I myself do not believe or value these things, I can see that it is appropriate that someone in the agent's circumstances should do so, and in consequence it was altogether sensible for the agent to have proceeded as he did.' It lies in the very meaning of the concept of rationality that if something is indeed 'the rational thing to do' then it must be possible in principle for anyone to recognize the rational sense of it once enough information is secured. Rational belief, action, and evaluation are possibly only in situations where there are cogent grounds (and not just compelling personal motives) for what one does. And the cogency of grounds is a matter of objective standards.The idea of rationality is in principle inapplicable where one is at liberty to make up one's rules as one goes along—to have no predetermined rules at all. The dictates of rationality are objective. If something makes good rational sense, it must be possible in principle for anyone and everyone to see that this is so. This matter of good reasons is not something subjective or idiosyncratic; it is objective and lies in the public domain. Both the appropriateness of ends (for a person of particular make-up, talents, tastes, and the like) and the suitability of particular means for pursuing those particular ends pose objective issues that are open to others every bit as much as to the agent himself. Indeed, with respect both to someone's needs and
The Universality of the Rational
159
to his best interests, other informed people (his doctor, his lawyer, his tax advisor, and so on) may well be in a position to make better and wiser—that is, more rational—judgements than the individual himself. Robinson Crusoe may well act in a perfectly rational way. But, he can only do so by doing what would make sense for others in similar circumstances. He must in principle be in a position to persuade others to adopt his course of action by an appeal to general principles to show them that his actions were appropriate in the circumstances. Rationality is thus something inherently general in its operations. To be sure, the circumstantiality of reason makes for an unavoidable aspect of person-relativity. Rational resolutions are indeed universal, but only circumstantially universal in a way that makes room for the variation of times, places, and the thousands of details of each individual and situation. What it was rational for Galen to believe—given the cognitive 'state of the art' of his day regarding medical matters—is in general no longer rational for us to believe today. The routines of training and practice that a young 'natural athlete' can rationally set for himself may not make sense for a young cripple or an active septuagenarian. Obviously, what it is rational for someone to do or to think hinges on the particular details of how he is circumstanced—and the prevailing circumstances of course differ from person to person and group to group. The rulings of rationality are indeed subject to person-relativity— but a person-relativity as regards objectively determinable circumstances. Consider an example. I am hungry; I go to the restaurant; I order a meal. Have I acted rationally? Of course. But why exactly? Well—because a long story can correctly be told about what I have done, a story in which all of the following play a significant role: My well-evidentiated beliefs that eating food alleviates hunger and that restaurants provide food; my sensible preference for the comfort of satiation over the discomfort of hunger; my custom of doing what I effectively can to alleviate discomfort. The whole chain: 'alleviate discomfort—proceed to secure food—go to a food supplier—order food' is part and parcel of the rationalitydictated rationale of my action. If the chain were severed at any point (if, for example, I realized that the restaurant had run out of food last week), then my action (ordering that meal) would cease to be rational in the circumstances.
160
The Universality of the Rational
One only proceeds rationally when what one does at each step is a particular instance 'covered' by a universal principle of rationality that holds good generally and for everyone. I study the menu and order steak. Was it rational of me to do so? Of course—because I was hungry, came to eat something at the restaurant, and found steak to be the most appealing entry on the menu. On due reflection, it could be (quite appropriately) said that I proceeded on the principle 'Presented with various options for food (and other things being equal), select that which one deems the tastiest.' (To be sure, other things may not be equal—my choice of beef might deeply offend my dinner guest, who deems cattle sacred.) Here we have a strictly universal principle—one that it makes perfectly good rational sense for anyone to act on. Though clearly not every sensible person would order steak, I nevertheless could be said to have done—under the aegis of the indicated principle—something that any sensible person would do. Similarly, any rational choice must be 'covered' by a universally valid desideratum. It must implement, in its particular context, a principle that is of strictly universal validity—although, to be sure, one that is of a conditional nature. Again, some things we desire for ourselves ('Mary as a wife'), others we see as universal desiderata that hold generally and for everyone ('having a good spouse if married'). Now, the crucial fact is that a personal want or preference qualifies as rational only in so far as it can be 'covered' by something that is an unrestrictedly universal desideratum (all else being equal). Only in so far as I am convinced that Mary will prove to be an instance of something that everyone can acknowledge as desirable (having one's marriage partner be 'a good spouse', 'a caring helpmeet', 'a desirable mate') will my own desire to have her for a wife be a rational one. Only those acts which instantiate in this way something that deserves the rather pompous title of a 'universal principle of reason' can qualify as rational. It is not 'being the last to cross the bridge to safety before its collapse' but 'managing to cross the bridge safely' that would be rationally advisable for anyone and everyone to opt for in relevant circumstances.1 Only those acts whose salient characterization is universally rational are rational at all. The ground of the universality of reason is not far to seek. It is 1 Cf. Derek Parfit, 'Providence, Morality, and Prisoner's Dilemma', Proceedings of the British Academy, 65 (1979), 555ff.
The Universality of the Rational
161
rooted in the nature of interests. Something can only be in my (real) interests by being of a generic type that is in everyone's (real) interests. It is in my interest to take a particular medicine because it is generally in anyone's and everyone's interest to care for their health. Any valid interest—any that merits the acknowledgement of reason—must inhere in a universal interest (as the validity of an interest in tennis rooted in a generic need for exercise or skill development). The contention 'What's rational for you need not be right for me' is certainly correct—within limits. Consider the medical analogy. You might do well to eat chocolate to provide the calories needed for the strenuous outdoor life you lead; for me, with my diabetes,it would be a very bad thing indeed. And so, low-level recommendations like 'Eat chocolate' indeed fall into the range of the juststated dictum. But with 'Eat the foods conducive to maintaining your health' the matter stands very differently. At this level there is no variability. What is right and proper here is right and proper for everybody. And, similarly, at the higher level of governing principles, rationality is absolute and universal. The uniformity of overarching rational principles transcends the variability of their particular cultural implementations. Different cultures do indeed implement a rational principle like 'Be in a position to substantiate your claims' very differently. (For one thing, there are different standards as to what constitutes proper 'substantiation'.) But, they cannot simply abandon it. If they convert to 'It's all right to maintain anything that suits your fancy' they do not have a different mode of cognitive rationality but rather, in this respect at any rate, are simply deficient in cognitive rationality. Anthropologists, and even, alas, philosophers, often say things like 'The Wazonga tribe deem it rationally appropriate (or even mandatory) to attribute human illness to the intervention of the rock-spirits.'2 But there are big problems here; this way of talking betokens lamentably loose thinking. For, compare: 1. The Wazonga habitually (customarily) attribute . . . 2. The Wazonga think it acceptable (or perhaps even necessary) to attribute . . . 3. The Wazonga think it rationally mandatory to attribute . . . 2 On 'alternative standards of rationality' see Peter Winch, 'Understanding a Primitive Society', American Philosophical Quarterly, 1 (1964), 307-24.
162
The Universality of the Rational
Now, however true and incontestable the first two contentions may be, the third is untenable. For compare (3) with 4. The Wazonga think it mathematically true that dogs have tails. No matter how firmly convinced the Wazonga may be that dogs have tails, thesis 4 taken as a whole remains a thesis of ours, and not of theirs! Accordingly, it is in deep difficulty unless the (highly implausible) condition is realized that the Wazonga have an essentially correct conception of what is, and, moreover, are convinced that the claim that dogs have tails belongs among the appropriate contentions of this particular realm. Analogously, one cannot appropriately maintain 3 unless one is prepared to claim both that the Wazonga have an essentially correct conception of what rationality is (correct, that is, by our lights), and furthermore that they are convinced that the practice in question is acceptable within the framework of this rationality project. And this concatenation is highly implausible in the circumstances. The anthropological route to a relativism of rationality, is, to say the least, highly problematic. There is no difficulty whatever about the idea of different belief systems, but the idea of different rationalities faces insuperable difficulties. The case is much like that of saying that the tribe whose counting practices are based on the sequence: 'one, two, many' has a different arithmetic from ourselves. To do anything like justice to the facts one would have to say that they do not have arithmetic at all—but just a peculiar and very rudimentary way of counting. And similarly with the Wazonga. On the given evidence, they do not have a different concept of rationality, but, rather, their culture has not developed to the stage where they have any conception of rationality whatsoever. To be sure, the question 'What is the rational thing to believe or to do?' must receive the indecisive answer: 'That depends.' It depends on context and situation—on conditions and circumstances. At the level of the question 'What is rational; what is it that should be believed or done?' a many-sided and pluralistic response is called for. The way in which people proceed to give a rational justification of something—be it a belief, action, or evaluation—is unquestionably variable and culture relative. We mortal men cannot speak with the tongues of angels. The means by which we
The Universality of the Rational
163
pursue our ends in the setting of any major project-—be it rationality, morality, communication, or nourishment—are 'culture dependent' and 'context variable'. Nevertheless, those projects themselves—in terms of the objectives and ideals that define them and of the basic principles that implement these objectives and ideals—have a uniform and universal validity. Greek medicine is something very different from modern medicine. But, the aims of the enterprise—'the maintenance of health', 'the relief of distressing symptoms', and the like—are similar throughout. It is, after all, these aims that define the issue; they indicate that it is medicine we are talking about rather than basket weaving. And this is so with rationality itself as well. Rationality is, after all, a definite sort of enterprise with a characteristic goal structure of its own—the pursuit of appropriate ends by appropriate means. Its defining principles make for an inevitable uniformity.
1 0 . 2 CULTIVATION HIERARCHIES
But how can the absolutistic universality of the denning principles of rationality, themselves rooted in the monolithic uniformity of 'what rationality is', be reconciled with the pluralistic diversity of appropriate answers to the question: 'What is it rational to do?' The answer lies in the fact that various intermediate levels, or strata, of consideration separate these 'basic principles of rationality' from concrete decisions about what it is rational to do. The tabulation of Display 10.1 depicts this descending hierarchy of principles, norms and standards, rules and (finally) rulings, which comprises the structure of rationale development. And it should be stressed that all of these stages of rationale development appertain equally to rational belief, action, and evaluation. There is a distinctive hierarchy of levels throughout. At the top of the hierarchy, the defining principles of rationality specify the ultimate aims of the enterprise. They explicate what is at issue: the giving of good reasons for what we do, the telling of a reasonable story {Koyov SiSdrai, rationem reddere). The characteristic mission of rationality is that of providing an account of our dealings, of committing ourselves in the context of our affairs to 'making sense', of rendering our dealings intelligible, of conducting our
164 DISPLAY
The Universality of the Rational 10.1 Stratification Levels of Principles of Rationality
1. Defining principles of rationality. The basic principles that determine the nature of the enterprise and specify what rationality is all about. (For cognitive rationality, for example, the pursuit of truth and the achievement of correct answers to our questions: 'The truth, the whole truth, and nothing but the truth.') These principles provide our criteria for assessing the acceptability and adequacy of rational norms and standards of rational procedure. 2. Governing norms and standards of rationality. Standards for appraising the 'rules of the game' governing the rational transaction of affairs. (For cognitive rationality these norms are afforded by desiderata such as coherence, consistency, simplicity, and the like.) These norms provide our criteria for assessing the acceptability and adequacy of our rules of rational procedure. 3. Rules of rational procedure. Rules for the rational resolution of choices. (In the cognitive case, rules like modus ponens in deductive inference or trend extrapolation in inductive inference.) These rules constitute our criteria for assessing the rational acceptability and adequacy of particular resolutions. 4. Rationally warranted rulings. Resolutions with respect to particular issues arising in particular concrete cases, such as: 'Do (or accept) X in the existing circumstances.' affairs intelligently. At the next level down, the governing norms and standards are our yardsticks of rational procedure: basic principles of logic, canons of inductive reasoning, standards of evidence, and the like, which already admit of some variation. Then, descending further, we encounter the 'rules of the game' that specify the procedures through which we implement ends and objectives of the enterprise in the concrete context of particular cases. Finally, at the bottom level, come the specific resolutions for particular cases achieved through the subsumption of concrete cases under the rules. These last clearly vary most of all. Such a 'cultivation hierarchy' (as we shall call it), characterizes any purposively oriented human endeavour. It takes the format: ultimate 'finalities': the ruling aims of the enterprise (governing principles)
The Universality of the Rational
165
implementing policies (guiding norms and standards; basic values and desiderata) methods of procedure (operating rules) specific rulings The top-level purpose is itself 'ultimate'; it defines and specifies what is at issue in the venture under consideration, the concerns that make it the sort of project it is (be it science or carpentry). The subsequent descending levels each address the matter of implementing the previously fixed aims and objectives. Justification at each subordinate level is thus purposive and turns on questions of efficiency and effectiveness in serving the needs of the next, higher level. There is a step-by-step descent from finalities (the characteristic aims inherent in the very definition of valid, need-meeting enterprises) through norms to rules and eventually to specific rulings. It is helpful to consider an analogy, turning from cognitive rationality to medicine: 1. Finalities (defining principles): 'Maintaining health', 'curing illness and disease', 'restoring and maintaining normal bodily functioning', 'removing painful symptoms'. (Note that if these things are not at issue, then medicine is not at issue. An enterprise not concerned with any of these, whatever it may be, is not medicine.) 2. Governing norms, standards, and criteria: 'How is one to assess 'health'? 'How is one to construe "normality"?' 'How is one to identify a "symptom"?' 'Just what constitutes an "illness"?' (Note that for the Greeks, unlike ourselves, the idea of an illness without subject-experienced symptoms was scarcely conceivable. At this level there is already some room for variation.) 3. Rules or procedures: the modus operandi of medical practices —surgery or chiropractic treatment, drugs or psychotherapy, and the like. (These of course differ drastically from age to age and culture to culture.) 4. Rationally warranted rulings: the specific interventions, prescriptions, and medical measures adopted in particular cases. (Take two aspirin and get some rest.') At the top level there is a fixity and uniformity based in
166
The Universality of the Rational
conceptual constraints inherent in the very definition of the nature of the enterprise. But, uniformity is achieved here at the price of an abstractness and generality that endows the principles with a conditional or hypothetical character. As we move downwards towards the level of particular cases, the situation is increasingly one of concrete detail, and this detail brings increasing scope for variation in its wake. For, while the top level is itself absolute and constant, there is 'slack' at each step down the ladder, leaving (appropriate) room for an increasingly large element of variability and differentiation. At each step there is some degree of underdetermination—scope for diversity and some degree of contextual variability. (In the cognitive case, variability arises with such issues as: What sorts of rules best implement the demands for cogent deductive and inductive reasoning? What sorts of solutions do schematic rules like 'Adapt theories to the data as well as possible' lead to?) At every step down the hierarchic ladder there is a further looseness of fit that provides for the adaptation of general principles to the specific characteristics of particular settings and circumstances. If there is not a thread of continuity—if a particular measure is not part of a long story that leads from our concrete choices all the way up to that fixed and stable top level of the principles that define 'the aims of the enterprise'—then it is really not a genuine medical measure that is in view. Throughout, justification of one's proceedings at lower levels involves an appeal to the higher. But the highest level is final in its defining role for the overall enterprise at issue. (After all, cognitive inquiry, like househunting, is a definite sort of business, different in its ideological nature from ventures like medicine or butterfly collecting.) For a particular resolution to qualify as rationally valid, the entire ascending chain of subordination that links it to those topmost, superordinated principles of reason must thus be appropriately validated. The whole rationale developed in terms of such an implementation hierarchy must be in good order if the particular ground-level decision or ruling at issue is actually to qualify as rationally appropriate. The rationale of a given ruling must be cogent 'all the way up' for the ruling itself to be rationally cogent. Such a 'cultivation hierarchy' of rationale-furnishing levels always plays a crucial role in the rational legitimation of what we
The Universality of the Rational
167
3
do. At its pinnacle there is some rationally valid (appropriately interest-serving) desideratum like health (or rationality itself) to furnish the 'ultimate' pivot point, but moving down the line we encounter the increasingly more concrete factors of rationalization, until ultimately we arrive at specific determinations about concrete items. Here, there is increasing room for context-supplied variation and dissensus. The bearing of that fixed, topmost—universally valid—desideratum transmits itself all down the line, albeit with ever-increasing qualifications. In the medical case, for example, we get: 1. maintaining health 2. maintaining health through nourishment (eating) 3. getting nourishment by eating healthy foods that one also happens to like Observe, however, that all of these are also universally appropriate modes of operation: Doing the things involved is rational for everybody when one proceeds down the ladder by appropriate steps.4 But, this universality becomes increasingly qualified in its conditions of application. (Clearly, not everyone happens to like meat in general or steak in particular.) And so, there is much room for variation in the concrete implementation of universals. As regards 2, different things are nourishing for different people, given their particular biomedical make-up. And as regards 3 it is clear that different people like different things. And so, we pursue common projects by person-differential means. But there can be no rationality without universality. The overall account (or 'rationale') that establishes the rationality of what we do (in action, belief, or evaluation) must be one in which universal needs and universalizeable standards play the ultimate determinative role. In being rational, we pursue universal desiderata in persondifferentialist ways—ways that we have good reason to deem effective in the peculiar conditions of our particular case. Not all of us eat what Tom does. But we can, all of us, (1) explain and 3 Which is not, of course, to say that other, 'higher' considerations may not countervail those at issue—that it may not make rational sense for me to endanger my health to save the life of another. 4 Note that it is not homogeneously a means-end hierarchy. (Steak is not a means to food, it is a kind of food.)
168
The Universality of the Rational
understand his eating kumquats once we realize that he happens to like them and (responsibly) believes them to be both hungerremoving and healthful, and (2) agree that the modus operandi involved in his case ('eating what one likes and responsibly believes to be nourishing') is one to which we ourselves do (and should) subscribe. Rationality is a matter of pursuing valid (and universally appropriate) desiderata as ends by appropriate means (but means that are individually appropriate and adjusted to the circumstances of one's personal situation). Such a perspective makes it clear that a uniformitarian absolutism at the top level of 'what rationality is'5 is perfectly consonant with a pluralism and relativism at the ground level of concrete resolutions regarding 'what is rational' in particular cases. The ruling principles of rationality never uniquely constrain their more specific implementations. At each step along the way we repeat the same basic situation: delimitation, yes; determination, no. The sought-for reconciliation between the universalistic absoluteness of rationality and the variability and relativity of its particular rulings is thus provided by the consideration that the absolutism of principles operates at the highest level of the hierarchy of rationale development, while there is ever more 'slack' and variability as one moves towards the lowest level of concrete determinations. The variability and relativity of good reasons at the level of our actual operations can indeed be reconciled with the absolutism of rationality itself by taking a hierarchical view of the process through which the absolutistic conception of ideal rationality is brought to bear on the resolution of concrete cases and particular situations.6 And so, indifferentism does not follow from pluralism. It may be a contingent matter that it is English that is at my disposal, rather than Polish. But that does not make it irrelevant how I go about using the language. It does not mean that there are no norms and standards—that I can throw words together any way I please. And it certainly does not mean that there are no higher, determinative, 5
Note that this comes to 'what rationality as we understand it is'. Note, too, that different top-level finalities can lead to priority conflicts through competing demands on resources. Health and knowledge, or family life and professional life, for example, may certainly conflict—not, to be sure, as abstract desiderata but in the competing demands that arise in the course of their practical implementation. In so far as such conflicts are rationally resolvable at all, still other finalities must be involved as arbiters. Resolutions can, in principle, always be accomplished in distinct yet still appropriate ways. 6
The Universality of the Rational
169
time- and place-indifferent principles that define and'delineate the process of communication as such. Irrationalism does not follow either. The fact that one can only implement the call to rationality in the particular way that one's historical context puts at one's disposal does not mean that rationality as such is something so protean and variable as to lose any meaningful identity altogether. People must feed themselves and shelter themselves—what is at issue here are universal human needs. But nature does not dictate any one single process for meeting them—we must proceed to make the best use we can of the materials that the conditions of place and time put at our disposal. And the same holds good for reason. We need to build a cognitive home for ourselves in this world—to create a viable thought structure for our beliefs, choices, and evaluations. Here too one must simply do the best one can. Neither the project nor its implementation is irrelevant, immaterial, or indifferent. A sensible relativism of situational variability is certainly not at odds with the fundamental and altogether absolute demand of rationality: that we pursue our ends intelligently—that we do the best we can with the limited means at our disposal in the restrictive circumstances in which we labour.
1 0 . 3 OBJECTIVITY AND CONSENSUS
Some theorists equate rationality with consensuality, deeming the prospect of attaining a consensus among all rational minds to be an indispensable requisite of reason. But there are deep problems here—not least in the cognitive case, where the idea that the lack of consensus undermines factuality is very questionable. As has been stressed, in characterizing something as rational we are making a claim whose bearing is universal. But it is perfectly conceivable that there might not be a universal consensus about it. No matter how plausible a contention on any significant issue— rationality included—may be, the prospect that some will (quite defensibly) dissent from it is ever present. The validity of judgements of rationality is not destroyed by finding that there are some who dissent from them. The abstract thesis that 'Other things being equal, all rational people choose recognizedly more effective problem-resolutions over less effective ones' is quite
170
The Universality of the Rational
correct—but of course people can (quite defensibly) disagree about what sorts of measures are effective. Failure of consensus is often used as a basis for impugning reason. The dismissive relativist argues: no consensus —> no objectivity His position is predicated on the principle that 'if different people can (justifiedly) think differently about some issue, then there just is no objective fact of the matter with regard to it'. He takes the line that where there is no relatively enforceable consensus—so that if two people disagree, then one of them is necessarily being unreasonable—then the issue is not an objective one: there is no 'fact of the matter'; the issue is one of mere opinion, of arbitrary decision, or a-rational (if not outright irrational) allegiance. This attribution of dire implications to unavailable consensus dates back to the teachings of the sceptics of classical antiquity—to the last of the ten tropes, the arguments for scepticism inventoried in classical antiquity by Sextus Empiricus in his Pyrrhonian Hypotheses.7 This argument pivots on the variation across the range of man's culturally diverse views in matters of custom, manners, laws, and above all beliefs. Throughout the sceptical tradition, this variation of customs has been invoked to support a deconstructionism that takes a lack of consensus to betoken an absence of the objective factuality needed for meaningful deliberation. The prospect of different constructions, different interpretations or opinions, is taken to annihilate the matter in view as an objective issue. The vitiating flaw of this position is that its salient mediating premiss is completely unjustified. To validate a move from the premiss 'People do not agree about Jf to the conclusion 'X does not represent a genuine issue of objective fact', what is required is clearly a thesis to this effect: If X is an authentic factual issue, then people will, of necessity, ultimately come to agree about it. But the closer one looks at this thesis the less plausible it appears. Where is the Moses who has come down from the mountain with a stone-graven guarantee that whenever there indeed are facts we 7
Seebk. 1, 145ff.
The Universality of the Rational
111
imperfect mortals can come to discern them with an accuracy that precludes disagreement? With many kinds of clearly objective issues, we can readily account for failures to achieve consensus in ways that do not at all countervail the existence of facts of the matter. And this is so, in particular, throughout the domain of rationality. In the case of factual information, the circumstance of evidential diversity, obtaining because different people have different bodies of evidence at their disposal, makes for a perfectly warranted— indeed rationally mandated—differentiation of belief. In the case of evaluation, the circumstance of experiential diversity (the fact that different people have different bodies of experience at their disposal) makes for a perfectly warranted—indeed rationally mandated—differentiation of appraisal. And the same holds for judgements about the appropriateness of actions. Sceptics through the ages have always failed to note how little actually follows from a lack of consensus. For one thing, one cannot argue: 'Consensus has not been achieved; therefore, consensus cannot be achieved.' More seriously yet, one cannot even argue: 'Consensus cannot be achieved by us humans, therefore, there is no truth or fact of the matter.' In the cognitive case, for example, there are clearly different defensible answers to questions like 'What sort of person was Napoleon?' or 'What motivated Caesar's decision to cross the Rubicon?' But de-objectification certainly does not follow—our inability to reach consensus in empirical inquiry certainly does not entail that there is no sort of objective 'fact of the matter' at issue, and that any set of opinions is as good as any other. Consider, moreover, the situation in the natural sciences. The history of science is a story of changes of mind. Where are the 'scientific truths' of yester-year—the absolute space of the Newtonian era, the luminiferous aether of turn-of-the-century physics? And there is no good reason to think this process will come to a stop. The scientists of the year 3,000 will think our theories every bit as inadequate as we deem those of our predecessors of 300 years ago. There is no reason to think that progress will ever come to a stop and that we will attain a definitive 'final solution', a definitive consensus enduring across the generations. But, this is no ground for seeing science as a misconceived activity dealing with non-factual matters.
172
The Universality of the Rational
The truth as such hinges on what the objective facts of the matter are. Consensus in the community of inquirers, on the other hand, is a matter of human doings and dealings—a question of how people think about things on the basis of the evidence at their disposal. It is clearly problematic to contend that whenever there is a fact of the matter we are bound to acquire enough of the right sort of evidence to find it out. The argument 'No consensus, therefore no objectivity' is thus deeply flawed. Dismissive relativism is predicated on the naive optimism of the notion that 'If we can get there at all, we can get there easily—certainly quickly enough to have arrived there by now.' Such a view is naive in its quest for the easy answer; its unwillingness to be patient in developing the complex distinctions needed for a reasonable and sensible appraisal of the implications of an absence of consensus. (Perhaps too it commits the fallacious slide from 'No facts, no consensus' to 'No consensus, no facts'.) No doubt, disagreement will (and should) engender caution, undermining any facile confidence that we have actually got it right. But there are no adequate grounds for construing disagreement—even stubbornly enduring disagreement—to mean that there just is no objective fact to disagree about. There is no reason to reject the idea that while in various areas of inquiry and deliberation we presently lack and indeed may never achieve consensus, we nevertheless operate in perfectly legitimate domains of deliberation where the idea of rational appropriateness applies.
1 0 . 4 OBJECTIVITY AND COGNITIVE CONSENSUALITY CONTRASTED: THE IMPORT OF THE IDEAL
Consensus among rational inquirers across the divides of time and culture is not something that often (let alone necessarily) happens in the real world. The link of consensus to factuality can only be established tightly at the level of idealization—only ideal or perfectly competent inquirers need reach a consensus on factual matters. In this imperfect sublunary dispensation, consensus is too much to expect—or to ask for. The sort of consensuality at issue with objectivity is a matter of ideal circumstances. It is not something on which we should insist here and now. Objectivity stands co-ordinate not with actual but with Meal consensus.
The Universality of the Rational
173
(Actual consensus/dissensus can be no more than a matter of providing relatively weak evidence for or against objectivity.) There is something deeply problematic about C. S. Peirce's view that if there is a real truth of the matter—if there is a fact of the matter at all—then continuing inquiry must eventually find or at least approximate it. And the idea of a transcendental argument— in the manner of Juergen Habermas—to an 'ultimate consensus' that must eventually emerge in the community of rational inquirers just does not work. For, Habermas is ill-guided in thinking that consensus as such is an ultimate objective of rational inquiry by way of being a realistic destination for the enterprise.8 The linkage between consensus and rationality is not descriptive or explanatory, but normative. With rationality, the crux is not what does work towards achieving consensus, but what should work to do so—what indeed would do so among fully rational people. How consensus is achieved is crucial for validity and thus for the whole issue of good reasons. Good reasons are not good because they lead to consensus, but they ought (ideally) to lead toward consensus because they are good. Only at the ideal level is there a linkage between rationality and consensuality. To say that matters of rationality are objective is not to say that people will reach agreement about them—it is to say no more than that they would reach agreement if they proceeded in a totally adequate way. Rationality is a matter of idealization. It gazes towards idealities and away from the actualities of an imperfect world. Different cultures will no more agree about the world's character than different eras will agree about the factual truths of science. And the reason for this in both cases is much the same— different groups have different bodies of experience. But, the evidential relativity of our contentions does not show that there are no facts of the matter on the topics to which they relate, and no objectively rational decisions to be made. The different views of those who have different data at their disposal no more destroy factuality and objectivity than the fact that different associates have different opinions of him annihilates a person. A salient demand of rationality is that we resolve the issues that face us in matters of belief, action, and evaluation as best we can on the basis of the experience at our disposal. But, the very fact 8
Juergen Habermas, Communication and the Evolution of Society, trans. Thomas McCarthy (Boston, Mass., 1979), 188 and 204-5.
174
The Universality of the Rational
that we are a plurality of rational beings with very different bodies of experience means that we must (rationally) resolve the issues before us differently. (It is at bottom this fact that makes the achievement of an actual consensus on substantive issues—and even on those procedural matters that reflect substantive commitments—something that is altogether dispensable for rationality.) But does abandoning a requirement for consensus not make rationality into something variable and culture dependent? Not at all. It means no more than that the ideal of rationality is pursued— within those cultures which happen to pursue it at all—-in variable and circumstance-conditioned ways. And this is surely quite harmless. After all, rationality is in this regard like communication. What communication is is the same everywhere and for everyone. But, of course different people in different places and times go at it rather differently. The cultivation-hierarchy perspective shows how different solutions reached in different circumstances can be justified by the same defining principles of rationality. The normative aspect of consensuality becomes crucial; for, to say that a rational resolution of an issue is universally binding on everyone is ambiguous. It can be construed descriptively as 'Everyone in fact considers themselves to be so bound', or prescriptively as 'Everyone ought to consider themselves to be so bound. It is this second, prescriptive mode of universality that is appropriate in our present context. No sensible rationalist has ever failed to recognize that the prescriptions of reason are nonuniversal in the former, consensus-oriented sense. These considerations support an important conclusion. Objectivity is essential to rationality, but consensus is not. Consensus is a matter of the development of people's views and hinges on such variable matters as evidence, education, and 'climate of opinion'. Consensus turns on what people do think: objectivity on what they ought to think.And the two converge only in the ideal limit—only where people do as they ought. Only 'ideal' consensus—consensus in an idealized community of perfectly rational agents with shared evidence and experience—bears on rationality as such. Consensus accordingly should be no more for us than a Kantian regulative ideal. No doubt, consensus is a good thing. When we have it, we can feel optimistically reassured of being on the right track—although St Augustine's dictum securus iudicat orbis terrarum doubtless overstates the matter. But, consensus is not something
The Universality of the Rational
175
on which we should insist so strongly that we dismiss as inappropriate or worthless ideas and views that lack the benefit of its reassuring presence. Objective and interpersonally operable standards are indispensably necessary for rationality to take hold: only where an issue is objective can rational deliberation come into it. But the realization of substantive consensus is not essential for rationality—at any rate not in the way of an achievement, or even of an expectation as contradistinguished from an aspiration and a hope. Consensus is not an eventual destination but merely a hopeful aspiration. The ideal demands of reason are absolute. But absolutes are never at our actual disposal. No language, no belief system, no thought framework is absolute—delivered in unchangeable perfection to mankind by the world spirit from on high. We can do no more and no better than to use the local, particularized, diversified instruments that come to hand. Thus far, relativism is both inevitable and correct. But, this emphatically does not engender an indifferentistic subjectivism. On the contrary, in rationally adopting a basis for the conduct of our affairs we become committed to trying for one that ought to be binding on everyone. It is the unifying cement of the ideal aspect of rationality—a crucial aspect built into our determinative notion of what 'rationality' is all about—that prevents rationality from becoming unravelled through the relativistic variability of our rational proceedings. Rationality itself is, after all, a project that we are bound to pursue by variable means amongst the varied circumstances of a difficult world, where the consensus that objectivity ideally involves may well be unattainable in practice. (The extent to which' reality co-operates with the demands of rationality is limited.) Rationality must, for us, remain something of an ideal which we can only realize to the limited extent that the circumstances of our situation permit.
11 The Rationality of the Real
SYNOPSIS (1) How is one to account for the fact that 'the real is rational' —that nature is intelligible to us in our man-made, conceptual terms? (2) The answer lies in part in the fact that we are ourselves products of nature and that our intellectual mechanisms—science included—fit nature because they are the intellectual instrumentalities of a creature that itself is the product of an evolutionary process, proceeding within nature's realm. (3) Nature for her part must also be co-operative—learnable enough for the development of intelligence to come about by evolutionary means. (4) Along these lines one can secure a strictly scientific explanation of the impressive effectiveness of natural science in comprehending nature's ways. There is no warrant—and no need—for postulating a metaphysical thesis of 'the rationality of the real' to accomplish this explanatory job. (5) Accordingly, maintaining the rational intelligibility of the real does not require one to accept a 'principle of sufficient reason'. To be sure, such a principle is available. But it is—in so far as plausible—of a strictly methodological and not substantive character.
1 1 . 1 THE I N T E L L I G I B I L I T Y OF NATURE
Why is nature intelligible to man? How is it that we insignificant humans, who play so small a part on the world's immense stage, can manage to unlock nature's secrets and gain intellectual access to nature's laws? And how is it that our mathematics—seemingly a free creative invention of the human imagination—can be used in science to characterize the operations of nature with such uncanny efficacy and accuracy? Why is it that the majestic lawful order of nature is intelligible to us humans in our man-devised, conceptual terms? This issue of why it should be that 'the real is rational' deserves consideration in the setting of a study of rationality in general. Like their theological predecessors, the philosophical theoreticians of ancient Greece were inclined to personify nature in various
The Rationality of the Real
111
ways. Accordingly, they assigned to reason (Sidvoia voi*;, in Latin ratio) an actively directive role not only in human affairs but in the world as well. As they saw it, like calls unto like, and reason in man resonates to reason in nature. We have long ago ceased to hold to such an anthropomorphic view of things, and deem it implausible to say (with Anaxagoras and Plato) that reason makes the world go around. Judaeo-Christian thought took another turn. It saw the world as the created product of a mathematicizing intelligence—as the work of a creator who proceeded more mathematico in designing nature. Such a view also renders nature's mathematical intelligibility unproblematic. God endows nature with a mathematical order and mind with a duly consonant mathematicizing intelligence. There is thus no problem about how the two get together— God simply arranged it that way. But, of course, if this is to be the canonical rationale for mind's grasp on nature's laws, then when recourse to God for the purposes of scientific explanation is abandoned our hold on the intelligibility of nature is also lost. Some of the deepest intellects of the day think that this secure hold is gone for ever. Various scientists and philosophers of the very first rank nowadays affirm commonly and without hesitation that we just cannot hope to solve this puzzle of the intelligibility of nature in a mathematically lawful manner. Erwin Schroedinger characterizes the circumstance that man can discern the laws of nature as 'a miracle that may well be beyond human understanding'.1 Eugene Wigner asserts that 'the enormous usefulness of mathematics in the natural sciences is something bordering on the mysterious, and there is no rational explanation for it'2 and he goes on to wax surprisingly lyrical in maintaining that: 'The miracle of the appropriateness of the language of mathematics for the formulation of the laws of physics is a wonderful gift which we neither understand nor deserve.'3 Even Albert Einstein stood in awe before this problem. In a letter written in 1952 to an old friend of his Berne days, Maurice Solovine, he wrote: 1
Erwin Schroedinger, What is Life? (Cambridge, 1945), 31. Eugene P. Wigner, 'The Unreasonable Effectiveness of Mathematics in the Natural Sciences', Communications on Pure and Applied Mathematics, 13 (1960), 1-14 (see p. 2). 3 Ibid. p. 14. 2
178
The Rationality of the Real
You find it curious that I regard the intelligibility of the world (in the measure that we are authorized to speak of such an intelligibility) as a miracle or an eternal mystery. Well, a priori one should expect that the world be rendered lawful only to the extent that we intervene with our ordering intelligence . . . (But) the kind of order, on the contrary, created, for example by Newton's theory of gravitation, is of an altogether different character. Even if the axioms of the theory are set by men, the success of such an endeavour presupposes in the objective world a high degree of order that we were a priori in no way authorized to expect. This is the 'miracle' that is strengthened more and more with the development of our knowledge . . . The curious thing is that we have to content ourselves with recognizing the 'miracle' without having a legitimate way of going beyond it.4 On this grand question of how natural science is possible at all, some of the best scientific intellects of the day thus avow themselves baffled and unhesitatingly proceed to shroud the issue in mystery and mysticism. According to all these theorists, we are confronted with a genuine mystery. We have to acknowledge that nature is intelligible, but have no real prospect of understanding why this is so. The problem of nature's intelligibility via man's mathematical theorizing is seen as intractable, unresolvable, hopeless. All three of these eminent Nobel laureates in physics unblushingly employ the word 'miracle' in this connection. Perhaps, then, the question is even illegitimate and should not be raised at all. Perhaps the issue of nature's intelligibility is not just intractable, but actually inappropriate and somehow based on a false presupposition. For, to ask for an explanation of why scientific inquiry is successful presupposes that there indeed is a rationale for this fact. And if this circumstance is bound to be fortuitous and accidental, then no such rationale can exist. This is exactly the line taken by Karl Popper, who writes: [Traditional treatments of induction] all assume not only that our quest for [scientific] knowledge has been successful, but also that we should be able to explain why it is successful. However, even on the assumption (which I share) that our quest for knowledge has been very successful so far, and that we now know something of our universe, this success becomes miraculously improbable, and therefore inexplicable; for an approach at an endless series of improbable accidents is not an 4
Albert Einstein, Lettres a Maurice Solovine (Paris, 1956), 114-15.
The Rationality of the Real
179
explanation. (The best we can do, I suppose, is to investigate the almost incredible evolutionary history of these accidents . . .)5 On such a view, the question of the intelligibility of nature is an illegitimate pseudo-problem—a forbidden fruit at which sensible minds should not presume to nibble. We must simply rest content with the fact itself, realizing that any attempt to explain it is foredoomed to failure because of the inappropriateness of the very project. Surely, however, such an approach is highly problematic. Eminent authorities to the contrary notwithstanding, the question of nature's intelligibility through natural science is not only interesting and important, but is surely also one which we can in principle hope to answer in a more or less sensible way. The present discussion, at any rate, is predicated on the supposition that this issue needs and deserves a strong dose of demystification. To be sure, when one presses the question 'Why does mathematics apply to reality?' the logical theorist seems to have a ready answer. He says: 'Mathematics must apply to reality. Mathematical propositions are strictly conceptual truths. Accordingly, they hold of this world because they hold of every possible world.' But this response misses the point of present concerns. Admittedly, the truths of pure mathematics obtain in and of every possible world. But they do so only because they are strictly hypothetical and descriptively empty—wholly uncommitted on the substantive issues of the world's operations. Their very conceptual status means that these propositions are beside the point of present purpose. It is not the a priori truth of pure mathematics that concerns us, its ability to afford truths of reason. Rather, what is at issue is the empirical applicability of mathematics, its pivotal role in framing the a posteriori contingent truths of lawful fact that render nature's ways amenable to reason. After all, it is perfectly clear that the fact that pure mathematics obtains within a world does not mean that this world's laws have to be characterizable in relatively straightforward mathematical terms. It does not mean that nature's operations have to be congenial to mathematics and graspable in terms of simple, neat, elegant, and rationally accessible formulas. In short, it does not 5
K. R. Popper, Objective Knowledge (Oxford, 1972), 28.
180
The Rationality of the Real
mean that the world must be mathematically tractable in being receptive to the sort of descriptive treatment it receives in mathematical physics. How then is one to account for the fact that the world is mathematics-congenial? The answer has to lie in a somewhat complex, two-sided story. The circumstance that X and Y stand in a condition of mutual affinity and consonance (that nature and mathematics are duly coordinated) is a two-sided affair in which both sides must be expected to have a part. For nature to be intelligible to us, there must thus be an alignment that requires co-operation on both sides—hers and ours. The analogy of cryptanalysis is helpful. If X is to break Vs code, there must be due reciprocal alignment. If Xs methods are too crude, too hit-or-miss, he can get nowhere. But even if X is highly intelligent and resourceful, his efforts cannot succeed if Vs procedures are simply beyond his powers. (The cryptanalysts of the seventeenth century, clever though they were, could get absolutely nowhere in applying their investigative instrumentalities to a high-level naval code of World-War-H vintage; the phenomena are too complex for their methods to gain any purchase-hold upon them.) If we are to understand nature— if there is to be a cognitive compatability between these two parties—then both parties, both nature's ways and man's mathematicizing, must be duly co-operative. Let us trace out this line of thought.
11.2
'OUR' SIDE
Man's side of this bilateral story is relatively straightforward. After all, homo sapiens is an integral part of nature—wired into its scheme of things as an intrinsic component thereof. So, the kind of mathematics that we are going to devise is pretty much bound to be the kind that is applicable. Our experience is inevitably an experience of nature. (That after all is what 'experience' is—our intelligence-mediated reaction to the world's stimulating impacts upon us.) So the kind of mathematics that we humans are motivated to develop in the light of our experience is bound to be the kind that is applicable to nature as we experience it. The mathematics of an astronomically remote civilization whose experiential resources differ from ours would surely be substantially
The Rationality of the Real
181
different from the mathematics we humans ourselves know and love. Their dealings with quantity might be entirely a-numerical— purely comparative, for example, rather than quantitative. Especially if their environment is not amply endowed with solid objects or stable structures congenial to measurement—if, for example, they were jellyfish-like creatures swimming about in a soupy sea—their 'geometry' could be something rather strange, largely topological, say, and geared to flexible structures rather than fixed sizes or shapes. Digital thinking might go undeveloped, while certain sorts of analogue reasoning might be highly refined. In particular, if the intelligent aliens were a diffuse assemblage of units comprising wholes in ways that allowed of overlaps, then social concepts might become so paramount in their thinking that nature would throughout be viewed in fundamentally social categories, with those aggregates we think of as physical structures contemplated by them in social terms. Communicating by some sort of 'telepathy' based upon variable odours, or otherwise 'exotic' signals, they might, for example, devise a complex theory of empathetic thought-wave transmittal through an ideaferous aether. The sort of 'structures' that underlie their mathematicizing might be very different indeed. Mathematics is the theory of imaginable structures. And 'imaginable' here is a matter of imaginable by a nature-evolved and nature-embraced creature. Admittedly, mathematics is not a natural science but a theory of hypothetical possibilities developed by theorists who care not one whit for 'applications'. Nevertheless, those structural possibilities with which a 'mathematics' deals are possibilities as conceived by creatures—by beings who do their possibility conceiving with a nature-evolved and nature-implanted mind and whose dealings with possibilities are conditioned by the fact that some of our imaginative projections are and others are not congenial to nature as we experience it. It is thus not surprising that the sort of mathematics we contrive is the sort we find to be applicable to the conceptualization of nature. Our human intellectual mechanisms—mathematics included— fit nature because they are themselves a product of nature's operations as reflected in the cognitive processes of an intelligent creature that possesses its intelligence thanks to the course of its emergence through evolutionary processes. Our mathematics is destined to be congenial to nature because it is itself the product of
182
The Rationality of the Real
a natural process; it fits nature because it reflects the way we ourselves are placed within nature as integral constituents thereof, as products of an evolutionary process proceeding within nature's realm. The intellectual resources we devise in coming to grips with the world—in transmuting sensory interaction with nature into intelligible experience—are themselves mechanisms of adjustment. It is no more surprising that our mind can grasp nature's ways than it is surprising that our eyes can accommodate nature's rays or our stomach nature's food. Evolutionary pressure can take credit for the lot. Still, it could perhaps be the case that we do well only in point of cognitive adjustment in the immediate local micro-environment that defines our particular limited ecological niche. The possibility is still open that we do not really do all that well on a wider scale— that we get hold of only a small localized part of a large and remote whole. Thus, man's own one-sided contribution to the matter of nature's intelligibility cannot be the whole story. For, even if we do reasonably well in terms of our own evolutionary requirements, this might be very inadequate in the larger scheme of things. Nature's wider receptiveness to our cognitive efforts— that nature is substantially amenable to reason and not just somewhat (and perhaps marginally) so remains to be demonstrated. To clarify this issue we must therefore move on to contemplate nature's contribution to the bilateral mind-nature relationship.
11.3
NATURE'S SIDE
If an inquiring being who is placed within nature and forms mathematicized conceptions and beliefs about it on the basis of physical interaction with it is to achieve a reasonably appropriate grasp of its workings, then nature must also do its part. Obviously, first, it must permit the evolution of inquiring beings. Moreover, it must present them with an environment that affords sufficiently stable patterns to make coherent 'experience' possible—enabling them to derive appropriate information from structured interactions. Finally, the fruits of this experience must be capable of mathematical representation. What thus needs to be shown for present purposes is that the employment of mathematics can
The Rationality of the Real
183
provide intelligent inquirers with an adequate and accurate grasp of nature's ways. The applicative success of mathematics involves a two-sided story whose second component—the 'co-operativeness' of nature—must also be addressed. Nature's own contribution to the issue of the mathematical intelligibility of nature must be the availability of a relatively simple law structure—one that encodes so simple a set of regularities that even a community of inquirers possessed of only rather moderate capabilities can be expected to achieve a fairly good grasp of them. But how can one possibly establish that nature simply 'must' afford a fairly straightforward law structure? Are there any fundamental reasons why the world that we investigate by the use of our mathematically informed intelligence should operate on relatively simple principles that are inherently accessible to mathematics? There are indeed! A world in which such intelligence evolves at all will have to be a world that is congenial to mathematics. The argument to this farreaching conclusion goes roughly as follows. For evolution to occur at all, even the humblest of creatures that possess only the most rudimentary anatomy (say, snails and algae) must so operate that certain types of stimuli (patterns of recurrently discernible impacts) call forth appropriate corresponding types of responses. Such creatures can detect a pattern in their natural environment and react to it in a way that works to their advantage in evolutionary terms. And this means that nature must be cooperative in a certain very particular way—it must be stable enough and regular enough and structured enough for there to be 'appropriate responses' to natural events that can be 'learned' by creatures. If such 'appropriate responses' are to develop, nature must provide 'suitable stimuli'. Accordingly, there must be structured patterns of occurrence in nature that even simple, single-celled creatures—the most rudimentary proto-intelligences —can embody in their make-up and reflect in their modus operandi. The world must encapsulate straightforwardly 'learnable' patterns of occurrence in its operating structures—relatively simple laws, in other words. The existence of such learnable 'structures' of natural occurrence means that there must be some useful role for mathematics, which, after all, is the abstract and systematic theory of 'structure in general'.
184
The Rationality of the Real
Moreover, a world in which intelligence evolves by anything like standard evolutionary processes must be one in which the exercise of intelligence affords an (at least modest) evolutionary advantage to its practitioners. For intelligence to develop at all, it must give an 'evolutionary edge' to its possessors. A world that did not comport itself 'as if an intelligent being designed it—which did not exhibit numerous simple patterns and regularities that can bring grist to the mill of intelligence—would simply not be a world in which organic, let alone intelligent, life could develop by evolutionary means. And so, we may conclude that a world in which intelligence can develop by evolutionary processes must be a world amenable to understanding in mathematical terms.6 A world in which evolutionary processes engender high intelligence must in many respects be a highly intelligible world: it must provide food for the mind even as it must provide food for the body. It lies in the nature of evolution that a world in which mathematicizing intelligence emerges must be a world in which intelligent beings find grist for their mill in endeavouring to 'understand' the world. It thus cannot properly be seen as surprising that mathematicizing intelligence arrived at through evolution should be substantially successful in comprehending nature's ways.7 The development of life and, thereafter, of intelligence in the world may or may not be inevitable; the existence of intelligent creatures in the world may or may not be surprising in itself and as such. But, once they are there, and once we realize that they got there thanks to evolutionary processes, it can no longer be seen as surprising that their efforts at characterizing the world in mathematical terms should be substantially successful. An intelligence-containing world whose intelligent creatures came by this intelligence through evolutionary means must be so structured as to be substantially intelligible by mathematical means. Only if we take a view reminiscent of Plato's Timaeus, and locate mind outside nature herself—as an external agency acting upon it from an external vantage-point—will it be surprising that 6 Conversations with Gerald Massey have helped in clarifying this part of the argument. Nothing, of course, follows about the completeness of this intelligibility, which could certainly be very partial. Nothing in our discussion eliminates the prospect that there is a continuum of complexity in nature's laws and that we have only got at the simplest ones so far.
The Rationality of the Real
185
mind should understand nature, that these two parties should be so well attuned to one another. But a mind that is an integral part of nature, and that came to be so through the process of natural selection, is a mind that must be in a position to understand nature on her own terms. On this line of deliberation, then, nature is mathematicsamenable not just in that it has laws (is a cosmos), but in that it affords a variety of relatively simple laws, and it has these relatively simple laws because if it did not it just would not be a potential environment for intelligent life. Given that we intelligent creatures are here, and got here by evolutionary means, the world must behave 'as if it were planned as a theatre of operations for rational calculation. If it were not so, the world would not provide a stage on which evolutionary processes could bring intelligence forth. The sort of processes on whose basis alone information transmission can issue in evolutionary development would be absent. The strictly hypothetical character of the preceding line of reasoning must be recognized. It does not maintain that nature's modus operandi has to be simple enough to admit of elegant mathematical representation by virtue of some sort of transcendental necessity. Rather, what it maintains is the purely conditional thesis that if intelligent creatures are to emerge in the world by evolutionary processes, then the world must be mathematicsamenable. And it must be stressed that this conditional story is quite enough for present purposes. For the question we face is why we humans, we intelligent creatures who are present on the world's stage, should be able to understand its operations in terms of our mathematics. The conditional story at issue fully suffices to accomplish this particular job. 8 Three points are thus paramount for the course of these deliberations: 1. Intelligence evolves within a nature that provides for life because it affords living creatures a good way of coming to terms with the world. 8 It is, of course, altogether possible that we should use the knowledge of nature facilitated by mathematicizing intelligence to destroy mankind. (And what price evolutionary advantage then?) But our present concerns relate only to the initiation of mathematics through the deployment of adaptive intelligence. The issue of eschatology, of the ultimate fate of mathematicizing intelligence and its evolutionary advantage in the long run of the future, is immaterial for present purposes.
186
The Rationality of the Real
2. Once intelligent creatures evolve, their cognitive efforts are likely to have some degree of adequacy because evolutionary pressures align them with nature's ways. 3. It should not be surprising that this alignment eventually produces a substantially effective mathematical physics, because the fundamental structure of the operations of an intelligence-providing nature is bound to be relatively simple. Let us now put the pieces together.The overall question of the intelligibility of nature has two sides: 1. Why is mind so well attuned to nature? 2. Why is nature so well attuned to mind? The preceding discussion has suggested that the answers to these questions are not all that complicated—at least at the level of schematic essentials. In telegraphic brevity, they are simply this: Mind must be attuned to nature because it is an evolved natural product of nature's operations. And nature must be accessible to mind because it must be so if mind is to arise at all. The solution of our problem of the mathematical intelligibility of the real thus roots in the combination.of two considerations: (1) a world which admits of the evolutionary emergence of a (relatively powerful) mode of intelligence must be (relatively) regular and simple in its modus operandi and thus amenable to mathematical characterization, and (2) a sufficiently powerful intelligence will be able to comprehend such a world in mathematical terms. The possibility of a mathematical science of nature is accordingly to be explained by the fact that, in the light of evolution, intelligence and intelligibility must stand in mutual coordination. If mind and nature were too far out of alignment—if mind were too 'unintelligent' for the complexities of nature or nature too complex for the capacities of mind—the two just could not get into step. It would be like trying to rewrite Shakespeare in a rudimentary language with a 500-word vocabulary. We just lose too much important information in such a case. The situation would be akin to trying to keep tabs on a system with ten importantly relevant degrees of freedom with a cognitive mechanism capable of keeping track of four of them. If something like this were the case, mind could not accomplish its evolutionary mission. We had
The Rationality of the Real
187
best turn to an alignment process that does not take the cognitive route. There may indeed be mysteries in this general area. (Questions like 'Why should it be that life evolved in the world?' and—even more fundamentally—'Why should it be that the world exists at all?' are perhaps candidates.) But, the presently deliberated issue of why nature is intelligible to man, and why this intelligibility should incorporate a mathematically articulable physics, does not qualify as all that mysterious, let alone miraculous. To be sure, the preceding account is highly schematic and demands a great deal of amplification. A long and complex tale must be told about physical and cognitive evolution to fill in the details of the present account. But there is good reason to hope and expect that such a tale can ultimately be told. And this is the pivotal point. Even if one has doubts about particular aspects of the preceding evolutionary story, the fact remains that some such story can provide a perfectly workable answer to the question of why nature's ways are intelligible to us humans in terms of our mathematical instrumentalities. The mere fact that such an account affords a hopeful explanatory prospect shows that the issue need not be painted in the black on black of impenetrable mystery. There is simply no need to join Einstein, Schroedinger, and company in viewing the intelligibility of nature as a miracle and a mystery that passes all human understanding. If we are willing to learn from science itself how nature operates and how man goes about conducting his inquiries into its workings, then we should be able increasingly to remove the shadow of mystery from the problem of how it is that a being of this sort, probing an environment of that type, and doing so by means of those evolutionarily developed cognitive and physical instrumentalities, should be able to arrive at a relatively successful account of how things work in the world. We should eventually be able to see it as plausible and only to be expected that inquiring beings should emerge and survive to project themselves through the corridors of time. We can thus look to science itself for the materials that enable us to understand how natural science is possible. And there is no good reason to expect that it will let us down in this regard.9 9 This discussion draws upon ch. 5, 'The Intelligibility of Nature', of the author's The Riddle of Existence (Lanham, Md, 1984).
188
The Rationality of the Real 1 1 . 4 A R E C O G N I T I O N OF LIMITS
It must be stressed, however, that nothing in the preceding discussion should be construed to assert that the extent of nature's amenability to the probes of mind is anything but limited. We can sometimes come to know the truth about the world's modus operandi, and perhaps in some contexts even nothing but the truth, but certainly not the whole truth. Based as they are on appeals to evolutionary considerations, the deliberations set out here can go no further than to show that we can achieve effective knowledge of the world in various matters that 'concern' us in one way or another—that in certain, particularly relevant respects nature must be accessible to the probes of mind. Nothing can be inferred from this about the overall extent of our knowledge. We cannot conclude from the apparent adequacy of our knowledge in certain limited domains (i.e., within those limited parametric ranges of temperature, velocity, pressure, etc., over which we can monitor nature's processes) that our knowledge of the world's processes is adequate overall. Nor can we infer from the statistical adequacy of our information for our purposes and interests to its unfailing correctness over an unrestricted range. The very fact that our knowledge of the world must be built up on the basis of interactions with it—that extrapolation from the perceived to the yet undiscerned is always involved—means that in principle the possibility of surprises can never be ruled out. On the contrary, the history of science clearly exhibits a continuing succession of revolutionary surprises. Such deliberations indicate that the real is only imperfectly rational—only partially open to cognitive penetration by realityimplanted intelligences. Considerations of theory and the lessons of experience combine to imply that the extent to which reality cooperates with the demands of rationality is emphatically limited.
11.5
A 'PRINCIPLE OF SUFFICIENT REASON'?
TWO very different versions of a 'principle of sufficient reason' can be invoked to implement the idea of the rationality of the real that
The Rationality of the Real
189
is at issue in nature's intelligibility: 1. Whenever p is indeed so in nature, there will be a 'good reason'—namely an operative physical cause—that produces it and thus leads to its being so rather than otherwise. 2. Whenever we say that/? is so, there should be a 'good reason' —namely an operative rational ground—that authorizes it and thus justifies us in saying that it is so rather than otherwise. Principle 1—the traditional 'law of causality'—is of course eminently problematic. The discovery of an increasingly substantial range of stochastic phenomena in nature has undermined it severely. Principle 2, however, is something else again—a merely methodological rather than substantive principle. And this weaker principle is eminently plausible. For, the general principle of practical rationality, 'If you do something, have a good reason for doing it', comes into play to assure 2, seeing that to accept or maintain a contention p is indeed to do something. We cannot flatly say that nature itself is rational. But of course we can—and should!—be rational in what we say about nature. This is not a matter of prejudging the principles on which nature operates, which of course is something that we would have to learn by investigating it. Rather, it is simply a matter of how we are to proceed in conducting our own affairs—our own inquiries about nature. Such an approach does not prejudge the 'rationality' of the real as a matter of substantive fact, but merely reflects a determination to proceed rationally in the conduct of our own business—our cognitive business included. And of course such a procedural process does not of itself prejudice the outcome. The features of the inquiry process do not project on to the objects of inquiry. Knowledge need not share the features of its objects: to speak of a sober study of inebriation or a dispassionate analysis of passions is not a contradiction in terms. A rational study of irrational component is perfectly possible. The rational study of nature could discover its essential irrationality—at least in theory, though not in fact if our preceding deliberations hold good. In sum, then, there is no need to postulate the 'rationality of the real' in order to validate rationality in our own proceedings. Cognitive rationality can and should stand on its own feet in the
190
The Rationality of the Real
context of our inquiries. It does not require the sustaining prop of a metaphysical rationality-in-nature postulate that pictures nature itself—rather than our picture thereof—as being the product of a creative intelligence.
12 Rationality and Humanity
(1) The presumption that people act rationally is standard in the explanation of human dealings. This presumption is not only convenient in explaining people's actions but provides the foundation for their social co-ordination with one another. (2) Yet, while we humans are by nature rational animals, rational comportment is certainly not inevitable for us. People are not always rational—non-rational thought and action are plainly commonplace phenomena in human life. (3) Aristotle to the contrary notwithstanding, weakness of will is possible: recognizing the better, people can yet do the worse. We often can act contrary to reason; the salient fact is not that people are rational, but that they ought to be. (4) Our commitment to rationality roots in an ontological obligation inherent in our dedication to the task of selfrealization through the full development of our potentialities. Rationality is thus required of us because it reflects an essential aspect of our very nature. SYNOPSIS
12.1 PEOPLE AS RATIONAL AGENTS: THE PRESUMPTION OF RATIONALITY
Humanity may not be important for rationality—we humans may well be only one of many sorts of rational creatures on the universe's wider stage—perhaps even a relatively non-standard one. But, rationality is crucial to humanity, an integral part of what defines us as the sort of creature we are—or take ourselves to be. The claim to rationality is a crucial aspect of our self-image. Of all our characteristics it is the most central and important. Does man have reason or does man merely think he has reason? The question is almost academic: it just does not matter all that much in the final analysis. What matters primarily is exactly that we do and must think ourselves to be rational—that this capacity is central in our self-definition. This circumstance that we invoke
192
Rationality and Humanity
reason to define ourselves as the beings we take ourselves to be is quite enough to give the concept of rationality its central place in the scheme of things. Yet, why fix on reason as our definitive trait? Why not another characteristically human capacity like imagination or foresight or worriment or deceitfulness or some such? Essentially, because reason is the crossroads where all these characteristic attributes come together. Either these other capacities are essential resources for rationality (like imagination or foresight), or the capacity for reason is involved in their operations (like worry or deceit). All of the 'higher' capabilities characteristic of man are bound up in one way or another with his capacity for reason. But are we humans really all that rational—might we not well use that intelligence of ours to destroy ourselves. Certainly. But consider. Wolves are never sinful; only a creature capable of moral action can act immorally. Similarly, plants are never foolish; only a creature that has the capacity for intelligent thinking can act unintelligently. There is thus a crucial divider between having the capacity for intelligent (rational) action and exercising this capacity intelligently (rationally). The very fact that we can use our intelligence unintelligently—self-destructively, for example—betokens our status as rational agents. In principle, of course, rationality is not an exclusively human characteristic. Other kinds of creatures may well be rational—if not here on earth, then elsewhere. But then the capacities that they must have to qualify as rational are just exactly those we ourselves lay claim to: intelligence (the ability to acquire knowledge through inductive learning and to process it through discursive, reason-governed thought) and free-will (the capacity for decision and action in the light of evaluation on the basis of information).1 To recognize someone as a rational being is to ascribe to him a capacity for the intelligence-guided pursuit of duly evaluated goals. This is an attribute that we standardly grant others—even as we wish them to grant it to us. It is bound up with the view that human beings as such have certain rights and claims, that by their very nature they have a peculiar status as rational agents (and morally responsible agents as well). In considering people's actions, we generally give them credit 1
Cf. Jonathan Bennett, Rationality (London, 1964).
Rationality and Humanity
193
for rationality, proceeding on the basis of a presumption of rationality with respect to our fellows. In the first instance—in the absence of any evidence to the contrary—we proceed on the supposition that people do what they do on the basis of reasons, granting them the benefit of the doubt in point of rationality. We take the stance that there is an explanation on the basis of good reasons as to why they proceed as they do—that they act as they do under the impression that some good will come of it, some benefit for themselves or others. When asked to explain why someone did something, it suffices that we establish that this was the rationally appropriate thing for them to do—at any rate, in the absence of indications that throw their rationality into question. Only as a reluctant last resort do we judge that someone has proceeded on the basis of forces or motives outside the range of their rational control. Thus here, as elsewhere, a presumption of normalcy prevails. Realizing full well that people are not always rational, we find that they (fortunately) are so generally and ordinarily, and feel free to proceed on this basis as an operating reason. In the ordinary course of things we grant people the benefit of any doubt and treat them as rational agents in the absence of convincing counter-indications. This presumption of rationality is not just a matter of generosity, but one of self-interest too. For it affords us an important laboursaving device by allowing us to explain people's actions simply by noting that they were, in the circumstances, rational. In so far as people's actions are of a sort that can be accounted for on principles of practical rationality, we may rest content and pose no further explanatory questions. Accordingly, we need not in general raise questions about someone's motives when they do the rationally indicated thing. In the ordinary course of events—when things go 'normally'—no further special account of the matter is called for. Inquiry can rest and explanation cease in the assurance that things are as they can be expected to be. In giving people the benefit of any doubt in point of rationality, we save ourselves an immense amount of explanatory labour. The assumption that others behave rationally is very useful in another regard. For, people need to be predictable to one another to coexist in fruitful interaction—we need to make ourselves intelligible to others to co-ordinate our activities. Thus, by demanding that people act rationally, and in conforming our own
194
Rationality and Humanity
actions to this expectation, we facilitate the alignment of human actions. (We cannot align on systematically unintelligent choices because the possibilities become unmanageable.) Even if the presumption of rationality did nothing else beyond rendering this service of making our activities intelligible to one another, it would be eminently worthwhile.2 Nor does explanatory convenience tell the whole story. A world in which we cannot communicate and collaborate with others is not a very safe world for us and our kind. And so, evolutionary processes dictate our impetus to increasingly complex communication and collaboration. The presumption of rationality is a crucial aspect of social interaction, since in its absence the basis of communication and collaboration is annihilated. We cannot communicate or collaborate with others in the absence of a supposition of (a fair degree of) rationality on their part. Its highly practical nature accordingly contributes importantly to validating the presumption of rationality. 12.2
ON F A I L I N G TO DO THE R A T I O N A L T H I N G
Are people by nature irrational? Recent psychological studies have sought to establish with experimental precision that people are generally inclined to reason in inappropriate ways. One investigation,3 for example, concludes that people systematically commit the well-known fallacy of denying the antecedent by reasoning: If p, then q Not-p Therefore: Not-q But, it is far from clear that an error is actually committed in the cases at issue. For, people often use 'If p, then q' in everyday discourse as abbreviation for 'if but only if p then q'. ('If you have a ticket, they'll let you board', 'If you pass the course, you'll get four credits'.) What seems to happen in the cases at issue is not 2
Cf. the discussion at the close of sect. 2 of ch. 3 above. L. Rips, 'Reasoning as a Central Intellective Ability', in R. J. Steinberg (ed.), Advances in the Study of Human Intelligence, ii (Hillsdale, 1984), 105-47 (see esp. pp. 134-7). 3
Rationality and Humanity
195
misreasoning, but mere conclusion-jumping by tacitly supplying that 'missing' inverse. Again, various investigators have suggested that people generally commit various fallacies in probabilistic reasoning.4 For example, they incline to the 'gambler's fallacy' of judging that in repeated trials an outcome that has not occurred in a long time is therefore the more likely to arise in the near future. But, observe that this can be seen not as a misinference but simply as a—potentially mistaken—association of the phenomenon at issue with those which (like an animal's need for sleep or food) involve a saturation process of some sort .Thus, the issue may well be one of a mistaken factual supposition rather than one of fallacious reasoning. Or again, experimenters maintain that statistically untrained subjects tend to overestimate the probabilities of conjunctive outcomes and to underestimate those of disjunctive ones. In particular, subjects often judge the probability of a conjunction to be greater than that of one of the conjuncts.5 For example, when subjects are presented with a description of a certain object (for example, as two-bladed), they may deem it quite probable that the object is a pair of scissors, but rather improbable that it is a cutting implement (which, if it is indeed a pair of scissors, the object of course will also be). But such misjudgements can also be accounted for as a matter of misinterpretation of the issue rather than one of misreasoning. The error at issue is presumably better diagnosed as reflecting a (perhaps unwarranted) preoccupation with the inverse-likelihood than as a failure in probability calculation. For one systemic weakness of all such claims of a general penchant to fallacious inference lies in the enthymematic character of much human reasoning. We frequently make substantive 4
See, for example, Amos Tversky and Daniel Kahnemann, 'Judgments of and by Representativeness', in Daniel Kahnemann, Paul Slovic, and Amos Tversky (eds.), Judgment under Uncertainty: Heuristics and Biases (Cambridge, Mass., 1982), 87, and also these authors' 'Belief in the Law of Small Numbers', ibid., p. 24, as well as their earlier paper 'Judgment under Uncertainty, Heuristics and Biases', Science, 125 (1974), 124-31, and cf. L. D. Phillips and W. Edwards, 'Conservatism in Simple Probability Inference Tasks', Journal of Experimental Psychology, 72 (1966), 346-57. 5 See M. Bar-Hillel, 'On the Subjective Probability of Compound Events', Organizational Behavior and Human Performance, vol. 9 (1973), 396-406. See also R. E. Nisbett and L. Ross, Human Inference: Strategies and Shortcomings of Social Judgment (Englewood Cliffs, NJ, 1980), 146-7.
196
Rationality and Humanity
assumptions about how things stand in the world on the basis of experience or of inculcation, and the incorrect conclusions people draw can stem from these assumptions rather than any error of inferential reasoning. After all, any mere inference, however fallacious, can be restored to validity by the addition of suitable supplementing premisses. To construe the data of these interesting experimental studies to mean that people are systematically programmed to fallacious processes of reasoning—rather than merely indicating that they are inclined to a variety of (occasionally questionable) substantive suppositions—is a very questionable step.6 For, when psychologists maintain that the commonplace and normal reasoning processes of people violate the logical and mathematical norms of standard rationality, they overlook the fact that their data can be explained, equally satisfactorily, by supposing that people incline to certain (often perhaps incorrect) factual suppositions about the issues with which they are confronted. But, in any case, even if the psychological data did indeed convincingly show that the inherent 'natural' tendencies of human inferential reasoning were in some respects systematically fallacious, this would not serve to undermine the importance of rationality in human affairs. (After all, the importance of measurement could not be undermined if human vision systematically distorted the relative size of certain objects as in the Mueller-Leier optical illusion.) The value of intelligence would not be diminished if it emerged that people generally proceed unintelligently in certain matters. Only someone who naively expected 'the real' (as reflected in human praxis) also to be highly rational could find this shocking. The key point is that the significance of rationality does not, ultimately, lie in its role as a descriptive characterization of human proceedings (in how people do function) but rather in its normative role, as an indication of how people should function in the best interests of their cognitive and practical concerns. Rationality, like morality, is of normative bearing, concerned with the correct, proper, intelligent way of doing things, and not with the merely usual or customary course of things. The norms of 6 For a cogent analysis of this situation see L. J. Cohen, The Dialogue of Reason (Oxford, 1986), 157-92. On the specifically probabilistic issues see the author's Forbidden Knowledge (Dordrecht, 1987), 58-77.
Rationality and Humanity
197
rationality—like those of morality—are in no way undermined or invalidated by the fact that people violate them. Failures of rationality deserve closer consideration. With the ancients, we continue to conceptualize man as the rational animal. Yet, we do so not because people indeed are always rational, but because they generally have the capacity for reason, and most of them actually exercise this capacity some of the time. Our rationality thus exerts only a statistical impetus: people often, but by no means inevitably, proceed as rationality requires. The presumption of rationality is defeasible and defeatable. We know full well that people's actions can derive from sources quite different from the rational impetus of good reasons: passion ('getting carried away'), suggestibility ('the influence of the crowd'), loss of control (through alcohol or drugs), or the like. To be sure, bitter experience teaches us that all people are sometimes irrational and some people usually so. Realists have long recognized—and psychologists insisted—that the dispassionate consideration of objective factors and the intelligent consideration of real interests are less prominent and powerful determinants of human belief, action, and evaluation than has often been thought—or than people are generally happy to acknowledge. We have at our disposal a whole armoury of special person-characterizing terms (irascible, jealous, suggestible, etc.) and of actcharacterizing qualifications ('out of spite', 'accidentally', 'by a whim') to mark departures from rationality. And a vast descriptive taxonomy is at our disposal to account for peoples' straying from the path of reason: anger, petulance, wilfulness, fear, ambition, and so on. The rational has diverse contrasts; it can be distinguished from what is a-rational, in being indifferent to reason, simply not concerned with it; irrational, in going against reason; extra-rational, in going beyond or outside the range of reason. Natural forces are a-rational; foolish and unreasonable people are often irrational; mystics and visionaries comport themselves in ways that are extra-rational in proceeding by 'purely intuitive' means when in circumstances where reason is rather silent than
198
Rationality and Humanity
negative. In particular, it warrants note that the a-rational is one thing and the irrational another. One can only be actually irrational if one has the capacity for reason and proceeds to neglect, misuse, or abuse it. The violations of rationality that are at issue in irrationality generally fall into three main groups: 1. the incorrect assessment of ends: adopting ends unsuited to one's needs or interests 2. the inadequate management of means to ends: adopting acts geared to means unsuited to ones ends 3. the seriously inappropriate allocation of resources: investing one's limited resources of time, effort, attention, money, or the like, in such a way that some interests are grossly overemphasized at the expense of others that are no less important. Even with otherwise intelligent individuals, it is by no means the case that actual decisions are always, or even mostly, for the best. Nothing, alas, is easier and more common than failing to do the best thing—even the best as seen by oneself. Unfortunately, we often 'see the better and do the worse,' doing what we want rather than what we judge to be best. Man is a frail creature. We are too lazy, too feckless, too careless to take proper care for what we ought to do—for what is good for us rather than what we want. Self-deception can all too easily becloud rationality, letting selfindulgence or convenience or wishful thinking get distortingly in the way of what we ought to do. Irrationality is pervasive in human affairs. While all (normal) people are to be credited with the capacity for reason, they frequently do not exercise it well. No doubt every human action has a motive of some sort, but this is not always a reason—not even a poor one. Not every action is geared to the facilitation of some putative good:—we do some things out of anger or sheer frustration or 'impulse' (spontaneous wilfulness). Indeed, various of our actions lack reasons of any sort, let alone good ones. There are certainly alternatives to using reason in adopting beliefs, values, or actions. All the same, the fact remains that most of us see ourselves as committed to rationality. We are rational much of the time—many or most of us do frequently apply our intelligence to guide us in
Rationality and Humanity
199
what we do by indicating what, in light of the available information, is for the best. Rationality thus exerts a powerful impetus upon how we conduct our practical affairs. And this circumstance endows rationality with an ideological and even political dimension. For, an element of uncertainty pervades all human affairs, and in uncertain situations reason counsels prudence, care, and tentativity. It calls for proceeding with 'calculated caution', rejecting drastic and implausible remedies and untested measures in favour of low-risk, 'pragmatic' approaches. It distrusts drastic innovations proposed on the basis of abstract general principles alone, and calls for trial, experimentation, and inquiry into the modus operandi of things. It demands care and moderation, and reserves drastic remedies for drastic situations where more cautious alternatives are not available. It holds to 'the middle of the road' and counsels empiricism and pragmatism, discouraging 'leaps in the dark' and incalculable risks. The caution 'One step at a time!' is generally reason's motto. There is something rather dull and unexciting about the life of reason. (Rationality is a middleage virtue, out of tune with both the fervent ardour of impatient adolescence and the inflexible conservatism of old age.) And so the call of unreason or anti-reason—of 'noophobia'— will find advocates in every era and every society. There will ever be those who have no patience for the life of cautious inquiry, experimental gradualism, and pragmatic accommodation to things that manifestly need improvement but cannot be mended overnight. The ways of reason are accordingly anathema to advocates of Utopian revisions and radically innovative leaps in the dark. And so, at the political level, in particular, rationality cuts against the tendencies which find expression in fascism, anarchism, and communism—drastically revisionary ideologies that ride roughshod over any suggestion of compromise with the prevailing realities of the human condition within the existing limitations of space and time. The devotees of such approaches reject reason in favour of the siren call of 'natural instinct', 'the spirit of the time', 'historical inevitability', or whatever. All reason can do is to shake its head in frustrated recognition of a facet of human reality that limits its own potential for amelioration. It recognizes and accepts, however reluctantly, the melancholy but significant truism that man's capacity for reason does not mean that people usually act in a rational way. The bearing of rationality is normative, geared less
200
Rationality and Humanity
to the descriptive portrayal of what agents do than to the evaluative analysis of what they ought to do. It is important, in particular, to distinguish between (1) the criteriological question of determining what is the rationally appropriate thing to do in particular circumstances, and (2) the psychological question of the sorts of motives or drives or dispositions that may impel people towards (or prevent them from) doing the rationally appropriate thing. And it is clear that the first of these questions is more fundamental than the second and conceptually prior to it. 7 We cannot sensibly deal with (2) at the level of concrete particulars until we have resolved (1)—that is, we have no prospect of addressing the issue of why someone fails to do the rational thing on some occasion until we have in hand the means of settling what the rational thing to do in those circumstances actually is. And here reasoning in the normative/ evaluative order is paramount. The theory of practical reasoning is not concerned with techniques of self-management as such, but only with identifying what is the rational thing to do. It issues in judgements, addressing the criteriological problem of what we are well advised to do, and leaving the rest to us. The matter of whether an individual is willing to implement such a determination in particular circumstances lies outside the domain of practical reasoning. The object of practical reason is to enable us to figure out what it is that we are well advised to do. The task of getting us to do it is something else again—a matter not primarily of reasoning but of the psychology of self-motivation and self-management. Different issues are at stake.
1 2 . 3 AGAINST THE GREEKS
The thesis that a freely acting agent does what he decides to do is an empty tautology, given what is at issue in 'free agency'. But such an agent does not necessarily do that which he believes to be right and proper for him to do. People not only fail to act for the best, they often even fail to act for the putative best—the best as 7 It thus seems odd that the interesting recent book by David Pears on Motivated Irrationality (Oxford, 1984) concerns itself solely with the second issue and ignores the first altogether.
Rationality and Humanity
201
they themselves see it. It is a fact of life that people often set about doing things they recognize full well to be just stupid. To be sure, philosophers since Socrates have often said that free agents only do what they deem to be for the best. But, to say this is to take too naive a view of human perversity. The scholastic maxim that we seek only what we deem to be good (nihil appetimus nisi sub ratione boni) does not hold always and automatically for people in general, but only for rational people. Much of the time many of us opt for what we want rather than for that we deem best for us. Accordingly, there is no logical inconsistency or incompatibility between the contentions: 1. X realizes that on balance (with everything taken into account) A is the best (optimally advantageous, most sensible, rational, reasonable) thing for him to do. and 2. X fails to do A, and wilfully chooses to do something else instead. There is no contradiction here, and no paradox—unless we are labouring under the mistaken impression that people inevitably 'must' do what they deem the best (most advantageous and beneficial). Of course people often do not do this at all—only rational people do. All too frequently, people in fact simply do what they choose to do—what they want to do in preference to what they deem best (to what they in reflective moments think they should do). This difference between what they ought to do because it is in their best interest (be it short or long term) and what they merely want to do is a crucial one for rationality. The pivotal issue for rationality is what ought to be done—and here, as elsewhere, there is a potential gap between is and ought—between the actual and the appropriate. Man is a rational animal only in that we are capable of reason—but we certainly do not always heed reason's call. The pivotal fact is not that people are rational, but that they ought to be. Actual preference must accordingly be distinguished from rational (duly value-guided) preference: what we want is one thing and what we take to be best is another. But, Aristotle is quite mistaken when he takes this distinction to run parallel to the
202
Rationality and Humanity
distinction between appearance (seeming) and reality.8 In holding that what we actually want is always the apparent good—the option we deem (perhaps mistakenly) to be the best—Aristotle gives us too much credit. Only the ideally rational man chooses something automatically when he evaluates it as the best; the rest of us often choose things simply because we want them, recognizing full well that they may not in fact be all that good for us. People are not always rational—non-rational thought and action are plainly commonplace phenomena in human life, Aristotle to the contrary notwithstanding. All too often we recognize the better but do the worse, letting wilfulness override judgement. Wrong choice is not necessarily a matter of weakness of will—of the will's infirmity in implementing the mind's recognition of the good; it can be a matter of perversity of will as well—of the will's acting in acknowledged contravention of the mind's recognition of the good. Philosophers still sometimes follow the Greeks who generally adopted Socrates view that a person is moved to act only by all he currently deems to be good—that no one ever knowingly does what he at the time takes to be the wrong thing to do. But to take this line is deeply problematic. For it claims that choice always issues from putative choice-worthiness and always lays claim to rational legitimacy. And (to re-emphasize) this holds only for strictly rational people, and not for people in general. Practical rationality accordingly will often not issue in action as such—one's correctly figuring out the appropriate thing to do through rational calculation does not assure one's doing it. In this regard, Aristotle, who saw action (praxis) as constituting the 'conclusion' of practical reasoning,9 was quite wrong. The conclusion of a practical inference, like the conclusion of any other inference, is a proposition—namely a thesis to the effect that a certain action or course of action is the appropriate thing to do (is optimal, proper, well advised, rationally mandatory, or some such). Practical reasoning is a matter not just of acting but of thinking—of deliberating, of figuring out what is to be done. An action as such may implement a course of practical reasoning, but 8 'For the apparent good is the object of desire (imBv/ua: appetite) and the real good is the object of rational choice (/SOUATJOW).' Metaphysics, xii. 1072a28-30. 9 De motu animalium, bk. vii. 701al-16. Cf. John Cooper, Reason and Human Good in Aristotle (Cambridge, Mass., 1975), ch. 1.
Rationality and Humanity
203
does not constitute it (either in whole or in part). With rational deliberation, as elsewhere, a gap may arise between theory and practice, opinion and deed. Not only external interventions but also such internal 'interventions' as passion, fear, or fecklessness can intervene to abort implementing the results of a course of practical reason. The usual linkage between cognition and attempted action—between accepting 'I ought (everything considered) to do A' and actually proceeding to do A—may be short-circuited. This can happen not only through outside interference (through prevention by others or by circumstances), but also through inside interference (through blockage by the force majeure of passion or wilfulness). 'I ought to do A' is a matter of judgemental appraisal—of the insight of rational reflection that (in the circumstances) A is the preferable thing to do. But there is a gap between preferability and preference, between acknowledging that preferring A is the reasonable and appropriate thing to do, and actually preferring A. For, reason may simply fail to prevail in a conflict with its competitors—it may be 'overruled', as it were, by such forces as passion, wilfulness, hedonism, and the like. There is no contradiction, no anomaly, in saying 'He realized perfectly well that A was the best thing for him to do, but still chose to do something else.' To be sure, we generally like to see ourselves as rational—to view what we believe as belief-worthy and what we want as wantworthy (that is, preferable). But, wishes are not horses, and things are not effected through our thinking that it would be nice for them to be so. It is de facto preference that is linked to (attempted) action, not de jure preferability. And the step from preference to preferability can only be accomplished under the supposition that the agent proceeds rationally, that he actively implements in action what rationality indicates in reflection—a supposition that is often plausible but by no means always true. The person who cannot 'bring himself to believe' something that he realizes to be true, or who cannot 'bring himself to do' something that he realizes to be optimal, or who cannot 'bring himself to prefer' something that he realizes to be preferable, does not commit an error of reasoning—on the contrary, his reasoning is (by hypothesis) perfectly in order. (The discrepancy at issue would disappear were this not so.) His problems lie on the side of the will rather than on that of the intellect—albeit generally in the
204
Rationality and Humanity
direction of perversity rather than in that of the infirmity of Aristotelian 'weakness of will' (aKpam'a). However, this sort of failure is immaterial from the standpoint of practical reasoning, which, after all, relates rather to the examining of conclusions than to the issue of their implementation. The 'irrationality' at issue lies not in an inadequacy of reasoning but in a failure to maintain coherence between thought and action. If, realizing that A is the rational thing to do, you choose nevertheless to do an incompatible action A' instead—be it out of wilfulness or perversity or weakness of will or whatever—this circumstance beings questions of the psychology of motivation upon the scene. You are an irrational agent all right, but the irrationality is one of implementation rather than inference. Your failure will not necessarily reflect on your capacity to perform practical reasoning. Quite the reverse! If you did not realize that A is the rational thing to do in the circumstances, the problem is mitigated—for then doing the contrary A' instead would not mark you as irrational. Inability or failure to figure out what the rational thing to do is takes the matter no further than incompetence. The actual irrationality at issue inheres in doing something at variance with an explicit recognition of what the rational thing to do is. Irrationality thus presupposes the ability to accomplish practical reasoning correctly. A creature that lacks this ability cannot be said to behave irrationally, however 'strangely' it may act.
12.4
RATIONALITY AS A D U T Y : O N T O L O G I C A L O B L I G A T I O N AND THE IMPERATIVE TO REASON
Does rationality issue commands or counsels? In the Summa Theologica, Thomas Aquinas sensibly observed that: 'a commitment implies obligation, whereas a counsel is left to the option of the one to whom.it is given' (Q. 14, a. 1). A command is imperative and definitely binding—its demands are framed in the language of 'you must'. A counsel, by contrast, is advisory and offers a recommendation governed by the formula, 'You are well advised to do it'—its demands are framed in the language of 'you should'. Which is at issue with the declarations of reason? To all appearances the voice of reason recommends rather than commands, and advises rather than obliges, so that rationality is a
Rationality and Humanity
205
matter of discovering and choosing that which, as best we can tell, optimally conduces to the realization of our overall interests. Doing what reason instructs is in general not a matter of duty—of some sort of obligation. Rather, it is a matter of prudential advisability. The person who violates the standards of rationality simply manages his affairs in a way that is less than fully intelligent, and thereby (in general) damages no one but himself by frustrating the realization of his own best interests. Thus, rationality's injunctions are seemingly matters of recommendation rather than obligation, of counsel rather than requirement. But the matter actually stands otherwise. For we have not only an opportunity but even a duty to be rational. The most fundamental injunction of reason is 'Be rational—act in line with reason's recommendations!' And this fundamental imperative is a command rather than a counsel. But whence does the advice of rationality obtain the imperatival force in virtue of which we ought to do what it indicates? How is it that the advice of reason obtains to deontic force—that one is somehow obliged to do what reason recommends? What is the basis of this 'obligation' to be rational? The answer here lies in the consideration of ontology. The pivot point is the situation of the individual and of the species in the world's scheme of things. The deontic impetus to rationality inheres in our very nature. It is rooted ultimately in the fact that rationality is part and parcel of the capacities that rationally define man as an intelligent agent. The deontic impetus to rationality lies in the basic ontological imperative to make the best of our given opportunities. The binding obligation to be rational inheres in the 'metaphysical' consideration that we 'owe it to reality at large' to realize ourselves as the sort of being we are—to take our proper place in the world's scheme of things. The factors of self-interest and of self-realization contrive to thrust the rationality project upon us as one in which we both self-interestedly should be and properly ought to be involved. We ought to comport ourselves rationally because rationality is an essential part of our self-definition as human persons. Rationality thus represents a crucial aspect of our deepest self-interest— our being able to maintain a proper sense of legitimacy and selfworth by being able to see ourselves as the sorts of creatures we claim to be. Our very identity as beings of the sort we can and should want to be is at stake.
206
Rationality and Humanity
An ontological or metaphysical imperative to rationality is thus at work. A creature that possesses the capacity for rational agency ought to realize this potential—it ought to act so as to develop itself as a rational being. For, to lose out on such an opportunity to realize the good is simply unintelligent—and thus contrary to the impetus of reason. The obligation to be rational is an ontological obligation that inheres in our capacity for self-development and self-realization—a commitment to the full development of our human potentialities. The imperative to rationality is a matter of the fundamental impetus to make good use of our opportunities for self-development—of doing the best we can with ourselves under the conditions in which we labour. We have here a rationale that grounds obligation in considerations of nature (that is, of a modality of existence)—by its very nature as such, a being that has the capacity for value realization ought to realize it. We here cross the boundary from an is of sorts (that is, of evaluative sorts) to ought. It is through this fundamental ontological imperative that mere 'counsels of reason' are transmuted into commands—commands issued by one side of our nature (the rational) to ourselves in general. We have to do here with an injunction issued by one part of our self to the whole. Our claim to be rational free agents of itself establishes our position in the world's scheme of things, with the result that rationality becomes a matter of duty for us, of ontological obligation. Francis Hutcheson saw morality as a matter of so acting that we can reflectively approve of our own character—of so acting that one need make no excuses for oneself towards oneself. Even so, the fundamental ontological impetus to self-development is a matter of acting in the light of what sort of person one ought to be—of so comporting oneself that one can unhesitantly approve of oneself as being that which one has, through one's own actions, made of oneself. No one is closer to us than our own self (egomet mihi sum proximus), and being on good terms with ourselves is perhaps the most fundamental and basic real and true interest that we have. The ontological imperative to full self-realization—and the rational and moral imperatives it carries in its wake—are simply part and parcel of this fundamental impetus and commandment. Our deepest nature calls on us to be on good terms with ourselves and thus, in turn, requires due heed of our rationality.
Rationality and Humanity
207
The crux here is the fundamental duty to make good use of the opportunities that come our way to realize ourselves as fully as possible—the fundamental duty of self-realization. In so far as one 'owes' it to anyone at all, one owes this duty to oneself and to 'the world at large' or, at any rate, to the community of conscious intelligences within it.The duty at issue is a duty at once to oneself and to the general scheme of things that brought one forth to develop one's highest potential as the kind of creature one is. It roots in the imperatives: 'Realize your highest potential as the sort of being you are!' and 'Develop yourself for the best as best you can!' This 'fundamental ontological duty of self-realization' appertains to any rational agent whatsoever. Any such agent is in a position—in so far as it actually is an agent—to realize its potentialities for providing good (for enhancing value). Selfrealization is the point of confluence where self-interest and obligation flow together. Human rationality is the product of a prolonged process of evolution. There are many ways for an animal species to make its way in the world—many diverse alternative routes for coping within nature present themselves to biological organisms: the routes of multiplicity, toughness, flexibility, isolation, and others. But, one particularly important pathway is afforded by the route of intelligence, of adapting by the use of brain rather than brawn, of cleverness rather than force, of flexibility rather than specialization. In a competitive, Darwinian world a creature that can understand how things work in its environment and exploit this understanding in action has an evolutionary edge—as the master himself already stressed.There is a promising ecological niche for a creature that makes its way in the world not by sheer tenacity or by tooth and claw, but by intelligence—by co-ordinating its own doings and the world's ways through cognitive foresight. We live and breathe and have our being in a natural environment that is not originally of our making and in which—at any rate—we must forge our own way by the use of our wits. For it is the route of mind, and not the route of tooth and claw, that is our evolutionary destiny. But rationality is not programmed into us like an animal's instinct. We are free creatures. And as such we do well to walk in the paths of reason, not because considerations of necessity dictate that we must, but because considerations of desirability indicate
208
Rationality and Humanity
that this affords the greatest real advantage for us as the sort of creatures we are. Reason is an imperfect guide—but it is the best we have. The point at issue is a fundamentally axiological one. Accordingly, there is yet another explanation for its centrality in human affairs. As members of the genus Homo sapiens we are creatures that have a capacity for at least partial self-construction —for making ourselves into beings of a certain sort by viewing ourselves in a certain way. In substantial part, we are what we are because of what we claim to be. Persons become persons through their capacity to see themselves as such. In particular, we value ourselves as members of a certain category (species, society, group). We have a sense of belonging—a preparedness to recognize others as instances of 'our type' accompanied by a sense that we ourselves deserve to be so recognized by them in turn. We see ourselves as bearers of value in a community of mutual recognition—as members along with others of an affinity-community of 'people like us'. It is a crucial part of our interests to maintain a proper sense of self-worth in such a setting of reciprocity. An injury to this sense of self-worth is one of the very worst things that can happen to a person. We are then deprived of the self-esteem that goes with membership in a group of which we are pleased and proud to be a part. In eroding one's sense of legitimacy, such an injury undermines one's sense of worth by degrading one where it counts the most—in one's own eyes. After all, we are in crucial part what we are because of what we claim ourselves to be in determining an identity for ourselves as members of a wider community. (The T one takes oneself to be is crucially conditioned by the 'we' of its associates.) The first-person plural ideal of 'we' and 'us' that projects one's own identity into a wider affinity-community is a crucial basis for our sense of worth and self-esteem. A feeling of self-worth is essential to one's sense of legitimacy—one's ability to feel at home in the world. (It is this threat, rather than a fear of its superior weaponry, that makes the idea of contact with a 'superior' civilization so intimidating.) The crux of our deontic commitment to rationality lies in the region of axiology—in the value that rationality has for us. If rationality were merely and only a matter of (true) prudence and self-interest (however much it is our real interests that are at
Rationality and Humanity
209
stake), the value of rationality would be less than it is. We do— and must—value our rationality not just because it helps to feather our nest but because we see it as a crucial component of our very nature.
13 Rationality and Happiness
SYNOPSIS (1) Does rationality pay off in terms of happiness? (2) To deliberate sensibly about this important question, we must distinguish between the affective (psychological) and the reflective (judgemental) modes of happiness—between emotive pleasure and judgemental contentment of mind. For, rationality has a direct bearing on the latter, but not on the former. (3) Yet, while rationality is not generally a promoter of affective happiness in its positive form, it can unquestionably greatly diminish—via science and technology—the affective unhappiness of pain and suffering in life. (4) Is rationality cold, passionless, inhumane—an obstacle that stands in the way of human happiness? By no means! Reason herself is quite prepared to acknowledge the validity of those many life-enriching, unreflective activities that call for little if any exercise of reason. (5) Given that rationality is a matter of intelligence, it is only natural that reason should be supportive of happiness when this is construed in the specifically reflective mode.
13.1
ARE RATIONAL P E O P L E H A P P I E R ?
Is rationality a good thing? The question has a rhetorical air about it. It is all too obvious that rationality is by its very nature a positive quality—a 'perfection' in the philosophical terminology of an earlier day. The issue cries out for such old-fashioned treatment, since its more contemporary-sounding reformulations strike the ear strangely. 'Other things equal, is a creature better off for being rational?' But, how could other things possibly be equal in rationality's absence? The very idea is ludicrous. And yet, when everything is said and done, the question which forms this chapter's central theme still arises: Are rational people happier? While earlier chapters have examined the inner workings of rationality, the present chapter turns in another direction to deliberate about the consequences of rationality. In particular, it will examine the bearing of rationality on that cardinal aspect of
Rationality and Happiness
211
the condition of man, the issue of human happiness, addressing one of those large questions that have preoccupied philosophers throughout the ages: Are rational people happier? Does rationality —the proper use of our intelligence—pay off in terms of advantage for this key aspect of the human condition? This theme harks back to deliberations of the philosophers of ancient Greece on the question: Is the wise man also happy?
1 3 . 2 TWO MODES OF 'HAPPINESS'
As often happens with philosophical problems, the pivotal issue is not just one of examining the facts, but predominantly one of clarifying concepts and issues. For, the'question of the linkage between rationality and happiness hinges critically on just how we propose to understand the idea of 'happiness'. Distinctions must loom large here. In particular, we face two crucially diverse alternatives, depending on whether we construe happiness in an affective or in a reflective sense—whether we conceive of it as a psychological state of subjective feeling, or as a judgemental matter of rational assessment and reflective evaluation. This distinction between affective happiness and reflective happiness— between euphoria and contentment, as it were—requires closer consideration. Figuratively put, affective happiness depends upon the viscera and reflective happiness on the brain. The difference turns on whether one responds to things positively by way of an emotive, psychological reaction, some sort of warm, inner affective glow, or whether one responds to them by way of a rationalized proappraisal, a deliberate, intellectual judgement of the condition of things. As a psychological state, affective happiness is a matter of how one feels about things—a matter of mood or sentiment. Primarily, it turns on what would commonly be called enjoyment or pleasure. It is the sort of psychic state or condition that could, in theory, be measured by a euphoriometer and represents the sort of physiologically engendered condition that might—and indeed can—be induced by drugs or by drink. (Think of the 'happy hour' at cocktail bars.) By contrast, reflective happiness is a matter of how one thinks
212
Rationality and Happiness
about things, it is reflected in appraisal and judgement—in how one assesses or evaluates the current situation rather than how one reacts to it emotionally or affectively or psychically. It is not a psychological state of feeling at all, but an intellectual stance of reflectively positive evaluation. It is a matter of being so circumstanced as to appraise one's condition with judgemental approbation. The issue is one of rational satisfaction rather than pleasure, of what Aristotle called evdai^ovia in contrast to -qSovyj. Happiness in this second sense consists in the reflective contentment of one who 'thinks himself fortunate' for good and sufficient reason. Its pivot is not 'pleasure' but 'contentment of mind'. The two sorts of 'happiness' accordingly also have very different temporal qualities. Affective happiness (pleasure) is generally something fleeting and short term—a thing of psychic moods and whims, of the feeling of the moment. By contrast, reflective happiness (rational contentment) is generally something deeper and less transient—a matter of understanding rather than feeling, of stable structure rather than transitory state. Very distinct issues are thus at stake. People may well take satisfaction (quite legitimately) in actions or occurrences which, like Kantian works of duty, do not at all promote their 'happiness' in any affective sense of that term—indeed, which may even impose a cost in this regard. It does not follow that the individual who prospers in happiness or welfare is thereby superior in 'quality of life'. (We come back to the cutting edge of John Stuart Mill's obiter dictum: 'Better to be Socrates dissatisfied than a pig satisfied.') And our attitude towards these matters tends to be very different. Toward people or nations that have—even to abundance —the constituents of affective happiness, we may well feel a certain envy, but our admiration and respect could never be won on this ground alone. Such issues are judgemental and accordingly hinge on reflective happiness. Recognizing that there are two very different ways of interpreting the idea of 'happiness', we must note that which of them one adopts will make all the difference for the question of how rationality and happiness are interrelated. If happiness is construed in the reflective sense as rational satisfaction, then rationality is indeed a means towards greater happiness. For one thing, people who proceed rationally are, thanks to their rationality, going to improve the chances that
Rationality and Happiness
213
things will eventuate favourably for the promotion of their real interests. And even when things do go wrong—as they doubtless often will—the rational person has the consolation of rationality itself, of the recognition of having done one's best. For, the rational person prizes reason itself and takes rational satisfaction in the very fact of having done what reason demands. Even when matters go awry due to 'circumstances beyond one's control', the rational agent has that contenting consolation of 'having done one's best in the face of the inevitable' which was so greatly prized by the ancient Stoics. Realizing the limits of his powers, the rational person avoids pointless regrets and futile recriminations, achieving the self-respect and justified self-satisfaction that goes with the realization that one has done all one can. Deliberations of this sort combine to indicate that if happiness is construed in terms of one's reflective contentment with the condition of things by way of intellectual appraisal, then there is certainly good reason to think that the rational person will indeed, by virtue of that rationality, fare better in the pursuit of happiness. On the other hand, if happiness is seen as an affective psychic condition—a matter of point-accumulation on the euphoriometer —then the thesis that rationality promotes happiness becomes very questionable. In the first place, there is the fact that we can gain ready access to euphoria through avenues not particularly endorsed by reason. For one thing, there is the prospect of drugs and psychic manipulation. For another, the very fact that we can speak of 'harmless pleasures' indicates that there are also harmful ones of which reason is bound to disapprove. By its very nature, reason is geared not to our pleasure but to what is in our best interests, and so there is no basis for thinking that a heed of reason's dictates will advantage us in the pursuit of affective pleasure.1 ' It is useful to observe the close parallelism of these ideas to discussions in the post-Aristotelian schools of Greek philosophy. The distinction between affective happiness and reflective happiness runs parallel to their distinction between pleasure or enjoyment (^Sourj) on the one hand, and genuine well-being (evSai/j-oCa) on the other. And if one identifies rationality with what those ancients called wisdom (aoCa), then their insistence that wisdom was a necessary (though not necessarily sufficient) condition for the achievement of true happiness (well-being = evdoaixovCa = humanflourishing)parallels our present conclusion that rationality is bound to facilitate reflective happiness. The discussions of those classical moralists are intimately relevant to our present deliberations, and point towards results of much the same general tendency. (Where wisdom rather than 'know-
214
Rationality and Happiness
Somebody is bound to object: 'Surely rational people are the happier for their rationality, even in the affective mode of happiness, because their intelligence is capable of benefiting them in this regard as well.' It would doubtless be very nice if this were so. But, alas, it is not. For, while intelligence can lead one to water, it cannot ensure that drinking produces any worthwhile effects. People being what they are, there is no reason to think that conducting their affairs intelligently benefits them in terms specifically of increased affective happiness. To be sure, there is the fact that rational people will be the more 'knowledgeable'—that they will (presumably) transact their cognitive and their practical affairs with greater success in the realization of their objectives. But this will not mean all that much for their specifically hedonic happiness. For experience teaches that people are not generally made affectively happier by 'getting what they want'.This very much depends on the kind of thing that they are after. And even if they indeed are after the things of which reason approves, this will not help them all that much when affective or hedonic happiness is at issue. Of course, people who proceed rationally will be disappointed less often than they otherwise would be. Their rationality can plausibly be expected to spare them sundry unpleasant surprises. But, by the same token, rationality may also possibly occasion its bearer some pain and dismay. For rational foresight and foreknowledge can also lead to painful apprehensions and gloomy forebodings with respect to things that will probably go wrong—of which life is bound to afford many instances. And there is yet another, less obvious, aspect of the matter. Rational comportment is a matter of the intelligent use of means towards realizing our appropriate ends. The region where it will prove productive is where intelligent action can be expected to bear good fruit. But happiness in its hedonic sense is not a good instance of this. For affective happiness is something too ephemeral and capricious to lend itself to effective manipulation by rational means. (Even—and perhaps especially—people who 'have everything' may yet fail to be happy; there is nothing all that paradoxical or even unusual about someone who says 'I know that how' is concerned, there is no technological obsolescence.) For an informative and interesting treatment of the relevant issues, see J. C. B. Gosling and C. C. W. Taylor, The Greeks on Pleasure (Oxford, 1982).
'
Rationality and Happiness
215
in these circumstances I should be happy, but I'm just not.') 2 Affective happiness is largely a matter of moods and frames of mind—easily frustrated by boredom or predictability. It is an ironic aspect of the human condition that affective happiness is inherently resistant to rational management. Judge this from your own experience! Among the people you know, are the rational ones—the intelligent and sagacious and prudent ones—any happier, on balance, affectively speaking, than their more thoughtless and happy-go-lucky compatriots? Most likely not. It would seem that an easy-going disposition and good sense of humour counts for more with affective happiness than intelligence and rationality. To be sure, it might seem on first thought that the singlemindedly efficient pursuit of affective happiness is bound to provide greater pleasure in the long run. But the facts of experience teach otherwise. John Stuart Mill's description of his own experience is instructive in this regard. In a striking passage in his Autobiography he wrote as follows: It was in the autumn of 1826. I was in a dull state of nerves, such as everybody is occasionally liable to. . . . I t occurred to me to put the question directly to myself: 'Suppose that all your objects in life were realized; that all . . . could be completely effected at this very instant: would this be a great joy and happiness to you?' And an irrepressible selfconsciousness distinctly answered, 'No!' At this my heart sank within me: the whole foundation on which my life was constructed fell down. The experiences of this period had two very marked effects on my opinions and character. In the first place, they led me to adopt a theory of life, very unlike that on which I had before acted, and having much in common with what at that time I certainly had never heard of, the antiself-consciousness theory of Carlyle. I never, indeed, wavered in the conviction that happiness is the test of all rules of conduct, and the end of life. But I now thought that this end was only to be attained by not making it the direct end. Those only are happy (I thought) who have their minds fixed on some object other than their own happiness; on the happiness of others, on the improvement of mankind, even on some art or pursuit, followed not as a means, but as itself an ideal end. Ainjing thus at something else, they find happiness by the way. The enjoyments of life 2
Recall Edward Arlington Robinson's poem about 'Richard Corey', the man who 'was everything to make us wish that we were in his place' and yet one night 'went home and put a bullet through his head'.
216
Rationality and Happiness
(such was now my theory) are sufficient to make it a pleasant thing, when they are taken en passant, without being made a principal object.3 f, Getting what we naively and unevaluatedly want can be a hollow business. And ironically, when hedonically affective happiness is pursued, however rationally and intelligently, it inclines to flee away. (This is yet one more way in which the project of 'the pursuit of happiness' faces substantial inherent difficulties.) For, as Mill's deliberations indicate, rationality itself teaches us in the school of bitter experience about the ultimate emptiness of this sort of thing—its incapacity to deliver on the crucial matter of real contentment by way of reflective happiness. Considerations of this sort combine to indicate that when happiness is construed in the hedonic terms of affective euphoria or pleasure it becomes implausible to hold that the rational man is the happier for his rationality.4
1 3 . 3 MORE ON THE AFFECTIVE R E W A R D S OF RATIONALITY
There is, however, a further, importantly relevant, aspect to the issue of the bearing of rationality on happiness in its affective dimension. For, our deliberations have to this point neglected an important distinction. The hedonic domain actually has two sides—the positive, which pivots on affective happiness or pleasure, and the negative, which pivots on affective unhappiness or pain. A negatively-oriented affective benefit is the removal or diminution of something bad. (It is illustrated in caricature by the story of the man who liked to knock his head against the wall because it felt so good when he stopped.) A positively-oriented affective benefit, on the other hand, is one which involves something that is pleasant in its own right rather than by way of contrast with a distressing alternative. This distinction bears importantly on our problem. For, there is no doubt that the state of human well-being has been, and can still 3 The Autobiography of John Stuart Mill, ed. J. J. Coss (New York, NY, 1929), 94-101. 4 To be sure, the fact remains that rational people will certainly be better off (reflectively speaking) on rationality's account—seeing that they are bound to take pleasure in rationality itself.
Rationality and Happiness
217
further be, greatly improved through the use of intelligence—in the area of science and technology—towards the negatively-oriented benefits of reduced human misery and suffering. Consider only a few instances: medicine (the prevention of childhood diseases (through inoculation), anaesthetics, plastic and restorative surgery, hygiene, dentistry); waste disposal and sanitation; temperature control (heating and air conditioning); transportation and communication; and so on. It would be easy to multiply examples of this sort many times over. Intelligence can certainly stand us in good stead in averting causes of distress and boredom. It can vastly improve the 'quality of life'. But the fact remains that, as the world turns, this diminution of the negative does not necessarily yield positive repercussions for affective happiness. Augmented well-being does not mean an increase in affective happiness; a lessening of suffering and discomfort does not produce a positive condition like pleasure or joy or happiness. For, pleasure is not the mere absence of pain, nor joy the absence of sorrow. The removal of the affectively negative just does not of itself create a positive condition— though, to be sure, it abolishes an obstacle in the way of positivity. And so, the immense potential of modern science and technology for the alleviation of suffering and distress does not automatically qualify it as a fountain of affective happiness. The harsh fact of the matter is that technical rationality is relatively powerless as a promoter of hedonic happiness in its positive dimension. To be sure, technical intelligence can indeed provide such enhancers of affective positivity as alcoholic beverages or drugs. But there is a big fly in this ointment. The affective pleasure of such euphoria-inducers becomes eroded by routinization. Habituation swiftly undermines the pleasantness of these 'pleasures', so that little if any real pleasure accrues from their merely nominal 'enjoyment'. In the end, it is not the pleasure of indulgence but the discomfort of deprivation that comes to prevail. Natural psychological and physical mechanisms soon transmute the 'benefits' at issue with these technically contrived euphoria-inducers from a positive to a negative character. Designed for the enhancement of pleasure, their potential is soon reduced to the diminution of pain. No doubt rationality pays. But, the irony of the human condition is that as far as affective matters are concerned the utility
218
Rationality and Happiness
of reason is vastly more efficacious at averting unhappiness than at promoting happiness in its positive dimension. 1 3 . 4 IS RATIONALITY INHUMANE?
Can one be simply too rational in the management of one's life? It is sometimes said that a person's rationality can actually impede the realization of happiness. After all, man does not live by reason alone, and many rewarding human activities—family life, social interaction, sports and recreations, 'light' reading, films and other entertainment, and so on—make little or no use of reason or reasoning. And so, people often say things like: 'Rationality is cold, passionless, inhumane. It stands in the way of those many life-enhancing, unreflective, spontaneous activities that have an appropriate place in a full, rewarding, happy human life.' One frequently hears such claims maintained. But they are profoundly mistaken. We must return here to the preceding chapter's distinction, between those actions and activities that are a-rational in that they involve little or no use of reason and those that are irrational in actually going against reason. Now, reason can and does recognize as wholly proper and legitimate a whole host of useful activities in whose conduct it itself plays little if any part—socializing, diversions, recreations, and so on. Reason itself is altogether willing and able to give them its stamp of approval, recognizing their value and usefulness. Accordingly, one cannot be too rational for one's own good. If, contrary to fact, there were such a defect—if this could be established at all—then reason herself could bring this circumstance to light. Intelligence does not stand as one limited faculty over against others (emotion, affection, and the like). It is an allpervasive light that can shine through to every endeavour—even those in which reason herself is not involved. Whatever human undertaking is valid and appropriate can be shown to be sound by the use of reason. It is the exercise of rationality that informs us about priorities. For that very reason it takes top priority. While man is indeed a rational being, he is not only a rational being. There is more to humanity than rationality. Our natural make-up is complex and many-sided—a thing of many strains and
Rationality and Happiness
219
aspects. We have interests over and above those at issue in the cultivation of reason. But there is no reason whatever why our reason should not be able to recognize this fact. To fail to do so would be simply unintelligent—and thus contrary to the very nature of rationality. The very fact that man is the rational animal means that there is a good deal more to us than reason alone—and nothing prevents reason from recognizing that this is so. People can certainly neglect those various valuable a-rational activities in favour of over-calculation, over-planning, and an inflated over-commitment to various uses of reason. However, the salient fact is that rationality itself dis-recommends this. In being 'too rational' one would, strictly speaking, not be rational enough, it is perfectly rational sometimes to do heedless or even madcap things in this life—'to break the monotony' and inject an element of novelty and excitement into an otherwise prosaic existence. All work and no play makes life go stale. People can sometimes take quite appropriate pleasure from 'irrational' actions—climbing mountains, betting on the ponies, dipping into a freezing river. To break the mould of a colourless rationalism is, within limits, not all that irrational—is not at all unintelligent. It is part and parcel of a deeper rationality that goes beyond the superficial. After all, rationality aims at goods as well as goals. It is clearly in a position to appreciate the values of enjoyment as well as those of achievement. Several among the ancient philosophers—Aristotle pre-eminently —insisted on the primacy of the strictly intellectual pleasures inherent in the exercise of reason. They maintained that only the purely rational intellectual activities—learning, understanding, reasoning—yield satisfactions of a sort worthy of a rational being. Only in the pleasures of the mind did they see true satisfactions. Accordingly, they suggested that only in the pleasures that are consequent upon the exercise of reason can rational creatures take appropriate satisfaction—everything else is a matter of dross and delusion. But this line of thinking is deeply problematic. Rationality does not demand that we seek satisfaction in reason alone and view the pleasures of reason as solely and uniquely genuine. Far from it! Reason can and does acknowledge the need for diversity and variation; she can and does recognize the importance of activities that call for little if any exercise of reason. The importance of a
220
Rationality and Happiness
balance of varied goods within a complex 'economy of values' is something which reason herself emphasizes—even though this complex must itself encompass various mundanely a-rational goods. To insist that rational satisfaction—reflective contentment —rather than mere 'pleasure' is the pivot of genuine happiness does not mean that commonplace pleasures have no legitimate place in a truly happy life. There is no sound reason why rational people need be spoil-sports. A deep distrust of reason is a constant leitmotiv of Spanish philosophy that runs from Francisco Sanchez and Gracian y Morales in the seventeenth century to Miguel de Unamuno and Jose Ortega y Gasset in more recent days. What the Spaniards have against reason is that, as they see it, she directs people to certain particular sorts of ends; namely, 'reasonable' and 'sensible' ends— paternalistic ends which can be objectively validated through the approval of others (experts). But the pursuit of such ends does not make people happy. 'Be reasonable!' is the eternal cry of disillusioned middle age against the enthusiasms of youth. And this eternal cry is destined to be eternally unavailing because youth realizes instinctively—and rightly—that the path to happiness does not lie in this particular direction. Reason aims at what is clear, fixed, cut-and-dry. Hers is the way of the well-ordered zoo, not the way of the jungle. She is at odds with the vibrant disorder of human life. The Spanish perspective combines a common-sensical view of the good life with a deep scepticism that reason can get us there. For the world—and in particular the social world in which we humans live—is changeable, chaotic, irrational. 'General principles' are of little help; the useful lessons of life are those people learn in the school of bitter experience. Reason as such is not a suitable guide to a satisfying life. So urge the Spaniards. All of this has a certain surface plausibility. But even in a difficult world those who do not examine it rationally—and refuse to profit by a reason-guided exploitation of the experience of others—create needless difficulties for themselves. To be sure, there are other guides to human decision than reason herself— custom, instinct, experience, and spontaneous inclination among them. But, only reasoned examination can teach us about their proper use. What is particularly ironic in the Spanish critique of applied
Rationality and Happiness
221
reason is its inevitable reliance on reason herself for its validation. That reason may have its limits as a guide to the attainment of a satisfying life may well be true. But only reason herself can inform us about this—only a rational scrutiny and reason-guided investigation of the matter can reliably inform us what these limits are. In developing their case for the limitations of reason, the Spaniards are (inevitably) constrained to make use of the resources of reason herself. Yet, does rationality not undermine the emotional and affective side of man—the uncalculating, unselfish, open, easygoing, relaxed side that is no less significant in the overall scheme of human affairs than the sterner enterprise of 'pursuing our ends'? Is reason not deficient in one-sidedly emphasizing the 'calculating' aspect of human nature? Not at all! There are good grounds for reason not to deny the claims of man's emotional and affective side. For life is infinitely fuller and richer that way! Reason, after all, is not our sole directrix. Emotion, sentiment, and the affective side of our nature have a perfectly proper and highly important place in the human scheme of things—no less important than the active striving for ends and goals. In so far as other valid human enterprises exist, there is good reason why reason can (and should) recognize and acknowledge them. To insist on reasoning as the sole and all-comprising agency in human affairs is not rationalism but a hyper-rationalism that offends against rationality as such. To say that reason is cold, inhumane, bloodless, and indifferent to human values is to misconceive rationality as purely a matter of means to arbitrary ends, committed to the approach of 'let's get to the goal but never mind how, with no worry about who or what gets hurt along the way'. But, such a 'mechanical' view of reason, regrettably widespread though it is, is totally inappropriate. It rests on that familiar fallacy of seeing reason as a mere instrument that is in no position to look critically at the goals towards whose realization it is being employed. It refuses to grant reason that which is in fact her definitive characteristic—the use of intelligence. Again, is reason not defective because—so it is said—she generally counsels a prudent caution that is at odds with righteous indignation, courage, bravery, and other manifestations of the 'spirited' side of human life? Shakespeare's Troilus made the point as follows:
222
Rationality and Happiness You know a sword employed is perilous, And reasonfliesthe object of all harm. * * * Nay, if we talk of reason, Let's shut our gates and sleep. Manhood and honour Should have hare-hearts, would they but fat their thoughts With this crammed reason. Reason and respect Make livers pale and lustihood deject.5
But here we have once more an overly narrow conception of reason. Reason is perfectly capable of acknowledging that 'sweet reasonableness' is not called for in all circumstances, and recognizing that there may be occasion for indignation and outrage in a just cause. There is little question that, as one recent author puts it, 'one can do harm to important human values by overemphasizing the values of theorizing and cognition.'6 Yet, to acknowledge this in no way undermines the claims of reason. On the contrary! It is reason herself that demands that we recognize the limited place of the virtues of cognition, inquiry, and the cerebral side of life. An adequate account of rationality must rightly stress its importance and primacy and still at the same time recognize that the intellectual virtues comprise only part of the overall picture regarding the good life. As we have seen, reason urges the intelligent cultivation of appropriate ends. And in so far as those various a-rational activities do indeed have value for us, reason herself is prepared to recognize and approve this. The life of reason is not all calculating, planning, striving. For us humans, rest, recreation, and enjoyment are very much a part of it. Accordingly, reason is perfectly willing to delegate a proper share of authority to our inclinations and psychic needs. It goes against reason to say that rational calculation should pervade all facets of human life. Reason does not insist on running the whole show by herself, blind to her limitations in being simply one human resource among others. As 5 Troilus and Cressida, II, ii, 40-50. " Stephen Nathanson, The Ideal of Rationality (Atlantic Highlands, NJ, 1985), 157. This book is well worth reading on the subject of our present concerns.
Rationality and Happiness
223
has been stressed from the outset of this discussion, rationality is a matter of the intelligent pursuit of appropriate ends. And it would be absurd to think that this is something that can be of doubtful propriety.
13.5
REASON AS A BASIS FOR R E F L E C T I V E H A P P I N E S S
The upshot of these deliberations regarding the bearing of rationality upon happiness is clear enough, even though subject to various complicating distinctions. If we construe happiness in the more reflective mode as an intellectual matter of rational contentment, then the rational person is bound to be the better off by way of improved chances for happiness. But, if happiness is construed in the affective mode as a matter of pleasure or euphoria, then there are no good grounds for thinking that rationality is profitable for happiness in its positive aspect— though even here it does have the merit of being able to help in averting affective unhappiness. The outcome to the question of rationality's claims as a supporter of happiness is thus indecisive. The answer will depend crucially on just which conception of happiness we propose to adopt. These deliberations accordingly lead to a result that is not perhaps all that surprising. Given that rationality is a matter of intelligence—of the effective use of mind—it is only natural and to be expected that rationality should be congenial to and supportive of that reflective, judgemental mode of happiness over which mind is the final arbiter. But, being what it is, rationality need not be unintelligent about it and overlook the importance for us of values outside the intellectual domain.
14 Conclusion: The Grandeur ahcl Misery of Reason
(1) The glory of reason is that she liberates her possessors from the control of 'external' forces, endowing them with a claim to at least partial self-determination. (2) Reason is autonomous: no wholly alien authority is in a position to lord it over her. Rational justification— including the rational justification of placing reliance on reason as such— admits of no court of appeal whose authorty is not endorsed by reason herself. (3) The misery of reason lies in the fact that, while always and everywhere insisting on obedience to her requirements, she nevertheless can provide no certified assurance that in following her counsels as best we can we may not actually damage rather than enhance the prospects of attaining our legitimate ends. While there is no intelligent alternative to doing reason's bidding, we must nevertheless do this in the absence of guarantees. SYNOPSIS
14.1
THE G R A N D E U R OF REASON
Though man is—as Pascal said—but a reed, he is a thinking reed. Though a mere pawn on the world's immense chess-board, he is— or can be—a knowledgeable pawn. We are weak and mortal creatures at the mercy of nature and of chance, but the exercise of intelligence affords us a glimpse into the realm of the infinite and the timeless. Reason is the source of human power—and of human nobility. Many of the great things we can achieve in this world we can accomplish only through her means. Such greatnesses as we have, we owe in large measure to the possession of reason. It is thought, intelligence—the exercise of rationality in the interests of knowledge, valuation, and right action—that sets humans apart from other creatures and renders us (mere animals that we are!) akin to the gods. Homo sapiens is a rational animal. The fact that we are animals
The Grandeur and Misery of Reason
225
places us squarely within the order of nature. But, the fact that we deem ourselves rational means that we see ourselves as exempted from the absolute rule of external forces and as endowed with some measure of self-determination. A rational creature is one capable of making its idealized vision of what it should be determine at least in part what it actually is. Our claim to rationality means that our nature is not wholly given—that we have the ability to contribute in at least some small degree to making ourselves into the sorts of creatures we are. What mechanisms can ensure the co-ordination of human behaviour for the general good? There is of course the path of force, constraint, and coercion. But, that is hardly an attractive option. Again, there is the path of rhetorical appeal, propaganda, subliminal suggestion, and the advertising arts. But, this way of eliciting agreement is ultimately unworthy of man in violating his dignity as a thinking being. Only through rational persuasion— through appeal to his reason—can we do justice to another's rational humanity. Kant put the matter well: Nothing is so sacred that it may be exempted from this searching examination [of reason] which knows no respect for persons. Reason depends on this freedom for its very existence. For reason has no dictatorial authority; its verdict is always simply the agreement of free citizens, of whom each one must be permitted to express, without let or hindrance, his objections or even his veto.1 Only an appeal to persuasive grounds in convincing people, only a recourse to their intelligence, is a truly worthy means of obtaining their compliance. It alone acknowledges the dignity of man and assures the treatment of others as we ourselves would wish to be treated—as people who can come to responsible decisions when presented with the relevant information. The possession of rationality—our capacity to act on the basis of good reasons whose normative force we ourselves recognize as such, rather than acting wholly under the constraint of 'external' compulsions and influences—is exactly what makes us free agents. Free will is the capacity to choose and 'do as we want'—be it to heed the call of reason or to ignore it. But, it is only in heeding reason's call that we are fully free: we are 'true to ourselves' precisely when we act on the basis of reasons whose cogency we 1
Kant, Critique of Pure Reason, A739, B767.
226
The Grandeur and Misery of Reason
ourselves acknowledge. Internal compulsion—the peremptory dictate of temperament or will—is as contrary to true freedom as external compulsion. Only in heeding reason's call do we exercise our freedom and our human intellectual potential simultaneously. If somebody is to be convinced or persuaded rationally, it must of course be on grounds that they themselves acknowledge as appropriate. One recent writer sees rational argument as coercive, complaining that 'a successful . . . argument, a strong argument, forces someone to a belief'.2 But, this gets it completely wrong. Probatively cogent rational argumentation does not—cannot— take us beyond acknowledged premisses. It is wholly noncoercive, and can only succeed in leading someone where his own beliefs and his own convictions naturally take him. Rational 'compulsion' is compulsion in name only—it is a 'compulsion' that proceeds from within the orbit of our own thought-processes and their commitments, and thus is no actual compulsion at all. Reason is inherently non-coercive because it can only take us where the ruling part of ourselves, the rational part, is prepared to go. The glory of reason is that she liberates her possessors from the control of 'external' forces, endowing them with a claim to at least partial self-determination. In viewing ourselves as rational beings we lay claim to an (at least partial) freedom from the impetus of external forces outside the range of such authority as we ourselves endorse. And in seeing others as proceeding rationally we credit them with a similar status, manifesting a respect for their status as persons that we would in turn expect and welcome from them. Viewing our fellows as rational beings is a matter of hoped-for reciprocity—of 'treating others as we would have them treat us'. As Spinoza rightly emphasized, reason provides for the essential basis of man's freedom—his autonomy as an agent. The fact that we can choose in the light of what we ourselves judge to be acceptable or desirable—that we can think and act as we deem fit by the standards that we ourselves acknowledge as rationally appropriate—frees us from the burden of deeming ourselves altogether subject to the impetus of external constraints. It is in exercising our reason—and not our will, whose moving impetus is 2
Robert Nozick, Philosophical Explanations (Cambridge, Mass., 1981), 4.
The Grandeur and Misery of Reason
227
generally some 'external' drive or influence—that we realize ourselves as self-determined beings for whom the only ultimately decisive authority lies in the internal forum of our own thought processes.
1 4 . 2 THE AUTONOMY OF REASON
Once we accept that A is the rational thing to do—which may take some showing!—there is no room for any further questioning about 'Why do AT For, at this stage, the best of reasons speak for A (by hypothesis). Any search for further and different reasons (over and above those that initially assured the now-uncontested fact that A is the rational thing to do) is at this point superfluous and otiose. Once the rationality of something is indeed established, there are no extra—or supra-rational—reasons to which one could appeal for validation. The autonomy of reason is exhibited in a striking way on the practical side. For consider some of the things greatly prized by evaluative reason: understanding, peace of mind, happiness. Suppose now a path to these ends that bypasses reason altogether —a pill, say, or hypnotic influence. Reason would be discontented at their use. The attainment of even valid ends in ways that bypass reason herself and wholly ignore her claims is generally disapproved by reason. It is a feature of her claims to autonomy that she herself insists on remaining in charge—on being herself the directrix who governs the pursuit of the ends that she enjoins. Reason is willing to delegate authority, but refuses categorically to abdicate her sovereign controlling power, seeing that the abrogation of intelligence is inherently unintelligent Reason herself is our best, our only reliable, guide. She is the only competent judge—even in her own case. The point is that reason is autonomous—that even on the question of which issues fall within her province she is sole appropriate judge. Like any court of final appeal, reason herself is in a position to decide what falls under her jurisdiction. She is, and must b^, her own arbiter. Whatever the limits and limitations of reason may be, this will be something about which we will have to learn through rational inquiry. Only reason herself can instruct us on these matters; no other authority is able to speak convincingly here.
228
The Grandeur and Misery of Reason
A sceptical objection yet remains: AH this emphasis on the value of reason is problematic. For, surely our own intellectual tradition—with its heavily rational orientation—is not the only viable one. Consider such alternatives as the mysticism of the Zen Buddhist, the other-worldly religiosity of monasticism, the aestheticism of the Bohemian, the utopianism of the political visionary. Rationality does not occupy a high place in such alternative valuehierarchies, since various other values (desirelessness, self-control, godliness, attunement to the march of history, or the like) would take a superior place. How then can you ultimately justify a determinative role for the value commitments of your own particular rationalistic tradition (with its emphasis on cognitive truth and pragmatic success), in contrast to the variant values of such reason-subordinating traditions? But, this line has its difficulties. Questions about the rational appropriateness of an appeal to reason are analogous in character to the question 'Can I ever pose meaningful questions?' By the time one poses this question, it is already too late to ask. The point of no return has been passed: the issue has become academic. One has already reached a juncture where no further observations on the issue can reasonably be demanded. If it is a rational justification for valuing reason that the objector demands of us, then this consideration gives rationality a special standing of (context-relative) pre-eminence from the very outset. After all, how can someone who is prepared to join the mystic or the Bohemian in subordinating reason to other values sensibly proceed? How can one cogently defend such priorities save by reasoned argumentation? And how can one intelligently implement them save by thinking their implications through? One can certainly live a life that does not grant prominent value to reason. (No doubt about that—instances abound on every side.) But, given reason's nature, one cannot do so intelligently. A mode of life can indeed be advocated by people from the standpoint of an a-rational tradition—mystical, say, or aesthetic, or hedonic. But such a 'justification' can be cogent only in so far as it is rational. Reason's autonomy from 'external' pressures means that there just is no 'greater' or 'higher' authority to which she answers, no court of higher jurisdiction to which appeal from the decrees of reason can reasonably be made. But, to say this is not, of course, to say that man lives by reasoning alone. To see reason as autonomous is certainly not to deny that there are important
The Grandeur and Misery of Reason
229
human goods and goals outside the domain of ratiocination. Rather, it is simply to say that in so far as other human enterprises have valid claims upon us reason is in a position to discern this and to value them for it. It lies in the nature of things that 'the rational thing to do' cannot be rationally overridden—that it is just what must win out in rational deliberation. It is never sensible to proceed unintelligently. Rationality is (rationally) indefeasible. One can certainly reject or neglect reason. But, one cannot do so in a sensible, rationally defeasible way. To produce an argument against reason is already to do it homage. Reason is autonomous: no wholly alien authority is in a position to lord it over her. Rational justification —incuding the rational justification of placing reliance on reason— admits of no court of appeal whose authority is not endorsed by reason herself.
1 4 . 3 THE MISERY OF REASON
The power of reason lies in the hypothetical: 'In so far as man can achieve great things, this is so largely by virtue of his reason.' Reason's weakness lies in the circumstance that this assurance rests on a mere hypothetical—that the prospect it holds out to us is far from being a matter of assured realization, as reason herself fully recognizes. This brings us to the theme of the misery of reason. The misery of reason is rooted in the state of affairs we have characterized (in Chapter 2 above) as 'the predicament of reason'. Reason can issue no absolute guarantees. We have and can secure no assurance that doing what reason urges is in actual fact the best thing to do—that her recommendations will not actually prove counter-productive. And this means that we must live the life of reason in the full recognition that, while always and everywhere insisting on obedience to her requirements, she nevertheless can provide no certified assurance that in following her counsels as best we can, we may not actually damage rather than enhance the prospects of attaining our legitimate ends. It is the misery of reason that she can issue no guarantees—and yet must nevertheless require us to obey her demands. And so, while reason demands the recognition of her own limits
230
The Grandeur and Misery of Reason
and limitations, she is also imperious and acknowledges no external authority. Difficult mistress that she is, reason mocks us by insisting that, while no other, superior resource is available to us, we must nevertheless recognize her incapacity ever to meet fully the demands we would make upon her. Beyond reason there lies hope—an attitude of trusting expectation that is encouraged by reason even though not validated by it. On the one hand, reason sees optimistic hopefulness in her own efficacy as something that she simply cannot warrant unqualifiedly. On the other hand, she sees such optimism as an eminently desirable attitude which deserves every possible encouragement and support. Her fully recognized impotence to do more is a source of unavoidable frustration. It is a fact of profound irony that assured confidence in the efficacy of reason requires an act of faith.
Name Ijjd§x Ackerman, Robert 76 Anaxagoras 177 Aquinas, St Thomas 204 Aristotle, 13, 73-5, 90, 92,104,202, 212,219 Arnauld, Antoine 45 Audi, Robert 6 Augustine, St 174 Baier,Kurtl8,33,34n,97 Bar-Hillel, Maya 195 n Benn,S. I. 93 Bennett, Jonathan 18 n, 192 Bentham, Jeremy 110 Bernstein, Richard J. 140 Blanshard, Brand 18 n Boas, George 18 n Butler, Joseph 36 Brandom, Robert 76, 88 Cicero 45,100 n Cohen, L.J. 196 Cohen, Morris R. 18 n Cooper, John 202 Corey, Richard 215 Dahl, Norman O. 92 Darwall, Stephen 5,18 n Descartes, Rene 37 n Diogenes, Laertuis 71 Durkheim, Emile 74,145 Edwards, Ward 195 n Einstein, Albert 177, 187 Evans-Pritchard, E. E. 146 Ewing, A. C. 18 n Ferrater Mora, Jose 18 n Feyerabend, Paul K. 45 n, 57 Foley, Richard 76 Freud, Sigmund 18 n Galen 159 Gert, Bernard 97
Gilson, Etienne 87 Gosling, J. C. B. 214n Gracian y Morales, Balthasar 220 Habermas, Juergen 173 Heraclitus 73 Hintikka, Jaakko 76 Hollis, Martin 146 Hume, David 66-8, 93-6,123 Huntford, Roland 65 Hutcheson, Francis 206 James, William 53,64,68 Kahnemann, David 195 n Kant, Immanuel2,225 Kekes, John 40, 72, 147 Leibniz, Gottfried Wilhelm 7,45 n Levy-Bruhl, Julian 74, 146 Locke,John 66 Lukes, Steven 146 MacClintock, Stuart 87 Maclntyre, Alasdair 44 Massey, Gerald 184 n Mauss, Marcel 74 McCarthy, Thomas 173 McCullagh,C. Behanl47 Mill, John Stuart 114,212,215,216 Mortimore, G. W. 93 Murphy, Arthur E. 18n Nagel, Ernest 18 n Nansen, Fridtjof 65 Nathanson, Stephen 18 n, 93,222 Needham, Rodney 74 Newton, Isaac 134,178 Nietzsche, Friedrich, 95 Nisbett,R. E. 195 n Nozick, Robert 226 Ortega y Gasset, Jose 220
232 Parfit, Derek 160 Pascal, Blaise 9,224 Pears, David 200 Peirce, Charles Sanders 62, 1?3 Phillips,L. D. 195n Plato 3, 7,177,184 Popper, Karl R. 178 Price, H. H. 59,60 Putnam, Hilary 144, 156 Fyrrfio 71 Rawls, John 103 Rips, Lance 194 Robinson, E. A. 215 Robson-Scott, W. D. i 8 n Ross, Lee 195 n Sanchez, Francisco 220 Santayana, George lg n Sartre, J. P. 101 Saxe, John Godfrey 134 Schick, Frederic 11,99
Name Index Schroedinger, Erwin 177,187. Sextus Empiricus 66,170 Shakespeare William 221-2 Sidgwick, Henry 96-7,103 Simon, Herbert A. 95 Socrates 202 Solovine, Maurice 177 Spinoza, Benedictus de 226 Swain, Marshall 76 Taylor, C. C. W. 214 Tversky, Amos 195 n Unger, Peter 70 Unamuno, Miguel de 220 Walsh, W. H. 18 n Warnock,G.J.18n Weber, Max 93 Whitehead, A. N. 18 n Wigner, Eugene 177 Winch, Peter 146,161
Subject Index absolutism 139-56 acceptance 123-4 affective happiness 211-23 akrasia 204 anthropological relativism 145-56, 161-3 Averroism 87-8,155 autonomy 42-7 autonomy of reason 227-9 axiology 208-9 Azande tribe 146 C-beliefvs. P-belief78-9 calculated risks 54-60,199 circularity 42-6 circumstantiality of reason 157-75 co-ordination 193-4 cognition 3,122-5 cognitive sources 52 communication 71 conceptual egocentrism 133-56 consensus 169-75 consistency 8 co-operation 225 cost-benefit analysis 6,98-9 contentment 211-13 criteria and standards 164-9 criteriological egocentrism 144-56 cultivation hierarchies 163-75
evidential cogency 8 expected-utility calculation 118 faute de mieux consideration 36 finalities 164-9 flourishing 103-6 free agency 225-9 free will 11 frustration of reason 229-30 gambling aversion 116 goals and ends 5-6 good reason 4-6 happiness 210-23 hierarchies of cultivation 163-9 hyper-inconsistency 75 ideal rationality 27-32 idealization 172-5 inconsistency 73-91 indexing 108 indifferentism 138-43, 155 irrationalism 155, 169 irrationality 199-200 instability of optima 22-7 intelligibility of nature 176-90 interests 98-106, 112 knowledge, limits of 188-90
dialectical reasoning 83-90 decision theory 118 deliberation 13 desires and wants 15-16 economic rationality 107-18 efficiency 8 ends and goals 5-6 ends, rationality of 92-106 enthymematic reasoning 195-6 eudaimonia 104-5,212-13 euphoria 211-13 evaluation 3,12,25-7,97-106 evolution 180-8, 207
limits of knowledge 188-90 local consistency 75 logical chaos 75 market 108 maximization 107-18 measurement of value 108-11 metacriterial monism 149-56 misfortunes 57-9 motives vs. reasons 101-2,198 negative vs. positive benefit 216-18 nature, intelligibility of 176-90
234
Subject Index
needs and wants 99-106 Nuer tribe 146-7 objectivity 169-75 ontological obligation 204-9 optimalization 6 - 8 , 1 0 , 2 2 - 7 , 2 9 - 3 1 , 36-42 overriding reasons 10 P-beliefvs. C-belief78-9 pleasure 211-16 positive vs. negative benefit 216-18 practical policies 122-5 practical rationality 27-32, 39-42,64-9 praxis 3,127,202 predicament of reason 19-32 preface paradox 79 preferability vs. preference 111-15 preference and preferability 97-106, 111-15,201-3 presumptions 50-3 presumption of rationality 191-4 principle of basis equality 139 principle of sufficient reason 188-90 probability 36-9 pursuit of happiness 215-16
reasons vs. motives 101-2, 198 reason, principle of sufficient 188-90 reflective happiness 211-23 relativism 133-56,148-9,175 requisites of reason 16 retrovalidation 53 risk 54-9 risk aversion 116 rock-bottom fallacy 45 scepticism 45-54, 137-8,155,228-9 Schadenfreude, 112 self-supportingness 205 sources for cognition 52 Spanish moralism 220-1 standards and criteria 164-9 stoicism 213 stratification levels 164-9 strong inconsistency 75 systematization 16-7, 83-7,126-32 taxonomy of reason 12 truth 80-3 universality of reason 157-75 utility 107-18
quality of life 217 questions 80-1 questioner's prerogative 144
vacationer's dilemma 117 value, measurement of 108 values 98-104
rationalization 6 real possibilities 117-18 realism 176-90 reasons, overriding 10
wants and desires 5-6,97-106 Wazonga tribe 161-3 weak inconsistency 75 welfare 103