Body, Mind and Self in Hume’s Critical Realism 9783110327076, 9783110326680

This essay proposes that Hume’s non-substantialist bundle account of minds is basically correct. The concept of a person

189 89 2MB

English Pages 512 [553] Year 2008

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Acknowledgments
Note
Table of Contents
Introduction
Chapter OneSelf as Substance
(1) The Substance Tradition1
(2) The Metaphysics of Morals
(3) Morality and the Substantial Self Untied
(4) Human Nature Defended
(5) George Grant: Aristotelian Moral Philosophy Made Modern
(6) Another Sort of Mind
(7) Minds as Bundles
Endnotes to Chapter One
Chapter TwoNominalism and Acquaintance
(1) Individuation and Nominalism
(2) The Principle of Acquaintance in Locke and Hume
(3) The Appeal to Acquaintance: Empiricism vs. Descartes
(4) Hume’s Nominalism
(5) Nominalism and Relations
(6) Nominalism, Causation, Substances and Things
Endnotes to Chapter Two
Chapter ThreeFrom the Substance Tradition through Locketo Hume:Ordinary Things and Critical Realism
(1) Up to Locke
(2) From Locke to Hume9
(3) Hume’s Causal Inference to Critical Realism
(4) The System of the Vulgar as False, Inevitable and Reasonable
(5) The World of the Philosophers
(6) Conclusion
Endnotes to Chapter Three
The Disappearance of the Simple Self: ItsProblems
(1) Substance and Self in Locke1
(2) The Contents of the Humean Mind
(3) Explaining Consciousness
(4) Privacy and Other Minds
(5) The Problem of the Self in Hume
Endnotes to Chapter Four
Chapter FiveHume’s Positive Account of the Self
(1) Mind and Body
(2) The Bodily Criterion
(3) Humean Persons
(4) Becoming Our Selves
(5) Conclusion – The Final One
Endnotes to Chapter Five
Bibliography
Recommend Papers

Body, Mind and Self in Hume’s Critical Realism
 9783110327076, 9783110326680

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Fred Wilson Body, Mind and Self in Hume’s Critical Realism

Philosophische Analyse Philosophical Analysis Herausgegeben von / Edited by Herbert Hochberg • Rafael Hüntelmann • Christian Kanzian Richard Schantz • Erwin Tegtmeier Band 22 / Volume 22

Fred Wilson

Body, Mind and Self in Hume’s Critical Realism

ontos verlag Frankfurt I Paris I Ebikon I Lancaster I New Brunswick

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.

North and South America by Transaction Books Rutgers University Piscataway, NJ 08854-8042 [email protected]

United Kingdom, Ireland, Iceland, Turkey, Malta, Portugal by Gazelle Books Services Limited White Cross Mills Hightown LANCASTER, LA1 4XS [email protected]

Livraison pour la France et la Belgique: Librairie Philosophique J.Vrin 6, place de la Sorbonne; F-75005 PARIS Tel. +33 (0)1 43 54 03 47; Fax +33 (0)1 43 54 48 18 www.vrin.fr

2008 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm www.ontosverlag.com ISBN 978-3-938793-79-4

2008 No part of this book may be reproduced, stored in retrieval systems or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work

Printed on acid-free paper FSC-certified (Forest Stewardship Council) This hardcover binding meets the International Library standard Printed in Germany by buch bücher dd ag

Dedication

To the memory of Richard Popkin, philosopher and Hume scholar

Acknowledgments I should acknowledge my immense debt to Gustav Bergmann, who was a great philosopher and a great teacher. He taught me what philosophy is and how to do it. I want to recognize, too, my debt to Richard Popkin: I never had him as a teacher, but learned much from his work in the history of philosophy; and, while I have disagreed with his reading of Hume, it was nonetheless incredibly stimulating. Let me note also that the work of L. Addis on the philosophy of mind and H. Hochberg on ontology have also contributed. On Hume, there is, as always, Páll Árdal, and also the encouragement of John Davis. Finally, let me note how useful I have found John Bricke’s Hume’s Philosophy of Mind and Julius Weinberg’s Abstraction, Relation, Induction.

I have used material from my contribution to the Bergmann festschrift , “Why I Do Not Experience Your Pains,” in The Ontological Turn: Essays in Honor of Gustav Bergmann, ed. M. Gram and E. D. Klemke (Iowa City: University of Iowa Press, 1974), pp. 276-300. Used with permission of the University of Iowa Press..

Note I should perhaps point out that the present essay that develops themes that I have discussed already in my The External World and Our Knowledge of It: Hume’s Critical Realism: An Exposition and a Defence, forthcoming from the University of Toronto Press. At the same time, many of the themes of the present study, particularly those concerning Hume’s critical realism and our knowledge of the external world, so called, find a more detailed defence in this earlier study. The two studies support one another. And it is worth adding that things in both these studies find their defence in turn in my earlier Hume’s Defence of Causal Inference, also from the University of Toronto Press. Relevant also are my The Logic and Methodology of Science in Early Modern Thought (University of Toronto Press) and The Logic and Methodology of Science and Pseudoscience (Canadian Scholars Press). References to other parts of my work which support various parts of the present work, for example arguments concerning the logic of scientific explanation and arguments concerning the logical structure of scientific theories, can be found in the Endnotes to the various chapters.

Table of Contents Introduction: What Is a Person?

1

Endnotes to Introduction

16

Chapter One: Self as Substance

17

(1) The Substance Tradition (2) The Metaphysics of Morals (3) Morality and the Simple Self Untied (4) Human Nature Defended (5) George Grant: Aristotelian Moral Philosophy Made Modern (6) Another Sort of Mind (7) Minds as Bundles

17 26 34 42 69 86 94

Endnotes to Chapter One

96

Chapter Two: Nominalism and Acquaintance

103

(1) Individuation and Nominalism (2) The Principle of Acquaintance in Locke and Hume (a) Locke (b) Hume (3) The Appeal to Acquaintance: Empiricism vs. Descartes (4) Hume’s Nominalism (5) Nominalism and Relations (6) Nominalism, Causation, Substances and Things (7) Resemblance, Abstract Ideas, the Distinction of Reason, and the Simplicity of Things

106 108 108 114 121 124 130 139 151

Endnotes to Chapter Two

182

Chapter Three: From the Substance Tradition through Locke to Hume: Ordinary Things and Critical Realism 191 (1) Up to Locke (2) From Locke to Hume (3) Hume’s Causal Inference to Critical Realism (4) The System of the Vulgar as False, Inevitable and Reasonable (5) The World of the Philosophers (6) Conclusion

191 196 214 224 231 241

Endnotes to Chapter Three

246

Chapter Four: The Disappearance of the Simple Self: Its Problems

251

(1) Substance and Self in Locke (2) The Contents of the Humean Mind ( i ) The Intrinsically Mental ( ii ) Neither Mental nor Material (3) Explaining Consciousness (4) Privacy and Other Minds (i) Parallelism: Hartley, Hume and Something of Its History: Why It Is Believable (ii) Parallelism and Interactionism (iii) Parallelism: Further Exposition (iv) Ownership of Mental States ( v ) Privacy (5) The Problem of the Self in Hume

255 257 258 273 283 312

Endnotes to Chapter Four

361

Chapter Five: Hume’s Positive Account of the Self

375

(1) Mind and Body (2) The Bodily Criterion

375 395

316 322 328 335 348 352

(3) Humean Persons (i) Hume on Norms: His Account of “Ought” (ii) Character and Personal Identity (iii) Knowledge of Oneself Conclusion; or, perhaps not ... (4) Becoming Ourselves (5) Conclusion – The Final One

403 405 424 431 463 471 488

Endnotes to Chapter Five

496

Bibliography

507

Index

535

Introduction What is a Person?

Hume gave a certain answer to the question posed by the title of this Introduction, an answer that has long been controversial, and rejected by many commentators. However, what we will be arguing is that this rejection has often been the result of a failure to read Hume’s Treatise carefully. Moreover, where it has been a reasoned rejection of Hume’s position, it has, unfortunately, often been based for the most part on premises deriving from the substance tradition, even though Hume offered a devastating critique of that tradition. There are certain commonsense things that everyone can agree with concerning persons. In the first place, persons are individuals. And secondly, they are natured individuals: every person is not just a person but has a distinctive character or identity as a moral person, as a Scot, and/or as a librarian, and/or etc. The substance tradition offers a certain account of these facts. Hume criticizes this account and offers an alternative. We must discuss the account, the criticism, and the alternative. We shall discover that the criticisms and the alternative are generally sound, and the substance account, and therefore the criticisms based upon it, unsupportable. This is not to say that Hume’s positions are problem-free: to the contrary, they are not. But other thinkers in the empiricist tradition offered solutions that help to defend Hume’s central claims. For example, John Stuart Mill made significant points about our knowledge of our selves and about the logic of our inferences to other minds, and Russell made points about relations that are absolutely central if Hume’s view of persons is to be defended. The central proposition about persons is that A person is an individual that has a certain character.1 A character is what a person is, that is, it is his or her identity. It is a configuration of capacities, that is, a configuration of passions, abilities, aptitudes, proclivities, powers, habits, virtues, vices, tastes, sentiments and temperaments, all understood as dispositions, that is, as analysable into ‘if...then’ causal connections. Among the important dispositions are the social roles that we play, butcher, baker, stock broker, professor, bank teller, bank depositor, borrower, lender, employer, employee, promisor,

Introduction

promisee, waiter, parent, gangster. These features constitute the self as a bearer of rights and privileges in civil society and in the subgroups that go to form civil society. Beyond this are other features of the person, e.g., his or her being a Scotsman, raised near Edinburgh, a speaker of English, a member of the Hume family, one capable of understanding, and maybe, to an extent, feeling the sentiments of a Calvinist – those features that give his or her self a continuing (though changing) identity, rooted in his or her social and psychological being. Often there are certain dominant traits that organize the others. Thus, the miser is dominated by the passion for money; his various capacities tend to be exercised in a way that helps him systematically to realize his passion. Each person can be identified as a certain unique configuration of dispositional traits. This unique character forms the nature of the individual; a person’s name picks him out not only as an individual but also as having a certain character, self and identity. This linkage of the individuality and the character of a person was noted by Hume when he pointed out that, for socialized persons, “our reputation, our character, our name are considerations of vast weight and importance” (T, p. 316).2 In the first instance a person’s character determines his or her response to events and experiences; in the first instance it is one’s character that determines one’s responses to social and environmental conditions, rather than those conditions determining one’s character. To know a person’s character is to know what sort of life is best suited to actualizing his or her real potentialities. But persons can also grow and develop. They can not only respond in character to events in their environment but their characters can themselves change in response to those events. Of course, some capacities are innate; these innate aptitudes, dispositions and instincts form the very core of a person. For it is within the framework of this core that the history of the individual is worked out. But these innate capacities are for the most part capacities to form capacities, capacities to acquire habits; and the history that is worked out includes the acquisition of further capacities and habits. Dispositions are, of course, actualized on the occurrence of certain events; and different events will lead to the acquisition of different habits. On the same core framework of innate dispositions, many different histories are possible, many different sorts of person could be produced, many identities, depending upon external events over which the mind or person often has no control. People thus come to have very different characters, and to be very different individuals, that is, as we said, unique

2

Body, Mind and Self in Hume’s Critical Realism

identities, unique configurations of dispositions. But dispositions as such can belong to different individuals. All the many samples of sugar are soluble, all bankers tend to do bankerly things; all misers share the disposition to be miserly, and so on. Even configurations that are unique to a person could, in principle at least, be had by more than one. A person not so much is but has a unique character, an individual character if you wish, but besides this is also an individual. What is it, then, that individuates these unique characters into the persons that we are and that we know? What is the principle of individuation of persons? What gives them their identity? Dispositions find their actualization in facts and events. What distinguish the dispositions that constitute one person from those that constitute another person, then, are the events which are their actualization. In other words, what distinguishes persons from each other, what individuates one from another, what gives them their identities, are the sequences of concrete facts and events that constitute their histories. When a disposition is actualized, there is an event which is the occasion of its exercise, and the event which is its actualization. In the case of first order dispositions, the event which is its actualization is a concrete event. This disposition may be physical-mental as when we come to perceive something. The capacity to come to know objects by sense is a capacity. The occasion of its exercise is (usually) a sensible event or object that causally interacts with our sense organs. The disposition is actualized with regard to a particular event or object, which event or object evokes a perceptual awareness of the event or object the object or event occurs at a particular time and place, as does the awareness of the event or object – the awareness is itself a particular event that occurs at a particular time and place, different from, but generally near to, the event or object of which it is an awareness. Or again, the disposition might be mental-physical. Thus, we have the capacity to raise our arms. The occasion for the exercise of this disposition is (often) a volition, and the exercise of this disposition is the arm’s actually going up. And there are also mental-mental dispositions, the occasion and exercise of which are both mental, as, for example, when I set out to solve a problem “in my head,” or strive, and then succeed, in remembering some forgotten date. We also have second-order dispositions, that is, dispositions to acquire dispositions. Some events in the history that constitutes a person

3

Introduction

consist in the acquisition of first-order dispositions. Such an event is the actualization of appropriate second-order dispositions. This may be a momentary process, the occasion a single event; cutting myself with a knife I become disposed to infection by germs. Or the occasion may itself be a long history, as when I acquire the dispositions characteristic of being a skilled typist or a skilled pianist. Among the capacities that a person has is that of rational deliberation. Given a certain end, a person can deliberate about the means to achieve that end, and choose the one that he or she prefers. Given several ends, a person can reason about them, and schedule their pursuit. A person can even choose among the ends which one is to be pursued. Persons can also adopt attitudes towards their own activities and those of others, preferring some, being repelled by others, morally approving some, disapproving others. One can even approve and disapprove of the desires that others have, and, even more importantly, of the desires and attitudes that one has oneself. One deliberates to answer not only question of “How shall I achieve this?” but also “Should I achieve this?” and even “what should I achieve?” and even “What sort of person am I if I decide to do this? Do I want to be that sort of person?” Our critical, reflective and deliberative capacities may thus lead to the revision or rejection of an end that we previously had. It can also lead to the acquisition of ends that we previously lacked. In deciding to become a parent, I come in the end to have a wide variety of ends, passions, sentiments, purposes, goals that I previously lacked. Some of these are acquired involuntarily, as consequences of the history I go through in becoming a parent; others are deliberately cultivated. Thus, “What shall I do?” is connected in the end with “How shall I live?” and, even more deeply with “What sort of person shall I be?” I not only have a character but in addition I can pose the question of my character in a normative way, “What ought I to be?” that is, what sort of person ought I to be. Or at least, some persons, some times, raise that sort of question. To be able to raise the question, deliberate on it, answer it, and act upon that answer is itself a capacity, and as such a feature of character. It is, moreover, a feature that some have to a greater degree than others. A person who is unable to reflect critically upon his or her own ends lacks a measure of depth and seriousness; such a person we think of as shallow. We see, thus, that some persons have the capacity to change their own characters; there can be self-development. Those who can do the latter have a special character trait, that of being able to reflect upon and to self-

4

Body, Mind and Self in Hume’s Critical Realism

consciously monitor their own selves and change themselves in ways that they self-consciously determine. At least some persons are individuals who are responsible for themselves. They are not merely the subjects of desires, of passions, of mental and physical abilities, and of traits of character. They are also beings who are able to reflectively monitor themselves; they are able to pose the de jure question about whether the self they discover themselves to be is the sort of being they ought to be; and if they are not, they are able to modify their characters to conform to the standard that this moral desire sets for them. In Heidegger’s terms, man is “das Seiende, dem es in seinem Sein um dieses selbst geht,”3 that is, a person is the sort of individual for whom the question arises about what kind of being he or she is going to realize. Or, as Frankfurt has put it, there is in persons a “capacity for reflective self-evaluation...manifest in the formation of second-order desires,” that is, desires to the effect that we want to be moved by certain desires, and in the formation of “second-order volitions,” volitions determining which first-order desires move one to action.4 As we noted above, the same core of native tendencies can provide the framework for many different histories, and therefore many different individual persons. Each of us can imagine how things would have been different if certain events over which we have no control had been different. What if that accident that almost killed one’s parents had actually proved fatal and that one had as a consequence been orphaned and then raised in a very different family? What if the person you married had after all decided to stay in his or her native city rather than moving away to yours? One can imagine the differences, and perhaps even in one’s day dreams live and enjoy them, the persons one might have been. In this context, one’s actual character may seem quite unessential, almost like one of the various social roles that one takes on and then gives up as the occasion and one’s interests require. Here one might compare Zeno Vendler’s point that I, in this case Zeno Vendler, this man whom I am, could not be the individual that I am if I had been born at any other period or of any other parents than my own. Yet I can imagine living in a different world, at a different time, e.g., as Hannibal watching the battle of Cannae, or as Claudius. But the impossible cannot be imagined, so it must be possible that I, Zeno, be Hannibal or Claudius. But Claudius has one individual essence, Hannibal another, and Zeno still a third. So the ‘I’ that is the subject of such a transference “has no content and no essence; it is a mere frame in which any picture fits; it is the bare form of consciousness.”

5

Introduction

The transcendental ‘I’ is not a thing...The transcendental aspect of my being consists in nothing else but in the realization that I, as a subject of experience, am only contingently tied to the senses of this body,...in one word, it consists in my ability to perform feats of transference.5

What is crucial in all this is the self that determines which role one takes on, which individual character one actually is. But this is not merely the set of innate dispositions with which we are born, but the self with the more or less well developed capacity for determining the self that one is and is to become. And it involves the capacity to imagine being other than one is, imagining as a possibility that one might become or imagining for the sheer joy of fantasizing. The self that one is, is created by oneself (though not solely by oneself) out of a past. This past is accessible to that self through memory – though we should also recognize three things: first, that, through causal factors that we cannot control, much of the past, and even parts of our own past, are inaccessible to us; second, that the past that we know is in part shaped by our present selves, that we are, in other words, very much responsible for how we regard the past, conceptualize it, and assimilate it; and third, that even when we disguise the past, it may nonetheless still influence us. But the self that fashions a self out of the past has a certain aim, a certain, more or less vague, vision of the self that it aims to be, or rather, to create. As Williams has put it, “an individual person has a set of desires, concerns, or as I shall often call them, projects, which help constitute a character,” and a person has a character “in the sense of having projects and categorical desires with which that person is identified.”6 In other words, a person can identify with certain ends, often second-order ends, in that he or she regards his or her pursuit of these ends, or pursuit in certain ways, as crucial to his or her life, to defining what he or she is to be as a person. And if the person is successful, then he or she becomes the sort of person defined by the ends with which they have identified. If what one is constitutes one’s identity – and that, after all, is the point of the traditional formula for identity: two things are identical just in case they have all their properties in common x = y : : (f)(fx fy) A – then the character that one identifies with becomes through the self acting out of a certain past the character which constitutes the identity of

i

. ‘– >’ represents the material conditional and ‘’ the material biconditional.

6

Body, Mind and Self in Hume’s Critical Realism

the self. The self creates itself out of its past and out of the vision that it has of itself. The self with the vision of itself provides the unity that links the past to the self that is created and for which it is responsible. The unity of the self is not simply something that we discover but also something that we create. The agency of self-consciousness, our capacity to reflect upon ourselves and deliberate about the sorts of persons we shall be, extends from the past into the future in the activity of creating the unified self that each of us is. Persons are thus self-determining individuals, changing but unified as one person throughout the process of change, responsible not only for the consequences of their actions but for themselves, shaped by a central core of being, a reflective self-consciousness, capable of determining exactly who it is they are. This concept of a person is, as Amélie Rorty has put it, the idea of “a unified centre of choice and action, the unit of legal and theological responsibility.” This concept of a person is central to Christian belief. Again as Rorty puts it, “If judgment summarizes a life, as it does in the Christian drama, then that life must have a unified location.”7 This concept had been wed for the mediaeval philosophers to the doctrine of substance that they inherited from the ancient world. This coupling of the two was effected by Boethius who held that “persona...est naturae rationabilis indiuidua substantia”: a person is the individual substance of a rational nature.8 Hume accepts this picture of the self or person as a natured and unified individual, determined to be the sort of person he or she is by his or her own reflective self-consciousness. But at the same time he rejects the substantialist account of the self that had been inherited by the modern world from the mediaeval world by way of Descartes. However, his own picture of the self leaves him curiously dissatisfied. He is reasonably satisfied with his account of causation. He is reasonably satisfied with his account of body. But when it comes to the self, he is, as he records in the famous “Appendix” to the Treatise, unable to arrive at a conclusion that he regards as fully satisfactory philosophically. The problem, in its basic outline is clear enough. What Hume argues is that the self is a “bundle or collection of different perceptions which succeed one another with an inconceivable rapidity and are in a perpetual flux and motion” (T, p. 252). This seems incompatible with the idea that the self is a unity, and that that unity derives from a centre of reflective self-consciousness. Unable wholly to resolve the difficulty he is left dissatisfied.

7

Introduction

Unlike many of his critics, however, Hume does not take this seemingly unresolvable inconsistency as a reductio of his position; more reasonably than those critics, he takes it instead as a problem that remains to be solved. His attitude is that one does not throw up one’s hands and abandon a well defended position at the least difficulty; rather, one puts it aside and hopes that as oneself or one’s students and successors work on the problem it will in fact prove to be solvable. Or, perhaps, since this is philosophy, dissolvable. In fact, as Hume sets it up, there are several problems. One is the problem of one’s identity in the sense of one’s character, the properties and dispositions that one and others take to define oneself as a person. The problem is that one is not born with these characteristics, most of them at least, though some of them appear naturally as one develops from an infant to an adult: and if one does not have these characteristics as part of one’s biological nature, how does one acquire them? through what sort of learning process do they come to form what one is? Then, second, there is the problem of the nature of the self as a being that is morally responsible for one’s actions. This, too, seems to be, in part at least, a matter of learning: infants are not responsible, we teach them so that they become beings responsible for their actions, and, one hopes, beings that accept responsibility for their actions. Third , there is the problem of the self as a centre of consciousness. What is the nature of this consciousness? and how does it come to be a centre? Fourth, related to this, there is the problem of self-consciousness, in the sense of reflective self-consciousness: how is it that one can reflect upon oneself, and one’s states of consciousness in particular? what is the nature of mind that allows for this? And how, in the light of such reflection, can one allow oneself to do certain things, and not do others, and indeed how can it be that one can shape one’s own character? (This last relates, of course, to problem one, above.) Fifth, there is the problem of continuity: what does one mean when one and others take one to be today the same person that you were yesterday? (This relates to the second problem, above: how can one today be morally responsible for what you, presumably the same person, did yesterday?) All these issues receive a simple solution in the substance tradition On this tradition, the person is a simple substance, an entity that endures through change. This substance is active: through this activity this

8

Body, Mind and Self in Hume’s Critical Realism

substance brings about for changes that occur in its history, these changes are actions for which it is responsible. The substance has a certain nature or form or essence: this nature determines the sort of changes that can occur in the substance. Its form or nature is its species; this species determines its natural ends, the way towards which it is striving to be, and the direction in which, by virtue of that form, it is striving. The form or essence explains the direction of change which occurs in the substance; the explanation is therefore teleological. The species that a substance has it has of necessity, if (per impossibile) a substance loses its species or form then that substance ceases to be (Socrates is human but if Socrates ceases to be human he ceases to be). The species which a substance has will fall under a genus, which gives those ways of being which it shares with other species under that genus. What distinguishes one species from that other species under a given genus is the specific difference that defines the species. The real definition of the species is given by its genus and its specific difference. The real definition is not only logical but gives the ontological structure of the substance, a structure which it has as a matter of necessity, ontological necessity. Thus, according to Aristotle, the world consists of substances of various sorts or species, and in particular a person is a substance the nature or essence of which is rational animal: the real definition of a person is rational animal, where animal is its genus and rational is its specific difference. A person grows or develops, as a plant grown or develops – that is how it actively moves itself. Just as, by its nature an acorn grows into an oak tree, so the infant through its nature grows into an adult. All animals share this feature of form with plants. But animals can further move themselves about; they are capable of local motion. This is the genus: substances capable of growth and local motion. Persons are animals in this sense. What distinguishes them from other animals is the fact that they are rational; this is their specific difference. To be rational means that such a substance can grasp the reasons for things, that is, the reasons for things being as they are. Since the form or essence of a substance is the reason why it acts as it does, persons as rational can grasp these forms which are the reasons for things behaving as they do. Indeed, this determines its natural end as an animal that aims to know: knowing reasons is its telos. Among substances, however, a person too has a form and this form can itself be grasped by a person. A person is therefore not only conscious insofar as it is a knower of things, it is also self conscious, aware of its own being, grasping the species which is its form, as it grasps the

9

Introduction

species or forms of other things. And thus, knowing its own natural ends, it not only acts to be in those ways determined by it nature, but knowing the reasons for other things, it can choose the means that is best for those ends. But being an imperfect knower it can sometimes so choose that it frustrates its natural ends. Descartes is part of this substance tradition. A person is selfconscious, Descartes discovers in the cogito, when he becomes aware of his own thinking, that which turns out to be the reason for his being, which turns out as it were to be simply thinking: I think, therefore I am. In fact, Descartes argues, thinking is the very form or essence of a person: I am, he discovers, a thinking thing (sum res cogitans), a substance the essence of which is to think, that is, to come to know the reasons for things. Thus, where Aristotle held that a person is a rational animal, Descartes argued that, whatever a person’s relation is to his or her body, the essence of that person is thinking, and thinking alone, pure thought or pure reason. Mind is thus sharply distinguished from body: it is consciousness, pure consciousness, which can also be self-conscious. For both Aristotle and Descartes, the problems that we located above receive a simple solution in terms of the substances which persons are. These philosophers do give different accounts of the essence of human being, but, important as those differences may be, this difference is a mere detail – that is a mere detail compared to what Hume has to say about persons. Hume argues that there are no substances; one does not find them in our ordinary experience of the world and so one cannot admit them into any ontology that can pretend to be reasonable.. There are therefore also no forms or essences, nor are these the reasons for things. Explanation ceases to be teleological. Things happen but there are no substances to make them happen; there are patterns in which things change, but these patterns do not arise from the activities of an underlying substance. There are, clearly, no natural ends, that is, ends determined as a matter of necessity by some form or essence. To be sure, acorns still grow into oak trees but this development is not a matter of the striving of a substance moving itself towards its natural end; it is rather just the way things happen in the world in which we find ourselves. The same holds for persons: there are conscious states, just as there are the sensible appearances of things, but they not to be accounted for by appeal to an underlying substance, Moreover, just as for ordinary things, so for persons, explanation ceases to be teleological: there are simply patterns in the way thing happen. To be

10

Body, Mind and Self in Hume’s Critical Realism

sure, a person often has a vision of what he or she ought to be, and acts to become and to be as this vision would have he or she be. But that vision of the self is not the same as an end determined by a metaphysical nature, as in the substance tradition. The aiming at such an end is simply an event among events, and the explanation of it as a cause or an effect is simply in terms of a pattern, and certainly is not teleological as in the substance tradition. What, then, accounts for the continuity of a person? What, indeed, is a conscious state? How is it that it can be a state of knowing? What is the connection it has to things it knows? How can it be self-conscious? If there is no substantial continuity, how can a person be held morally responsible, responsible for, say, something done yesterday? Does not moral responsibility disappear along with the substances that Hume so easily eliminates? Is that not an argument against such elimination? If explanation is by patterns, then of course the question is, what are those patterns, what is the learning theory that describes the acquisition of those traits of character that defines the person’s identity? But more deeply, if explanation is by patterns then what actually links the past to the present and the present to the future, links them is such a way that one can be held responsible for what one did in the past and responsible for what one becomes and what one is in the future? And if there is no linkage then why should you be concerned (as you in fact are concerned) with the pain or the pleasure that will be suffered by the being that simply follows but is not really the same as you, that is, the you as you are at present? Though Hume rejected the doctrines of the substance tradition, the answers he gives to questions such as these cannot be understood apart from the substance philosophy which he is rejecting. His approach to the problem of the self and of personal identity was in fact shaped by the long tradition of substance philosophy. In the immediate background are Locke and Malebranche. Beyond them are Hobbes and Descartes, and the Cambridge Platonists who reacted to both of the latter. And beyond all of these are those from whom the substance tradition derived, as well as the Christian tradition concerning the soul that had come down to the early modern period. These sources are, of course, Plato in the first instance, and then Aristotle who perfected the substance tradition, and then Plotinus who added a very important point concerning the self. It is this tradition, the substance tradition, with which we must begin. Many of the problems that we shall encounter as we try to defend Hume’s account of the self are ontological. For example, in the earlier

11

Introduction

tradition, the enduring substance provides the relation that unites the various states of consciousness which a person has into a whole. But if there are no substances what, then, provides the tie or ties that bind together the states of a person so that they are the states of one and the same person? Hume, however, has problems articulating an ontologically tenable account of relations. In defending Hume’s account of the self we shall therefore have to spend considerable time dealing with the ontology of relations. Since there are no substances, one cannot divide the events of the world into patterns according to the species of substance that underlie and account for the patterns among these events. What, then, is the distinction between mind and body? Descartes, in one sense, has no problem with this distinction: there are minds, which are thinking things, and there are bodies, which do not think but are rather simply extended things. But he does have a problem with regard to how one substance effects a change in another substance, and therefore a problem in particular with regard to mind-body interaction: how can mind affect body and body affect mind? Hume does not have the clarity and sharpness of the distinction between mind and body that one finds in Descartes. We shall therefore have to spend considerable time on this issue: what precisely are mental states? Nor can we avoid the other Cartesian problem, how do mind and body interact? – though Hume does avoid the problems about interaction created by the substance doctrine for those who accept that tradition, we shall have to discover how he actually avoids those problems. And there is another problem which the substance tradition never did really address. That is the problem of the privacy of our mental states and, with that problem the problem of how we know the mental states of another person, which are as private to that person as our own mental states are private to us. We shall try to look at this issue from the perspective of the Humean account of mind. As for the patterns that describe and explain human behaviour, these are essentially those of the psychological theory of learning that Hume defended, the theory of associationsim.9 We shall have rather a lot to say about this theory. Suffice it here to say that it explains as any empirical scientific theory explains, by subsumption of the events to be explained under a general pattern; or, in other words, human behaviour and human consciousness is explained in a way that is in effect logically equivalent to the way in which science explains stars or stones or rainbows or oysters, by way of general patterns (and not teleologically).

12

Body, Mind and Self in Hume’s Critical Realism

Here in outline is how our discussion will go. Chapter One discusses the substance tradition and attempts to separate the concept of a self from the notion of a substance. It is argued in particular, through an examination of some aspects of the thought of the Canadian philosopher George Grant, that the issues are not simply metaphysical, but go well beyond these to connect with some current moral controversies: the differences in these controversies, while apparently moral, are in fact metaphysical, and are not to be settled at the rather superficial level of consequentialist vs. deontic considerations – what would Mill say? what would Kant say? we ask our undergraduates – but rather can only be dealt with seriously at the deeper level of ontological discourse. It makes a real difference then, one with practical relevance, whether one accepts a substantialist or a Humean view of the self. Now, nominalism is an ontological view that the substance tradition passed on to its successors. This includes Hume. But the nominalistic account of relations deriving from the substance tradition is ontologically inadequate, or at least it is not adequate for the philosophy of mind that Hume aims to defend. Chapter Two examines nominalism and the account of relations that it passed on to early modern philosophy. The chapter attempts to show why it is inadequate but also to show that there is an alternative ontology of relations that avoids the problems created by the nominalistic account but is also compatible (as the nominalistic account is not) with the basic empiricist principles on which Hume’s attack on substances, and his account of the self, both depend.10 Chapter Three discusses the development of the empiricist account of ordinary things out of the substance tradition. The line of thought goes from Locke to Hume, and culminates in a critical realism that is defended by Hume. This is, of course, a contested interpretation of Hume, so some time is spent on simply defending that reading of Hume.11 But the main point of the discussion is to give Hume’s account of body, the sort of body to which Hume argues that mind is related. Chapter Four discusses some of the problems that arise for the philosophy of mind and any account of the self once we have the disappearance of the simple, substantial self. What are the contents of our

13

Introduction

minds? how do we account for privacy? how can inferences to other minds be justified? are among the issues discussed. Chapter Five suggests that Hume’s discussion of the self is not merely negative: he also has a positive account. It is the task of this chapter to uncover and defend this positive account of the self and of what it is to be and to become a person. A person is, of course, a body, or so we shall see Hume arguing, but also more than a body: one’s identity as a self or person is in large part a matter of one’s character. This chapter discusses how Hume has the person acquire these structures definitive of his or her identity through the mechanisms of association and through the mechanism of sympathy. The chapter ends with an examination of Hume’s reply to the sceptic, one who in particular calls radically into question the existence of a self.

This will no doubt leave undiscussed many issues both in Hume and in the philosophy of mind But I do think that it will emerge that Hume’s view of the self, what is commonly referred to as the “bundle” view, can be given a reasonable defence. It is to be hoped that at the very least the usually dismissive allusions to Hume and the “bundle” view such as the following will become a thing of the past. Hume, like Locke, is an empiricist. He observes that we have a concept of a person and spends his efforts trying to explain how we could come to have this fiction through ideas produced ultimately from sensory input. That’s right; he does not think there are persons in Locke’s sense. For Hume, there is just a powerful fiction that persons exist. He reduces the metaphysical problem about persons and personal identity ... to psychological problems about how we could possibly be so deceived as to think persons exist and endure through time. It is hard to beat Hume for iconoclasm... 12

As we shall see, Hume does not deny that there are persons nor that these persons endure through time. What he denies is that persons are simple substances. He denies the views of such philosophers as Plato and Descartes, but not the commonsense that there are persons and selves, that these have certain identities, and that they can be said to endure through time. It is one thing to deny an ontological account of the self, another to deny the commonsense that there are selves; Hume does the former, not

14

Introduction

the latter. The denial of the ontological claim that the self is a simple substance is based on his argument that we have no simple idea of such a substance, which we would have to have if the notion of mind as a simple entity were to make sense. (All our ideas are derived from impressions, he tells us, and uses this principle as a guide to ontology: we have no impression of the self as a simple substance so we have no idea of such a self.) But to deny that we have no simple idea of the mind or of the person is not to hold that we have no idea. To the contrary, though we have no simple idea we nonetheless still have an idea of the self; it is just that it is a complex idea. But, having a rough notion of what philosophers are like, I expect there will continue to be those who hold that Hume denies that there are persons, that privacy is a real problem, that it is reasonable to be sceptical of other minds, that we can’t say what are the experiences of a bat, and so on. I do hope, however, that in what follows I have made it a bit harder to defend those views.

15

Introduction

Endnotes to Introduction

1.On a number of the points in the next few paragraphs, see C. J. Ducasse, Nature, Mind and Death, Ch. 17. I cite Ducasse on this topic because his views are contrary to Hume’s. If one can handle Ducasse’s notions about a person within a Humean framework, then the defence of the Humean concept of a person will be that much stronger. 2.References are to David Hume, Treatise of Human Nature, ed. L. A. Selby-Bigge, revised P. H. Nidditch.. They will be given in parentheses in the text, by “T” and page number or Book, part and section. 3.M. Heidegger, Sein und Zeit, in his Gesamteausgabe, Band 2, Sec. 9, p. 57. 4.H. Frankfurt, “Freedom of the Will and the Concept of a Person,” p. 7. 5.Zeno Vendler, “A Note to the Paralogisms,” pp. 117-118. 6.B. Williams, “Persons, Character and Morality.” 7.A. Rorty, “A Literary Postscript: Characters, Persons, Selves, Individuals,” p. 309. 8.Boethius, The Theological Tractates, p. 92. 9.For this theory, see F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill; also “Some Controversies about Method in Nineteenth-Century Psychology,” and “Mill and Comte on the Method of Introspection.” 10.More on these ontological points, about empiricism and about relations, is explored in greater detail in F. Wilson, Acquaintance, Ontology and Knowledge. 11.In fact, the interpretation has been given an extended defence in F. Wilson, The External World and Our Knowledge of It: Hume’s Critical Realism, an Exposition and a Defence. 12.Warren Bourgeois, Persons: What Philosophers Say about You, Second Edition, p. 145.

16

Chapter One Self as Substance (1) The Substance Tradition1 The substance metaphysics, as Vlastos2 and Turnbull3 have argued, can be found in outline in the “natural philosophy” that Socrates (Plato) outlines in the Phaedo. In this dialogue, Socrates expresses dissatisfaction with the natural philosophy of the physicists. In its place he proposes an alternative scheme for explaining changes in the sensible world. The problem is that sensible events, so far as sense can tell, are distinct and separable. Within themselves, so far as sense can tell, there is no reason why one sensible event should follow another; because they are distinct, that one event should be followed by such and such rather than so and so is apparently a matter of accident, apparently reasonless and irrational. But as Socrates argues, there is a reason why his being imprisoned is being followed by his drinking hemlock rather than by his running off to Thebes: it is because he is striving to be just, that is, striving to imitate in his outward life the ideal form of justice; it is his striving to be just that necessitates that his being imprisoned will be followed by his drinking the hemlock. The striving to imitate the form provides the connection between the sensible events that is not apparent in the events themselves. Moreover, to the extent that Socrates’ actions really do resemble the form, they constitute an instance of justice; they resemble justice and are therefore named after it. There are, then, three sorts of entity in this metaphysics of explanation. There are, first, the sensible characteristics. There are, second, the forms. And there are, third, souls. Sensible characteristics are in souls; souls strive after forms; and finally, sensible characteristics resemble or participate in the forms. Souls are individual or particular; and sensible characteristics are as particular as the souls they are in. (Sensible characteristics are thus what some have called “tropes.”4) Forms, in contrast, are such that several souls can strive after any one given form, and such that the sensible characteristics in several souls can all resemble the same form. In this respect the forms are universal, rather than particular. Sensible

One: Self as Substance

characteristics are in time, and come to be and pass away. All these entities are simple; in themselves they do not change. Change is understood in terms of the relations amongst these simple entities. Change is constituted by the coming to be and the passing away of sensible characteristics in souls; souls are thus simple entities which are continuants. As simple and therefore unchanging in themselves, souls are permanent; they endure through change. Forms, in contrast to sensible characteristics, are not in time; they are eternal, timeless, unchanging beings. And souls, while they are, like sensible characteristics, particular, are like forms in that they are outside the sensible world of time and change; like forms, they are unchanging and eternal, that is, while in their extrinsic sensible appearances they are in motion or change, nonetheless in their intrinsic nature as the souls that move things they are unchanging and eternal. They are simple unchanging centres of eternal activity, eternal motionless activity, that accounts for the motion and change that we observe by means of our senses in the temporal world of ceaseless flux in which we carry on our ordinary existence. That souls must be simple and unchanging is clear, given the pattern of explanation. Complexity implies separability, and separability requires explanation. Explanation is in terms of a soul striving to imitate a form. If, then, the activity of a soul were separable from part of itself, the union of the two would have to be explained by appeal to the activity of another, deeper soul. Since the activity of a soul is, upon the supposition, separable from the soul, this deeper soul would be separable from its activity, requiring an explanation in terms of a still deeper soul. And so on. Thus, allowing that the soul is separable from its activity leads to an infinite regress in which nothing ever gets connected to anything else. That is, it would be a regress in which nothing was ever explained. Thus, if the metaphysics of explanation is to succeed as a scheme upon which something actually is explained, then the soul cannot be separable from its activity. Plato recognized this much; it is, in effect, the basis of the argument for immortality in the Phaedo. But a similar argument from a regress of unsuccessful explanation attempts results from the separation of the form from the soul. These are separate, according to Plato, and are therefore separable. But then that separation requires explanation. There are two men, the sensible Socrates and the ideal form man. The sensible man is explained by the soul Socrates striving after the ideal form. But that form is separate from the soul. Why, then, is it that Socrates strives after the ideal form of humanity rather than

18

Body, Mind and Self in Hume’s Critical Realism

the ideal form of doggieness? If this is to be explained, then one must introduce a further form, a third man; the soul Socrates striving after the third man explains why Socrates strives after the ideal of humanity. And so on. Again one has a regress in which nothing is ever explained, unless, that is, one ends the separation of the forms from souls. This was seen by Aristotle, who made the soul inseparable from the form; the telos or goal of an activity is always an intrinsic part of that activity. Socrates apart from his humanity simply does not exist. In fact, in Plato the forms have a dual role. As Aristotle puts it,5 “the Forms are causes both of being and of becoming” (Metaphysics, 991b3). Sensible things resemble each other because they resemble the forms. This solves the ontological problem of sameness: sensible things are called by the same name because they both resemble one and the same form. But forms also contribute to solving the causal problem of order: sensible events occur together or in the sequence they do because the soul in which they exist is striving after a certain form. Aristotle rejects forms as the solution to both problems. The causal role of the separate forms is replaced by the causal role of the inseparable natures or forms of intrinsically active substances. The ontological role of forms in the problem of sameness is replaced by the assumption that the resemblances among sensible characteristics are simply primitive. The likeness and unlikeness of substances is a matter of the qualities that are predicated of them (Categories, 11a15-19), and qualities may resemble each other in various degrees (Cat, 10b25-28). On this scheme, sensible characteristics are intrinsically individual, that is, they are, as for Plato, tropes, and two of them, say the white in Phaedo and the white in Socrates, are called by the same name because there is a simple relation of resemblance, either exact or inexact, between them. For Aristotle, the soul, or, as he calls it in his modified ontology, the substance, that is, the simple continuant, and its categorial feature of being an informed or natured activity, are not given in sensible experience. Their introduction is justified by Aristotle by what has subsequently come to be called a transcendental argument. Such arguments have as their conclusion that such and such non-presented entities exist; the argument for this is to the effect that if these entities did not exist certain things that we all know to be true could not be true. Substance as a simple continuant must exist because otherwise we could not speak of change: ...if change proceeds from opposites or from intermediaries, and not from all opposites (for the voice is not-white [but it does not therefore change to white]), but from the contrary, there must be something underlying which changes into

19

One: Self as Substance

the contrary state; for the contraries do not change. Further, something persists, but the contrary does not persist; there is, then, some third thing besides the contraries, viz., the matter (Met, 1069b3-8).

Genuine change also requires that the substance have an active potentiality for change. This Aristotle argues against the Megarian school. “There are some who say, as the Megaric school does, that a thing ‘can’ act only when it is acting, and when it is not acting it ‘cannot’ act, e.g. that he who is not building cannot build, but only he who is building, when he is building” (Met, 1046b28-31). But, ...if that which is deprived of potency is incapable, that which is not happening will be incapable of happening; but he who says of that which is incapable of happening either that it is or that it will be will say what is untrue; for this is what incapacity meant. Therefore these views do away with both movement and becoming. For that which stands will always stand, and that which sits will always sit, since if it is sitting it will not get up; for that which, as we are told, cannot get up will be incapable of getting up. But we cannot say this, so that evidently potency and actuality are different (but these views make potency and actuality the same, and so it is no small thing they are seeking to annihilate)...(Met, 1047a 1020).

The form or nature is not present in a substance, as the sensible characteristics are, but rather it is predicated of the substance. The nature or form is therefore similar categorially to qualities, except that it is not separable from the substance nor is it known by sense. It is known, rather, by a kind of non-sensory intuition: it is intuition which is “the originative source of scientific knowledge” (Posterior Analytics, 100b15). Now, if reason is the faculty that enables us to understand, then it is clear that reason must be the faculty that grasps the form or nature of the thing whose motion we understand, since to understand is to be able to explain. Reason must therefore grasp the reasons for things, that is, upon the Aristotelian scheme, the natures of the changing substances. Thus, If thinking is like perceiving,...[t]he thinking part of the soul must therefore be, while impassable, capable of receiving the form of an object; that is, must be potentially identical in character with its object without being the object. Mind must be related to what is thinkable, as sense is to what is sensible (De Anima, 429a13-18).

The Aristotelian metaphysics of explanation, defended a priori by transcendental arguments, in this way imposes upon philosophy a certain

20

Body, Mind and Self in Hume’s Critical Realism

corresponding account of reason, an account in which reason at its most basic consists in the intuitive grasp of the natures of things. With one further twist, it was this view of the world and mind that came down to the early modern period. It was, for example, the view of Descartes. The difference between Aristotle and Descartes did not lie in whether there were substances or not. Nor did it lie in whether substances had forms or natures or essences. Here the difference was the philosophically minor one of how many forms there were – a great many as Aristotle said, or but two (or three, if you include God) as Descartes said. Nor was the disagreement over whether reason consisted in the rational intuition of the forms or natures of things. Rather, the disagreement was over the process which had such intuitions as its product: was it one of abstraction, as the Aristotelians held, or were the intuitions innate, to be called to consciousness when the appropriate stimuli were present, as Descartes held? These differences were, as it were, “within the ring” as compared to the later critics like Locke and Hume who challenged the whole substance tradition. The crucial twist to this tradition to which I referred was given by Plotinus.6 The materialists argued that the world consists of matter in motion. The soul is the origin of the motion of the body, but it too is material. Plotinus argued that this was impossible. Matter, as lifeless, cannot move itself; in fact it could not even stay together in a particular configuration: “...body in itself could not exist in any form if soul-power did not; body passes; dissolution is in its very nature; all would disappear in a twinkling is all were body.” Indeed, “Matter itself could not exist [without soul]: the totality of things in the sphere is dissolved if it be made to depend upon the coherence of a body which, though elevated to the nominal rank of ‘soul’, remains air, fleeting breath..., whose very unity is not drawn from itself” (Enneads, IV, 7, 3). Soul, as cause, contains within itself the forms or reasons of the things the temporally ordered diversity of which it produces. This produced order reflects in time the timeless order implicit in the cause: “If the leading principle of the universe does not know the future which it is of itself to produce, it cannot produce with knowledge or to purpose; it will produce just what happens to come, that is to say by haphazard. As this cannot be, it must create by some stable principle; its creations, therefore, will be shaped in the model stored up in itself; there can be no varying...” (Enn, IV, 4, 12). Moreover, our consciousness is in the first instance, as indeed the materialists hold, a series of events in time. However, each of these events is related to the others in the series; and they are

21

One: Self as Substance

related moreover to each other as modifications of a single consciousness, a consciousness which is a consciousness of each of them and all of them. “There can be no perception without a unitary percipient whose identity enables it to grasp an object in its entirety” (Enn, IV, 7, 6). The conscious self of which the events are the modifications cannot be the series as a whole, since within the series the events are successive; but at any moment the consciousness of those events does not involve a succession: “...prior and past are in the things it [soul] produces; in itself nothing is past; all...is one simultaneous grouping of Reason-Principles” (Enn, IV, 4, 16). That is, while the consciousness of the series is a consciousness of a before and after, within that self-consciousness there is no before and after; within the consciousness there is no such relation between the components as there is between the events of which it is the consciousness. The consciousness which a self has of itself must lie outside the temporal sequence of events of which it is conscious. Self-consciousness must therefore be an entity outside the temporal changes of the self, an eternal entity to which the events in the temporal sequence are related as modifications. There were already more than hints of this in the Phaedo: the soul which accounts for changes is Socrates, that is, the self that is conscious of its own self, and which is, in the light of that consciousness and in the light of its knowledge of the forms, is determining the changes which take place within it. There were also more than hints of it in Aristotle. A substance is a continuant but never simply that: it is also a continuant of a certain sort, i.e., it has a certain nature or form. This form has implicit within it, so far as the substance determines its own history, what has happened to it and what will happen to it: the form or nature is “the source from which the primary movement in each natural object is present in it in virtue of its own essence” (Met, 1014b19-20). Thus, the form as the active potentiality of the substance is the whole of the substance as eternally present to its various parts, the events the sequence of which constitute its history. In particular, the soul must be a substance. Some hold that the soul is divisible, and that one part thinks, another desires. If, then, its nature admits of its being divided, what can it be that holds the parts together? Surely not the body; on the contrary it seems rather to be the soul that holds the body together; at any rate when the soul departs the body disintegrates and decays. If, then, there is something else which makes the soul one, this unifying agency would have the best right to the name of soul, and we shall have to repeat for it the question: Is it one or multipartite? If it is one, why not at once admit that ‘the soul’ is one? If it has parts, once more the question has to be put:

22

Body, Mind and Self in Hume’s Critical Realism

What holds its parts together, and so on ad infinitum (De Anima, 411b5-14).

And so the soul, in this respect like any substance, is the atemporal source of, and eternally present to, those events that constitute its history: “...the soul must be a substance in the sense of the form of a natural body having life potentially within it” (De An, 412a20-1). Aristotle’s soul, as a simple active potentiality or form, thus contains within it in an atemporal way the series of events that constitute its history in time, as least insofar as it is self-determining; so also self-consciousness contains this history within itself. Not surprisingly, Plotinus, and the substance tradition that followed, identified the centre of self-consciousness with the atemporal active substance which causes the temporal series of events unified in that self-consciousness. These patterns remain appealing. They can be found, for example, in a study by Albert Shalom.7 He points out that when one tries to think of the relation of mind and body, one has not only the two relata, mind and body, but also a third entity, to wit, the consciousness that is thinking the two. It is the ignoring the latter that creates the insolvable problem of the self. In thinking of the mind and body as two but related entities, “...I have failed to realize...that in the very act of setting out the problem in this way, it is I, the subject, who am formulating it in these terms, and that this has implications of its own. The implications are that when I, the subject formulate the problem in this way, I have projected myself into the referents of the terms which I have used in that formulation – and I mistakenly assume that I, who am doing this, am absorbed within the framework of those referents themselves” (p. 411). There is a point to this remark, as we shall argue below (Chapters 4 and 5): when one is conscious of x, the centre of consciousness is not among the contents of that conscious state. In that sense, as Shalom also insists, the I, the centre of consciousness, is not “an observable entity, element, or process in the flow of material or psychological processes” (p. 414). At the same, time, however, what Shalom does not see, it does not follow from this that the centre of consciousness, that which makes x the content of a conscious state cannot become itself the content of a further conscious state, and to be, in that sense, observable. But Shalom makes a different inference, that of Plotinus, that the problem of multiplicity, that is, as he sees it, the duality in appearance of mind and body, cannot be solved without the appeal to a unitary subject: “...the compound called ‘body and mind’ can only itself exist as a theoretical derivation of the continued intellectual activity of that ‘I’ which does not merely think itself as ‘I’ but which constantly expresses itself as ‘I,’ thereby indi-

23

One: Self as Substance

cating its own priority relatively to the construct ‘body and mind’” (p. 412); or, as he puts it later, “...there is a basic unitary reality – called, indifferently, ‘I,’ ‘person,’ or ‘human entity’ – and that terms like body and mind are derivations from the basic reality and by that basic reality” (p. 419). As we shall argue, following Locke, Hume, and Mill, this Plotinian inference does not follow: there is no reason why the unity of consciousness cannot be due to a part of the whole. Nonetheless, for all that, Shalom makes the Plotinian inference that this is not possible: “...the constant ability of an ‘I’ to be aware of ‘itself’ throughout all bodily changes and throughout all the changes of subjective experience, is a problem which simply cannot be dealt with in terms of that framework.... ‘identity’ must be understood not as ‘sameness’ in the course of change, but as ‘identity’ throughout change” (p. 449). So there is a self that is outside the entities that it unites. It is for this reason that the self is not observable; it is not just that we are not conscious of it at a moment, but that it is the sort of entity that cannot be among the contents of consciousness: it is something “nontemporal” (p. 466) rather than something which is “the same over a span of time” (p. 480). This non-temporal self unites the contents of consciousness – it provides “a permanence of timelessness” (p. 481) –, and gives them structure, through its own activities: “...the operations of the constituents [the contents of consciousness] are determined by an existential dependence of which we are unaware” (p. 431). This activity is the actualization of an active “potentiality of subjectivity” (p. 445); what we have is a process of “constant change subtended by personal identity” (p. 447). Mind is not “a mere succession of subjective states” but rather “the unceasing development of subjective potentialities” (p. 451). The self “is actualized as a process of self-conscious awareness from its initial existential status of quasi non-temporal potentiality” (p. 466). Shalom thus sees himself, quite correctly, as following in the line of “an older and, I believe, wiser tradition [in which] persons were sometimes referred to as ‘souls’” (p. 413): his views on the self are of a piece with those of Plotinus. So Shalom follows Plotinus in identifying the centre of selfconsciousness with the atemporal active substance which causes the temporal series of events unified in that self-consciousness. One further implication of this ought to be noted. On this identification, the selfconsciousness which contains past and future within it is a consciousness of the active potentiality, the form of the self, which is the causal source of that past and future. But that active potentiality or form is the reason why those events occur in the sequence they do; it is that form that constitutes

24

Body, Mind and Self in Hume’s Critical Realism

the necessary connection between those events. Thus, the selfconsciousness is the consciousness that grasps the reasons of things. But the faculty that grasps the reasons of things is reason. The selfconsciousness that grasps the reason for its own being and becoming is thus reason – Reason – itself. Reason thus comes to be identified with the atemporal and substantial centre of self-consciousness. Shalom holds this position. The patterns that we observe in the world of appearances are to be explained in terms of the underlying reality of subjectivity, that is, active potentialities the actualization of which is the “explanatory key” for understanding those empirical patterns (pp. 433-5). So subjectivity explains, it constitutes the reason why the patterns are what they are. But at the same time, this subjectivity is consciousness (p. 443), it is the “actual experiencing” of things (p. 439), it is a self that “conceives itself” in certain ways (p. 437), it “thinks” itself as a permanent self (p. 412), and this is an “intellectual activity” (ibid.) – the subjectivity which is the self is, in short, that which attempts to understand things, to grasp the reason for things. That reason, however, that which explains appearances, is itself. The self as the reason that aims to understand must grasp itself as the reason that explains, yields understanding. Reason – capital “R” – is thus the atemporal and substantial centre of self-consciousness. It is precisely this tradition that we find summarized in the Cartesian cogito, which is at once reason and also a centre of self-consciousness and also a simple substance the activities of which cause the events, e.g., the doubtings, which are the events in the history of that consciousness. Of course, not everyone in the tradition accepted the separation of mind and body upon which Descartes insisted. Shalom, for one, so argues against Descartes. Consciousness is not something that a body has, something merely added on; rather, consciousness is expressed in behaviour (p. 410). Body and mind, that is, the inter-related structures of bodily and mental states, are both the common expression of the subjectivity that underlies and creates both in their inter-relation: “...the compound called ‘body and mind’ can only itself exist as a theoretical derivation of the continued intellectual activity of that ‘I’ which does not merely think itself as ‘I’ but which constantly expresses itself as ‘I,’ thereby indicating in act its own priority relatively to the construct ‘body and mind’” (p. 412). Thus, Shalom disagrees with Descartes in holding that mind as we experience it is embodied or incarnate, and body as we experience it is activated by consciousness. Nonetheless, behind this complex “body and mind” lies the subjectivity, the active potentiality, that is at once the reason that explains

25

One: Self as Substance

this complex and also the reason that aims to grasp this reason, structure the movement of “body and mind” toward this end. So the Reason as understood by Shalom is of a piece with that of Descartes; the only difference between them is whether this Reason creates only mind (Descartes) or (Shalom) “body and mind.” The disagreement here is still within the ring of the Plotinian framework. It is this framework that Locke and Hume are going to challenge in its totality. Specifically, they will argue, Locke rather tentatively, Hume decisively, that there are no substances. This means, in the first place, that the notion of what it is to understand the empirical changes that we observe will change. One will no longer search for the reasons for things by searching for entities. It means, as part of this change, that the very notion of reason itself will shift. Reason will not be a capacity to grasp entities, the entities – substances – that constitute the reasons for things. And it means, second, that the account of self-consciousness, that identifies this with a soul or power that transcends time and place, will be abandoned. Persons will still be rational animals, but what a person is and what rationality is will have completely changed. It is that change that we want to examine. (2) The Metaphysics of Morals One aspect of the substance tradition still needs to be discussed. That is how the tradition provides in the nature of forms of things an ontological basis for morality. To this we should now turn. But it is already clear that if substances, and therefore forms or natures, disappear, then so does any objective ontological ground for morality. Not only will there be no objective necessary connections, so also there will be no objective basis for morality. Now, it is necessary to be clear on the point that the explanation scheme of the substance tradition is anthropomorphic. It is, to be sure, not crudely anthropomorphic; Aristotle is not committed to saying, for example, that stones deliberate, and in fact he explicitly denies that (Physics, 199a20ff). Nonetheless, he takes activity to be the basic explanatory category, and the model for activity is clearly volition and the goal-directed activity of men. As one commentator has put it, “Appeal to a dunamis is fundamental, for Aristotle, in the way that appeal to a law of nature is taken to be fundamental by modern philosophers of science such as Carl Hempel”8 or, for our purposes, David Hume. Here the idea of the empiricist is that one explains events by subsuming them under a covering law or regularity. The regularity is a general fact about events in the world of

26

Body, Mind and Self in Hume’s Critical Realism

sense experience; it constitutes a timeless pattern among those entities. Thus, where Aristotle explains the events of the world of sense experience by appeal to a timeless entity outside the world of sense experience, Hume (and Hempel) explains the events of the world of sense experience by appeal to a timeless pattern in the world of sense experience. For Hume, since the activity which is explanatory for Aristotle is in itself totally outside the realm of sense experience, explanations in terms of it are vacuous. Indeed, the whole language of capacities is for Hume to be so reconstructed that it turns out, as Molière also said, that capacities – dormitive powers or whatever – never in themselves explain, and that they seem to do that only because laws are associated with their use. One defender of the Aristotelian position has stated that “No objection ought to be raised to the idea that a capacity can be explanatory.”9 But the empiricist does raise such an objection, and it is in fact the crucial point of disagreement between him and Aristotle. We have thus far largely concerned ourselves with events the explanation of which consists of their being produced by the natural activity of the substance in which they occur. This is the case of immanent causation. But events can also occur in a substance which are caused by another substance acting on the first. This is transeunt causation. All art in which an artist modifies a second substance involves transeunt causation. Now, some cases of transeunt causation may involve one substance bringing about in a second a motion or change that is contrary to the natural motion of the latter. Such “unnatural” motion is sometimes called “violent.” Thus, each of the simple bodies called elements has a natural motion; e.g., earth moves naturally in a straight line towards the centre of the universe. But such substances may also exemplify upward motion; in such a case it is not so much that the substance moves, i.e., moves itself, but that it is moved, i.e., is moved by another (cf. De Caelo, 300a20ff). Such motion is unnatural or violent, or, as Aristotle also says, “constrained”: “the constrained is the same as the unnatural” (De Caelo, 300a24). It is in this context that Aristotle introduces his two senses of ‘necessary’: Necessity...is of two kinds. It may work in accordance with a thing’s natural tendency, or by constraint and in opposition to it; as, for instance, by necessity a stone is borne upwards and downwards, but not by the same necessity (Post Analytics, 94b38-95a3; cf. Metaphysics, 1015a20ff).

The sense of ‘necessity’ ties in with Aristotle’s view of ‘possibility.’ For something to be possible, it must be more than merely “conceivable”

27

One: Self as Substance

or “logically possible”: it must have the capacity to exist. That is, a “possibility” is a “natural tendency” or a “natural power”, that is, one of the tendencies, powers, or capacities that is part of the nature or form of a thing. Aristotle conceives these strivings as successful unless prevented by some external force. Thus, Aristotle writes of a man who potentially knows a certain subject: “When he is in this condition, if something does not prevent him, he actively exercises his knowledge; otherwise he would be in a contradictory state of not knowing”; and he goes on immediately to make a similar point about ordinary physical objects: “In regard to natural bodies also the case is similar.” Thus what is cold is potentially hot: then a change takes place and it is fire, and it burns, unless something prevents and hinders it. So, too, with heavy and light: “light is generated from heavy, e.g. air from water (for water is the first thing that is potentially light), and air is actually light, and will at once realize its proper activity as such unless something prevents it” (Physics, 255b3-11). He puts it more abstractly elsewhere, that where there is a capacity for something to occur, it will occur; “he who says of that which is incapable of happening either that it is or that it will be will say what is untrue; for this is what incapacity meant” (Met, 1047a12-14). For, he asks, what is there to prevent a possibility from coming to pass, unless it is impossible? (Motu Animalum, 699b29). Thus, if p is an unnatural event brought about by an external force, it is necessary, in the second sense of necessary. But if p is unnatural then it is contrary to some natural state q which is prevented by the external force from happening. So not-q is also necessary. But if it is necessary that not-q, then, since the necessary is that which it is not possible that not, it follows that it is not possible that q, i.e., that q is impossible. In the other sense of ‘necessary’, however, it is the natural motion that is necessary. For it flows from the nature of the substance, and the nature or form of a substance is inseparable from it; that is, the substance has its form necessarily, and therefore has necessarily the characteristics which that form or nature causes to come to be in the substance. Motion which is natural is free, unconstrained motion. In one sense, therefore, motion which is free is necessary: it is free because it is caused by the unconstrained striving of the very substance which is moving; in such motion, the substance is a self-mover and in moving itself unconstrained by anything external its motion is free. Thus, when an earthy object is unsupported, it moves in a straight line downwards towards the centre of the universe; such motion is free fall. Conversely, motion which is unnatural is forced, constrained motion;

28

Body, Mind and Self in Hume’s Critical Realism

that is why it can be said to be violent. In this sense, then, unnatural motion is necessitated, forced upon the substance and contrary to the goal which the substance itself, through its active nature, is striving to achieve: in such motion the substance does not move itself but is moved by another. In modern science, the notion of force is a dead metaphor. It means mass x acceleration, and any suggestion of coercion plays no role; nor does the notion of a self-mover, as in the case, still spoken about in this way, of the free fall of heavy objects. For Aristotle, however, there is nothing metaphorical at all about these ways of speaking: the basic anthropomorphism of dynamic natures ensures that the talk of freedom and spontaneity in the case of natural motion, on the one hand, and force, constraint and violence in the case of unnatural motion, on the other hand, has a perfectly legitimate place. Aristotelian science must be contrasted to, and not confused with, modern science, or, what is the same, science in the empiricist sense. The central point of contrast, however, is perhaps the fact that the regularities that are to be reckoned lawful by the Aristotelian criterion of being regularities necessitated by the nature of the thing, that is, regularities caused by the striving of a thing in accordance with its nature, can have exceptions. Violent motion is always such an exception.10 It is this category of violent motion that establishes the contrast to modern science and to empiricism. For what modern science aims to do is to explain matters of empirical fact by appeal to exceptionless regularities. The point of difference is not merely one in the philosophy of science, whether explanation is by mere matter-of-fact regularities or by regularities that are grounded in natural necessities. From the perspective of Aristotle, there is no point to the general concern of modern science to discover exceptionless regularities. From the perspective of Aristotle, where explanation consists in appeals to natures, the central aim has to be the discovery of those natures, and not the discovery of exceptionless regularities. Thus, given the aim that the Aristotelian metaphysics lays down for science, much of the activity of modern scientific research is useless. To be sure, it may serve other aims – e.g., manipulative – but it does not yield scientific knowledge and therefore does not serve the natural cognitive goal of man of coming to know the natures of things. From the viewpoint of the empiricist, it is, of course, Aristotle’s anthropomorphism that opens up the way to this divergence between Aristotelian explanations and those of modern science. It is this same anthropomorphic content that opens the way to the le-

29

One: Self as Substance

gitimacy of the use of the legal analogy in the case of natural laws. In modern science this, too, is a dead metaphor: what science seeks in order to explain are exceptionless regularities about matters of empirical fact. Since they are exceptionless, they are not lawful in the Aristotelian sense, for, as we have just seen, when he introduces the notion of violent motion Aristotle allows that regularities which he would reckon necessary and therefore lawful can have exceptions. Moreover, scientific laws are mere regularities; they have no normative implications. For Aristotle, in contrast, a regularity which is a natural law, that is, one caused by the activity of the nature of the thing, is also normative, describing how the thing ought to behave. Greek legal terminology was already applied to nature by Anaximander, who said of the elements that “they make amends and pay reparation to each other for their injustice, according to the ordinance of time” (fragment 1). Here a certain general rule is implied – encroachments by one element will be counteracted. But, in the first place, this rule is not exceptionless, nor, in the second place, is there any attempt to represent nature in all its detail as subject to law. The Sophists, however, were later to contrast nomos (law or convention) and phusis (nature); the former is subject to cultural variation, and is therefore less regular than the latter. When, still later, Plato combined the words nomos and phusis to speak of a law of nature (nomos tes phuseas), he was speaking with no doubt deliberate paradox. In the Gorgias (483e), Callicles argues that when the stronger subjugate the weaker, though they are not following a law made by us, they are nonetheless following a law, namely, a law of nature. And in the Timaeus (83e), Plato remarks that sometimes our blood is replenished not from food and drink, as happens for the most part, but otherwise in a way contrary to the laws of nature. It is the Aristotelian notion of nature that can make sense of this. The second Platonic example would simply be a case of violent motion. In the first case the appeal would be to a human nature – except, of course, both Plato and Aristotle would argue that Callicles, like Thrasymachus, has a false view of human nature, that there are more impulses to human nature than the urge to acquire power in order to be able to satisfy as many of one's lower desires as possible. This example makes clear how the link-up between law in the normative sense and nature is effected: it depends upon nature in the Aristotelian sense being teleological, that is, an activity intrinsically directed towards a goal. For, “the good [is]...that at which all things aim” (Nichomacean Ethics, 1094a2-3; cf. 1076a3).

30

Body, Mind and Self in Hume’s Critical Realism

Now, there is for persons more than one end, and we choose some of these for the sake of others, the final ends. It is these that define the “chief good” (1097a27). This chief good is sought for its own sake (10974a34), and happiness consists in the achievement of this final end, or, what is the same, the achievement of what is desired for its own sake (1097b1). Human activities serve a certain function; ultimately they serve the final ends which man has (1097b25ff). These ends include the nutritive (in common with plants and animals), sensitive (in common with animals) and rational, which is peculiar to humans (it is the difference which defines the species). In man the other ends function to serve the differentiating end, which regulates them (1098a1ff). One may serve these ends poorly or excellently; virtue consists in serving them excellently (1098a15-17; 1106a15-16). Thus, “the virtue of man...[is] the state of character which makes a man good and which makes him do his work well” (1106a22-3). Since in regard to human appetites one can have an excess, a defect, and a correct intermediate (1106b15-17), virtue is the habit, or state of character, of acting for the good, and where this is a matter of choice, it is the choosing of the means which is determined by rational deliberation to be the one which best satisfies our ends, i.e., as it turns out, the mean or intermediate option in the case of each appetite (1106b36-11072). As for deliberation, “We deliberate not about ends but about what contributes to ends” (1112b12); we deliberate about means, not about ends. We may, of course, deliberate about intermediate ends, but if in our search for means to bring these about we find one cannot be achieved, we give it up: “if we come upon an impossibility, we give up the search, e.g. if we need money and this cannot be got; but if a thing appears possible we try to do it. By ‘possible’ things I mean things that might be brought about by our own efforts...” (1112b24-27). The reason for this is simple enough: “choice cannot relate to impossibles, and if any one said he chose them he would be thought silly...” (1111b21-22). Now, the final ends we desire for their own sake. But, as G. E. Moore taught us to ask, are these ends desirable, not in the sense of being capable of being desired but in the sense of being worthy of desire? Ought we to desire those ends? Ought they to constitute the good of human being? To raise this question is to begin to deliberate about these ends. That is to treat them as intermediate ends relative to some further standard which is final. It is also to treat them as things which it is possible, through our efforts, to change. But the final ends for which human being strives, that is, the ends at which the human telos is directed, are determined by human nature itself. And this nature or form is something that each of us

31

One: Self as Substance

has as a matter metaphysical necessity. It is therefore not possible to change the final ends of man. And if it is impossible, we give up trying to find out how to achieve it. It is simply pointless or silly to pretend to have ends beyond the final ends determined by human nature. So the question, are these desirable, in the sense of being worthy of desire is a question for which it would be silly to answer in any way but the affirmative. Kant has made us aware that “ought implies can”; if we can’t be something or somehow it is pointless to say that we ought not to be that way. The converse of this principle is that “must implies ought”.11 Its point, not surprisingly, is the same as that of Kant’s principle: it is pointless to say about what we must be anything other than that we ought so to be. Aristotle’s claim that human nature determines a human’s good is, in effect, a conclusion based upon this “must implies ought” principle. As a matter of metaphysical necessity, we have the nature we have. That is how we must be: we must be the sort of person who aims at the ends which, by our nature, we desire for their own sake. Since we must be the sort of person who has those final ends, it is pointless to hold other than that we ought to be that sort of person, i.e., that the ends that we naturally desire for their own sake are also desirable for their own sake in the sense of that they are worthy of desire. If virtue is, as Aristotle says, the habit or state of character of acting for the good, then striving as we must for our natural ends is virtuous. In effect, then, the “must implies ought” principle is none other than the piece of common wisdom that it is only reasonable to make a virtue of necessity, and that it is only reasonable in the face of necessity to make of it a virtue because not to do so – to say that what must be is either forbidden or an alternative is permitted – would be at best pointless and at worst frustrating and painful – as trying to do the impossible so often is. It should be emphasized that what we do of necessity in the sense in which we are now speaking, is not done under compulsion – the other sense of ‘necessity’ – but freely. As we emphasized before, moving oneself to achieve one’s natural ends is a necessary motion – since it is a motion that derives from the nature or from that which is necessarily and inseparably part of us – but it is also a free motion – since it is not constrained by any external force. It thus turns out that natural patterns of behaviour are also patterns of behaviour that we ought to exemplify. Regularities which have natural necessity are also patterns of how we ought to behave. Regularities which are natural laws in the descriptive sense and which are therefore relevant

32

Body, Mind and Self in Hume’s Critical Realism

to explanation are also natural laws in the legal sense and are thus also relevant to moral evaluation. Aristotle’s metaphysics of explanation thus not only justifies the legal metaphor but in fact shows that the evaluative terminology is more than metaphor. The metaphysics not only justifies certain patterns of explanation but also provides the metaphysical foundations for the natural law position in ethics. This is the central point of the metaphysical ethics of Aristotle. Two points are important. First. It establishes the connection between the necessary ontological structure of the universe and the moral order of things. Given the connection between being and value that is here established what we established is the identity of (perfect) being and value. Furthermore, the form or nature is expressed in the sensible appearances of things, and is therefore the standard as to whether those appearances represent what the thing truly is in its essential nature or form. The form or nature is that standard of truth. So we also have the result that (perfect) being and truth coincide, as do truth and the good. Value and truth are rooted in the very being of things. It is worth noting that Aristotle’s teleological vision of the universe in which each entity in it has a soul with its own necessary telos, enables the legal terminology to be extended quite reasonably from the human case through animals and plants to even the elements. The hierarchy of souls/forms is thus also a hierarchy of values. Second. In establishing the connection between being and value, Aristotle’s metaphysics establishes that virtue is not a matter of convention, but is grounded in the ontological structure of the universe. We have in the natures of things, and in the nature definitory of human being, an objective standard of value by appeal to which we can decide which social conventions are true, that is, truly express the natural law. The relevant social conventions define human relationships, declaring which relationships are those that we ought to cultivate and declaring which are those that we ought not to cultivate and, indeed, ought to suppress. Since most virtues are acquired habits, the social conditions for their development cannot be ignored. This the Nichomachean Ethics notes at its beginning (Bk.I, ch.2) and it ends (Bk.X, ch.9) with a transition to the Politics, and Aristotle’s ethics is, as is Plato’s, in the end inseparable from his political science. But this is surely true of any moral philosopher. In this respect Hume surprises us no more than Aristotle or Plato.

33

One: Self as Substance

(3) Morality and the Substantial Self Untied In the early modern period it was Hobbes that first challenged the Aristotelian picture of the universe. Hobbes argued that the norms of civil society – the norms of property, of contract or promising, and of allegiance to the chief magistrate – are purely prudential. The standard view, just examined, deriving from Plato and Aristotle, was that these norms are rooted in a human essence or form: they are part of human nature. The point is metaphysical: any individual with this essence aims, as a matter of metaphysical necessity, at the ends determined by that essence, where this essence or form is a metaphysical reality not given in ordinary sense experience of the things of the world but presented in a sort of rational intuition. And since the essence or form determines how we necessarily aim, it determines what we ought to do. The essences are not only descriptive but normative: the values required by civil society are built into our very being and into the ontological structure of the universe. All things have a built in teleology determined by their essence or form, but human beings have further capacity, that of becoming aware of or conscious of essences: we are rational creatures and are able to become aware of these reasons for things. Among the essences of which we can become aware are our own; this is an awareness of our necessarily determined ultimate ends. About these, we, as rational beings, do not deliberate, but we can deliberate about means to these ends. Among the ultimate ends, sanctified by the ontological structure of the universe, are the norms of civil society: simple self-awareness will tell us that much. This had been the background in Richard Hooker. But Hobbes invoked what is in effect the empiricist principle, that if something is not given in sense experience or explained in terms that are, then that thing does not exist, it is not to be part of one’s ontology. This eliminates the essences or forms of things from the furniture of the universe. But in doing that there is also an elimination of the human essence or form, and of course our awareness of such things. That is, there is the elimination of both the moral structure of the universe, including the norms of civil society, and that part of human nature that aims us towards those things as ultimate ends. Human beings simply are not fit for civil society, nor do they have any moral sense. All action is prudential, towards ends that have no intrinsic moral value. This includes the rules of civil society: we conform ourselves to their dictates and the dictates of the sovereign because by so doing we will jointly rescue ourselves from the worse fate of the state of nature, of the “war of all against all.” Civil society is justified by the fact

34

Body, Mind and Self in Hume’s Critical Realism

that it is useful, not because it is intrinsically good, justified by the moral structure of the universe. Such was Hobbes’ vision of human being. It immediately drew criticisms from those with, as they saw it, a vision of human beings as beings which are no more than animals.12 The relativity of the rules of the social order, according to Hobbes, was clear. Thus, Bishop John Bramhall, in The Catching of the Leviathan, argued that Hobbes got himself in effect into a contradiction in arguing that propositions about what is right and wrong may be “true” in one society and “false” in another. He prayed to God to save us from such times as would entertain such a doctrine, arguing with greater sarcasm than cogency: “into what times are we fallen? When the immutable laws of God and Nature are made to depend upon the mutable laws of mortal men: just as if one should go about to control the sun by the authority of the clock.”13 The problem was Hobbes’ inaccurate account of human nature: for Bramhall, Hobbes “could not have vilified ... human nature more than he doth.”14 George Lawson argued that Hobbes “makes all men brutes, nay wild and ravenous beasts, and birds of prey, until they have made themselves into slaves unto some absolute sovereign, and such they must be, either beasts by the law of nature, or slaves by the law of a civil state.”15 Lawson to the contrary accepted the older metaphysical view of human being and human nature, that it includes an inclination towards civil government: the quarrels which we find in civil society are not due to nature but to its corruption.16 William Lucy argued that Hobbes, by fixating on the worst characteristics of human beings, rather than their best, in effect turned each person into “an incarnate devil, acting on those things which we abominate.”17 There are, to be sure, moral monsters, just as there are physical monsters; but one ought not to base generalizations on the worst cases. The overwhelming majority of people can be friends, one to another, just as Adam and Eve were friends (p. 38) and as persons marooned on an island would be companions (p. 48), in each case drawn together by their “common interest ... in humanity” (p. 148), following principles imprinted in the hearts of human beings by God independently of, and prior to, the activities of any sovereign (p. 148, p. 158). James Lowde attacked the Hobbesian concept of human nature, accusing Hobbes of making the anti-social parts of human nature for the whole; he suggested that Hobbes was like a troublesome fly “always busy about the sores of human nature.” The problem lay with Hobbes himself; his introspective analysis made his own case a model for all: “because the bloodshot eye of one man’s mind represents all things in red colours; therefore [for Hobbes] must cruelty [appear to] be the ... universal dictate of na-

35

One: Self as Substance

ture.”18 Important among the Hobbes critics that Hume was addressing in his moral philosophy were the Cambridge Platonists. These philosophers had gone back to ancient neo-Platonism to re-introduce notions of sympathy into British moral thought to reply to the egoism and social atomism of Hobbes. Moral philosophical responses to Hobbes were proposed by the Cambridge Platonists who found the tools for their reply in Plotinus. Central here was the doctrine of the self as a simple substance. This had entered the Christian world through Boethius. It has its roots in Plato’s Phaedo, however. In this dialogue, as we have seen, Socrates is asked to provide an account, in a way that Anaxagoras cannot, why his (Socrates’) being in prison is going to be followed by his drinking the hemlock rather than running off to exile in Thebes. Socrates is going to drink the hemlock, he has explained in the Crito, because of his duty as a citizen. The structure of his actions is derived from the teleological unity that his activity provides. The activity is that of a simple entity, the soul, which in Aristotle became the substance. The activity of the soul is determined by the form. Socrates is acting to exemplify as perfectly as he can in his outward appearances the form that determines his ends as a human being. Striving for these metaphysically determined ends defines the moral being of Socrates. The teleological unity is achieved by the form of the activity on the one hand and the fact that the activity is that of a simple entity, the soul or substance, on the other. As Lucretius makes clear,19 the ancient Epicureans held a view of the self that is of a piece with that of Hume: on empiricist grounds, that is, by appeal to ordinary experience of sense and internal awareness (a Principle of Acquaintance), they argued that the self that we know through inner awareness is a changeable whole consisting of a series of parts each of which is causally tied to a state of the body – in Hume’s metaphor, a “bundle of perceptions.” The mind, argued Lucretius, “which we often call the intellect, the seat of guidance and control of life, is part of a man, no less than hand or foot or eyes are parts of a whole living creature” (p. 99). The mind is a material thing, he argues; this is proved by the interconnection of mind and body: what happens in the one always influences what happens in the other (pp. 100-101). Perception consists in “replicas or insubstantial shapes of things (p. 131) being present to the mind. Plotinus, in the Enneads, argued that this view of the self and of knowledge is mistaken. He argued that without a unitary soul, there can be

36

Body, Mind and Self in Hume’s Critical Realism

no perception: the Epicurean doctrine that held that perception consisted in having images failed to account for the unity of consciousness. There can be no perception without a unitary percipient whose identity enables it to grasp an object as an entirety.

The various senses yield different perceptions, any one object by itself gives different perceptions. However, “there must be some central unity to which [all senses] report.” The lines of perception converge on a point which unifies them: This there must be, as there is a centre to a circle; the sense-impressions converging from every point of occurrence will be as lines striking from a circumference to what will be a true centre of perception as being a veritable unity (Enneads, IV, vii, 6, p. 346).

What is perceived is a multiplicity. If the awareness of this multiplicity is an idea which manifests the same multiplicity as the object, then consciousness will be divided into the multiplicity of the idea. There will be a consciousness of each part of the object of perception, but no consciousness of the whole. In order to achieve a unity of consciousness, in which the perceiver is conscious at one moment of each and all of the parts, the perceiving consciousness must be a unity. We cannot have an isomorphism of idea and impression, as the Epicureans would have it, but rather we must have a coordination of the complex impression into a unity. We do not have lines running from the various parts of the perceived to the various parts of the perceiver but rather lines from the perceived to a point, as all the radii of a circle converge on the dimensionless and single unified central point. The unity of consciousness in perception requires a simple entity, the soul, as perceiver. The argument for the simplicity of the soul from its own selfawareness of the multiplicity within itself is simply a variation on this point about the unity of consciousness. The argument for the unity of consciousness generates the argument for the identity of the self throughout the complexity of its own being. Plotinus held, quite correctly, that we are beings capable of reflection upon ourselves, and that when we are in this way self-reflective we are aware of the whole that is the self. But, he then argued, a part that is within the series cannot be aware of the whole series. If a perception is not the whole soul, then it is not soul, soulless. However, in that case,

37

One: Self as Substance

Would not such a soulless addition be subject to just such loss and gain of substance, in fact to the non-identity, which marks the rest of our material mass? And if this were so how explain our memories or our recognition of familiar things when we have no stably identical soul? (Enn., IV, vii, 5, p. 345).

The self-reflective self which is aware of itself as a whole must there, he concluded, be a simple entity that is not part of the series: “no one of the parts can be identical with the entire being” (IV, vii, 5, p. 345). Rather the soul that wholly knows the whole of itself is outside the parts of the series that constitute itself. ... surely no honest mind can fail to gather that a thing in which part is identical with whole has a nature which transcends quantity, and must of necessity be without quantity: only so could it remain identical when quantity is filched from it, only by being indifferent to amount or extension, by being in essence something apart (IV, vii, 5, p. 346).

The soul, in other words, is outside the world of time and change that we know in ordinary experience. The self-reflective self that is aware of the self as a whole is an unchanging simple entity, eternally present to each part of the whole that constitutes the self. Since the soul must be a simple unity, it may be inferred that it is immortal. For, the simplicity of the soul implies that it is indestructible. ...the Soul ... is not a mass, not a quantity. May it not change and so come to destruction? No: the change that destroys annuls the form but leaves the underlying substance: and that could not happen to anything except a compound. If it can be destroyed in no such ways it is necessarily indestructible (IV, vii, 12, p. 356).

And if indestructible, then, contrary to the Epicurean, immortal. The Epicurean position was stated or re-stated by Lorenzo Valla in his dialogue On Pleasure.20 When this was first published in 1431, Valla was widely accused of unorthodoxy and immorality. If the three Books of the dialogue are read together, then the charges are probably wrong; Book III re-states the Christian ethic and the Christian point of view, arguing that it supersedes the purely Epicurean position of Book II, as the latter superseded the Stoicism of Book I. Even so, the Christian point of view of Book III has an Epicurean twist: it is love of God that at once redeems us from

38

Body, Mind and Self in Hume’s Critical Realism

mortality and is the highest and most enduring pleasure. Be that as it may, it is in Book II that we find a considered re-statement of the Epicurean ethics, and, as part of this, the claim that the soul, like the body, is material and perishable. “...[N]othing remains after the dissolution of the living being... ,” Valla, or, at least, Valla’s Epicurean spokesperson, proposes (p. 219). Animals and humans are alike is sufficiently many respects that it is safe to say that, just as they are mortal, so are we. We are like them [other animals] in almost everything: finally, they die and we die – both of us completely (p. 221).

Reason, our capacity to know, is not different in kind from the reason of animals; and it, too, dies with the body – it, too, is mortal. Marsilio Ficino replied to Valla’s arguments in his Platonica Theologia de Immortalite Animorum, written between 1469 and 1474, and first published in 1495.21 This massive compendium in XVIII Books of arguments for the immortality of the soul includes a re-statement of the arguments of Plotinus. In particular, Ficino re-states Plotinus’ argument that if we assume that perception consists of images that are extended or complex in any way, then we cannot account for the unity of consciousness (VII, iv, p. 268). The soul, in knowing a whole, must be a simple unity (VIII, p. 311ff). This applies to the soul’s knowledge of itself. But Ficino did not end the debate. The Epicurean position was restated once again but from another perspective by Michel de Montaigne in his Essais, which, in effect, argued that what we are aware of as ourselves is a series of events joined by nothing more than the settled habits that constitute the self as the unique individual he or she is. It was Descartes who in this case was to re-state once more the neo-Platonic argument: in the cogito we become aware of the simple unchanging core of our selves, the simple centre of reflective self-awareness. Montaigne, unlike Epicurus and Lucretius, and unlike Valla’s pagan Epicurean, was a Christian, but there was nothing in his essays designed to defend as compatible with Christianity his implicit Epicurean account of the self. This was important since it was generally held that the Epicurean doctrine, in asserting the complexity and temporality of the self, implied that the soul was not intrinsically immortal. The neo-Platonic self, in contrast, provided a metaphysical basis for the Christian doctrine. Thus, to hold the Epicurean view of the self was in effect to accept the heresy of mortalism. Hobbes re-stated the Epicurean position in the English context. His account of perception and thought in terms of images is of course that of

39

One: Self as Substance

the Epicureans. Consciousness of things is constituted by images of those things. Hobbes was also the most notorious of the English mortalists, but the doctrine had been stated earlier by the Leveller Richard Overton22 in his pamphlet on Mans Mortallitie (first edition 1643).23 This pamphlet argues for the materiality and therefore the morality of the soul by appeal to both scripture and reason. The latter argument is for the most part a series of variations on the Epicurean theme, found already in Lucretius, that the state of one’s mind is dependent in a wide variety of ways upon the state of the body; and, since the body is mortal, so must the soul be mortal. The Cambridge Platonist, John Smith, provided, in his Select Discourses,24 the first English reply to the Epicurean and mortalist doctrines. The Discourses are probably too early for Hobbes to be his target, but it is entirely possible that he knew of Overton’s essay. In any case, he re-stated Plotinus' argument for the claim that the self, when it is aware of itself as a whole, is aware of a simple, unchanging self, eternally present to itself and to all its parts. He argued on the basis of the principle that “no substantial and indivisible thing ever perisheth” (p. 66). He then went on to appeal to the neo-Platonic arguments to establish the simplicity of the soul. In particular, he argues that we find ... a faculty within our own souls as colects and unites all the perceptions of our several senses, and is able to compare them together; something in which they all meet as in one centre...

In support, he quotes Plotinus: ... ‘that in which all those several sensations meet, as so many lines drawn from several points in the circumference, and which comprehends them all, must needs be one’ (p. 82).25

Here Smith is arguing from the unity of consciousness; he also argued from the unity of personal identity, in which the self recognizes itself in its past and future: a part of the series cannot know the whole series. ... if our souls were nothing else but a complex of fluid atoms, how should we be continually roving and sliding from ourselves, and how soon forget what we once were! The new matter that would come in to fill up that vacuity which the old had made by its departure, would never know what the old was, nor what that should be that would succeed... (p. 84).26

This was again re-stated by his junior colleague, Ralph Cudworth, in his

40

Body, Mind and Self in Hume’s Critical Realism

True Intellectual System of the Universe,27 as an argument against Hobbes. Following Smith, Cudworth repeats Plotinus’ argument: ... the soul conceives extended things themselves, unextendedly and indivisibly; for as the distance of a whole hemisphere is contracted into a narrow compass in the pupil of the eye, so are all distances yet more contracted in the soul itself, and there understood indistantly; for the thought of a mile distance, or of ten thousand miles, or semidiameters of the earth, takes up no more room in the soul, nor stretches it any more than does the thought of a foot or inch, or indeed of a mathematical point (vol. II, p. 827).

Self knowledge, too, requires the simplicity of the soul: ... mind is not different from itself: and indeed were it so, it could not be one thing (as it is) but many; every conceivable part of distant and unextended substance being a substance by itself (p. 828).

The political doctrines of Hobbes and Overton were very different, but, from the viewpoint of both the tradition and the establishment, equally radical. Both argued that their doctrines followed from their metaphysics of human being. Given these politically radical doctrines, it is evident that the argument concerning mortalism and the simplicity of the self was not merely metaphysical but also political. It is this metaphysical, religious, and political context, that is in the background to Hume’s discussion of the self. This background is important, for it should caution us not to import into Hume’s discussion points of view of more recent philosophers. It is the claim of Plotinus, Ficino, Descartes, Smith and Cudworth, that when we are, in reflective self-awareness, aware of our self, then we must be aware of a simple entity eternally present to all its parts. The simple self that endures through change is not something we “posit”: it is something of which we are aware. To introduce the neo-Kantian notion of “positing” into this context, as is now too often done, is totally misleading, however popular the term 'posit' has become in the context of the new relativism. Moreover, the doctrine of the simple self is not something introduced in order to “lend credence” to our belief in personal identity. Nor is the substantial self introduced, as Dennett has recently suggested, as a kind of hypothetical entity in psychology which serves as some sort of inner homunculus that is supposed to explain outer intentional behaviour.28 Rather, the thesis that the self of which we are aware in our reflective self-awareness is simple and eternally present to all its parts is a conclusion drawn from the

41

One: Self as Substance

argument that self-awareness is of such a nature that a part cannot be aware of the whole of which it is a part. This argument establishes the nature of personal identity – personal identity is constituted by a simple eternal substantial self – but it does not support or “lend credence to” any belief in personal identity nor imply that the whole point of such a self is that of an explanatory scientific entity (as Dennett has suggested). The Cambridge Platonists who revived the neo-Platonic argument to attack the radical Hobbesian view of the self and argued for the simplicity of the perceiver were not the only ones to employ the argument deriving from the ancient world via Plotinus. The argument was in fact very much “in the air” in the 17th and 18th centuries. Pierre Bayle in his sceptical Historical and Critical Dictionary29 equally used the argument to attack materialism (Article, Leucippus, Remark E). He argued that a soul which was a multiplicity could not feel pain since the pain would have to divide itself among the many parts; the soul could therefore not be material. However, Bayle was Bayle, and was therefore also willing to be sceptical about the possibility of a simple soul, or at least the doctrine of the simple soul as it had been developed by Leibniz (Rorarius, Remarks H and L). Since the soul both must be simple and cannot be simple, it follows that the notion is contradictory – just as the notion of extension is contradictory, if Bayle’s argument is successful (Zeno, Remark G) – and that soul, like extension, cannot exist. Hume, however, was not impressed by the argument for the simplicity of the soul, and turned it about in order to defeat it by ridicule. In the Treatise (I, IV, 5, p. 238ff) he argues that if what we immediately experience are impressions visible and tangible consisting of minima arrayed spatially, then if the soul were simple it would be too small to know more than a very small part of them or else it would have to be either spread out over them without being extended or be in more than one place at the same time. Hume thus simply rejects with a joke the attack on Hobbes’ imagistic account of thought based on the argument that the soul must be simple. It is clear that he held that the unity if you wish of the self was compatible with one’s holding in contrast to the substance tradition that the soul is not a simple continuing entity. (4) Human Nature Defended Hume’s argument regarding human nature was in fact a criticism of, and reply to Hobbes, but it was also a reply to those who criticized Hobbes for failing to ground morality objectively in the ontological structure of things.

42

Body, Mind and Self in Hume’s Critical Realism

Hume accepts Hobbes’ argument that there are no natures or forms, no necessary connections, causal or moral, in the ontological structure of things. But he also argues that morality has roots in human nature – empirical human nature – that Hobbes missed. In that sense, while accepting Hobbes’ moral relativism, he rejects to the Hobbesian reduction of morality to prudence and self-interest, and to that extent sides with Hobbes’ rationalist and Aristotelian critics, that Hobbes’ philosophy omitted something crucial about human being. What is this feature of human nature to which Hume directs our attention? It is the mechanism of sympathy. As it turns out, this same mechanism, as we shall see,30 plays a role in Hume’s account of personal identity. We shall return to the latter later. But now we must at least sketch how Hume roots morality in sympathy. In fact, many now argue that Hume does really adopt the Hobbesian position, that morality is a matter of conventions arrived at by prudential considerations. This is the now perhaps standard reading of Hume that he is a “contractarian,” that is, a philosopher who proposes, in Hobbesian fashion to obtain a civil society from the chaos of a lawless “state of nature” by prudential reasoning based on self-love leading to a sort of contract that all will conform to the rules of justice. But there is another side to Hume’s ethics, the side that is based on sympathy, that removes him far from Hobbes. This aspect of Hume’s moral theory and moral psychology has been explored first by Páll Árdal31 and more recently by Annette Baier.32 In the present chapter we shall attend to this side of Hume’s position, to try to bring out clearly the distance he is from Hobbes – though, to be sure, the Hobbesian side will not be neglected. Hobbes presents his case based on a certain model, that of the “state of nature.” Hume makes use of this model, but he introduces a second model. This latter brings out the differences between his view of human nature and that of Hobbes. Hobbes’ model builds in an assumption that there is a scarcity of resources. Hume’s second model replaces this assumption with an assumption in which there is a plenty of resources. Hume’s second model is introduced in the context of a wealth of classical and Biblical allusions that would make it plausible to his readers, more plausible perhaps than the Hobbesian model. It is this second model, Hume’s model with its assumption of plenty, upon which this chapter concentrates, exploring its own role in Hume’s thought and how it contrasts to the Hobbesian model. But as we shall see, Hume uses it not only to contrast his own position with that of Hobbes but also to contrast his position

43

One: Self as Substance

with that of such advocates of an objective morality as Samuel Clarke (of whom, more shortly). While the model shows that Hume agrees with Clarke in holding that there are certain native or “natural virtues,” it also shows that, contrary to Clarke but in agreement with Hobbes, the virtues of justice, contract and loyalty to the civil magistrate are “artificial virtues” which are the result, in the first instance, of convention. Hume was prepared to take over from Hobbes the notion of a “state of nature” which is “describ’d as full of war, violence and injustice.” (T, p. 493) This he regards as a “fiction” (ib.), a situation created by the imagination of the philosopher. It is contrary to fact – there never was such a state of human being as Hobbes describes in his state of nature –, but it is a construct which it is useful to consider as a situation revealing something about human nature. Even as physicists can usefully ask how a body would move if acted upon by one force when it is in fact acted upon by several, so the moral philosopher can usefully ask how persons would interact were they moved by their affections alone, unguided by reason. The state of nature is created by assuming three conditions, namely, scarcity (T, p. 488), limited generosity or benevolence (p. 486f), and rough equality among participants. Now assume further that, in this sort of situation, persons are moved by their affections alone, unguided by reason. The result is the state of nature, that is, the state in which there is a “war of all against all.” Hobbes concluded from this sort of hypothetical reasoning that there is no natural virtue in the sense that virtues such as generosity are not native to the human constitution but arise solely from social convention. This view, that justice and virtue is a matter of convention, was contrary to accepted Christian belief. For Paul had held that “...when the Gentiles, which have not the law, do by nature the things contained in the law, these, having not the law, are a law unto themselves” (Romans, II, 14). Hobbes’ account of justice and virtue was, therefore, shocking, as we have noted, to many in the 17th and 18th centuries. One who took offence was Samuel Clarke, who polemicized against Hobbes at some length in his Boyle Lectures.33 Thus, “Wars, hatred, and violence,” according to Clarke, “can never arise but from extreme Corruption” (II, p. 85). That entirely lawless State, which Mr Hobbs calls the State of Nature, is by no means truly Natural, or in any sense suitable to the Nature and Faculties of Man; but on the contrary, is a State of extremely unnatural and Intolerable Corruption (II, p. 79).

44

Body, Mind and Self in Hume’s Critical Realism

There are, Clarke argues, or, more exactly, asserts, objective moral truths: “...there is a Fitness or Suitableness of certain Circumstances to certain Persons, and an Unsuitableness of others; founded in the nature of Things and the Qualifications of Persons, antecedent to all positive appointment whatsoever; Also...from the different relations of different Persons one to another, there necessarily arises a fitness or unfitness of certain manners of Behaviour of some persons towards others...” (II, p. 30). In particular, it is fit that we should conform our will to these objective standards: “...originally and in reality, ’tis as natural and (morally speaking) necessary, that the Will should be determined in every Action by the Reason of the Thing, and the Right of the Case; as ’tis natural and (absolutely speaking) necessary, that the Understanding should submit to a demonstrated truth” (II, p. 40). As for our specific moral duties, the first is equity (II, p. 53). This includes the allocation of goods – it is “contrary to the eternal Reason of things, [that a Man] should desire to gain some small profit to Himself, by doing violence and damage to his Neighbour” (II, p. 54) – and the duty of allegiance – “the same Proportion is to be observed, in deducing the Duties of Parents and Children, of Masters and Servants, of Governours and Subjects, of Citizens and Foreigners; in what manner every person is obliged by the Rule of Equity, to behave himself in of these and all other Relations” (II, p. 55). The second of our moral duties is “universal Love or Benevolence.” The latter requires of us not only what is just and right in our dealings with other persons, but “also a constant indeavouring to promote in general, to the utmost of our power, the welfare and happiness of all men” (II, p. 57). And this, of course, provides the way out of the Hobbesian war of all against all: ...if Men are obliged by the original reason and nature of things to seek terms of Peace, and to get out of the pretended natural State of War, as soon as they can; how come they not to be obliged originally by the same reason and nature of things, to live from the beginning in universal Benevolence, and avoid entering into the State of War at all? (II, p. 80).

I don't claim that this is sound argument against Hobbes; certainly, it does not meet Hobbes on his own ground. But however that may be, it nonetheless makes clear the attitude of many towards the Hobbesian position. Hume argues that Clarke is in part correct against Hobbes – there are in fact virtues that are in a way native to persons – but he also argues that Hobbes’ central point is also correct – the virtues that define civil society, such as property, promise-keeping and allegiance, are not rooted in the on-

45

One: Self as Substance

tological structure of the universe nor even native to human being but derive from conventions fashioned by artifice, that is, by our affections guided by reason. Hume constructs this argument against Clarke in a way that is parallel to the state of nature argument. It proceeds hypothetically on the assumption that persons are moved by affections that do not derive from convention and the artifices of reason. But it makes different contrary to fact assumptions. Specifically, it assumes a situation in which there is no scarcity. This is the hypothesis of a “Golden Age”: ...this is painted out to us, as the most charming and most peaceable condition, that can possibly be imagin’d. The seasons, in that first age of nature, were so temperate, if we may believe the poets, that there was no necessity for men to provide themselves with cloathes and houses as a security against the violence of heat and cold. The rivers flow’d with wine and milk; The oaks yielded honey; and nature spontaneously produc’d her greatest delicacies. Nor were these the chief advantages of that happy age. The storms and tempests were unknown to human breasts, which now cause such uproar, and engender such confusion. Avarice, ambition, cruelty, selfishness, were never heard of: Cordial affection, compassion, sympathy, were the only movements, with which the human mind was yet acquainted. Even the distinction of mine and thine was banish’d from that happy race of mortals, and carry’d with them the very notions of property and obligation, justice and injustice (T, pp. 493-4).

This golden age was not the creation of philosophers, but of poets, who, however, have the better insight: they “have been guided more infallibly, by a certain taste or common instinct, which in most kinds of reasoning goes farther than any of that art and philosophy, with which we have been yet acquainted” (T, p. 494). Avarice, and the other passions that now animate the human breast, are not native to us, but a consequence of scarcity. If scarcity were removed, or if persons were universally benevolent, there would be no need for the virtues of justice and property. In such a context, “you render justice useless, by supplying its place with much nobler virtues, and more valuable blessings. The selfishness of men is animated by the few possessions we have, in proportion to our wants; and ’tis to restrain this selfishness, that men have been oblig’d to separate themselves from the community, and to distinguish betwixt their own goods and those of others” (T, p. 495). Hume makes his point in terms of a thought experiment about a Golden Age that parallels the thought experiment of Hobbes concerning the state of nature. But he then proceeds to argue the same point induc-

46

Body, Mind and Self in Hume’s Critical Realism

tively (T, p. 495). He indicates that in groups, e.g., the family, where we know generosity does extend to all the members, the distinction between mine and thine does not exist; and where there is no scarcity, e.g., in the air we breathe, we again find that there is no mine and thine. The assumptions defining a golden age do not have universal validity, as they are taken to have when considering such an age. But they do have restricted validity in certain circumstances, and what is inferred would hold in a golden age does in fact hold in the restricted circumstances where the assumptions do have empirical validity. The consequences inferred from the assumptions defining a golden age thus receive empirical confirmation. We can thus safely conclude that, contrary to Hobbes and in agreement with Clarke, certain virtues are native, and do not presuppose the institutions of civil society, while, contrary to Clarke and in agreement with Hobbes, the virtues of justice, promise-keeping, and allegiance, arise only when the conventions of civil society are instituted. Hume’s thought experiment about the Golden Age has an odd ring to us; we accept it as a model, but the model has plausibility only through what we antecedently know about human nature – we know from our own case and from those of others near to us, that we are motivated at certain times by something more than self-love. It is this latter that Hume’s inductive argument aims to establish. We tend to be much more persuaded by the inductive arguments that he offers than by the appeal to the notion of a Golden Age. Yet it is the latter than Hume presents first, and to which he gives much more space. This, I believe, is not just a matter of rhetorical organization, placed that way, perhaps, to put the notion of a Hobbesian state of nature on a par with the fiction of the poets called a “Golden Age.” It is instead, I would like to suggest, that what we and Hume take to be thought experiments would have been far from odd in his own age. Indeed, I submit, many of his readers would not have taken these stories, as Hume does, to be contrary to fact; they would, to the contrary, have taken them to be things actually true historically. And this suggests part of the rhetorical strength of Hume’s appeal to the notion of a Golden Age: Historical facts can have a stronger appeal than inductive arguments based on casual observation of what happens in certain restricted circumstances – just as the breakdown of law and order in Britain during the civil wars gave real force to the Hobbesian idea of a “war of all against all.” Hume’s History of England fits our notion of what an historical narrative should try to do. In good part, no doubt, Hume’s History helped determine that model. But Hume’s contemporaries had models that do not

47

One: Self as Substance

wholly mesh with ours: they were prepared to find in the historical narratives that they read many ostensible facts that for us are simply not part of history. Take another example from the centuries of Hobbes and Hume: Sir Walter Raleigh’s History of the World.34 This was a very popular history, and was re-printed many times, and abridged and epitomized in the 17th and 18th centuries. For its age it was no doubt reasonable, as it would not be for ours, that a history of this sort begin its narrative with the events of the Garden of Eden. What for us is myth and metaphor is for that age part of history, part of the historical narrative. And just as part of the role of historian, according to Hume, is to reconstruct from documents the events of history, so also it is part of the role of the historian, according to Raleigh. He therefore provides a lengthy and scholarly essay in which he attempts to determine for his readers the precise location of the Garden of Eden. A concern with this issue is not confined to Raleigh but runs through the historical thought of the 17th and 18th centuries. A hundred years after Raleigh we find Pierre-Daniel Huet concerned with the same issue of the location of the Garden of Eden. Huet was of course the sceptical critic of Cartesianism. In his Censura philosophiae Cartesiana35 he had transformed the cogito, ergo sum into “I may have thought, therefore perhaps I was.” In his Traité philosophique de la foiblesse de l'esprit humain36 Huet defended academic scepticism (which he argued was not significantly different from Pyrrhonism [pp. 139-50]). He advanced a wide battery of sceptical arguments designed to show that “l’Homme ne peut connoître la Vérité par le secours de la Raison avec une parfaite & entiere Certitude” (p. 22). There are many degrees of certitude, and faith alone leads to perfect certitude (p. 16ff). In the task of living, the correct rule is the only rule that turns out to be available to us, to wit, the rule of probability. “Il faut suivre dans l’usage de la view les choses probables, comme si elles étoient véitables” (p. 207ff). It is the senses that provide the rule of truth (p. 208). In particular he recommends the sceptical and non-dogmatic cognitive attitude of the Royal Society, “cette nouvelle Societé de Philosophes Anglois, qui a élevé tant d’excellens Esprits, condamne l'arrogance des Dogmatiques...” (p. 221). Hume was no doubt familiar with these sceptical writings of Huet. But Huet was also a scholar and an historian. His sceptical scruples did not hinder him from writing works in history. In particular, he wrote a long scholarly essay, De la situation du paradis terrestre.37 For our purposes, Huet’s discussion is of considerable interest. In the first place, the issue of the location of the Garden of Eden, the

48

Body, Mind and Self in Hume’s Critical Realism

Terrestrial Paradise, was, for Huet, not a matter of faith but rather part of the common sense of history; the essay on the terrestrial paradise took its place alongside Huet’s rather less mythical Les origines de la ville de Caen, and his book on the origin of novels, Traité de l’origine des romans,. Huet’s discussion of the terrestrial paradise is throughout dependent upon taking the written text of Genesis as providing an historical narrative that needs merely to be properly read. And secondly, what amounts to much the same point, Huet only questions the interpretation of the Mosaic texts, not the basic assumption that the historical truth can be found in them, if only they are properly read. Huet recognises that there is a divergence of opinions about the interpretation of the texts, a “contrarieté de sentimens” concerning the location of the terrestrial paradise (p. 7). But whereas such a contrariety is used in the case of perception to argue to a sceptical conclusion, no such inference is allowed in the case of the Mosaic texts. Far from arriving at a sceptical conclusion about the existence, let alone location, of the terrestrial paradise, Huet is able to conclude quite categorically that ...le Paradis terrestre estoit situé sur the canal que forment le Tigre & l'Euphrate joints ensemble, entre le lieu de leur jonction, & celui de la séparation qu'ils font de leurs eaux, avant que tomber dans le Golphe Persique (p. 19).

Huet’s appeal to evidence is straightforward, and far from that of one who would question that the Mosaic texts were, though subject to interpretation, a source of historical truth. There are two sorts of evidence to which Huet appeals. One is to the Mosaic texts. These state that “...a river went out of Eden to water the garden; and from thence it was parted, and became into four heads” (Genesis, II, 10). These four rivers are then named: “The name of the first is Pison..../And the name of the second river is Gihon: the same is that compasseth the whole land of Ethiopia./And the name of the third river is Hiddekel: that is it which goeth toward the east of Assyria. And the fourth river is Euphrates” (II, 12-14). To identify the rivers is to identify the location of the Garden of Eden. But to say this is to take for granted the worth of the Mosaic text. Huet does precisely this: Mais comme l’unique fondement surquoy l’on puisse s’appuyer, ce sont les paroles que Moyse a employées pour décrire la situation du Paradis terrestre, il faut les rapporter avant toutes choses, en les traduissant mot à mot sur l'original (p. 13).

49

One: Self as Substance

This provides the basic evidence for the attempt to discover where the terrestrial paradise or Garden of Eden was located. These words provide the direct evidence for two of the four rivers: the Euphrates is clear, and the Hiddekel is evidently the Tigris. But what are the Pison and the Gihon? It was the view of Arian that the Pison was the Ganges (p. 97), but Huet rules this out (p. 98), arguing for his conclusion on the basis of the other sort of evidence that he invokes, namely, geographical (p. 100ff). Similar arguments rule out the Nile (p. 104), and the Danube (p. 105f) – the latter had been championed by Gregory of Nazianus (p. 105). Huet concludes that “On fait voir que le Phison est le canal occidental des deux [Euphrates and Tigris] en quoi se divisent le Tigre et l’Euphrate joints ensemble” (p. 95, p. 107-9). Huet draws a similar conclusion about the Gehon: Le Gehon est le canal oriental des deux qui divisent l’assemblage de l’Euphrate & du Tigre (p. 146ff).

But then Huet must explain away the fairly clear identification in the Mosaic text of the Gihon with the Nile (Ch. XII). The point is that for Huet, as for Raleigh, the existence of the Garden of Eden is taken for granted. For the book of Genesis clearly and unambiguously asserts that this Garden did exist. The only issue is where precisely this Garden was located, an issue that arises because of the rather vague description given in the book of Genesis. And this issue is to be decided by the tests available within the framework that Huet recommended, that of probabilistic reasoning of the sort practiced by the members of the Royal Society. The point to be made is that those who professed sympathies to a modest scepticism, and were willing to defend a critical probabilistic account of knowledge, thinkers like Huet, were also quite prepared to take the existence of the Garden of Eden to be a matter of fact. Even a sceptic like Huet, if you wish, is prepared to do this. Thus, the world for which Hume wrote was one in which even sceptics like Huet take for granted the existence of a terrestrial paradise. For such a world, Hume’s discussion of the morality that would be present in a Golden Age would have the ring, not of mere fantasy, but of a serious concern about what went on during a period that was both clearly real and about which certain historical texts give us reliable information as a basis for such reflections. In his assumption of the reliability of the inference from the Mosaic texts to the conclusion that the Garden of Eden did in fact once exist, Huet simply followed the members of the Royal Society itself, those whose sci-

50

Body, Mind and Self in Hume’s Critical Realism

entific procedures Huet recommended. These English scientists also took the notion of the Garden of Eden as truth, and assumed that their probabilistic methods of investigation lead to that conclusion. Thus, according to Samuel Clarke, again in his Boyle Lectures,38 the existence of the Garden of Eden (II, p. 189) is one of the “Doctrines, which the true, simple and uncorrupted Christian Religion teaches” (II, p. 183), doctrines which have “all the Marks and Proofs of being actually and truly a Divine Revelation” (II, p. 168). As we have said, Hume’s political thought is often placed in the context of thinkers such as Hobbes, and rightly so. But it can also be placed in the broader context formed by views such as those of Raleigh and Huet concerning the Earthly Paradise and the early history of mankind. We can begin to see why Hume could take for granted for certain of his assumptions a plausibility that is less evident to us. We can begin to see how Hume can take for granted that his readers would find his thought experiment, his model of the Golden Age, entirely plausible once we put it in the context of the works of perfectly respectable thinkers such as Raleigh and Huet. Hume’s moral philosophy is in fact redolent of themes familiar to a readership that was well versed in the stories of the Greeks and Romans and of the Bible and in fact at times took both sets of stories to be, so tospeak, Gospel truth. But what we find in Hume is that these themes are removed from their religious context; historical facts become thought experiments, and the moral philosophy associated with those themes is humanized and secularized. Consider the work of Raleigh. Raleigh in his History of the World canvases a variety of opinions on the location of the Garden of Eden (Bk. I, Ch. III, sec. ii). Unfortunately, as we have noted, the Mosaic account in Genesis provides few details. In fact, all that we are told is that it had boundaries, that there were at the centre two trees, the tree of life and the tree of knowledge of good and evil, that there was gold in one of the adjacent countries, and that four rivers flowed from it, the Euphrates, the Hiddekel or Tigris, the Phison (Ganges?) and the Gihon (Nile?) (Genesis, II, 814). Scant details leave much room for division of opinion concerning the real location. The “strange opinions concerning Paradise” that Raleigh surveys include the opinion of Tertulian, Bonaventure, and Durandus that it had a location in the southern hemisphere beyond the torrid zone of the equator; the opinion of Postellus that it is at the North Pole; the opinion of Origen that it is but myth; the opinion of Strabus that it is located above this world at the lunar sphere or at least in the air; and the opinion of Novi-

51

One: Self as Substance

omagus that it included the whole earth. Raleigh argues against these views, concluding, as Huet was later to conclude, that “there was a true local Paradise eastward in the country of Eden,” where “Eden is...a country near unto Charon in Mesopotamia...” (p. 23). Hume’s thought experiment concerns a land of plenty, without scarcity. The point is that the Garden of Eden was, according to the Mosaic account, a terrestrial paradise, just such a land of plenty. The Biblical story tells us that “out of the ground made the Lord God to grow every tree that is pleasant to the sight, and good for food” (Gen. II, 9). The trees were clearly fruitful, and always so, from which one can infer that it was perpetually Spring in the Garden. Man’s original diet consisted of seeds and fruit from these trees: “God said, Behold, I have given you every herb bearing seed, which is upon the face of all the earth, and every tree, in which is the fruit of a tree yielding seed; to you it shall be for meat” (Gen. I, 29). The animals ate the same food: “And to every beast of the earth, and to every fowl of the air, and to every thing that creepeth upon the earth, wherein there is life, I have given every green herb for meat...” (Gen. I, 30). Peaceful relations among the animals can be inferred, and also that man lived at peace with the animals; there was an inequable but pacific relation between man and the other species, established when God brought the animals into the Garden for Adam to name (Gen. II, 19-20), and commanded them to obey him (Gen. I, 26). Raleigh’s judgement is that “Paradise was a terrestrial garden, garnished with fruits, delighting both the eye and taste...it was watered and beautified with a river...” (History, p. 23). Samuel Clarke is of the same opinion: “God, after the Formation of the Earth, created Man at first upright and innocent, and placed him a happy and paradisiacal State, where he injoyed plenty and abundance of things without labour or sorrow” (Discourse, II, p. 189). The Mosaic account of a terrestrial paradise has parallels in other stories that would have been familiar to the 17th and 18th centuries, stories provided by Homer, Virgil and Ovid. Thus, Homer described the delights of the perpetual spring in the garden of Alcinous, where, in a large orchard of four acres: “Their fruit never fails nor runs short, winter and summer alike, it comes at all seasons of the year, and there is never a time when the [soft] West Wind’s breath is not assisting, here the bud, and here the ripening fruit; so that pear after pear, apple after apple, cluster on cluster of grapes, and fig upon fig are always coming to perfection.”39 But the Greeks further distinguished Arcadia, a primitive rather than an innocent place, where the inhabitants were notorious for their rugged virtue and their igno-

52

Body, Mind and Self in Hume’s Critical Realism

rance, their rustic hospitality, their skill in singing, and their low standard of living, and further yet Elysium, the land of the dead, where, according to Homer, the shades of those whom the gods favoured lived in perfect happiness on the banks of the Oceanus river at the end of the earth. Virgil, however, made Arcadia into a land of luxuriant vegetation and perpetual spring, while Pindar made the entrance into Elysium depend upon having lived a good life, so that it was easy to merge these two places into the general notion of a once real earthly paradise. Ovid’s description of the Golden Age fits the same pattern. Here again we have the fertile soil and the perpetual spring: “The earth itself, without compulsion, untouched by the hoe, unfurrowed by any share, produced all things spontaneously, and men were content with foods that grew without cultivation. They gathered arbute berries and mountain strawberries, wild cherries and blackberries that cling to thorny bramble bushes: or acorns, fallen from Jupiter’s spreading oak. It was a season of everlasting spring, when peaceful zephyrs, with their warm breath, caressed the flowers that sprang up without having been planted. In time the earth, though untilled, produced corn too, and fields that never lay fallow whitened with heavy ears of grain. Then there flowed rivers of milk and rivers of nectar, and golden honey dripped from the green holm-oak.”40 It is likely that it is Ovid whom Hume has in mind when he speaks of the “poets” describing a “Golden Age”: certainly, Hume’s description and that of Ovid are very close. The Mosaic and the classical pictures of an earthly paradise do not all fit exactly one to another, but in fact many considered them to be of a piece. One way or another later thinkers, mostly Christian, succeeded in ignoring these differences. If Hume was alluding to Ovid when he described a land without plenty, his readers would have likely taken him to be referring also to the Mosaic account of an earthly paradise. Thus, for example, his readers would likely have understood Virgil to be something of an Honourary Christian at the least. For they would have known that Virgil had, in his Fourth, the so-called “Messianic,” Eclogue,41 prophesied the return of the Golden Age, to be ushered in by a boy. Virgil was likely referring to events at the time of Augustus, but later ages thought differently and interpreted the Eclogue as indicating the coming of Jesus as the Christ. The first to do so apparently Constantine the Great.42 He was not the last, and Virgil became in effect an unofficial Church Father. Augustine refers to Virgil often, and with great veneration, in the Civitas Dei, suggesting, however, that the actual prophecy of Christ was to be attributed to the Sibyl to which the poem refers and that Virgil himself had no knowl-

53

One: Self as Substance

edge of the person whose coming was prophesied.43 This view was accepted rather nearer to Hume’s time by Alexander Pope, who, in the preface to his Messiah, records his agreement with the view of Augustine that the prediction was taken by Virgil from the Sibyl. If it be said that Pope was simply following the tradition of his own church (Pope was Catholic), it should be recorded that the indubitably Protestant Samuel Johnson did not in his Life of Pope demur from the view.44 Many influential thinkers, and no doubt many more ordinary readers, held that the stories of the Greeks and Romans that we call myths could in fact be fit into the historical scheme established by the Biblical narratives. Lactantius was one of these.45 Another was Tertullian, who held that Homer had received his idea of such a garden from Moses,46 and he was followed in this by Raleigh: “...whence had Homer his invention of Alcinous’s gardens...but out of Moses’s description of Paradise” (History, p. 24). In this way, Christian writers came to quote the classical authors as if they enjoyed Biblical authority; their stories were taken to be as veracious as the Mosaic. As the examples of Raleigh and Pope make clear that there were many in Hume’s age who were prepared to take for granted the traditional assimilation of classical themes about a Golden Age to the Christian doctrine of a terrestrial paradise. The classical stories of Ovid and Homer were thus more than myth, more than poetic fiction for many in that age. Hume’s appeal, then, to such stories would be for these readers an appeal to historical fact and not just to poetry. Now, if we take as given the fruitfulness of the Garden of Eden, we can infer directly that our forebears who lived there had no need to labour for their food. As we saw Clarke put it, in the Garden of Eden man was “in a happy and paradisiacal State, where he injoyed plenty and abundance of all things without labour or sorrow” (Discourse, II, p. 189). Saint Ambrose elaborates this notion in his Hexaemeron, where he gives us a description of the world as it existed directly after creation and before the Fall, combining themes from Genesis and from the Golden Age of classical poetry: Spontaneously earth bore all fruits; though it could not be plowed in the absence of a plowman – for no farmer yet existed – nevertheless it abounded in the richest harvests, and, I do not doubt, with an even larger yield, since the slothfulness of the husbandman could not rob the soil of its richness. For now the fertility of a piece of land is in proportion to the labor expended upon it, and neglect, or the injury caused by heavy rains or by the aridity of the land or by hailstones or by some other cause, is punished by the barrenness of the soil. In those days, however, the earth of itself everywhere brought forth its fruits, since

54

Body, Mind and Self in Hume’s Critical Realism

He so commanded Who is the fulness of all things. For the word of God fructified upon the earth, nor had the soil as yet been laid under any curse. For the time of the birth of the world is more ancient than our sins, and the guilt because of which we have been condemned to eat our bread in the sweat of our brow, to know no food without sweat, is more ancient.47

It was only after sin that there was scarcity and the need to labour: “...cursed is the ground for thy sake,” God told Adam, “in sorrow shalt thou eat of it all the days of thy life; Thorns also and thistles shall it bring forth to thee; and thou shalt eat the herb of the field: In the sweat of thy face shalt thou eat bread till thou return unto the ground...” (Gen. III, 1718). Clearly, for Adam there was no need for government, given the population of the Garden of Eden. But some went further in their claims about the Golden Age. In such a place as the Garden of Eden or Golden Age was, with plenty for all, there was no need for property, commerce or government. Thus, Ovid tells us that during the Golden Age, “...men of their own accord, without threat of punishment, without laws, maintained good faith and did what was right. There were no penalties to be afraid of, no bronze tablets were erected, carrying threats of legal action, no crowd of wrongdoers, anxious for mercy, trembled before the face of their judge: indeed, there were no judges, men lived securely without them. Never yet had any pine tree, cut down from its home on the mountains, been launched on ocean’s waves, to visit foreign lands: men knew only their own shores. Their cities were not yet surrounded by sheer moats, they had no straight brass trumpets, no coiling brass horns, no helmets and no swords. The peoples of the world, untroubled by any fears, enjoyed a leisurely and peaceful existence, and had no use for soldiers” (Metamorphoses, p. 32). It was only after the overthrow of Saturn by Jupiter that the Silver Age was ushered in through the institution of four seasons by Jupiter; only then did people need shelter and have to labour for their food: “It was in those days that men first sought covered dwelling places: they made their homes in caves and thick shrubberies, or bound branches together with bark. Then corn, the gift of Ceres, first began to be sown in long furrows, and straining bullocks groaned beneath the yoke” (p. 32). In short, Hume’s claim against Clarke that in a Golden Age there was no need for justice, property, commerce, i.e., promise-keeping, or allegiance, is taken by Ovid to be a fact of history, or, at least, many were prepared so to take Ovid. Thus, Saint Ambrose agrees that in the Golden Age no boundary stone separated field from field; all things were held in common. “Nature has poured forth all things for all in common,” and God has ordained that “the earth be com-

55

One: Self as Substance

mon property for all.” It was avarice that broke down communal ownership. Private property is therefore “unnatural” and has to be controlled by law for that reason alone: Natura...jus communs generavit, usurpatio jus fecit privatum.48 Further, even after labour became necessary on account of scarcity, government was not yet necessary. Familial relationships and paternal guidance sufficed to maintain order. It was only when men and vice increased that obedience fell away and, as Raleigh put it, “the soft weapons of paternal persuasion became in all over-weak” (History, p. 102). It was this that led for the formation of government. Raleigh continues: ...when the hearts of men were only guided and steered by their own fancies, and toss’d to and fro’ on the tempestuous seas of the world, while wisdom was severed from power, and strength from charity; necessity (which bindeth every nature but the immortal) made both the wise and foolish understand at once, that the estate of reasonable men would become far more miserable than that of beasts, and that a general flood of confusion would a second time overflow them, did they not by a general obedience to order and dominion prevent it. For the mighty, who trusted in their own strengths, found others again (by interchange of times) more mighty than themselves: the feeble fell under the forcible; and the equal from equal received equal harms. Insomuch, that licentious disorder (which seemed to promise a liberty upon the first acquaintance) proved upon a better trial, no less perilous than an unendurable bondage. These arguments by necessity propounded, and by reason maintained and confirmed, persuaded all nations which the heavens cover, to subject themselves to a master, and to magistracy in some degree (p. 102).

The absence of the plenty of the Garden of Eden, and, as part of that, the necessity of labour, does not by itself create the necessity of government. So long as persons are part of a single family, the sentimental ties that bind the family together suffice to maintain order. But as society grows, those ties gradually cease to extend to all. Once this happens a sort of “war of all against all” will take place unless people subject themselves to a magistrate. Recognizing that their state will be worse in the absence of such an agreement than if it is instituted, people do agree to the institution of a magistracy. We can note here a major disagreement between Raleigh and Clarke. The arguments that persuade persons to subject themselves to government are, according to Raleigh, prudential: they appeal to “necessity.” There is no sense in which they are a priori. Nor, therefore, is the duty of obedience anything more than prudential. There is, in other words, no sense in which

56

Body, Mind and Self in Hume’s Critical Realism

it is innate or native to us. This is in contrast to the sentimental ties that bind persons together into their family groups. In contrast, Clarke argues that not only the mutual obligations of those in families but also those of subjects and magistrates are native to us, given a priori, to be deduced from the a priori principle of equity. To deny equity “either in Word or Action, is as if a Man should contend, that, tho’ two and three are equal to five, yet five are not equal to two and three” (Discourse, II, p. 54); and “the same Proportion is to be observed, in deducing the Duties of Parents and Children, of Masters and Servants, of Governours and Subjects...; in what manner every person is obliged by the Rule of Equity, to behave himself in each of these and all other Relations” (II, p. 55). So for Clarke, in contrast to Raleigh, the duties of civil society are native. On this point, of course, Hume agrees with Raleigh, arguing as he does that allegiance is an artificial virtue based on a convention that is instituted to satisfy the deep need to have an external mechanism for enforcing contracts or promises (Treatise, III, II, vii). Raleigh’s story concerning the formation of civil society and the institution of magistrates is similar to that of Hobbes: the obvious potential effects of a threatened war of all against all leads men as a matter of selfinterest to agree to obey a common magistrate. The war of all against all and hence also the potential effects thereof, are a consequence of selfinterested attempts in the context of scarcity to acquire whatever one wants. This same self-interest recognizes that this war and its consequences would be disastrous for its own aspirations. So it agrees to a magistracy. For Raleigh all this is part of history. That is, the Hobbesian state of nature is alongside the Garden of Eden or Golden Age a fact of history. There is, however, a very crucial difference between the pictures of the human condition given by Raleigh and Hobbes. For Hobbes, persons have no motives other than self-interest. For Raleigh, in contrast, the war of all against all arises from vice; this means that, for Raleigh, unconstrained self-interest is a vice. But where there is a vice, there is also potentially a virtue. Thus, for Raleigh, persons are, potentially at least, other than simply selfish; they are, potentially at least, moved by passions other than self-interest. It is these moral sentiments that originally bind humankind together in communities after the expulsion from Eden: the sentimental ties of family members for each other suffice, at least in these small communities, to establish social order. These provide motives for conduct other that self-interest, and serve to curb and constrain the latter. It is, of course, this that the standard critics of Hobbes complained about: to char-

57

One: Self as Substance

acterize persons as moved solely by self-interest is to characterize them as immoral, as devoid of all those moral sentiments, e.g., benevolence, that distinguish those who are virtuous from those who are vicious. As we saw Clarke put it, Hobbes’ state of nature “is a State of extremely unnatural and Intolerable Corruption” (Discourse, II, p. 79). Raleigh is in agreement with Clarke that there is an objective moral law knowledge of which is native to the human breast. According to him there is an objective law of nature, and it is imprinted on our souls: it is “...the impression of God's divine light in men, and a participation of the law uncreated and eternal. For without any law written, the right reason and understanding, which God hath given us, are abilities within ourselves, sufficient to give us knowledge of the good and evil...” (History of the World, p. 154). The war of all against all arises from vice, the failure to act benevolently when the ties of family are broken by the growth of numbers, or, as Clarke would later put it, the failure to act on the recognition of the objective fitness of benevolence. Hume of course agrees with Hobbes that there are no objective moral standards, and that property, commerce or contract, and government arise to serve self-interest; they are not native to the human mind. On the other hand, he also holds with Raleigh and Clarke that there is a natural morality, one that would exist in the absence of the institutions, e.g., a magistracy, that define civil society. Except, for Hume, unlike Clarke, this morality does not include the notion that it is fit to respect property, that it is fit to keep promises, and that it is fit to conform one's own behaviour to the judgements of the magistrate. That is, for Hume as for Raleigh, the rights and duties of civil society are not native, not given a priori. For Hume, some virtues produce their goods only if we assume the existence of a social convention. Such are the virtues of justice, promise-keeping, and allegiance. These are not native. Nonetheless, Hume does agree with Raleigh that there are certain natural virtues that precede the artificial conventions that establish civil society. If some virtues produce their good only upon the establishment of a social convention, there are other virtues produce their goods in the absence of convention (Treatise, p. 574f). These are the natural virtues. Some of these would hardly count as virtues in a Golden Age, since they produce goods only in the context of scarcity, e.g., the virtues of industry and frugality (T, p. 587), or that of charity (T, p. 578). But they are natural in the sense that they would produce goods in a social situation where the conventions of civil society were absent provided that generosity were unbounded. Charity, along with generosity and benefi-

58

Body, Mind and Self in Hume’s Critical Realism

cence are often called “social virtues, to mark their tendency to the good of society” (T, p. 578). Yet unlike charity, the virtues of generosity and beneficence would tend to the good of all even during a Golden Age where there was no scarcity and no need to labour. So would the private virtues of prudence and dexterity (T, p. 587). So would the natural virtues of humanity, compassion, gratitude and friendship (T, p. 603). All this, of course, fits with the story of the earthly paradise. Thus, as Ovid said of the Golden Age, “men of their own accord, without threat of punishment, without laws, maintained good faith and did what was right” (Metamorposes, p. 31), while according to the account of creation in Genesis II, Eve was created precisely so Adam could have a companion with certain virtues: “And the Lord God said, It is not good that man should be alone; I will make him a help meet for him” (Gen. II, 18). Clearly, given the function that Eve was to serve, it was the expectation of their Creator that Adam and Eve would exercise at least the virtues of friendship and generosity. Where there is virtue, there is also, of course, the possibility of vice. In the Mosaic account, vice did enter the earthly paradise; that was the cause of the expulsion of humankind from the garden, and of the need for them to labour for their food and shelter. No doubt Hume would take this to be poetic fiction too, but there is a point to it, just as there is a point to considering the innovations of Jove that brought about the transition from the Golden Age to the Silver and with it the necessity of labour. The point can be made in terms of another thought experiment: would the vice that, according to Genesis, brought about the Fall bring about a transition from a Golden Age to a Hobbesian state of nature? Unfortunately, while Genesis makes clear what was the act that led to the Fall and the expulsion of Adam and Eve from the Garden of Eden, the account is less than forthcoming about what exactly was the vice this act exemplified. Some, including Tertullian and Basil, laid emphasis upon Eve's eating the apple, and attributed the Fall to greediness. As Tertullian put it, Adam “yielded more readily to his belly than to God, heeded the meat more than the mandate, and sold salvation for his gullet”49; as Basil put it, Adam’s vice was the “lust of the belly.”50 A second group looked to the consequence of eating the apple, that Adam and Eve recognizing their nakedness, and held that the story was an allegory for the beginning of sexual relations. This group also includes Tertullian, who tended to identify the vice of Adam with the sin that he happens to be denouncing at the moment.51 Before the Fall, Adam and Eve were emotionally and physically pre-adolescent, and the attainment of the knowledge of good and evil coin-

59

One: Self as Substance

cides with the attainment of sexual maturity.52 A third group, including at times Ambrose, was more inclined to see hunger, gluttony and lust as consequences rather than causes of the Fall; this group pointed to the name of the tree, and interpreted the story in terms of man trying to rise above his station – the creature had already been endowed by God with wisdom (sapientia), but he aspired to knowledge (scientia) of the Creator.53 Raleigh was among this group that attributed the Fall to pride and ambition (History, p. 42). A fourth group, likely a majority, and including Theophilus54 and Basil,55 interpreted the Fall quite simply in terms of disobedience. In due course the church, without formally deciding between the four views, impartially recommended as virtues to its members fasting, chastity, humility, and obedience. There is considerable question, however, whether any of these could be natural virtues in the Humean sense, nor therefore whether any of the natural Humean vices could have precipitated the Fall. For Hume, fasting and chastity are more Christian than correct, “monkish” virtues as Hume would express it; in the context of the earthly paradise of abundance they are hardly conducive to human well-being and therefore are hardly to be reckoned among the natural virtues. As for obedience, this presupposes something to be obeyed. For Raleigh and Clarke, as for St. Paul, this means in the first place the natural law imprinted in the human breast, and in the second place the Creator that decreed this law and imprinted it upon His works. For Hume, however, there is no innate knowledge of virtue, nor any morally superior creator that has implanted it within us. For all that, the natural virtues are natural, in a basic way native to us, in the sense at least that they are prior to all conventions of civil society. The natural virtues are native, not in the sense that there is innate to our reason a knowledge of these principles, nor in the sense that the creator has directly implanted them within our soul, but in the sense that they arise directly from the functioning of the native mechanism of sympathy: it is this mechanism that naturalistically places moral principles within our soul. We respond to the pleasures and pains of others with similar feelings of our own. “When I see the effects of passion in the voice and gesture of any person, my mind immediately passes from these effects to their causes, and forms such a lively idea of the passion, as is presently converted into the passion itself” (Treatise, p. 578). This is the mechanism of sympathy. Thus, when I see someone in danger, their distress at their situation creates a distress about their situation that I feel. And, often enough, I act upon the distress that I feel, extracting them from

60

Body, Mind and Self in Hume’s Critical Realism

the situation of danger. Now, to judge something a virtue is to feel a certain moral sentiment. Such a sentiment of virtue arises from a character or passion “because of their tendency to the good of mankind” (T, p. 578); these qualities are either agreeable to others or agreeable to the person who possesses them (T, p. 590). Hence, the mechanism of sympathy which leads me to relieve the distress of others or moves me to give them pleasure produces actions which we judge to be virtuous and deters us from actions which are vicious. It is sympathy which leads me to act decently towards others; “when I relieve persons in distress, my natural humanity is my motive” (T, p. 579). Sympathy thus produces the actions that are naturally virtuous. However, these native tendencies to virtue arise not from God but from a naturalistic psychological mechanism. Since sympathy is an innate mechanism, the natural virtues are in this sense innate, and native to us. This, however, is not a matter of innate knowledge, but rather a matter of innate tendencies to have sentiments productive of certain kinds of action. On this view, then, to sin in the Garden of Eden or during the Golden Age, would consist in acting contrary to what our natural human sympathy inclines us to do. But is this sin disobedience? Surely for the latter we must be conscious of what is right and wrong. For there to be disobedience we must have made moral judgements. These judgements are, Hume argues, perhaps notoriously, not matters of reason but of feeling. Not only do we perform naturally virtuous actions, we also feel moral sentiments about them: we have feelings of moral approbation towards the virtuous and of moral disapprobation towards the vicious. These feelings are themselves “certain peculiar sentiments of pain and pleasure” (T, p. 574). In other words, we take pleasure in, or pain in, the pleasure or pain that persons produce for others or for themselves. These pleasures and pains, our moral sentiments, are themselves produced by the mechanism of sympathy; our “strong sentiment of morals...can proceed from nothing but our sympathy...” (T, p. 580f). Since sympathy is an innate mechanism, we are “naturally” led to approve of virtuous qualities and disapprove of the vicious (T, p. 578, p. 579). This is true of the natural virtues (T, p. 578f). It is also true of the artificial virtues, once the conventions that lead to their producing pleasure are instituted (T, p. 577f); though the virtues are artificial, our moral approval of them, like our moral approval of the natural virtues, is natural, the product of an innate mechanism (T, p. 579f). Sin in the Golden Age, then, would consist in acting contrary to our natural human virtues and the sin of disobedience would consist in acting contrary to the way in which

61

One: Self as Substance

we are inclined to act by our natural judgements approving of those virtues, that is, contrary to the way in which our moral sentiments are inclined to move us. So, can there be disobedience in the Garden of Eden? Disobedience would consist in deliberately acting in a way contrary to the way in which our (conscious) moral sentiments are inclined to move us. A deliberate action requires a motive. Our moral sentiments reinforce our natural virtues; we would be acting contrary to our sense of duty only if we were acting contrary to some natural virtue. But since there is no scarcity, there is nothing to move us to deny pleasure either to ourselves or to others. In the Golden Age, neither self-love nor hunger could do this. And since there is nothing that can move us contrary to our natural virtues, there is nothing that can move us in a way that is contrary to our sense of duty. Is this disobedience? Surely only in a metaphorical sense.56 For, what the sinner would be disobeying would be nothing other than him- or herself. Still, there are those who would suggest that it is nonetheless disobedience to the (to be sure, self-imposed) moral law. The real issue is whether we would find instances of this sin in the earthly paradise, a world without scarcity. This eliminates as not virtues after all three of the qualities to which the Church suggested Adam and Eve were acting contrary when they precipitated the Fall. This leaves only the virtue of humility. Even here there must be some question whether this is a virtue. Pride is the contrary of humility, and pride is not only always agreeable to ourselves at least (T, p. 597) but further it “makes us sensible of our own merit, and gives us a confidence and assurance in all our projects and enterprizes” (T, p. 597). Since “the utility and advantage of any quality to ourselves is a source of virtue” (T, p. 596), it follows that we must reckon a “due degree of pride” (T, p. 596) a virtue. Humility will be a virtue only when it is opposed to undue pride. Thus, alone among the four possibilities humility can be seen as a natural virtue, that is, as something productive of human good even in a Golden Age, but even this is true only if humility is taken in the sense in which it is opposed not to pride per se but to excessive or inappropriate pride (T, p. 594ff). This, though, would fit with the views of the Christian theorists: Ambrose attributes the Fall to pride but it is an excessive – superbia or insolentia – pride, a desire on the part of man to make himself the equal of God.57 The Hobbesian state of nature is a consequence of a conflict of self-interests, however, and not to conflict of prides in the sense of contrariety in judgements of merit. To remedy the former we establish

62

Body, Mind and Self in Hume’s Critical Realism

the rules of justice that secure property; to remedy the latter “we establish the rules of good-breeding, in order to prevent the opposition of men's pride, and render conversation agreeable and inoffensive” (T, p. 597). The effect of the vice of excessive pride in the Golden Age would not be a Fall into a Hobbesian state of nature, but a Fall into a society of disagreeable people. And there is a remedy that is consistent with the condition of the Golden Age. The remedy consists in instituting a set of conventions, but these are not the conventions property, contract, or allegiance. They are, rather, as Hume says, the rules of good-breeding that thou shalt not, for example, brag and shalt not put down others. As do Raleigh and Clarke, Hume allows that there are natural virtues and that our moral approval of them is native. Unlike Raleigh and Clarke, however, Hume does not attribute this to God. In place of God we have the natural human mechanism of sympathy. In the Mosaic story, Adam and Eve sin, commit a vicious act, and as a consequence the source of moral law, God, expels them from the Garden to a world of scarcity where they must labour for food and shelter. In Hume’s story of the Golden Age, taken over from the poets, persons are capable of vice, acts contrary to the natural virtues. But the sorts of acts that Christians suggested precipitated the Fall are either not naturally vicious in the Golden Age, or, if they are, they do not lead to a Fall into a state of nature. The sympathy that teaches morality is humane enough not to punish us by forcing us to live in situations of material deprivation. Which is but another way of saying that the Humean picture of humankind is far less harsh than the Christian picture that he had known in his youth. The Christian myth of an earthly paradise helps him to make this evident. Nonetheless, if Hume agrees with Raleigh and Clarke against Hobbes that there are certain natural virtues, this part of morality proves to be fairly limited. The remedy for moral defects in the Golden Age does not consist in the sturdy if artificial virtue of justice but in the rather more anaemic artificial virtue of good breeding. The sturdy virtue of justice arises only in the context of adversity, not in the context of plenty. Interestingly enough, this too has a long tradition. Lactantius held that the first man had lived without the need to labour; he had been placed in a garden ut espers omnium laborum deo patri summa devotione servit [that free from all labour he might serve God the Father with the greatest devotion].58 But he also holds that “Virtue can neither be discerned unless it has vices opposed to it, nor can it be perfect unless it is exercised by opposition” (ibid., p. 38). From this point of view, from the beginning the

63

One: Self as Substance

Creator must have refrained from making life too easy for man and must have subjected him to difficulties and trials that arise both from external circumstances and from his own inner constitution (p. 41). This implies that the happy existence of the garden must have been a temporary state, an infantile prelude to man's real history: since “wisdom cannot exist apart from evil...the first of the human race, so long as he expressed only good, lived as an infant, ignorant alike of good and evil” (p. 39). We find Raleigh making the same point. As he had himself led a voyage to Guiana, he could state with authority that “if there be any place upon the earth of that nature, beautie, and delight, that Paradise had, the same must be found within...the Tropicks.” There were good rivers and stately cedars in the tropics, and “so many sort of delicate fruites, every bearing, and at all times beautified with blossoms and fruit, both greene and ripe, as it may of all other parts bee best compared to the Paradise of Eden.” If any part of the world had a claim to be regarded as the earthly paradise, it was the lands near the equator with their “perpetual Spring and Summer.” Raleigh did not conclude, however, that it was in fact the earthly paradise: the advantages fruitfulness and the temperate climate came with many disadvantages. “Lay downe by those pleasures...the fearfull and dangerous thunders and lightnings, the horrible and frequent earthquakes, the dangerous diseases, the multitude of venomous beasts and wormes.” In any case, it was hardly a country conducive to virtue; to the contrary, Raleigh agrees with Lactantius, such a country was morally corrupting: “for nature being liberal to all without labour, necessity imposing no industry or travel, idleness bringeth forth no other fruits than vain thoughts, and licentious pleasures.” It was for this reason, Raleigh holds, that countries near the equator “are called Terrae Vitiosae, vicious countries” (History, p. 28). Bacon had made the same point in his famous essay “Of Adversity”: “The virtue of Prosperity is temperance, the virtue of Adversity is fortitude, which in morals is the more heroical virtue....virtue is like precious odours, most fragrant when they are incensed or crushed, for Prosperity doth best discover vice, but Adversity doth best discover virtue.”59 The contrast is of the garden to the wilderness. The latter has its own Christian resonances. Israel had wandered for forty years in the wilderness. This was a time of temptation, testing, and sifting out the immoral and backsliding. It was a time for warning a wayward generation not to falter, murmur and backslide, to perish as had the six thousand who died when they defected.

64

Body, Mind and Self in Hume’s Critical Realism

And thou shalt remember all the ways which the Lord thy God led thee these forty years in the wilderness, to humble thee, and to prove thee, to know what was in thine heart, whether thou wouldst keep his commandments, or no. And he humbled thee, and suffered thee to hunger, and fed thee with manna, which thou knewest not, neither did thy fathers know; that he might make thee know that man doth not live by bread alone, but by every word that proceedeth out of the mouth of the Lord doth man live (Deuteronomy, VIII, 23).

Humankind learns of its moral duty in the context of testing in a land of hardship and labour, but at the same time, precisely because humanity here knew clearly the word of God and was close to Him, life in the wilderness is to be remembered with joy: “...the word of the Lord came to [Jeremiah] saying,” Go and cry in the ears of Jerusalem, saying, Thus saith the Lord; I remember thee, the kindness of they youth, the love of thine espousals, when thou wentest after me in the wilderness, in a land that was not sown (Jeremiah, II, 1-2).

Time and again the prophets turned for inspiration to the experience of Sinai. Elijah withdrew to the wilderness and the cave of Horeb, Elisha to that of Jordan60; Isaiah referred to the “voice of him that crieth in the wilderness.”61 John the Baptist retired to the wilderness, and Jesus himself, after being baptised by John, was led into the wilderness to be tempted by the devil for forty days and forty nights. It is not within the luxuriant and fruitful garden, leading an effortless life, that one finds God and virtue. These are to be found, rather, in a life that requires labour of one, and, if necessary, self-deprivation. Here the Garden as a place of ease and softness changes place with the Wilderness as a place of labour, trial and hardship; the Garden becomes a moral wilderness, while the wilderness, the forest or the desert, becomes the place in which virtue flourishes, the moral paradise. The theme of the trials of the wilderness as necessary for the production of virtue was familiar indeed in English, Scottish and French religious thought in the period prior to Hume. Thus, in England, the dissenter Richard Baxter (1615-1691)62 wrote in The Saint’s Everlasting Rest: If they [the Israelites] had not felt their wilderness necessities, God should not have exercised his wilderness providences and mercies. If man had kept his first rest in paradise, God had not had the opportunity to manifest that far greater love to the world in the giving of his Son. If man had not fallen into the depth of

65

One: Self as Substance

misery, Christ had not come down from the height of Glory, nor died, nor risen, nor been believed on in the World. If we were all well, what need we the Physician. And if all were happy, and innocent, and perfect, what use were there for the glorious works of our sanctification, justification, preservation, and glorification: what use for his ministers, and word, sacraments, and afflictions, and deliverances?63

In France, with the revocation of the Edict of Nantes in 1685, the Huguenots who did not emigrate took flight to the caves, forests and desolate places. They called their forbidden conventicles “les églises du désert.” Relying upon texts such as Revelation, XII, 6, And the woman fled into the wilderness, where she hath a place prepared of God, that they should feed her there a thousand two hundred and threescore days.

The fugitive church had to keep revising its eschatology but never lost confidence in itself as the true Church temporarily hiding in the wilderness.64 And in Scotland, one finds similar themes among the Covenanters. Charles I had attempted to impose Anglicanism on the Scots, and this effort was renewed after the Restoration of Charles II. These attempts to eliminate the Presbyterian Church of Scotland were resisted by the Covenanters.65 Among the latter, the same image that moved the Huguenots, of ancient Israel wandering and covenanting in the wilderness, played a powerful role. One finds it, for example, in the letters of Samuel Rutherford66 collected in his Joshua Redivivus,67 and in the thought of one of the greatest of the later Covenanting prophets, Alexander Peden,68 who preached to the conventicles of the moors and caves69 on the themes to be found in his The Lord’s Trumpet Sounding Alarm Against Scotland.70 Thus, Rutherford sees his own virtue as being the greater for having been tested through a series of trails: “I doubt not but my Lord is preparing me for heavier trials. I am most ready at the good pleasure of my Lord, in the strength of his grace, for any thing he shall be pleased to call me to...” (Joshua Redivivus, p. 78). But it goes beyond the individual to the community: the virtue that resides in the community, in this case, Scotland, is not native, but emerges only out of its own trial in the wilderness: ...people shall be stolen off their feet with well-washen, & white-skinn’d pretences of indifferency; but it is the power of the great Antichrist working in this land. Woe, woe, woe be to Apostate Scotland: there is wrath, & a cup of the red wine of the wrath of God Almighty in the Lord’s hand, that they shall drink

66

Body, Mind and Self in Hume’s Critical Realism

and pue and not rise again. The star called Wormwood & Gall is fallen in the fountaines, and rivers, & hath made them bitter: the sword of the Lord is flourished against the Idol-worshippers of the land; women shall bless the barren womb & miscarrying breasts; all hearts shall be faint, and all knees shall tremble, an end is coming, the leopard and the lion shall watch over our cities, houses great and fair shall be desolate without all inhabitant: the Lord hath said, Pray not for this people, for I have taken my peace from them; yet the Lord’s third part shall come through the fire, as refined gold for the treasure of the Lord, & the out-casts of Scotland shall be gathered together again, & the wilderness shall blossome as the flower, & bud, & grow as the rose of Sharon, & great shall be the glory of the Lord upon Scotland (pp. 77-8).

Peden makes the same points. Persecution, hardship, the wilderness, are required for individual virtue: “But now persecuted people of God, fear not at the Cross, for it is the way to the Crown, Trouble and Suffering hath always been the Lot of Saints, and began as soon as Grace died” (The Lord’s Trumpet Sounding and Alarm against Scotland, p. 30). The same is true of the community: “...Ease is never good for the Kirk of GOD, and the People of GOD; For the Kirk and People of GOD Thrives ay best under sadest persecution; it hath been the Experience of the Kirk and People of GOD in all Ages” (p. 25). This holds in particular for Scotland (and the “other Two Lukewarm Indifferent Lands” [England and the Netherlands]), which shall not escape, but shall drink oft the Dregs of the Cup of His Anger: but indeed the Blood of the Persecuted Remnant that hath been shed to Witness against Defection for a Testimonie speaks best to thee, O Scotland, a Remnant of thee shall escape, and shall be Monuments of His Mercie in all this Flood of Wrath that is coming on these Lands (p. 15).

Near his death and again hiding in a cave, Peden still preached the birth of virtue in the adversity of the wilderness, prophesying to some friends that God shall make Scotland a desolation. 2dly, There should be a remnant in the land, whom God should spare and hide. 3rdly, They should lie in holds and caves of the earth, and be supplied with meat and drink; and when they come out of their holds they should not have freedom to walk for stumbling on dead corps. 4thly, A stone cut out of the mountain should come down, and God should be avenged on the great ones of the earth and inhabitants of the land, for their wickedness, and then the Church should come forth with a bonny bairntime at her back, of young ones.71

67

One: Self as Substance

Hume undoubtedly was familiar both with these ideas and with much of the literature, and his story of the emergence of the conventions and virtues of justice and promise-keeping echo these notions. It is not in the Golden Age that we find the virtues of justice and covenanting. They arise, as Lactantius had said, only where life is rendered difficult by external circumstances and by the problems of our own internal constitution, under the condition of scarcity which requires labour and commerce and under the condition of limited generosity. These conditions of hardship would lead to the Hobbesian state of nature were it not for human being’s reason which shows that people can, by an artifice, rescue themselves from such a state. This artifice is the convention first of property and then of covenanting or promising. Long-range self-interest leads to the institution of these conventions. The felicity that actions in conformity to these conventions create for humanity means that they are, given general conformity to the conventions, virtuous. Our awareness of this felicity means in turn that the mechanism of sympathy produces feelings of moral approbation towards actions that conform to the conventions. Note, however, that while Hume agrees with Raleigh and the covenanters like Peden, that people acquire the capacity virtuously to covenant in conditions of the wilderness, he in contrast holds that what redeems humankind from corruption and misery is not the grace of God but rather our own reason and our human capacity to react sympathetically to others. Once again, Hume’s moral philosophy is redolent of themes familiar to a readership that was well versed in the stories of the Greeks and Romans and of the Bible and in fact at times took both sets of stories to be as it were Gospel truth. But also once again, the themes are humanized and secularized. The picture of humanity that Hume was creating in turn became a powerful force, one that meant that once it was embraced no one could any longer take seriously the themes from the Greeks and the Romans and from the Bible that had lent considerable force to Hume’s argument. The picture is still compelling, but it is now compelling more by force of argument than by echoes from a mythical past. Certainly, Hume has provided a clear case against the critics of Hobbes, that one can on the one hand reject any account of value as residing in a metaphysical human nature and the ontological structure of the universe while also holding, on the other hand, that there is a basis in our empirical human nature for both the natural virtues and also the artificial ones that make civil society possible.

68

Body, Mind and Self in Hume’s Critical Realism

(5) George Grant: Aristotelian Moral Philosophy Made Modern Political philosophy began in the classical world and began there with a priori arguments. These arguments were made systematic in Aristotle. They were criticized by empiricist philosophers such as Hobbes and Hume. The classical substance tradition has continued to have its defenders, however. It has been ably presented in particular in the Canadian context by the person who is perhaps the best known among those who have reflected on politics in Canada, George Grant. His presentation of classical political philosophy shows the connection in the modern world of the arguments we have found in Aristotle. In particular, Grant locates value, the standards towards which we ought to strive, in the ontological structure of the universe. According to Grant,72 morality is given to us in the form of Natural Laws. “The theory of natural law is the assertion that there is an order in the universe, and that right action for us human beings consists in attuning ourselves to that order” (p. 28). This Natural Law is an objective moral order: “We do not make this law, but are made to live within it” (p. 29). There are two assumptions. The first assumption is that the universe “conforms to law; and to conform to law is to be held in being by reason” (p. 29). The order that we discern in the universe is not merely matter of fact regularity, not merely the patterns that we perceive by means of our sense experience, but patterns that reflect the moral structure of the universe. All things live “within” this moral order, and are “attuned” to it: this is how they are “made.” It is thus in the natures of things that each strives to reflect in the world of sense the divine transcendent and objective moral order. But this striving is not an arbitrary attachment. It is a natural “attunement” between the natures of things and how they act, on the one hand, and, on the other, the objective law. Things naturally move towards the form required by the law because the law itself moves them in that direction. The moral order thus has the power to move things, and is the reason why things are as they are. It is in this sense that things are “held in being by reason.” The order that we perceive in the universe is thus a teleological order deriving from the power of the transcendent moral order to draw things according to their nature into certain forms of being, to wit, those forms required by that divine law. Thus, things have natures. These natures are such that things strive to reflect in their being the moral law. The moral law which so moves a thing that by its nature it strives to achieve, to be, provides the reason why the

69

One: Self as Substance

thing is as it appears to be; the thing is thus “held in being by reason.” The second assumption is that in particular there is a universal human nature and that “this universal human nature is to be a rational creature” (p. 32). Other beings are held in being by reason, but human being alone has the capacity to grasp the reasons for things. The capacity to grasp the reasons of things is reason; rationality just is this power. But the moral provides the reasons for things. These include the reasons for its own being. Thus, it is the essence of human being to grasp the moral law which constitutes the reasons of things and in particular the reason for its own being. The natural law so moves things that they are held by it in being, in its own image. But there is also a natural attunement between this natural law and the reason that grasps it. There is a natural law for human reflection, a standard or norm to which it must conform: this is the standard provided by the natural law itself. Here, too, as everywhere else, there is a natural “attunement” between being in the world and the transcendent law of that being. Human reason thus naturally moves to grasp the transcendent moral order which itself draws the human mind toward it. The human reason, like all other things, is to be understood as part of a teleological order, in which what moves reason is the reasons for things. Most things strive to bring into to being the transcendent order through a non-reflective, non-conscious motion. In the case of human being, the motion is, or can be, consciously directed. Since to grasp the reasons of things is to grasp the objective moral order, and, since we, like all things, strive to achieve that order, it is our nature as rational beings to choose our ends so that they conform to this objective moral order, and so bring about in our own being a reflection of this moral order: “justice is only the living out in time of a transcendent eternal model of justice” (p. 19). There is an order in the universe which human reason can discover and according to which the human will must act so that it can attune itself to the universal harmony. Human beings in choosing their purposes must recognize that if these purposes are to be right, they must be those which are proper to the place mankind holds within the framework of natural law. We do not make this law, but are made to live within it (p. 29).

We are not born fully attuned to the moral law. That we become so attuned, that we become fully moral beings, is the task of social and educational institutions. They must make us moral.

70

Body, Mind and Self in Hume’s Critical Realism

Education is seen as the process by which a person comes to think clearly about the proper purposes of human life....In the old theory of education, when a man began to see what was the ultimate purpose of human life, he was said to be wise – to have the virtue of wisdom. Wisdom was then the purpose of education (p. 32).

When wisdom is achieved we achieve an identity between our reason and the divine reason, when our own reason becomes one with the objective moral order, the objective reasons that hold things, ourselves included, in being. Thus, indubitable certainty of the moral reason that holds us in being is the consequence of an education the point of which is to cultivate reason. To live according to nature is then the supreme good for a man, as it was for any being in the universe. But, of course, our nature being higher than that of an animal demands of us a higher way of life than that which is good for an animal. The good life for man includes within it the perfection of his rational nature. Man is to be perfected and brought to his highest possibility through the union of his reason with the divine reason. Thus, the logical completion of natural law is to pass beyond a simply practical life to a life of mysticism. Practice passes into adoration (pp. 32-3).

Since humans are by nature rational animals, there is a natural tendency within us to grasp the divine truth, the objective moral law to which we must strive to conform our behaviour. This natural tendency explains our behaviour. If we do not conform to the objective standard, then that is because the natural tendency to grasp the truth has somehow been thwarted in its strivings. Grant argues that in fact persons in modern western society, the United States and Canada in particular, do not strive after virtue; they seem in fact to have forgotten it. This is, he suggests, because modern technological society has so managed to produce goods that satisfy our bodily wants and indeed has so managed to produce wants in us that demand even more material goods to satisfy them that we are totally involved with the satisfaction of wants to the exclusion of the search for justice and wisdom. The development of technological society and capitalism has succeeded in using us for its own material ends and has done so by causing us to hide from ourselves the truth that, as natural beings, we are inclined towards an objective justice and divine law that lies beyond, and indeed should control, the satisfaction of our material needs and wants. This modern predicament is all the fault of the empiricist philoso-

71

One: Self as Substance

phers, starting, Grant argues, with Bacon. ...what meaning is there in Bacon’s attempt to cut off truth humanly discoverable from the revealed dogmas of religion? Its only implication is that the truths of religion are not rational but arbitrary. Therefore, it leads to an exaltation of the truths of natural science, and such an exaltation, coupled with man’s original sin, leads straight to the griming mask of scientific humanism at Hiroshima.73

The empiricists – Bacon, Locke, Hume – attacked the claim that there are forms or essences, and in so doing they attack the notion that values are built into the ontological structure of the universe. But those forms are also the reasons for things, reasons both for what things are and for what things, especially ourselves, ought to be. That attack opened the way for the common acceptance of acts that are as a matter of objective fact morally reprehensible. It is the attack on the classical tradition in metaphysics that has made the world the evil place that it now is. So, at least, George Grant. Grant stated the issue more completely, with reference to the modern world, in the twentieth century, in response to one whom he took to be a leading representative of the empiricist tradition, Bertrand Russell.74 Russell defended the Hobbes-Hume view of the world, in which there are no forms or natures and, in particular, no metaphysical human nature, nor therefore any objective values. At the same time Russell vigorously defended a variety of moral positions, including the claim which he apparently shared with Grant, that we ought to abandon nuclear weapons. Grant argues that Russell’s position is clearly inconsistent. As Grant puts it, Russell is acting as a moral leader when he issues his injunctions; but he does this without attempting “to discover the principles underlying these [Russell’s] activities” (p. 323). But Russell proposes that such principles cannot be discovered; he cannot invoke any principles because, when he develops his empiricist picture of the world, there is no place for objective moral standards for reason to know. “Russell cannot have it both ways. He cannot be sceptical about the positive role of the philosopher in discovering ethical principles and also expect us to take with any seriousness his statements about how we ought to live” (p. 323). This, however, is misleading indeed, and quite unfair to Russell and to his philosophical position. Russell is in fact not a sceptic about objective moral values. He argues that such standards do not exist: this is something that he knows. But knowing that, he also rejects any account of reason – human reason – that requires the latter to be able to discern objective val-

72

Body, Mind and Self in Hume’s Critical Realism

ues. But it does not follow from the fact that there are no objective values that therefore there are no values: for, there are relative values. Grant is suggesting that one who rejects any claim that there are objective values has no right to pronounce moral judgments. He gives no reason for accepting that inference. In fact, if there are no objective values in the ontological structure of the world, then one cannot turn to them to justify his moral judgments; but it does not follow that one must refrain form making moral judgments. In fact, one can quite well argue, as Hume does, and as Russell also does, that there are grounds for some positions as more morally justified than others. To be sure, such a justification will not be an appeal to objective standards; but it does not follow that there is no justification available to justify virtue and oppose vice. We are told that “if Russell is right and reason cannot speak about the fundamentals of conduct, then he is using the world ‘should’ with no rational significance” (p. 324). But this is merely a colossal begging of the question; it is not rational argument. Russell has argued that there are no reasons in Grant’s sense, in the sense of Plato and Aristotle. It does not follow that there cannot be good reasons for and against various proposal about what we ought to do. To be sure, they rest upon human feelings and passions, our subjective moral sentiments. But that does not prevent them from being reasons. Not reasons of the sort Grant would like, but there are none of the latter, so we had best make do with what we do have. Grant suggests that relying upon our subjective moral sentiments means that our moral judgments are “no more than expressions of like and dislike or of command...” (p. 325). There is, however, a difference between our likes and dislikes, on the one hand, and our moral sentiments, on the other. The latter cannot be reduced to the former. Hobbes thought that could be done, but Hume argued that Hobbes was wrong, and Russell agrees with Hume. Likes and dislikes are arbitrary, where our moral sentiments are not in that sense arbitrary. To be sure, they are arbitrary if one is an Aristotelian, but Hume and Russell deny that there is such an objective basis for our moral values. The latter therefore require a different sort of justification. Hume suggested such a justification for the norms of civil society, and upon this reasoning these norms are far from arbitrary. Grant does not argue against Russell, he merely asserts. But until he goes beyond assertion to present some arguments we have no reason to accept his rhetoric. Again, Grant suggests that making his moral judgments and recommending them to others means Russell is simply insincere and misleading. For, upon Russell’s view those principles are ones which he “rejects as

73

One: Self as Substance

false” (p. 326). But Russell does not reject them as false; but neither are they true. There is simply no objective basis for saying that they are either true or false. It does not follow that there are no grounds for supposing that some moral judgments are better than others, and in that sense more reasonable. It is equally misleading for Grant to suggest that Russell, in making and recommending his values as justified is “fostering illusion” (p. 326). For Russell, as for Hume, however, it is Grant who is fostering an illusion, the illusion of objective values. Again, it is misleading and simply question begging to assert that Russell’s defence of certain forms of action as virtuous and others as vicious “can only lie in a completely irrational acceptance of one tradition as against another” (p. 327). Russell’s defence is irrational only if one accepts Grant’s position that there are objective values. But Russell, following Hobbes and Hume, is arguing that there are no such objective values, no such objective reasons to ground our moral judgments. It does not follow, however, that there are no grounds for preferring one set of moral judgments against another, one tradition as against another; and in that sense the acceptance of a moral stance is far from irrational: what can be backed by reasons, the only sort of reasons that there are, is not to be condemned as irrational. Grant suggests that he has “never met a human being who does not hold some conception of the highest good, however, imperfectly formulated and imperfectly followed. That is, I [Grant] have never met a human being who was not capable of giving some faint semblance of order to his [or her] desires” (p. 330). One such as Hume and Russell can only agree. But that ordering is not based on a set of objectively grounded set of values. It does not follow, however, that it is not grounded. To the contrary, our moral sense is grounded, however imperfectly, in our common sense of humanity, a sense that we derive from our very human capacity to respond sympathetically to others. Is that not something itself to be valued, particularly when contrasted to Grant’s illusion of objective values? Certainly, it is evident that it is to argue that on the Hume-Russell view of human being, people “are not rational animals, but clever beasts with a facility for mathematics” (p. 334). This is to suggest, wrongly, that if one rejects the claim – the illusion – of objective moral values, then one is holding that human beings are beasts with no moral sense, clever beasts but still essentially without moral values. We are back to the Anglican critics of Hobbes. Grant has not passed beyond them and certainly has not recognized how Hume made a serious advance on Hobbes, recognizing our moral sense and the rational basis in our common human sense of sympa-

74

Body, Mind and Self in Hume’s Critical Realism

thy for others for our moral reasons for action. One can make similar comments about another remark of Grant’s. Commenting on Locke’s moral philosophy, which like that of Hobbes and Hume, denies that there are objective values, he states that “the untruth of the traditional teaching means that there is no ... highest good given to human beings in their recognition of the way things are.”75 The characterization of Locke is just: there are no objective values, and the tradition which says so, Grant’s tradition, is untruth. The foundations of the rules of justice in civil society must be elsewhere than in the ontological structure of things. Locke’s answer, like that of Hobbes and Hume, is that “justice is contractual, not natural” (p. 17), that is, not in the objective natures or forms of things. “Justice is those convenient arrangements agreed to by sensible men” (p. 18), or, rather, sensible persons. But Grant objects to the use of the term ‘person’ here. To call human beings ‘persons’ is clearly not a scientific description. What is it then? If the word is a true description but not scientific, is it not then one of these supra-scientific metaphysical terms which analytical philosophy would have us eschew? (p. 32)

Grant is onto something here. If we do give up the traditional forms or natures, and in particular the form or nature of human beings, then we can no longer give the traditional account of human being: human beings are no longer to be construed as substances with a certain nature or form. But then to call human beings persons is something “mysteriously ... brought in to cover up the inability to state clearly what it is about human beings which makes them worthy of high political respect” (p. 32). Clearly, ‘person’ as implying that that to which it is applied is something worthy of respect can no longer be construed as a descriptive term, scientific as Grant would have it, but is rather normative: human beings are those beings whom we ought to give political respect. As Locke was to put, the term ‘person’ is a forensic notion, not a metaphysical notion. ‘Person’ ... is a foresnsick term appropriating actions and their merit, and so belongs only to intelligent agents capable of a law, and happiness and misery. (Locke, Essay concerning Human Understanding, bk. II. Ch. 27, sec. 26).

Grant suggests that since the term no longer refers a metaphysical reality, it is therefore wrong for the empiricist, the analytical philosopher, to use the term: values have disappeared. But only objective values have disappeared.

75

One: Self as Substance

Locke and the other empiricists have not given up, however, the moral sense of human beings, nor the capacity of human beings to make moral judgments. What are human beings, these being worthy of our respect? As Locke says, just those beings with whom we form civil society. To the extent that a member of the species homo sapiens is capable of living with others in civil society to that extent we deem him or her worthy of respect and a carrier of human rights, we deem him or her, that is, a person. ‘Person’ is thus a normative term, introduced to designate those beings whom we ought to treat as bearers of rights in civil society. Grant is correct: this term is not one of science, which uses only descriptive terms. It is indeed normative. But those norms are rooted in ordinary human empirical needs and being, not in the ontological structure of the universe. Or at least, Grant provides no argument for there being such a metaphysical grounding of value. And until he does the empiricist need not worry about his rhetoric. But Grant is surely correct in one thing: What one holds with regard to objective metaphysical natures or forms of things has profound effects on how one defends or criticizes various moral claims. For a substance, e.g., a person, to fulfil its nature is for it to be as it ought to be; the nature determines that at which it aims and so to be in that natural state is its good. To frustrate its movement towards its good is to bring about in it an unnatural or violent change: it is an evil, an objective evil grounded in the natures of things. Such an injury is the worse the greater the attainment of the goal of its natural motion is frustrated. Death is therefore the greatest evil, while murder, as Aquinas says, is an “injury to one’s neighbour against his will,” and is in fact the “greatest injury.”76 To be sure, lawful execution can be justified. The “imperfect are for the perfect”, and from this it follows that it is “not unlawful to use plants for animals, and animals for man” (SS Q 64 A1) For the same reason it is generally unjustified to kill a man: “to kill a man is evil in itself, since we are bound to have charity towards all men...”. (SS Q 64 A2 Obj 3) This charity that binds us to all men (and women) is built into our natures, and through that we are parts which form and contribute to the social whole. The part has as an aspect of its essential being that it is part of the whole. The whole is the perfection of the part, and the social the perfection of the individual: “every part is directed to the whole as imperfect to perfect, wherefore every part is naturally for the sake of the whole.” (ib.) Now, sometimes parts are diseased and threaten the well-being of the whole: the disease of a part can threaten the whole. Thus, “if the health of the whole

76

Body, Mind and Self in Hume’s Critical Realism

body demands the excision of a member, through its being decayed or infectious to other members, it will be both praiseworthy and advantageous to have it cut away.” (ib.) And equally, then, when a person who threatens the well-being of the whole community, it is praiseworthy and advantageous for a person “to be killed in order to safeguard the common good...”. (ib.) Thus, if a person is deemed by duly constituted authority to have harmed or to have threatened to harm the social whole, the common good, it is lawful to execute him as it is lawful to excise a seriously infected tooth. In short, criminal or various sinful acts are punishable by death, but otherwise killing is wrong: it brings about the non-being of a substantial entity whose nature aims at it being, that is, being in its natural state. This provides Aquinas’ argument against suicide. It is in effect selfmurder, and is forbidden for the same reason that killing another is forbidden. Suicide is forbidden because it is contrary to its natural inclination of a substance to bring itself into being: ...everything naturally loves itself, the result being that everything naturally keeps itself in being, and resists corruptions so far as it can. Wherefore suicide is contrary to the inclination of nature, and so charity whereby every man should love himself. Hence suicide is always a mortal sin, as being contrary to the natural law and to charity.77

As we noted before, this argument clearly depends upon the substance metaphysics deriving from Socrates’ model of explanation. When the soul grasps the Forms, or as Aquinas would put it, the Natures of things, it grasps the reasons for things; and in grasping its own Form or Nature, it is moved to imitate that Form in its own being in time in the world of sense experience. The soul in grasping its Form is active in establishing its own being in the world. For it to move towards its own non-being by committing suicide would, in effect, be for it to turn away from reason and the natural law which ought to move it. For Aquinas, such turning away is a sin. In a case of suicide, save perhaps in exceptional circumstances, the soul, the seat of reason, would have to be moved by, say, the passions, by something in any case other than reason, that is, the reason that grasps the reasons for things. Any reason that dictated suicide would at once be moved by a form or nature towards both its being and its non-being. So it contradicts itself, formally, in obliterating itself. It is thus formally self-contradictory to say that it is right and in conformity to reason that one commit suicide. So the

77

One: Self as Substance

suicide not only violates that natural law that each seeks to be, but also the rule that actions ought to be guided by reason. Aquinas also argues that suicide is forbidden because it injures the community: Suicide is forbidden ...because every part, as such, belongs to the whole. Now every man is part of the community, and so, as such, he belongs to the community. Hence by killing himself he injures the community (ib.).

Socrates already in the Phaedo had made this point. Socrates there argues that “this seems to me well expressed, that the gods are our guardians and that men are one of their possessions.” His interlocutor Cebes agrees, and Socrates goes on: And would you not be angry if one of your possessions killed itself when you had not given any sign that you wished it to die...?

Again Cebes (we are not surprised) agrees. Perhaps then, [Socrates continues,] put in this way, it is not unreasonable that one should not kill oneself before a god had indicated some necessity to do so, like the necessity now put upon us (Phaedo, 62b4-c5).

The necessity, of course, turns out to be moral necessity, that which is required by duty. As for the gods, that which is divine, these turn out to be the forms, those entities that not only explain why things turn out as they do but also tell us how things ought to be. They, the forms = the divine, guide us in the sense of informing us of what is for the best; and we are their possessions in the sense that when we grasp them it is they who determine how we behave and act. We receive a sign from them precisely when we grasp them. In grasping them we understand, and act to realize, our moral duty. And so, for the gods to indicate that there is a necessity to kill oneself is simply to know that it is one’s duty to do so. Aquinas repeats Socrates’ argument that “life is God’s gift to man, and is subject to His power, Who kills and makes to live. Hence whoever takes his own life, sins against God, even as he who kills another’s slave, sins against that slave’s master, and as he who usurps to himself judgment of a matter not entrusted to him” (op. cit.). But Aquinas makes clear that he intends more than does Socrates: for Aquinas it is never the case that one ought to commit suicide: “It is altogether unlawful to kill oneself...” (ib.).

78

Body, Mind and Self in Hume’s Critical Realism

Thus, he quotes Augustine78 to the effect that suicide is forbidden by the divine commandment, which we are bound to obey, that “Thou shalt not kill”: Hence it follows that the words “Thou shalt not kill” refer to the killing of a man; – not another man; therefore, not even thyself. For he who kills himself, kills nothing else than a man (ib.).

Hume’s essay “On Suicide”79 is perhaps the most notorious of the defences of the rationality of suicide. It can be seen as a reply to Aquinas. Aquinas’ position depends, as we have seen, on the natural law ethics deriving from the substance tradition and the Aristotelian model of explanation. Hume in this simply takes for granted that this position is mistaken; he makes the assumption that the arguments that he has deployed elsewhere against forms and natural necessities are sound. Thus, Hume understands the ‘nature’ of ‘natural law’ to mean nothing more than the empirical universe, and the ‘law’ of ‘natural law’ to refer to matter-of-fact regularities in the universe. And it is quite true that once the metaphysical basis for Aquinas’ arguments concerning suicide is eliminated, those arguments lose most of their force. When Hume examines these Thomistic arguments from his own naturalistic and non-substantialist perspective, he finds them to be, not surprisingly, uncompelling. Thus, to the Thomistic argument that “Life is God’s gift to man, and is subject to His power,” Hume replies that God exercises His power through natural causes, and that taking one’s own life is simply one natural cause among others; it cannot, therefore, run counter to God’s power. In fact, the argument is absurd, since it would equally argue against the protection of our lives from threatening natural causes. Were the disposal of human life so much reserved as the peculiar province of the Almighty that it were an encroachment of his right for men to dispose of their own lives; it would be equally criminal to act for the preservation of life as for its destruction. If I turn aside a stone which is falling upon my head, I disturb the course of nature, and I invade the peculiar province of the Almighty by lengthening out my life beyond the period which by the general laws of matter and motion he had assigned it (“On Suicide,” p. 107).

Where you have a distinction between natural causes, those that flow from the Natures of things, and unnatural causes, those that act contrary to the Natures of things, then Aquinas’ point can be made; but no such distinction is available in the empiricist philosophy of Hume and Epicurus that denies

79

One: Self as Substance

that there are any such entities as the Natures of things. Hume, therefore, from this perspective rightly rejects this Thomistic argument against suicide. The next argument that Hume considers is the Thomistic argument that in suicide a person harms the community. But, Hume argues, “a man who retires from life does no harm to society; He only ceases to do good; which, if it is an injury, is of the lowest kind.” Hume is prepared to allow that we do have obligations to the society in which we live; but there are limits, and other considerations, in particular those of self-interest, may on occasion override those duties towards others. All our obligations to do good to society seem to imply something reciprocal. I receive the benefits of society and therefore ought to promote its interests, but when I withdraw myself altogether from society, can I be bound any longer? But, allowing that our obligations to do good were perpetual, they have certainly some bounds; I am not obliged to do a small good to society at the expense of a great harm to myself; why then should I prolong a miserable existence, because of some frivolous advantage which the public may perhaps receive from me? If upon account of age and infirmities I may lawfully resign any office, and employ my time altogether in fencing against these calamities, and alleviating as much as possible the miseries of my future life: Why may I not cut short those miseries at once by an action which is no more prejudicial to society? (ib., pp. 109-10).

For Aquinas, of course, removal from society is contrary to human nature which establishes us, through our natural charity, to be essentially social, and suicide therefore constitutes an injury to society; but that argument assumes the doctrine of Natures which Hume rejects. Hume, therefore, from the empiricist perspective rightly rejects the Thomistic argument that suicide is always, of its nature, injurious to society. Once that general conclusion is withdrawn, it is easy for Hume, to counter, on the basis of considerations of utility and self-interest, that suicide does not always constitute a blameable harm to others with whom we are bound up through our social relations into larger social units. Indeed, Hume goes on, when life has become an intolerable burden, suicide itself may acquire social utility. For, in that way I relieve society of the burden of supporting me: ...suppose that it is no longer in my power to promote the interest of society; suppose that I am a burden to it; suppose that my life hinders some person from being much more useful to society. In such cases my resignation of life must not only be innocent but laudable (ib., p. 110).

80

Body, Mind and Self in Hume’s Critical Realism

Indeed, in such circumstances suicide is “...the only way that we can then be useful to society, by setting an example, which, if imitated, would preserve to everyone his chance for happiness in life and would effectually free him from all danger or misery” (ib.). In general, Hume concludes, “That suicide may often be consistent with interest and with our duty to ourselves, no one can question who allows that age, sickness, or misfortune may render life a burden, and make it worse even than annihilation....If suicide be supposed a crime ’tis only cowardice can impel us to it. If it be no crime, both prudence and courage should engage us to rid ourselves at once of existence, when it becomes a burden” (ib.). Thus, contrary to Aquinas death is not always an evil, and we cannot conclude, as some have suggested, that “a bad end is in store for us all.” Hume’s point is that we do have aims, and dying prevents, often, the fulfilment of those aims and projects. But if we have completed our projects, if we have more or less satisfied our life’s aims, then death is no “bad end”: in dying nothing is lost. And in other contexts, it may be that there are projects unfinished but the loss that occurs through death is outweighed by some gain or other, say the avoidance of pain. There is no metaphysically grounded end that aims simply at our being, and which, by virtue of the metaphysical necessity, has our continued being trumping all other ends. But Hume also notes also that we often have reasons not to commit suicide, reasons for wanting to go on living, things we want to do and ought to do. “I believe that no man ever threw away life while it was worth keeping. For such is our natural horror of death that small motives will never be able to reconcile us to it...” (ib.).80 A premature death that prevents one from fulfilling his or her potential or aims or commitments may well yield grounds for regretting that death. To that extent, death may at times amount to a “bad end.” But where a person has fulfilled his or her hopes, as for the most part Hume himself had when he felt death approaching, there is nothing to regret. Once again, it is simply not true that death is a “bad end” that awaits us all. There is a disagreement between Hume and Aquinas. In one sense, it is over suicide. But beyond that it is over the arguments that Aquinas’ used to condemn suicide and the arguments that Hume proposed could be used to justify it. They are arguing in different ways that simply bypass each other because there is a deeper difference. This is over the issue whether there are natured substances or not: Aquinas argues that there are and relative to that position his arguments against suicide are sound; and Hume ar-

81

One: Self as Substance

gues that there are no substances and no natures or forms and relative to that position his arguments in defence of suicide are sound. Before the issue of ethics can be settled, it is first necessary to settle the metaphysical issues. But of course Hume has done just this: he has argued on empiricist ground that there are no substances and no objective necessary connections, that is, no natures or forms. This defeats Aquinas’ case, removing the presupposition upon which the soundness of his arguments rests. And it establishes his own metaphysical position, allowing his arguments that suicide can be justified to stand as sound. There is a similar point to be made about the morality of abortion. Assuming that the ovum when fertilized becomes as the foetus a substance, then at that point it has a nature or form, in this case the form of being human that it shares with human beings; it is as such a substance a person. Since to abort the foetus would prevent that person coming to be what, by virtue of its form, it aims naturally to be, an evil befalls it, in fact the greatest evil of death. Abortion is therefore, like suicide, and for the same reason, forbidden as contrary to law – and contrary to our natural charity towards other persons. As one of the defenders of the Roman Catholic position on abortion has put it, “The greatest physical evil one can suffer is to lose one’s life and existence, not just because one’s ‘person’ or ‘personal’ capacity to relate with others is lost, but because the very existence itself of the individual is lost in death,”81 Now, as we have seen in discussing the morality of suicide, death need not be the greatest evil; indeed, there is no reason at to think that non-being is, in itself, any sort of evil, and every reason to think that in certain contexts our non-being, our death, is a relief from evil. At least, that is so if one rejects, as Hume and Humeans do, the notion that persons are, metaphysically, natured substances. It is central to the claim that abortion from the moment of fertilization on is immoral that the foetus is a substance which has, of necessity, a form or nature, where this form or nature is an active potentiality the activity of which creates the being throughout of the foetus as a maturing person. Thus, Catholics, in defence of their position on abortion, argue that The person is the being which causes the development of these states [the states of consciousness that are successively the stages in life’s history] as well as human bodily states, and personhood is not the consequence of these states but is their cause. (ibid., p. 161)

The teleological nature of the explanation is clear: the cause is the end itself, pulling as it were from in front, rather than pushing from behind. That

82

Body, Mind and Self in Hume’s Critical Realism

nature is conceived to be an active form the striving of which for actuality is the cause of the development of the adult which is, normally, also the person, that at which the thing, the substance, aims to be and which, because that is its aim, it also ought to be. Natural potentialities need nothing but the proper matter to be reduced to action, and the potentiality of the fetus to become an adult is an active natural potentiality or tendency which is guaranteed as long as no external factors intercede. (ibid., p. 164)

The difficulty is, or ought to be, the metaphysics, the ontology of natural or substantial forms. It is the assumption that there are in the structure of things forms that determine objectively moral value. It is those forms that the empiricist rejects. For such a philosopher there is no objective standard of value or of moral worth: “is” does not imply “ought.” As Hume had made the point, In every system of morality which I have hitherto met with, I have always remarked, that the author proceeds for some time in the ordinary way of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when of a sudden I am surprized to find, that instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible; but is, however, of the last consequence. For as this ought, or ought not, expresses some new relation or affirmation, it is necessary that it should be observed and explained; and at the same time that a reason should be given, for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it. (T, p. 469)

For the empiricist, then, the fact that a fertilized ovum will, if left to normal processes, will develop into a being of the sort that we reckon a person and therefore something that is a bearer of moral rights, does not, as a mere fact imply that it, too, as a fertilized ovum, ought to be treated as a bearer of rights. The fact that a fertilized ovum has in some sense the potentiality to become a person does not make it into a person. The potentiality is nothing other than the fact that if such and such happens then so and so eventuates. The issue is, ought a fertilized ovum be treated as a bearer of rights. The fact that it is a fertilized ovum does not by itself establish this norm. To establish that the fertilized ovum is the bearer of rights because it is already a person requires a defence of the Aristotelian metaphysics of objective value. One does not validate one’s position on abortion by simply tak-

83

One: Self as Substance

ing that ontology for granted.82 Those who argue for what is if you wish a more liberal view of abortion have a very different view of human being. It is often argued that the relevant moral stance is that of “welfarism” which is the view that “the ultimate point of morality is to ensure that individuals’ lives go well rather than badly, thus to protect and promote their well-being.”83 Now, welfare is “subjective”: what is relevant is how well life is going for him or her, how satisfying and fulfilling it is. But “the bare minimum necessary for having a welfare is sentience...” (ibid., p. 208; his italics) It is concluded that the criterion for being (barely) a person is sentience. “The threshold of moral standing for a developing human being is therefore the stage when it first acquires the most primitive form of conscious feelings or affect – the capacity to experience pleasure and pin.” (p. 208) Since neither the newly fertilized ovum nor the growing foetus for some time is sentient it is not a person and so has no moral standing. Abortion is therefore permissible at that time, at any time until the foetus becomes sentient. It is clear that “welfarism,” taking welfare alone as a criterion of the morality of human acts, neglects the moral sense of human beings. Preserving and promoting the welfare of human beings provides prudential reasons for instituting and conforming to the norms of civil society. But those norms become binding morally and not just prudentially by virtue of the sympathetic response of human beings to one another’s pain and pleasure, our natural charity towards others. This suggests an alternative to sentience as a criterion of personhood. This would be the capacity of the foetus to elicit a sympathetic response from members of civil society. An entity ought to be taken as a person when we can respond to it as we respond to members of our moral community. This has implications for what one says about the morality of abortion. Sentience is not the point at which there are such sympathetic responses from members of the mother’s moral community. Since before birth such a response to the behaviour and actions of the foetus could come only from the mother who is bearing it, she alone determines the point at which the foetus acquires moral status. The role of our moral sense is important, but whether one adopts the criterion of sentience or adopts the role of our moral sense as determining moral status, both presuppose a certain metaphysics of persons. Both reject the doctrine that a human form or nature or essence forms the basis for moral status in civil society. One can imagine the response of the Anglican critics of Hobbes to this view of human being. Like Hobbes’ view, the ac-

84

Body, Mind and Self in Hume’s Critical Realism

count of human being presupposed by “welfarism” does not acknowledge that there is a basis in human nature for moral awareness, where it is that moral sense that separates us from lesser creatures. Hume’s defence of a moral sense remedies this defect. But there is something deeper, something that confronts both “welfarism” and the Humean view of morality. For the critics of Hobbes, this moral sense finds its objective basis in a metaphysically rooted human form or nature. The criticism that this is ignored confronts the other views on abortion. Any defence of these views requires antecedently a rejection of the metaphysics that implies the rejection of abortion. The issue about abortion is in the end a metaphysical issue: are there or are there not metaphysical forms or natures determining the personhood and moral status of human beings. Here George Grant saw things clearly. The need to justify modern liberal justice has been kept in the wings of our English-speaking drama by the power and strength of our tradition. In such events as the decision on abortion it begins to walk upon the stage. To put the matter simply: if ‘species’ is an historical concept and we are a species whose origin can be explained in terms of mechanical necessity and chance, living on a planet which can also be explained in such terms, what requires us to live together according to the principles of equal justice?84

The issue is one of “theoretical differences in ‘world views’.” (p. 74) Before any “moderate” view of abortion can be established the metaphysical issues must first be settled. The issue is not merely one of morals; it is also one of ontology: the metaphysics of the substance tradition must be rejected. The metaphysics has its roots in the ancient world, but the issues it establishes as important remain, as George Grant makes clear, to this day.

85

One: Self as Substance

(6) Another Sort of Mind The substance account of persons and of minds had been denied already in the ancient world, by the Epicureans. As we have noted, among the entities that, according to Lucretius,85 one finds in the world is oneself and one’s understanding. These are entities that we experience as part of the world in which we live, and they are subject to the same sorts of explanation as other entities in that world, thunder, for example, or celestial phenomena. Again, the details need not detain us; many of them are now simply quaint. The point is that persons and selves, like other entities, are conceived of as composed of changing parts: just as plants are patterned sets of changing parts, so too the mind and spirit are patterned sets of changing parts. As Lucretius puts it, ...the mind, which we often call the intellect, the seat of the guidance and control of life, is part of a man, no less than hand or foot or eyes are part of a whole living creature (On the Nature of Things, p. 99).

The same is true of what Lucretius calls the “spirit” or living force that moves the body (p. 100), and, in particular in the case of humans, moves it under control of the mind (p. 108). Lucretius argues in detail that, so far as one can tell from one’s experience of the world and of oneself in the world, there is nothing permanent or unchanging about mind and spirit. Thus, he concludes, “minds of living things and the light fabric of their spirits are neither birthless nor deathless” (p. 108). It may be that something of mind or spirit survives the death of the body, but that is not to be identified with the self: each person is a body guided and animated by a mind and spirit. If any feeling remains in mind or spirit after it has been torn from our body, that is nothing to us, who are brought into being by the wedlock of body and spirit, conjoined and coalesced (p. 121).

We die when our body dies; if anything survives, it is not us. Certainly, it is not an immortal soul – a notion for which there is no basis in experience, and which, in fact, is rather absurd. For, as Lucretius asks, What can be imagined more incongruous, what more repugnant and discordant, than that a mortal object and one that is immortal and everlasting should unite to form a compound and jointly weather the storms that rage about them? (p. 120).

86

Body, Mind and Self in Hume’s Critical Realism

In the modern world, it was Locke who spoke most clearly in defence of this Epicurean position on the nature of the self. What is important to personal identity is our consciousness of ourselves, which is not a consciousness of a simple enduring substance: ...it being the same consciousness that makes a man be himself to himself, personal identity depends on that only, whether it be annexed solely to one individual substance, or can be continued in a succession of several substances. For as far as any intelligent being can repeat the idea of any past action with the same consciousness it had of it at first, and with the same consciousness it has of any present action: So far it is the same personal self. For it is by the consciousness it has of its present thoughts and actions, that it is self to itself now, and so will be the same self, as far as the same consciousness can extend to actions past or to come; and would be by distance of time, or change of substance, no more two persons, than a man be two men by wearing other clothes to-day than he did yesterday, with a long or a short sleep between: The same consciousness uniting those distant actions into the same person, whatever substances contributed to their production. (Essay concerning Human Understanding, Bk. II, Ch. 27, sec. 10)

With this judgment – a judgment of commonsense – though not of a theologically determined metaphysics – Locke in fact transformed the discourse about persons. To be sure, Locke was much criticized, and much of the traditional doctrine continued. Thus, Butler criticized Locke for presupposing what was to be proven; as Butler put it, Locke ignored what seemed to him (Butler) to be “self-evident,” that “Consciousness of personal Identity presupposes, and therefore cannot constitute, personal Identity.”86 But then in stating his own position Butler simply dogmatically asserts the very notion of substance that Locke had challenged. ...this he, the person or self, must either be a substance, or the property of some substance. If he, if person, be a substance; then consciousness that he is the same person, is consciousness that he is the same substance. If the person, or he, be the property of a substance; still consciousness that he is the same property is as certain a proof that his substance remains the same, as consciousness that he remains the same substance would be: since the same property cannot be transferred from one substance to another (ibid. pp. 395-6).

But many recognized that this was no longer enough. Shaftesbury87 recognized the problem that Locke had raised. “That there is something undoubtedly which thinks, our very doubt itself and scrupulous thought

87

One: Self as Substance

evinces. But in what subject that thought resides, and how that subject is continued one and the same...this is not a matter so easily or hastily decided by those who are nice self-examiners or searchers after truth and certainty.” One must distinguish between saying that I exist and saying that I continue to exist, that is, that I persist. It is not enough, Shaftesbury continued, “to use the seeming logic of a famous modern, and say, ‘We think, therefore we are.’...But the question is, ‘What constitutes the We or I?’ and ‘whether the I of this instant be the same with that of any instant preceding or to come?’” (Characteristics, II, p. 275). It even entered into the discourse of the general public. “There has been very great Reason on several Accounts,” Addison wrote in the Spectator in 1714, “for the learned World to endeavour at settling what it was that might be said to compose, personal Identity.”88 Locke’s view challenged the traditional Christian position – as it continues to challenge that position as defended by more recent philosophers such as George Grant. This is perhaps not so clear in Locke’s own writings, for Locke was firmly wed himself to Christian beliefs, at least with regard to the soul and to eternal life, though likely enough he denied the Trinity. But it comes out in a straightforward way in a polemical exchange between Anthony Collins and Samuel Clarke.89 In the Christian tradition, an individual person was a composite of both mind and body, material and immaterial substances. The immaterial substance or soul is certainly not the whole person according to this tradition, but equally certainly it is that indivisible and immortal aspect of the person which assures personally continuity and ontological permanence. As Étienne Gilson once put it, “firmly based...on the substantiality of the intellect and the immortality it carries with it, the Christian individual [was] invested with all the dignity of a permanent being, indestructible, distinct from every other in his permanence...”.90 If the substantial self provides us with a priori metaphysical certainty of our continuing existence as a “permanent being,” it also assures us that we will be held accountable, in both this world and the next, for our actions. The substantial self therefore plays an important role in our system of morality. Locke himself noted this “forensic” nature of the concept of a person, and its connection with a system of rewards and punishments, both now and in the hereafter. ‘Person’, he says, as we noted previously, “... is a foresnsick term appropriating actions and their merit, and so belongs only to intelligent agents capable of a law, and happiness and misery.” (Essay concerning Human Understanding, Bk. II. Ch. 27, sec. 26). And as Samuel Clarke was later to put it to Collins, the doctrine of a

88

Body, Mind and Self in Hume’s Critical Realism

of a substantial self “facilitates the Belief of...a future Retribution, by securing a Principle of Personal Individuality, upon which the Justice of all Reward and Punishment is entirely grounded.”91 That the notion of personal identity plays an important role in our ways of thinking about the world is clear. In the first place, it is to the self that we primarily attribute mental states. As we ordinarily speak we distinguish mental states such as doubtings, believings, willings, intendings, hopings, desirings, fears, etc., not only by their species (as we just did), by their contents or intentions, and by the time at which they are experienced, but also by the person who holds them. Moreover, actions are also attributed to persons; while mere movements are attributed to bodies, actions are attributed to persons. This, of course, is a consequence of mental states being attributed to persons. The answer to Wittgenstein’s famous question, what is left over if we subtract the fact that my arm goes up from the fact that I raised it, is – it takes a Wittgenstein not to notice this – the intention that caused the arm to go up92; actions are bodily motions caused by certain mental states, and the connection of the latter to persons serves to connect actions to persons also.93 With respect to both knowledge and action, we presuppose that persons persist or endure through time. In order for one to have knowledge of anything, the person who perceives must surely be the same as the person who draws inferences from these perceptions, who imagines what else could be the case, who remembers what was the case, who doubts that some of these things are reasonable, and who draws conclusions from all this and gains knowledge. Similarly, not all actions are simple or basic, like raising my arm. There are also complex actions like seducing my neighbour’s spouse or going to a movie or acquiring an education. Such complex actions require the uniting of many basic actions into tactics and strategies that work out over time to bring about the end intended. These complex actions as they proceed to completion themselves extend through time; so do the persons to whom they are attributed. The connection of persons and actions leads into the connection of persons with our system of moral principles; here too we presuppose the notion of personal identity. Thus, it is persons who are the bearers of moral rights. If a promise is made to someone that a book will be returned to him or her at a later time, then he or she has a right later on to expect that the book will be returned. It is no excuse to argue that the earlier he is not the same person as the later he; an identity of person is assumed. Similarly for our notions of reward and punishment: Someone cannot excuse himself from punishment now for an act he or she performed earlier on the grounds

89

One: Self as Substance

that he or she is not the same person. And even if we step aside from our system of morality and look only at the area governed by self-interest, here again the notion of personal identity enters: in acting to maximize our selfinterest over the long run we presuppose that it is one and the same person the endures from now to the long run. Finally, of course, there are the simple facts of experience. We usually experience simultaneous perceivings, rememberings, intendings, emotions, etc., as being ours, and therefore as belonging to one person. Moreover, we remember past experiences, and past actions as being ours as well. The traditional substance doctrine accounts for all this rather well. The attribution of mental states and actions to persons fits in easily. So does the unity of self over time that is needed for knowledge and for complex actions, as well as for our moral concepts of rights and responsibilities. It would seem, too, that it could account for the facts of the experience of unity very well. As indeed it could, if only we were to in fact experience the substance that is supposed to provide the unity of the self. But alas! as Locke saw and as Hume was to see, we do not experience this substance. And in fact the tradition recognized that fact: its case is never based only the appeal to experience but to transcendental arguments. When Locke appealed to the “historical plain method” to understand our ideas, he established that if one was going to follow the empirical method of Newton, then the resulting world would not contain substances. When Locke argues that “self is not determined by Identity...of Substance...but only by Identity of consciousness,” he is indicating that the concept of a person that we use, or ought to use, is to be determined by pragmatic considerations94: is the concept that we have useful for purposes of reporting the empirical facts of human psychology and for purposes of moral discourse? The age understood this. The Memoirs of Martinus Scriblerus95 record this. These were largely the work of Dr. John Arbuthnot, with help from Alexander Pope, John Gay and Jonathan Swift.96 The memoirs are by a professed philosopher, Martin, who is a man of great learning. Martin is a thorough-going nominalistic empiricist. He uses only one operation of the intellect, that of forming single apprehensions of perceptions. Consequently, “Martin’s understanding was so totally immers’d in sensible objects, that he demanded examples from Material things of the abstracted Ideas of Logick” (p. 119). Also consequently, it turns out that Martin’s judgment and argumentation are very weak. Martin’s father finds him a

90

Body, Mind and Self in Hume’s Critical Realism

companion, Conradus Crambe. Crambe is less a nominalist and rather more a Lockean than Martin. Martin can only form the perception of a particular Lord Mayor; for him, a Lord Mayor is always that Lord Mayor and never a concept or idea that might include several individuals. But Crambe, “to shew himself of a more penetrating genius,” frames a notion of the mayor not only without his official appurtenances (horse, gown), but even without stature, feature, colour, or, indeed, any body – “which he suppos’d was the abstract of a Lord Mayor” (p. 120). One of the concerns of Martin as his education nears completion is abnormal psychology, the “Diseases of the Mind” (Ch. xii). As part of this concern he endeavours to find the “Seat of the Soul,” that is, the principle of human nature that defines the person. During this search, Martin is addressed by the Society of Free-Thinkers. This group aims to demonstrate that “since you cannot find it, there is no such thing” as the Soul, but that there is an “easy mechanical Explication of Perception or Thinking” (p. 138). The views of this group are in fact those that Anthony Collins defended in his controversy with Clarke. For the Free-Thinkers, a human being is a mere body, a material thing, where thought is simply the effect of purely mechanical functions and whose identity is that of a collection of particles governed by laws: “Consciousness, with its several modes of sensation, intellection, volition, &c. [does not] inhere in [any one part of the person], but is the result from the mechanical composition of the whole Animal” (p. 139). The sense of identity is “only a fallacy of the imagination, and is to be understood in no other sense than that maxim of the English Law, that the King never dies” (p. 139). As for the forensic aspect of being a person, the Free-Thinkers consider it when they consider the objection “that Punishments cannot be just that are not inflicted upon the same individual, which cannot subsist without the notion of a spiritual substance”, to which they reply that “this is no greater difficulty to conceive, than that a Corporation, which is likewise a flux body, may be punished for the faults, and liable to the debts, of their Predecessors” (p. 140). In support of their view that body is the seat of both bodily and mental functions, the Free-Thinkers report how one of their members devised an “artificial Man,” a hydraulic engine with mechanical blood, heart, lungs, and muscles “who will not only walk, and speak, and perform most of the outward actions of the animal life, but (being wound up once a week) will perhaps even reason as well as most of your Country Parsons” (p. 141). The satire suggests that, while a purely mechanical parson is possible, a purely mechanical person is not. The Scriblerians thus use their comic wit to raise a

91

One: Self as Substance

genuine philosophical challenge to the thesis of Collins that personal identity is to be explained through bodily identity. The problem of personal identity is raised in another way through Martin’s love for the Double Mistress, a beautiful Siamese twin called Lindamira-Indamora who is on display in a circus. Martin is in love with the passive, relatively inert Lindamira; Indamora is her mirror image, dark, inflamed with love for Martin, inviting rape by a Manteger, and marrying the Black Prince. Martin finally marries Lindamira-Indamora and is promptly sued by the circus manager, Randal, who wants his property back; Randal contrives to marry Indamora to the Black Prince. Since we have in Lindamira-Indamora the possibility mooted by Locke of two persons in one body, Martin has opened himself up to a strange variety of crimes: rape, bigamy, and incest (p. 155). Moreover, according to law, a wife was not obliged to live with a concubine (ibid.). The Scriberlians make the most of the comic possibilities, but the points are more than just comic: the ridiculous nature of these implications from the premise that there are two persons in one body performs the philosophical task of calling into question that premise. The Scriblerians not only mock Collins and Locke through the situation that Martin is put into, but also through their description of Martin’s way of thinking. The philosophical problem of personal identity is first noted when Crambe discovers the results of Lockean philosophy: Martin follows Collins in denying the traditional doctrine of personal identity. Martin would tell his Instructor, that All men were not singular; that Individuality could hardly be praedicated of any man, for it was commonly said that a man is not the same he was, that madmen are beside themselves, and druncken men come to themselves; which shews, that few men have that most valuable logical endowment, Individuality (p. 119).

The Scriblerians are here giving a comic twist to an argument used by Clarke in response to Collins. What Clarke said was this: A Man, you say, who, during a short Frenzy, kills another, and then returns to himself, without the least Consciousness of what he has done; cannot attribute that Action to himself; and therefore the mad Man and the sober Man are really two as distinct Persons as any two other Men in the World, and will be so considered in a Court of Judicature. Extraordinary Reasoning indeed! Because in a figurative Sense a Man, when he is mad, is said not to be himself; and in a forensick Sense, is looked upon as not answerable for his own Actions, but an-

92

Body, Mind and Self in Hume’s Critical Realism

other Person’s; and the same Man is really two distinct Persons! (Clarke, Works, III, p. 902).

This is an attack on the obvious implication that Collins drew from Locke’s view on personal identity; Clarke attacked Locke directly on the same point (ibid., pp. 903-4). Collins had stated the implication in an extreme way. As he put it in 1708, “no Man has the same Consciousness to Day that he had Yesterday.” We “are not conscious,” he held, “that we continue a Moment the same individual numerical Being” (ibid., p. 870). Clarke’s response sums up the early reaction to Locke’s view of the person: “You make individual Personality to be a mere external imaginary Denomination, and nothing at all in reality” (ibid., p. 902). Butler makes the same point, referring explicitly to both Locke and Collins in footnotes; Locke’s “hasty observations,” he says, have been carried to a strange length by others, whose notion, when traced ...to the bottom, amounts, I think, to this: ‘That personality is not a permanent, but a transient thing: that it lives and dies, begins and ends continually: that no one can any more remain one and the same person two moments together, than two successive moments can be one and the same moment: that our substance is indeed continually changing; but whether this be so or not, is, it seems, nothing to the purpose; since it is not substance, but consciousness alone, which constitutes personality; which consciousness, being successive, cannot be the same in any two moments, nor consequently the personality constituted by it.’ (Dissertation of Personal Identity, p. 392)

What Locke did when he insisted that one use the empirical method, the “historical plain method,” to understand the acquisition of concepts and of knowledge, was destroy the illusion that we have any legitimate concept of substance. This in turn attacked the traditional and Christian foundations of the theory of self. When Locke argues that “self is not determined by Identity...of Substance...but only by Identity of consciousness,” he is seen by his critics like Clarke and Butler as destroying the notion that there is a unitary core of personality persisting through time, replacing it by a notion of a human being as existing only from moment to moment. The important implication of Locke’s position is, as Tuveson has put it, that no “unchanging soul is necessary to constitute personality....” As a result, the “personality itself” becomes “a shifting thing; it exists, not throughout a lifetime as an essence, but hardly from hour to hour.”97 Hume again takes up problems where Locke leaves off. Just as

93

One: Self as Substance

Locke for all practical purposes left the notion of body as that of a bundle of qualities, so he has left the notion of the self or person as that of a collection or bundle of events, only contingently connected together. Collins had recognized this point and Hume pursued it. The mind is a kind of theatre where several perceptions successively make their appearance; pass, re-pass, glide away, and mingle in an infinite variety of postures and situations.... The comparison of the theatre must not mislead us. They are the successive perceptions only that constitute the mind (T, p. 253).

The problem was to find a basis for the stability of personality that was compatible with this ontological view of the self as a bundle of perceptions but yet sufficed to answer the concerns of Clarke and Butler, that the very notion of personality had been eliminated by Locke’s empiricism. Or rather, Hume had to find a view of personality that sufficed for everyday purposes to meet those objections. Hume saw no need to answer the theological concerns of the Anglican divines nor the theological concerns of George Grant. (7) Minds as Bundles The case can be made that the concept of a mind or person as a “bundle of sensations”, as Hume put it, is satisfactory to commonsense. To be sure, philosophers such as George Grant or Albert Shalom have challenged this Lockean-Humean – or Lucretian – point, but they take the issue a step beyond commonsense in their argument that there is a feature of the self, with which we are supposed to be intimately familiar, which is outside time and the world of ordinary experience. Whatever one wants to say about the arguments of Plato or Plotinus, it is certainly true that the modern world is the offspring of the mediaeval, and the reaction to Hume is often as it was to Hobbes, one of moral outrage. Goronwy Rees has reported98 “what I think is the experience of commonsense, or at least the experience of many, that never having “had that enviable sensation of constituting a continuing personality, of being something which, in the astonishing words of T. H. Green ‘is eternal, is self-determined, and which thinks’.” (pp. 9-10) Many can say, with Rees, that he or she is “quite certain that I had no character of my own, good or bad, that I existed only in the particular circumstances of the moment, and since circumstances were always changing, so fast, so bewilderingly, so absorbingly, how could it not follow that I must change with them?” (p. 10) The empirical fact that circumstances impinge

94

Body, Mind and Self in Hume’s Critical Realism

upon us and that we learn, perhaps, but certainly are changed by them implies a continuity – there is the I that changes – and a distinction – that I is separate form natural objects and other I’s – but no simple and unchanging Cartesian ego that endues through all change. Rees points out that the notion of a simple self is part in fact of the very grammar of our language, indeed, “so much a part of the structure of our language that to describe the bliss of existence in which ‘I’ has no part is to invite self-contradiction.” (p. 11) “And yet,” he continues, “life is not governed by the conventions of grammar.” (p. 10) Rees reports how he was liberated as an undergraduate at Oxford upon reading in the Treatise Hume’s statement of the commonsense about the self: Setting aside some metaphysicians ..., I may venture to affirm of the rest of mankind that they are nothing but a collection of perceptions which succeed each other with inconceivable rapidity and are in perpetual flux and movement. (T, p. 252)

Rees was, however, inclined to keep his acceptance of Hume’s view of the self largely private. Hume has always been regarded, particularly at Oxford, as a kind of rogue elephant which has run amok in the fields of Speculation, and anyone who chose to imitate his example laid himself open to the severest censure. I was soon taught that Hume’s error had been finally exposed by Bradley’s retort to the psychologist Bain: “Mr. Bain collects that the mind is a collection. Has he ever thought who collects Mr. Bain?” This seemed to me witty but not conclusive; yet I soon realised that if, in spite of Bradley, I persisted in following Hume, I should inevitably involve myself in the most serious trouble with my tutor, H. W. B. Joseph, who for three years wheedled, cajoled, threatened, and bullied me into holding the philosophical views which he thought appropriate to a scholar of New College; that is to say, into a curious form of Platonism which seemed to have received the personal blessing of William of Wykeham. (A Bundle of Sensations, pp. 12-13)

For Joseph, one who accepted Hume’s view was either mad or bad or both. There was certainly no place in Joseph’s philosophy for Hume’s scepticism about the self, and when I was tempted to wonder ... whether Hume may not after all have been right, I put the thought away with a guilty feeling that I had been near to committing not merely an error but a sin; for Joseph they were really indistinguishable. (p. 13)

95

One: Self as Substance

But Rees took some secret delight in thinking that Joseph might well be subject, as Hume would have it, to a natural propensity to accept the illusion of a substantial self. He rather relishes Hume’s ironic comment regarding the metaphysician, that “He may, perhaps, perceive something simple and continu’d which he calls himself, though I am certain there is no such principle in me” (T, p. 252) But, Rees goes on, ...even as I write these words my hand trembles, and I hear the high-pitched, admonitory querulous voice of Joseph, sitting in his armchair beneath his Indian fretwork over-mantel, as ugly and persistent as Socrates, and know that once again I have fallen into error and sin. (A Bundle, p. 15)

Rees might have found some relaxation of the tension had he gone on – had Joseph let him go on – to read Hume on the basis in human nature of a human morality. He might then have seen that it is possible for a self which is a bundle of perceptions to recognize moral distinctions and to comport him- or herself in ways that recognize the importance of virtue and the dignity of human beings. But if Hume’s view of human being does allow for the reality of virtue – though, to be sure, not the metaphysical reality of virtue – , it is also true that it is not quite so clear that he has an adequate account of the experienced unity (though not simplicity) of the self. To refer to the self as a “bundle” may be, and indeed is, a good argument against those who, like Plato, Descartes, or Joseph, suggest that the self is a simple entity continuing through change. Our experience, however, is that the self is not a “mere” bundle: if there is no simplicity, there is a unity, an identity through change. The issue to which we must now turn is precisely this. Can Hume present us with an adequate account of personal identity?

Endnotes to Chapter One 1.For a more detailed discussion of the substance tradition, see F. Wilson, The Logic and Methodology of Science in Early Modern Thought: Seven Studies, Study One; and also F. Wilson, Socrates. Lucretius, Camus – Two Philosophical Traditions on Death, Chapter Three. 2.G. Vlastos, “Reasons and Causes in the Phaedo.” It is the “physicist” Anaxagoras

96

Body, Mind and Self in Hume’s Critical Realism

against whose “natural philosophy” Socrates is arguing in the Phaedo. It is usually taken for granted that Socrates wins the argument, a clear look at the fragments of Anaxagoras’ work that remains certainly can raise questions about this easy assumption. The arguments of both Anaxagoras’ and Socrates’ are discussed in greater detail in F. Wilson, “Bergmann’s Hidden Aristotelianism,” in his Acquaintance, Ontology, and Knowledge, pp. 485-496. 3.R. G. Turnbull, “Aristotle’s Debt to the ‘Natural Philosophy’ of the Phaedo.” 4.For discussion of these issues, see F. Wilson, “Universals, Bare Particulars, and Tropes: The Role of a Principle of Acquaintance in Ontology,” in his Acquaintance, Ontology and Knowledge, pp. 363-398. 5.References are to R. McKeon, ed., The Basic Works of Aristotle. 6.References are to Plotinus, The Enneads, trans. Stephen MacKenna, by Tractate, Chapter and Section. For discussion of Plotinus and his relation to the substance tradition, see F. Wilson, Socrates, Lucretius, Camus – Two Philosophical Traditions on Death, Chapter Three. 7.A. Shalom, The Body-Mind Conceptual Framework and the Problem of Personal Identity. This book provides an important re-statement of the neo-Platonic case about consciousness, and has been unfortunately neglected by those concerned about the problem of personal identity. 8.R. Sorabji, Necessity, p. 171. The reference is to C. G. Hempel and P. Oppenheim, “Studies in thee Logic of Explanation”; see also C. G. Hempel, “Explanation in Science and in History.” Hempel’s discussion of the “covering law” or “deductivenomological” model of explanation is essentially Humean, that is, the explanation of individual facts or events by subsumption under a matter-of-fact regularity. 9.R. Sorabli, Necessity, p. 170. Compare the discussion in F. Wilson, The Logic and Methodology of Science in Early Modern Thought, Study One; this provides a systematic critique of the Aristotelian model of explanation and a defence of the empiricist, i.e., Humean, model of explanation. See also F. Wilson, The Logic and Methodology of Science and Pseudoscience. 10 Cf. F. Wilson, The Logic and Methodology of Science in Early Modern Thought, Study One. Also F. Wilson, The Logic and Methodology of Science and Pseudoscience. 11. For discussion of this principle, see F. Wilson, Hume’s Defence of Causal Inference, Chapter Two; and also F. Wilson, “Mill’s Proof that Happiness Is the Criterion of Morality.”

97

One: Self as Substance

12. John Bramhall, The Catching of the Leviathan (1658), p. 544. For a good discussion of the critics of Hobbes, see J. Bowle, Hobbes and His Critics. 13. Bramhall., p. 567. 14. George Lawson, An Examination of the Political part of Mr Hobbs his Leviathan, p. 83. 15. Ibid., p. 3. 16. William Lucy, An Answer to Mr. Hobbes his Leviathan with Observations, Censures, and Refutations of Divers Errours, p. 138. 17. James Lowde, A Discourse concerning the Nature of Man, p. 3. 18. Lucretius, On the Nature of Things. 19. Lorenzo Valla, On Pleasure: De voluptate, trans. A. Kent Hieatt and Maristella Lorch, Intro. by Maristella Lorch (New York: Abaris Books, 1977). 20. See Marsile Ficin, Théologie Platonicienne de l’Immortalité des Ames, texte critique établi et traduit par Raymond Marcel. All references are to Tome I, and are by Book, Chapter and page. 21. For Overton’s leveller views, see, for example, his Vox Plebis: or, The Voice of the Oppressed Commons of England against Their Oppressors (London: s.n., 1653); and Foundations of Freedom [also attributed to John Lilburne] (London: published for the satisfaction of all honest interests, 1648) . 22. Richard Overton, Mans Mortallitie (Amsterdam [ = London]: John Canne, 1643); second edition Man Wholly Mortal (London: s.n., 1655). 23. John Smith, Select Discourses (1660); fourth edition, ed. H. G. Williams (1859). References are to the latter. 24. The reference is to Plotinus, Enneads, IV, 7, 6. 25. He refers to Enneads, IV, 7, 5. 26. R. Cudworth, The True Intellectual System of the Universe, first edition 1678; second edition 1743. References are to the second edition. 27. See, for example, D. C. Dennett, Brainstorms, p. 119ff. For discussion of the appropriateness of this reading of traditional empiricists, see F. Wilson, “On the Hausmans’ ‘A New Approach’,” in his Acquaintance, Ontology and Knowledge, pp. 161194.

98

Body, Mind and Self in Hume’s Critical Realism

28. P. Bayle, The Dictionary Historical and Critical, Second Edition (1736). 29. In Chapter Five. 30. Cf. D. Gauthier, “David Hume: Contractarian,” for a classic statement of this position. 31. P. Árdal, Passion and Value in Hume’s Treatise. 32. A. Baier, A Progress of Sentiments. 33.Samuel Clarke, A Discourse concerning the Being and Attributes of God, the Obligations of Natural Religion, and the Truth and Certainty of the Christian Revelation, Sixth Edition (1725). 34.Pages references are to the 11th edition, London: J. J. and P. Knapton etc., 1733. 35.Paris, 1694. 36.Amsterdam, 1723. 37.Paris, 1691. 38.Samuel Clarke, A Discourse concerning the Being and Attributes of God. 39.Virgil, The Odyssey, trans E. V. Rieu, p. 115. 40.Ovid, Metamorphoses, trans. M. Innes, pp. 31-2. 41.See Virgil’s Messianic Eclogue: Its Meaning, Occasion, and Sources, Three Studies by Joseph B. Mayor, W. Warden Fowler, and R. S. Conway, with the Text of the Eclogue, and a Verse Translation by R. S. Conway. 42.Ibid., pp. 22-3. 43.Augustine, City of God, Bk. X, chapt. 27. 44.Samuel Johnson, Lives of the Poets, ed. M. Arnold, p. 419. 45.Quoted in George Boas, Essays on Primitivism and Related Ideas in the Middle Ages, p. 35. 46.A. Bartlett Giamatti, The Earthly Paradise and the Renaissance Epic, p. 68. 47.Hexameneron, III, x; quoted in Boas, p. 42.

99

One: Self as Substance

48.Quoted in Boas, p. 94; cf. A. O. Lovejoy, “The Communism of Saint Ambrose.” 49.Quoted in Boas, p. 17. 50.Quoted in Boas, pp. 31-2. 51.Boas, p. 18, n. 7. 52.Ibid., p. 18f. 53.Boas, pp. 43-4; and also G. H. Williams, Wilderness and Paradise in Christian Thought, p. 49. 54.Boas, p. 16. 55.Boas, pp. 31-2. 56.Pace Kant we cannot in reality divide ourselves from ourselves. Obeying the dictates of conscience is nothing other than acting upon our sense of duty. 57.Boas, p. 43. 58.Quoted in Boas, p. 33. 59.Francis Bacon, Essays, Advancement of Learning, New Atlantis, and Other Pieces, selected and Edited by Richard Foster Jones, pp. 15-16. 60.I Kings XIX, 8; II Kings IV, 38. 61.Isaiah, XL, 3. 62.For Baxter, see the “Life and Times” and “Life and Works” by William Orme in vol. I of The Practical Works of the Rev. Richard Baxter. On Baxter’s place in the thought of his time, see also W. M. Lamont, Richard Baxter and the Millennium: Protestant Imperialism and the English Revolution. 63.The Saint’s Everlasting Rest, in vols. 22 and 23 of The Practical Works of the Rev. Richard Baxter; see vol. 22, p. 341. 64.Williams, Wilderness and Paradise in Christian Thought, p. 92. 65.For a brief discussion of this movement, see David Stevenson, The Covenanters: The National Covenant and Scotland (Edinburgh: The Saltire Society, 1988). 66.On Rutherford, see the essay in Alexander B. Grossart, Representative Nonconformists (London: Hodder and Stoughton, 1879).

100

Body, Mind and Self in Hume’s Critical Realism

67.Rotterdam, 1664. 68.See Patrick Walker, Six Saints of the Covenant: Peden, Semple, Welwood, Cameron, Cargill, Smith, ed. with an Introduction by D. Hay Fleming, and a Forward by S. R. Crocket, in two volumes (London: Hodder and Stoughton, 1901). 69.See Sir Walter Scott, Old Mortality, Note 18. 70.First published 1682. References are to the Glasgow (?) edition of 1720. 71.Walker, pp. 95-6. 72.George Grant, Philosophy in the Mass Age. 73.George Grant, “The Philosophy of Francis Bacon,” p. 322. 74.George Grant, “Pursuit of an Illusion.” 75.George Grant, English-Speaking Justice, p. 17. 76.Thomas Aquinas, Summa Theologica, SS, Q 64, out. 77. Thomas Aquinas, Summa Theologica, Part II-II, Q. 64, Art. 5, “Whether It Is Lawful to Kill Oneself.” 78. Augustine, City of God, Bk. I, chapt. 20. 79. Reprinted in T. Beauchamp and S. Perlin, Ethical Issues in Death and Dying. Page references are to this publication. 80.We would likely, now after Darwin, add that our natural horror of death is biologically determined by natural selection: such an impulse makes us biologically more fit. There are no doubt similar good biological reasons why, as we become old, this natural horror of death becomes attenuated, why towards our natural ends we no longer fear such an end. 81.Robert Berry, “The Roman Catholic Position on Abortion,” p. 160. 82.Thus, R. Barry, “Personhood: The Conditions of Identification and Description,” simply begs the question on the issue that divides him from his opponents. 83.L. W. Sumner, “Moderate View of Abortion,” in Rem B. Edwards, ed., Advances in Bioethics: New Essays on Abortion and Bioethics, at p. 208. 84.Grant, English Speaking Justice, p. 73. 85.Lucretius, On the Nature of Things. References are by page to the translation of

101

One: Self as Substance

Ronald Latham (Harmondsworth, Middlesex: Penguin, 1951). 86.Butler, Dissertation of Personal Identity, p. 388. 87.Anthony Ashley Cooper, Third Earl of , Characteristics. 88.Addison, The Spectator, ed. D. F. Bond (Oxford: Clarendon Press, 1965), IV, p. 575. 89.The entire Clarke-Collins controversy, nine long pamphlets in all, is reprinted in Samuel Clarke, Works (London: 1738), vol. III, pp. 719-913. 90.Étienne Gilson, “Christian Personalism,” pp. 201-4. 91.Clarke, Works, pp. 851-2. 92.To speak in this way is not to commit oneself to interactionism; see below. For a more extended discussion, see F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, Chapter 8, and L. Addis, The Logic of Society. 93.For a discussion of the concept of action from this perspective, see L. Addis, The Logic of Society. 94.Cf. F. Wilson, “The Lockean Revolution in the Theory of Science.” 95. J. Arbuthnot, J. Swift, A. Pope, and J. Gay, Memoirs of the Extraordinary Life, Works, and Discoveries of Martinus Scriblerus, ed. Charles Kerby-Miller. 96.See Kerby-Miller’s introduction to his edition of the Memoirs, ibid. 97.Tuveson, Imagination, pp. 27-8. 98.Goronwy Rees, A Bundle of Sensations.

102

Chapter Two Nominalism and Acquaintance Locke and Hume base their case against the substance tradition on an appeal to the empiricist Principle of Acquaintance (PA). This chapter examines how these philosophers deploy this principle. Part of the argument consists in claiming on the basis of PA that events are in fact logically separable where the substance tradition claims there is an objective necessary connection. This includes causation in general, and in particular the connections among the events in the history of a material body or the history of a person. But there is another stream of argument to which Hume appeals in making the case for the separability of the events in question. This is the nominalist tradition. This tradition also has its roots in the thought of Plato and Aristotle, but Hume does not question its validity, as he does the validity of the notions of substance and necessary connection. This is unfortunate, for, as we shall see, the nominalism that he uncritically accepts raises (unnecessary) doubts about the views on objective necessary connections that he defends by appeal to PA. The present chapter therefore also introduces the nominalistic premises that Hume deploys alongside the appeal to PA. There are deeper reasons for our concern. These have to do with Hume’s account of the self. The nominalism has profound implications for the analysis of relations. In effect, it has the consequence that there are no relations. In the case of causation, this is no great problem. The appeal to PA by itself, and without appeal to nominalism, disposes of objective necessary causal connection. This appeal to PA leads to an account of science in which causation, objectively understood, is no more than empirical matter-of-fact regularity. The resulting account of natural science raises no problems: its method of experiment for the determination of causes can be shown to be reasonable. Hume outlines this method and its defence in his account in the Treatise of the “rules by which to judge of causes and effects.” But there are other relations. For example,– and this is the crucial one for what we are about –, there is the relation that binds the impressions

Two: Nominalism and Acquaintance

in the bundle that constitutes the self into an ordered series of impressions. In the traditional ontology deriving from Plato and Aristotle the connections that bind states of mind into the mind that is the mind of a single person are provided by the substance that as a continuant underlies those various states of minds and which, through its activities, explains those occurrences. When that substance is eliminated by Hume by appeal to the empiricists’ PA, and mind is argued to be no more than a bundle of impressions, one needs an alternative ontological account of the relations that structure the elements of the bundle into a unified whole. The nominalism, however, leads to a rejection of any objective status for any and all relations. As we just said, this is fine for causal relations, but the other ties that bind, including the relations that tie together the self or person and give it its continuing identity, lose their being, their reality. They turn out to be nothing more than subjective appearances. In reality, then, the mind is not only a bundle, but a structureless bundle. Bishop Butler saw clearly this consequence for the empiricist account of personal identity. Locke argued that personal identity was not constituted by a substance. But he also accepted the nominalist account of relations. That left him with no objective ties to bind the events in an ordered bundle, in spite of his suggestion that there are in fact such ties. Locke argued that in analogy with a tree or a vegetable, the bundle can be taken to have an identity. A tree viewed after 10 years is the same tree: a mind, experienced after a period of time, is the same mind (Locke, Essay concerning the Human Understanding, Bk. II, ch. 27, sec. 3). There is no problem: the parts do change, but those parts are structured into ordered wholes. And so, in the case of persons, there are the ties of consciousness that bind the bundle into an ordered whole. Butler objects to this: there is a difference. ... when a man swears to the same tree, as having stood fifty years in the same place, he means only the same as to all purposes of property and uses of common life, and not that the tree has been all that time the same in the strict philosophical sense of the word.1

Butler is saying that the identity of the tree is constituted by the social relations in which it stands to other things, mainly people. Locke, of course, was saying that in fact this is what for practical purposes one wants to say about persons: that is the thrust of his point that concept of a person is a “forensic” notion. This forensic notion is the notion of a being that is susceptible to rewards and punishments, that is, more generally, of a being

104

Body, Mind and Self in Hume’s Critical Realism

ceptible to rewards and punishments, that is, more generally, of a being that is fit for entering civil society, and that the relations which make a bundle of perceptions fit for participation in civil society is that they are bound together by ties of memory: it is this which makes it reasonable to punish this person now for what he or she did then, in the past, for that punishment will render this being one who will be, once again, fit for civil society. The details of Locke’s analysis of the notion of a person need not detain us; we will examine them in due course. The point here is that Butler is arguing that the parts of the bundle which is the person require stronger ties than Locke allows. His argument is that subjective ties do not provide an adequate ontological analysis of how parts are tied into a whole. He argues that “in a strict and philosophical manner of speech, no man, no being, no anything can be the same with that with which it hath nothing the same.” (p. 281) For, it is “a contradiction in terms, to say that [two things] are [the same] when no part of their substance, and no one of their properties, is the same.” (pp. 280-81) Even if the parts of the bundle that forms the self are felt to be a whole, that does not make those parts into objectively a whole – no more than a felt union of cause and effect implies an objective necessary connection obtains between them. The distinguishability of the parts implies an objective separability. Butler concludes that, since, on Locke’s (and Hume’s) view, personal identity is only the “loose and common” variety of identity, and not the “strict and philosophical kind,” it would follow on the view that “our present self [the one now entering the room] is not, in reality, the same with the self of yesterday, but another self or person coming in the room, and mistaken for it; to which another self will succeed tomorrow.” (p. 282) Of particular concern to Butler is the fact that it follows from this that any inquiry or interest concerning a future life is of no consequence at all to the person who is making it. A future life is of little concern to our purposes, but the point is more general: it follows, if Butler is correct, that “the person of today is really no more interested in which will befall the person of tomorrow, than what will befall any other person.” (p. 283) But of course, our future feelings are more intimately tied to our present feelings – and to our past feelings – than are the feelings of other persons. So Locke must be wrong in his account of what constitutes a person. The problem here is the principle that distinguishability is taken to imply separability. If the self is a bundle, that is, a bundle of distinguishable parts, then those parts are separable, not really or objectively united together into a whole. Now, the point about the self being a bundle and not

105

Two: Nominalism and Acquaintance

a substance, is a consequence of the empiricism of Locke and Hume. But the principle that distinguishability implies separability is a feature of the nominalism of Locke and Berkeley. Only if we accept the latter does Butler’s objection have any force against Locke and Hume. For us, the issue is whether the empiricism implies the nominalism. If it does not then one can accept the empiricist point of Locke and Hume about the substantial self, while rejecting the supposed consequence that there are no objective ties that bind the distinguishable parts of the bundle into a whole. Thus, to defend Hume’s account of the person we have to separate his nominalism from his empiricism, and argue that the latter does not require the former. So to separate them is the task of the present chapter. As it shall turn out, there are no good empiricist grounds for accepting nominalism – Hume simply takes it over uncritically from past tradition. It is thus possible to pursue, as we shall, the empiricist programme of Locke and Hume regarding the self while rejecting the nominalism that calls the empiricist conclusions into question. (1) Individuation and Nominalism Objects, as we perceive them, are complex entities. Thus, I now perceive two objects, a red disc and a green disc. I recognize them as two, that is, distinguish one from the other, identifying it as different. But the two are not simply different, not simply two. They are rather, each of them, of a certain kind, or kinds. Thus, when I distinguish an object, there are, as it were, two parts to this act. There is, first, an identification of a sortal element, which determines the kind to which the object belongs, and an indexical element, which identifies the particular object that in the context is being identified. As Aristotle put it, an individual object is a this such.2 Given the traditional criterion of identity, x = y : : (f)(fx fy) every individual will be a certain individual sort or essence, or, more precisely, to distinguish this from the “real essences” that Aristotle also talked about, every individual will have a definitional essence.3 This definitional essence is the set of the properties, relational and non-relational, that may truly be predicated of the object. The such in a this such is, ultimately, the definitional essence of the individual. The aspect – to speak as neutrally as possible – of a this such that is the this is that which makes the object the individual that it is; however one analyzes it, the this is that which individuates.

106

Body, Mind and Self in Hume’s Critical Realism

At the same time, however, the this also accounts for the unity of the object. For, the properties that form the definitional essence of the object are part of that essence only by virtue of being jointly individuated by the this. If the this individuates, it also unifies by tying the various properties together to form the complex which is the object. Many objects endure through time. Among the objects of this sort with which we are familiar are persons. Persons are individuals. But like all objects, they are also natured. Each person has a certain configuration of traits, that is, a configuration of passions, abilities, sentiments and temperaments, dispositionally analyzed. This configuration is unique and constitutes, as we say, the person’s self, character. and identity This unique character forms the nature or definitional essence of the individual person. But of course, persons are not the only natured objects that endure. There are also material objects such as tables, chairs, trees, and stones. For such objects, material objects and persons alike, there will be features of the definitional essence such that we can identify the this such now with the this such then. For objects that endure through time, there are features of the such that allow for re-identification of the object as the same. Thus, if the this has a unifying function, so too does the such, at least for those objects that endure through time. The problem of individuation is this: what sort of entity constitutes the this in the this suches that we perceive? Hume has a straight forward solution to the problem of individuation: everything that exists, all impressions and ideas, are individual or particular. This nominalism is based on the principles that whatever is distinct is distinguishable (P1) and (P2) whatever is distinguishable is separable. This nominalism has the consequence that only individuals, only particulars exist. This position is one that Hume shares with Aristotle and his successors. This part of Hume’s philosophy has been less explored than that for which he is most well-known: his argument against the rationalist position concerning causation. Using arguments based on a Principle of Acquaintance (PA) Hume develops the empiricist case that, contrary to the rationalist tradition of Aristotle and his successors like Descartes, causal connections are contingent. But Hume also appeals to his nominalism, in addition to his empiricist appeal to PA, when he attempts to justify his views on causation. It has been suggested by some modern Aristotelians like Harré and Madden4 that Hume's whole case against Aristotle depends

107

Two: Nominalism and Acquaintance

upon his nominalism. This is not so, as we shall see, yet the point that is made by other critics of Hume, like Edwards,5 that his nominalism violates PA, is, we shall also argue, correct. One of the aims of the present essay is to examine Hume’s nominalism and its consistency with his appeal to PA. We shall conclude that there is a real tension between these two positions. Indeed, we shall conclude that, if PA is the distinguishing characteristic of an empiricist, then an empiricist need not be, and probably shouldn't be, a nominalist. Hence, if one is going to be an empiricist, one had best not turn to a nominalism of Hume’s sort to solve the problem of individuation. But as we proceed we shall also discover that the nominalism has internal tensions: the account of relations that is implied by (P1) and (P2) understood nominalistically is inconsistent with the solution to the problem of individuation. Not only is Hume’s nominalism not a reasonable alternative for an empiricist, it is an alternative that the empiricist, if he or she is reasonable, will reject. But let us turn to look more closely at PA, the empiricist’s Principle of Acquaintance. (2) The Principle of Acquaintance in Locke and Hume (a) Locke It is Locke’s position that “If we can find out how far the understanding can extend its view, how far it has faculties to attain certainty, and in what cases it can only judge and guess, we may learn to content ourselves with what is attainable by us in this state” (Essay concerning Human Understanding, Introduction, sec. 4). His purpose is “...to inquire into the original, certainty and extent of human knowledge, together with the grounds and degrees of belief, opinion and assent” (Intro, 2), and the method that he proposed to use, as we all know, was the “historical, plain method” (ibid.), where ‘historical’ meant experimental or observational, i.e., the method of clinical medicine and of empirical science. Locke uses this method to explore the origins of the ideas that we use to form the propositions to which we assent and which, when evidence renders them certain, constitute human knowledge. Thus, if he can systematically describe the origins of our ideas, then Locke will have circumscribed the limits of human knowledge. The conclusion that Locke defends using the “historical, plain method” is no doubt well known.

108

Body, Mind and Self in Hume’s Critical Realism

Whence has it [the mind] all the material of reason and knowledge? To this I answer in one word, from EXPERIENCE. In that all our knowledge is founded; and from that it ultimately derives itself. Our observation employed either, about external sensible objects, or about the internal operations of our minds perceived and reflected on by ourselves, is that which supplies our understandings with all the materials of thinking. These two are the fountains of knowledge, from whence all the ideas we have, or can naturally have, do spring (Essay concerning Human Understanding, Book II, ch. i, sec. 2).

All our concepts (that are meaningful) derive from sensation and reflection, our ordinary ways of experiencing the world. Hume put the same point this way: “all our simple ideas in their first appearance are deriv’d from simple impressions, which are correspondent to them, and which they exactly represent” (T, p. 4). Other ideas are complex, and have simple ideas as their parts (T, p. 2). Our (empirically meaningful) concepts thus all either refer directly to entities with which we are acquainted or are defined in terms of concepts which so refer. This is PA, the empiricist Principle of Acquaintance. Locke and Hume use this principle to conclude that the Natures or necessary connections to which the rationalists constantly appealed do not exist. ‘Real essence’ is Locke’s term for those Natures that provide the necessary connections among the observable qualities of the things of which they are natures; and he distinguishes between “real essence” and “nominal essence.” The measure and boundary of each sort or species, whereby it is constituted that particular sort, and distinguished from others, is that we call its essence, which is nothing but that abstract idea to which the name is annexed; so that everything contained in that idea is essential to that sort. This, though it be all the essence of natural substances that we know, or by which we distinguish them into sorts, yet I call it by a peculiar name, the nominal essence, to distinguish it from the real constitution of substances, upon which depends this nominal essence, and all the properties of that sort; which, therefore, as has been said, may be called the real essence: v.g. the nominal essence of gold is that complex idea the word gold stands for, let it be, for instance, a body, yellow, of a certain weight, malleable, fusible, and fixed. But the real essence is the constitution of the insensible parts of that body, on which those qualities and all the other properties of gold depend (Essay, III, vi, 2).

If we let Ax

109

Two: Nominalism and Acquaintance

be the essence of a thing, and if Px –> Qx and Rx –> Sx are two dispositions defining that essence, e.g., malleable and fusible, then the proposition (*) (x)[ Ax : Px –> Qx. & .Rx –> Sx ] will have a different logical status depending upon whether A represents a real or a nominal essence. If A is a nominal essence then (*) will simply be vacuous, a tautology, true by definition. It will, of course, have no explanatory power. In contrast, if A is a real essence, then (*) is not vacuous but substantive and explanatory. That means that the dispositions on the right hand side of (*) do not define what it is to be A; rather, A is a simple, undefined concept, an unanalyzable power that explains the presence of those dispositions. In the case that A represents a real essence, (*) represents a real definition, not a nominal one. As a real definition it is a necessary truth, a self-evident principle such that the contrary is not possible. This means that the real essence of a thing is not separable from the dispositions which it explains. The real essence is, moreover, not separable from the individuals of which it is the real essence; if Socrates ceases to be human, he ceases to be Socrates, and this piece of gold, if it ceases to be gold, ceases to be. If a is an individual of which A is the real essence, then a and A are inseparable; the proposition that (+) a is A is also necessary. Given the necessity of both (*) and (+), the rationalist who accepts real essences has no problem of knowledge of laws. The necessity of (+) will guarantee that this proposition will continue to be true of a, and the necessity of (*) will in turn guarantee that the following empirical regularity holds (@) Whenever a is P then it is Q. In general, empirical regularities are explained by deriving them from necessary truths about real essences. Or rather, those empirical regularities which can be so explained have the status of laws; the necessity of the essential truths attaches to the laws for which the real essences account. Moreover, when the essential truths are known as self-evident truths that certainty attaches also to the empirical regularities explained by the real essences; the inductive uncertainty that attaches to mere empirical regularities disappears if those regularities are laws, that is, patterns rooted in the real essences of things.

110

Body, Mind and Self in Hume’s Critical Realism

What we observe are certain patterns of sensible qualities; these, according to Locke, are grouped together to define the nominal essences of things. Men observing certain qualities always joined and existing together, therein copied nature; and of ideas so united made their complex one of substances.

The grouping of characteristics together into nominal essences is thus a matter of discovering certain patterns of coexistence. These laws are matter-of-fact regularities, to be discovered in their details only with much research effort: ...it requires much time, pains, and skill, strict inquiry, and long examination to find out what, and how many, those simple ideas are, which are constantly and inseparably united in nature, and are always to be found together in the same subject.

Characteristics are grouped together into the definitions of nominal essences because we find that to be a convenient way to express our knowledge of the regularities of coexistence. In the terminology of G. Bergmann, we expect the concepts that we use to be significant,6 that is, useful in stating the regularities that we find in nature. We form these concepts which are the nominal essences; in that sense, it is us and not Nature that determines the boundaries of species. But where we draw the line is not arbitrary, since which characteristics regularly coexist is a matter of fact, not dependent upon us. We may define as we wish, but given that we wish to define concepts that are useful in recording the regularities that we discover in the world, then it is not arbitrary what concepts we define. Of course, even though people may share the desire to state regularities, their knowledge of what are the regularities may well differ, especially given the difficulty of coming to know, with the consequence that definitions of kinds, that is, of nominal essences, will vary from person to person. But in general, we may say, to search after nominal essences is to seek to come to know matter-of-fact regularities, and that such knowledge as we come to obtain is thoroughly fallible. Real essences are not known to us: “Our faculties carry us no further toward the knowledge and distinction of substances, than a collection of those sensible ideas which we observe in things...”. Real essences, were we able to know them, would provide us will a criterion for separating regularities into mere regularities and those which are lawful:

111

Two: Nominalism and Acquaintance

By...real essence I mean, that real constitution of anything, which is the foundation of all those properties that are combined in, and are constantly found to coexist with the nominal essence...

But for us this is no criterion: “...being ignorant of the real essence itself, it is impossible to know all those properties that flow from it...” Those, therefore, who have been taught that the several species of substances had their distinct internal substantial forms, and that it was those forms that made the distinction of substances into their true species and genera, were led yet further out of their way by having their minds set upon fruitless inquiries after “substantial forms”; wholly unintelligible, and whereof we have scarce so much as any obscure or confused conception in general.

The result is that for us there is no knowledge with respect to substances. We must rely upon experience for what we know of the laws of coexistence: “...we can by intuition or demonstration discover the coexistence of very few of the qualities that are united in substances: as we are left only to the assistance of our senses to make known to us what qualities they contain.” This sense experience yields only probability, not certainty: it is always fallible. But then it is not [in Locke’s terminology] knowledge, “...because the highest probability amounts not to certainty, without which there can be no true knowledge.” On the other hand, to introduce discourse about real essences does nothing but hinder our attaining even that degree of probability to which we, limited as we are to the world of sense experience, may reasonably aspire: To suppose that the species of things are anything but the sorting of them under general names, according as they agree to several abstract ideas of which we make those names the signs, is to confound truth, and introduce uncertainty into all general propositions that can be made about them.

Locke’s attack on real essences involves more than simply pointing out that they violate PA. To be sure, as the above quotations make clear, this appeal is there. But there are also other ways that Locke deploys PA in this context in order to exorcise real essences, in order to show that in fact these supposed entities play no role in such knowledge as we do have. Real essences can play the role of making our causal judgements certain, that is, overcoming the logical gap between sample and population, only if there are real meaning connections among the observable properties

112

Body, Mind and Self in Hume’s Critical Realism

of things. Locke appeals to PA to exclude such real meaning connections as would raise our experiential knowledge of regularities to the level of genuine knowledge or certainty. ...the names of substances, when made use of as they should be, for the ideas men have in their minds, though they carry a clear and determinate signification with them, will no yet serve us to make many universal propositions of whose truth we can be certain. Not because in this use of them we are uncertain what things are signified by them, but because the complex ideas they stand for are such combinations of simple ones as not carry with them any discoverable connection or repugnancy, but with a very few other ideas.

Real essences can play their role only if they are tied by real meanings to the observable properties of things; the connections like (*) must have the status of necessary truths. But in fact for most at least of the observable qualities of things, there is no such necessity, no such “connection or repugnancy.” This is a matter of the natures of the properties that we observe; in their nature there is, contrary to the Aristotelian, no connection that we can discover in experience that links these qualities to real essences. In vain, therefore, shall we endeavour to discover...what other ideas are to be found constantly joined with that of our complex idea of any substance: since we neither know the constitution of the minute parts on which their qualities do depend; nor, did we know them, could we discover any necessary connexion between them and any of the secondary qualities: which is necessary to be done before we can certainly know their necessary co-existence.

The trouble lies in the secondary qualities: ...our minds not being able to discover any connection betwixt...primary qualities of bodies and the sensations that are produced in us by them, we can never be able to establish certain and undoubted rules of the consequence or coexistence of any secondary qualities. There is no conceivable connection between the one and the other.

Locke is here challenging the notion that (*) can be a substantive necessary truth: the characters P, Q, R, and S are wholly presented to us, and acquaintance presents us with nothing that would constitute a logical or ontological tie, i.e., a necessary connection, with the underlying nature or real essence A. Locke’s inference is clear: we are acquainted with noth-

113

Two: Nominalism and Acquaintance

ing about these characters that would render it contradictory to suppose that they were not regularly connected; therefore, there is no such connection. We see here Locke using PA to exclude the real meaning connections necessary to the rationalization of the Aristotelian position and any position that introduces primitive nomological connectives to overcome the fallibilism of our causal judgments. (b) Hume Hume went further than Locke. He argued that in reality we have no idea of an objective necessary connection. Indeed, the only idea we have of such an entity is the confused idea that merges the definitions that he proposes. Hume’s argument is to a large extent a re-statement of Locke’s. We are presented with sensible qualities, but not with any necessary connections among these. “...the power lies not in the sensible qualities of the cause,” and “there being nothing but the sensible qualities present to us...” (T, p. 91), there are no (unanalyzable) powers, no necessary connections: There is no object, which implies the existence of any other if we consider these objects in themselves, and never look beyond the ideas we form of them. Such an inference wou’d amount to knowledge, and wou’d imply the absolute contradiction and impossibility of conceiving any thing different (T. p. 87).

Hume continues: But as all distinct ideas are separable, ’tis evident there can be no impossibility of that kind (T, p. 87).

Hume thus holds, like Locke, that given the natures of the sensible qualities that are presented to us, we may conclude directly that there are no necessary ties among things, and, in particular, that there are no necessary truths like the Aristotelian’s (*) about the qualities with which we are presented. Hume also adds that, if we consider an object of perception to be a collection of sensible qualities, then there can be no basis for holding that statements like (+) are necessary truths: there are no grounds in the sensible qualities that constitute the object a for assuming that the real essence or power A is necessarily tied to those qualities. And if the real essence can change, if it is not a possibility excluded by necessary ties, then appeal

114

Body, Mind and Self in Hume’s Critical Realism

to statements like (+) cannot help to overcome the logical gap between sample and population. ...at the utmost [you] can only prove, that that very object, which produc’d any other, was at that very instant endow’d with such a power; but can never prove, that the same power must continue in the same object or collection of sensible qualities; much less, that a like power is always conjoin’d with like sensible qualities (T, p. 91).

This case eliminates the idea of an objective unanalyzable real power or necessary connection. There is another aspect that should be noted. This deals with the causal principles that Locke and, before him, philosophers like Henry More and Descartes had thought could be known to be true a priori. Hume accepts the Lockean critique of innate ideas (T, p. 160). That critique eliminates the source that Henry More, and Descartes before him, had thought they had available when they claimed to have the idea of an objective real essence. From Locke and Malebranche came the claim that the idea does not derive from experience. Malebranche was quite categorical on this; we do not find it even in the case of volition. For Locke, however, while we for the most part do not find it in experience, volition is the exception; when we exercise our will we experience a necessary tie between cause and effect. Hume treats this case in detail in Book II, Part iii, and Book III, Part i of the Treatise. This is awkwardly placed relative to the main argument about the nature of causation which occurs in Book I, Part iii of the Treatise. (The parallel discussion in the Enquiries is better placed.) Hume attempted to remedy this awkwardness in the Appendix to Treatise where he added a new paragraph to be inserted into the discussion of I, iii, 14 which deals with our “idea of necessary connexion.” Here he argues on the basis of the principle that what is distinguishable is separable. For, if a volition and the action that it causes are indeed separable in thought then there is no perceivable necessary connection between them. ...the will being here consider’d as a cause, has no more a discoverable connexion with its effects, than any material cause has with its proper effect. So far from perceiving the connexion betwixt an act of volition, and a motion of the body; ’tis allow’d that no effect is more inexplicable from the powers and essence of thought and matter. Nor is the empire of the will over our mind more intelligible. The effect is there distinguishable and separable from the cause, and cou’d not be foreseen without the experience of their constant conjunction. We have command over our mind to a certain degree, but beyond that lose all

115

Two: Nominalism and Acquaintance

empire over it: And 'tis evidently impossible to fix any precise bounds to our authority, where we consult not experience. In short, the actions of the mind are, in this respect, the same with those of matter. We perceive only their constant conjunction; nor can we ever reason beyond it. No internal impression has an apparent energy, more than external objects have. Since, therefore, matter is confess’d by philosophers to operate by an unknown force, we shou’d in vain hope to attain an idea of force by consulting our own minds (T, p. 632-3).

Hume applies the same distinguishability/separability principle to establish that there are no causal principles that can be known a priori. He considers variations on the theme that the contingent events that we observe must, because of their contingency, have a cause (T, p. 80ff). Hume does not dispute the truth of the principle that every event has a cause (T, p. 82); indeed, it is one of the rules by which to judge of causes and effects, that is, those rules that define scientific rationality (T, pp. 173-4). What Hume does argue is that the principle is not self-evidently true; its truth is, rather, a matter of fact, a principle that we discover in our experience to be, so far as we can tell, true. It could not be self-evidently true because for any event it is always possible to conceive that it exists when other events do not. ...as all distinct ideas are separable from each other, and as all the ideas of cause and effect are evidently distinct, ’twill be easy for us to conceive any object to be non-existent this moment, and existent the next, without conjoining to it the distinct idea of a cause or productive principle. The separation, therefore, of the idea of a cause from that of a beginning of existence, is plainly possible for the imagination; and consequently the actual separation of these objects is so far possible, that it implies no contradiction nor absurdity; and is therefore incapable of being refuted by any mere reasoning from mere ideas; without which ’tis impossible to demonstrate the necessity of a cause (T, pp. 79-80).

This makes the negative case against the Aristotelians and the rationalists. But there is a positive case that must also be made. Locke did not do this; it remained for Hume to carry out this part of Locke's empiricist programme. Aristotelians claimed to be able to distinguish lawful regularities from those which are accidental by virtue of the former being explained by the real powers of things. This was satisfactory as long as one could claim acquaintance with those real powers, as indeed the Aristotelians did. But in the early modern period this claim that one had rational intuitions of these real essences was given up. Henry More, for example, and Locke both de-

116

Body, Mind and Self in Hume’s Critical Realism

nied the claim, but they were of course not alone. This forced philosophers to find another basis for our ordinary distinction between laws and accidental generalities. And for those who followed Locke in insisting that philosophy must adopt the empirical Newtonian method, this meant that the criterion had to be one rooted in our ordinary sensible experience of things. What Hume did was develop certain suggestions in Locke to provide just this sort of account of the distinction. This approach to laws and causation would be required even if one allowed, as Locke did, that, even though we cannot know specifically what those essences or powers are, we can nonetheless infer that there are such real essences in things. But so long as one thought one could infer the existence of real essences, and also of substances, then there was no need to rethink the traditional account of predication and the traditional account of ordinary material objects. It might be necessary to supplement that account, in order to show how we manage to cope with the world that we experience even though we have no knowledge of certain entities that are crucial for determining what occurs in it. These criteria for lawfulness and for real powers, and for predication, will be second-best, but they are necessary faut de mieux. But so long as the basic ideas of substance and real essence or power are held to be intelligible, these empirical criteria will continue to be reckoned second best, and it will never be thought that any fundamental revision of our metaphysics of bodies will be required. When Hume went on to attack what More and Locke were still defending he was forced to undertake just this re-thinking, and, in fact, to re-characterize as fundamental what earlier thinkers had thought of as “second best”. For the Aristotelian, the order that we observe in the properties of things is to be explained by reference to the nature or form or real essence of that thing. The same holds true for More and Locke, though they give up the claim that the Aristotelians themselves made, that they knew by rational intuition the entities that provided the explanation. For More and Locke all that one could know was that such an explanation exists, not what it is. Hume’s critique removed even that. In fact the whole pattern of explanation shifted: no longer is explanation to be conceived in terms of unobserved real essences, explanation rather consists in subsumption under matter-of-fact generalities, those generalities that Locke points out we must in any case rely upon for the inferences that we must all make as we go about pursuing our this-worldly purposes – including the pursuit of matterof-fact knowledge – in our everyday life. The implication of this is that the very concept of reason itself has

117

Two: Nominalism and Acquaintance

undergone a radical change. In the substance tradition of Aristotle, reason is the capacity to grasp the reasons of things, that is, the forms or natures that explain why things are as they are. But this concept of explanation in terms of simple natures has been rejected, to be replaced by the notion that explanation consists in showing that the fact to be explained is an instance of a regular pattern. That in turn means that reason is the capacity to come to know regular patterns. But all that we are acquainted with are individuals, not the patterns as such. So reason consists in the capacity to infer from observed particulars to the general pattern. As Hume points out, given the separability of all particulars, inferences of this sort are always hazardous. This makes the new reason of Locke and Hume, that is, the reason of the empiricist, very different from the reason of the substance tradition. The latter had it, as in Aristotle and as in Descartes, that reason was infallible. Not so the reason of Hume. But as Hume also argues, this implies a general scepticism about the ability of reason to achieve knowledge only if one accepts the traditional standard of infallibility. But if one rejects substances and natures and the concept of reason that they determine, then why should one accept the cognitive standards that they employ? why not accept, as Hume recommends, a standard that fits the fallible reason which is the only reason that we have?7 Locke is content to exorcise the objective necessities of the Aristotelians and rationalists. The result, as both Locke and Hume saw, is that casual laws are, ontologically, nothing more than matter-of-fact regularities. As Hume put it in his first definition of ‘cause,’ a cause is “an object precedent and contiguous to another, and where all the objects resembling the former are plac’d in a like relation of priority and contiguity to those objects that resemble the latter” (T, p. 172). But if this is so, then there is no objective basis for the distinction between causal laws and, as we say, “mere” regularities. The crucial point is that we do, as a matter of fact, draw a distinction between these two things, and, it would seem, reasonably so: there is a difference between “water, when heated, boils” and “all the coins in my pocket are bimetallic”! How, then, do we distinguish between a regularity like (@) when it is a casual law and when it is a “mere” regularity? As Hume put it, “There is a NECESSARY CONNEXION to be taken into consideration” (T, p. 77). There is no objective necessary connection, but from that it does not follow that there is no necessary connection, that there is no difference between a matter-of-fact regularity that is a law and one that is not. The answer that Hume defends8 is that this difference is not objective but subjective: in the case of a causal regularity, in

118

Body, Mind and Self in Hume’s Critical Realism

contrast to that of a “mere” regularity, an association has been established in the mind between cause and effect in such a way that we use the causal regularity to predict and to support contrary-to-fact inferences. As Hume put it in his second definition of ‘cause’, a cause is “an object precedent and contiguous to another, and so united with it in the imagination, that the idea of the one determines the mind to form a more lively idea of the other” (T, p. 172). As for when it is justified, that is, rationally justified, for this attitude to be taken towards a given regularity, Hume argues9 that it is rational to have such an association only if it has been formed in conformity to the rules of the scientific method of Bacon and Newton, that is, in conformity to what Hume refers to as the “rules by which to judge of causes and effects” (T, p. 173). But be the latter as it may, what is important for present purposes is the Humean point that there is no objective relation of causal necessity; objectively there are only separable events. But there is a subjective tie, in which cause and effect are inseparably associated in thought. Thus, to use Hume’s terminology, we must consider causation in two different ways, “either as a philosophical or as a natural relation; either as a comparison of two ideas, or as an association betwixt them” (T, p. 170). Hume also makes explicit the account of dispositions in terms of significant predicates that is implicit in Locke’s discussion. Given Hume’s critique of Aristotelian powers, there can be no unanalyzable powers that could provide necessary connections in nature. It follows that a disposition concept, say soluble, is simply definitional short hand for a hypothetical statement: ‘soluble’ means by definition ‘if in water then dissolves’ or ‘x is soluble’ means ‘x is in water –> x dissolves’ As Hume puts the point, “the distinction...betwixt power and the exercise of it, is...without foundation” (T, p. 171) and “entirely frivolous” (T, p. 311). It does not follow, however, that the language of powers is without any use, and that we should dismiss it as meaningless. For Hume goes on to point out that “tho’ this be strictly true in a just and philosophical way of thinking, ’tis certain it is not the philosophy of our passions...” (T, p. 311). The language of “powers” has a certain place in our discourse, and the philosopher should be able to account for this use. Hume proceeds to propose that “power has always a reference to its exercise, either actual or probable, and...we consider a person as endow’d with any ability when we find from past experience, that ’tis probable, or at least possible he may exert it” (T,

119

Two: Nominalism and Acquaintance

p. 313). Hume’s suggestion is that ‘P –> Q’ represents a power just in case that there is a regularity in which it appears, in the case of the passions a regularity to the effect that For any time, a is P –> a is Q and more generally a regularity to the effect that (x)[Fx –> (Px –> Qx)] where ‘F’ represents some categorical property.10 Hume is here using the Lockean criterion of significance, to again use Bergmann’s term, arguing that the concepts that we define are those that are useful in the sense of appearing in statements of empirical regularity. More generally, we can say that if Dx is defined to be shorthand for Px –> Qx then a person A can (properly) predicate ‘D’ as a disposition of an object m just in case that (i) ‘Dm’ is true (ii) A knows that m is D (iii) There is a categorical kind K that A knows (iv) which is such that m is K (v) and such that A knows that m is K, (vi) and there is a law (x) (Kx –> Dx) (vii) such that A knows (within the Humean limits of inductive inference) that this law is true (viii) and such that A knows that m being D follows from this law and m’s being K. Thus, with ‘soluble’ defined as ‘in water –> dissolves’, we can predicate ‘soluble’ of a pile m of white powder just in case that we know that m is of some kind, say sugar, such that we have a law like Whatever is sugar, if it is in water then it dissolves so that, knowing that m is of this kind, namely sugar, we can deduce that if m is in water then it dissolves, i.e., that m is soluble. This contextualistic account of disposition-language enables Hume to account for our ordinary usage while rejecting the Aristotelian claim that dispositions are unanalyzable and in themselves explanatory. It is not the disposition as such that explains but the law that renders it significant. And it is this significance which renders disposition concepts useful in discourse about persons and, for that matter, ordinary objects like lumps of

120

Body, Mind and Self in Hume’s Critical Realism

sugar.11 (3) The Appeal to Acquaintance: Empiricism vs. Descartes The case that Locke and Hume make against objective necessary connections is based on PA. Yet in Hume it is intertwined with arguments based on nominalistic principles, and these must be disentangled: that is our project. Before turning to this task, however, there is a further important point that should be noted about the case, based on PA, that Locke and Hume make against necessary connections. In making their case, both appeal to the principle that if things appear distinct then they are distinct. More accurately, they appeal to the rule, justified by PA, that if a distinction among properties is given in acquaintance, then that distinction is a real distinction: the properties thus distinguished are logically and ontologically separable. It is clear that this sort of appeal, arguing that properties presented as distinct are logically separable, is one that has its origins (at least so far as modern philosophy is concerned) in Descartes, and, in particular, in Descartes’ argument concerning the separability of mind and body. The Cartesian appeal is of undoubted historical importance – it is the point at which the rationalist Descartes comes into his closest contact with empiricism – but nonetheless there are important non-empiricist aspects to it. A brief look at the Cartesian arguments will serve to highlight the nature of the Locke-Hume appeal to PA. The argument that Descartes presents in the Second of the Meditations about the distinction of mind and body turns on a conceivability argument. In the cogito I have become acquainted with a property of myself, viz., doubting, which is a species or mode of the genus thinking. The insight that is the product of the cogito is that this fact, that I, who am, or exist, have this property, is one that is indubitable, one that I know with certainty. I then bring before my mind other things that I had previously thought myself to be, e.g., a body, say an animal, or even a rational animal, but one that takes nourishment, that senses, and so on. The hyperbolic doubt, the possibility of a malignant demon, shows me that I can consistently deny all these things of myself while still knowing indubitably that I am a thinking thing. Mind and body are therefore distinct. I am ...a real thing and really exist; but what thing? I have answered: a thing which thinks.

121

Two: Nominalism and Acquaintance

And what more? I shall exercise my imagination in order to see if I am not something more. I am not a collection of members which we call the human body: I am not a subtle air distributed through these members, I am not a wind, a fire, a vapour, a breath, nor anything at all which I can imagine or conceive; 12 because I have assumed all these were nothing.

Descartes allows that we can form abstract ideas of attributes apart from substances. “But the genus,” he says, “can be conceived without the species,...Thus, I can conceive figure without conceiving any of the attributes proper to the circle.”13 We find the same doctrine in the Port Royal logic, The Art of Thinking of Arnauld and Nicole. The authors consider abstract ideas like that of a triangle, or of extension, or of thought, and point out that “in these abstractions...each lower degree is some particular determination of the higher degree:...equilateral triangle, a determination of triangle; and triangle, of rectilinear figure. The higher degree, being less determinate, stands for more things. Through such abstractions we move from the idea of a particular to a more general idea, then to an even more universal idea, and so on.”14 The issue is not whether thought and extension can be conceived separately, that is, whether the one can be thought of apart from the other, but whether the one can be conceived to exist apart from the other. For an attribute to exist, it must be in a substance.15 The issue, then, is whether, when I think of thought as in a substance, is it necessary also to think of that substance as extended? Descartes can answer this question because in cogito he is clearly and distinctly presented not only with thought but with himself, i.e., a substance.16 He can therefore clearly and distinctly conceive of thought existing while supposing, via the demon hypothesis or hyperbolic doubt, that nothing else exists. Berkeley and Hume attack the Cartesian notion of abstract ideas; the notion is inconsistent.17 In this they no doubt have a strong case. When they do mount this attack, the distinction between separability in the mind and separability in reality disappears. As Hume puts it, “to form an idea of an object, and to form an idea is simply the same thing...” (T, p. 20). The distinction between conceiving A apart from B and conceiving A as existing apart from B thus disappears: to conceive A apart from B and to conceive A as existing apart from B becomes the same thing. We shall discuss this below, but for present purposes the difference does not matter. The point that is relevant for what we are now about is that Descartes takes himself to be presented with a property that as presented is wholly distinct from all other properties. And from this he concludes that it is not self-contradictory that

122

Body, Mind and Self in Hume’s Critical Realism

a thinking thing exists apart from all other sorts of (finite) thing. That is, there is no necessary connection between thinking and any other sort of attribute that would require of any thinking thing that it also have that other attribute. This is parallel to Locke's argument that we are wholly presented with secondary qualities as wholly distinct from primary qualities and that there can therefore be no necessary connection linking the two. Or, it is almost parallel. After presenting the case for the distinction of thought from all other attributes, Descartes proceeds: But perhaps it is true that these same things which I supposed were non-existent because they are unknown to me, and not really different from the self which I know.

“Perhaps,” he tells us, perhaps there are after all necessary connections which render thought after all not distinct from other sorts of attribute. To be sure, I am not presented with them – thought is presented as distinct – but perhaps there are after all such connections, really there but not presented, really there but unperceived by us. At this point Descartes does not dispute these doubts: I am not sure about this, I shall not dispute about it now; I can only give judgement on things that are known to me.18

Nonetheless they are real doubts, and come to be dispelled only in the sixth meditation, when the Divine Guarantee is at hand. Descartes warns the reader already in the Preface to the Meditations that the Divine Guarantee is required to complete the proof of the distinction of mind from body. ...so far as I was aware [the reference is to Descartes’ treatment of the point in the Discourse on Method], I knew nothing clearly as belonging to my essence, excepting that I was a thing which thinks, or a thing that has in itself the faculty of thinking. But I shall show hereafter how from the fact that I know no other thing which pertains to my essence, it follows that there is no other thing that really does belong to it.19

The Divine Veracity guarantees the truth of evidence. Hence, if the appearance that mind is complete in itself is evident, if it is presented with full clarity and distinctness, so that no contradiction is apparent if I suppose mind can exist apart from all other things, then mind really is distinct from all other things.20 This role of God’s Veracity in guaranteeing the real dis-

123

Two: Nominalism and Acquaintance

tinctness of the apparently distinct comes out most clearly in a letter to P. Gibieuf: Of all those things of which we have diverse and complete ideas...it implies a contradiction that they are inseparable. Yet I doe not deny that there can be in the soul nor in the body many properties of which I have no idea; I deny only that there are any which are inconsistent with those I have of them, and, among the others, inconsistent with the idea I have of their distinction. Otherwise God would be a deceiver, and we would have no rule to assure us of the truth.21

This is something that has no place in the appeal of Locke and Hume. For these philosophers, if a property is presented as complete, logically and ontologically self-contained, then it is complete – period; if it is presented as having no necessary connection to any other property, then that settles the issue: it is not necessarily connected to any property. It is not possible in any sense that there are connections beyond those which are presented. For Descartes there is a sense in which it is possible that there are such connections, and that is why the Divine Guarantee is needed to exclude them. Descartes must admit this sort of possibility because for him there is a world beyond that which is presented. Locke and Hume exclude such a possibility because for them there is nothing beyond what is presented: for Locke and Hume, experience, that is, the Principle of Acquaintance, defines not only the limits of our knowledge but also the limits of the world. The Cartesian cogito marks the introduction of acquaintance into a central place in ontology. But it does not mark the beginning of empiricist ontology, an ontology in which the Principle of Acquaintance determines what is to be admitted into one’s ontology. Locke makes the crucial turn to an empiricist ontology; Hume carries the programme through to its natural conclusion. What we want now to do is to return to the relation of Hume’s nominalism and his empiricist appeal to PA; does the former conflict with the latter? (4) Hume’s Nominalism When Hume defends the claim that there are no objective necessary connections, he argues that “there is no object, which implies the existence of any other if we consider these objects in themselves, and never look beyond the ideas we form of them.” If there were a necessary connection between a cause and its effect, it would be impossible to conceive the one

124

Body, Mind and Self in Hume’s Critical Realism

without the other. “But as all distinct ideas are separable, ’tis evident there can be no impossibility of that kind” (T, p. 87). Here, as in Descartes and Locke, the notions of “distinct” and “separable” are connected with the notions of “contradiction” and “impossibility of conceiving.” The latter we may take to be, not merely psychological impossibility, but rather the test of contradictoriness. In making inconceivability the test of being contradictory, Hume is, of course, merely following a long line in the history of philosophy that stretches well back through the Middle Ages.22 Hume makes the same connection between the notion of “separable” and that of “contradiction” elsewhere: The separation...of the idea of a cause from that of a beginning of existence, is plainly possible for the imagination; and consequently the actual separation of these objects is so far possible, that it implies no contradiction nor absurdity...(T, p. 76-80).

That logical possibility is the test of conceivability is made clear elsewhere too. Thus, Hume adopts the principle ...that whatever the mind clearly conceives includes the idea of possible existence, or in other words, that nothing we imagine is absolutely impossible (T, p. 32).23

And he makes the same point a little later this way: ’Tis vain to search for a contradiction in any thing that is distinctly conceiv’d by the mind. Did it imply any contradiction, ’tis impossible it cou’d ever be conceived (T, p. 43).

This is not a psychological but a logical point: what the imagination can do is limited to what is logically possible, not the converse. Logical possibility determines psychological conceivability, not the other way around.24 What Hume does, then, to establish that a causal law like (x)(Fx –> Gx) is not necessary but merely a contingent matter-of-fact regularity is to argue that one can conceive of something being F and ~G. That is, he argues that it is possible to conceive (@) (›x)(Fx & ~Gx) A i

. “(›x)(Fx)” means that there is at least one x such that this x is F.

125

Two: Nominalism and Acquaintance

And given that ‘conceivable’ is taken to mean ‘logically possible,’ the argument is that (@) is logically possible or consistent. And (@) is logically possible because the referents of the predicates ‘F’ and ‘G’ are properties presented to us in acquaintance and are presented as containing within themselves no connection to each other (nor to any other property with which we are acquainted). Hume’s defence, therefore, of his thesis about causation is based on the claim that predicates like ‘F’ and ‘G’ are logically separate, where this in turn is based upon the claim grounded in PA that the properties to which these predicates refer are ontologically separate. The argument of Hume and Locke on this point is, in other words, of a piece with the later arguments of the logical atomists like Russell and G. Bergmann to establish the same point.25 Given this, it is clear that distinction and separability in this logical sense are compatible with matter of fact inseparability. Hume was clear about this too. Thus, he tells us that ...the taste and smell of any fruit are inseparable from its other qualities of colour and tangibility; and which-ever of them be the cause or effect, ’tis certain they are always co-existent (T, p. 237).

There are times, however, where Hume introduces stronger senses of ‘distinct’ and ‘separable.’ To be separable in this stronger sense means that the one object can both remain distinctly itself and exist spatially and/or temporally separate from the other. The numerically distinct are distinguishable and the distinguishable are separable. Thus, in his discussion of simple and complex impressions and ideas, Hume tells us that Simple perceptions, or impressions and ideas, are such as admit of no distinction nor separation. The complex are the contrary to these, and may be distinguished into parts. Tho’ a particular colour, taste, and smell are qualities all united together in this apple, ’tis easy to perceive they are not the same, but are at least distinguishable from each other (T, p. 2).

and later that ...there are not any two impressions which are perfectly inseparable. Not to mention, that this [viz., the liberty of imagination to transpose and change its ideas] is an evident consequence of the division of ideas into simple and complex. Where-ever the imagination perceives a difference among ideas, it can easily produce a separation (T, p. 10).

126

Body, Mind and Self in Hume’s Critical Realism

And as he later tells us, ...as all distinct ideas are separable from each other, and as ideas of cause and effect are evidently distinct, ’twill be easy for us to conceive any object to be non-existent this moment, and existent the next, without conjoining to it the distinct idea of a cause or productive principle (T, p. 79).

The test of the distinguishable is freedom from contradiction in the statement that the one exists and the other does not. It is clear that when Hume says that the one object could exist while the object to which it is related should cease to exist, he means that the object that continues to exist remains the same in the sense that what can be truly predicated of it remains truly predicable of it; the object retains its identity in the sense of Leibniz and Russell, that (I) if there is an f such that fx and ~fy, then it is false that x = y Conversely, as Leibniz put it, the indiscernible are identical. Weinberg has argued in detail that this is the correct historical use of ‘distinct’.26 Thus, Ockham asserts that “Wherever there is any distinction or non-identity, contradictories can be verified of whatever are distinct...if A and B are not the same in all ways, then the following are true: ‘A is the same as A in all ways’ and ‘B is not the same as A in all ways.’ Thus ‘to be the same as A in all ways’ and ‘not to be the same in all ways’ are verified of A and B.” And later he adds that “it is impossible for contradictories to be verified of any unless those for which they stand are distinct things.”27 This notion of identity (and its converse, distinguishability) goes as far back as Aristotle: “...examine them in the light of their accidents or of the things of which they are accidents; for any accident belonging to the one must belong also to the other, and if the one belongs to anything as an accident, so must the other also. If in any of these respects there is a discrepancy, clearly they are not the same.”28 It continued down to the period of Locke and Hume. Leibniz of course was explicit in adopting it. It also appears in Hobbes: “two bodies,” he says, “are said to differ from one another when something may be said of one of them, which cannot be said of the other at the same time.”29 Thus, given the notion of separability, that if two objects are separable then the one can exist and the other cease to exist, it follows that if x and y are distinguishable, then they are distinct, and if y is separated from x by ceasing to exist, then x remains the same, it does not become distinct from itself, what has been truly predictable of it continues to be so.

127

Two: Nominalism and Acquaintance

This point also has a long history in philosophy. Suarez, for example, asserts that ...whenever two objective concepts are capable of separation in nature and in the concrete individual, either in such a way that one can remain in the sphere of reality without the other, or in such a way that one is really disjoined from the other and the real union between them is dissolved, we have a sign that the distinction between them is greater than the distinction of reasoned reason.30

and later that ...a distinction is real if both extremes can simultaneously and actually be preserved apart from a real union between them...[And] a distinction is real if one extreme can be preserved immediately and by itself without the other, and vice versa, to the exclusion of any ordination to or necessary connection with a third thing.31

The doctrine can be traced back to Aristotle: “moreover, see whether the one can exist without the other; for, if so, they will not be the same.”32 Descartes, as we saw, argues that mind and body both appear to be and are distinct. It is this which enables him to argue for the possibility of immortality. As he says in the sixth meditation: ...because I know that all things which I apprehend clearly and distinctly can be created by God as I apprehend them, it suffices that I am able to apprehend one thing apart from another clearly and distinctly in order to be certain that the one is different from the other, since they may be made to exist in separation at least by the omnipotence of God....just because I know certainly that I exist, and that meanwhile I do not remark that any other thing necessarily pertains to my nature or essence, excepting that I am a thinking thing, I rightly conclude that my essence consists solely in the fact that I am a thinking thing....And although possibly (or rather certainly, as I shall say in a moment) I possess a body with which I am very intimately conjoined, yet because, on the one side, I have a clear and distinct idea of myself inasmuch as I am only a thinking and unextended thing, as, on the other, I possess a distinct idea of body, inasmuch as it is only an extended and unthinking thing, it is certain that this I...is entirely and absolutely distinct from my body, and can exist without it.33

Descartes is holding more than the thesis that thought and extension are logically separable in the sense that one can consistently conceive a substance as a thinking thing while not being an extended thing. To be able to conceive the latter consistently is compatible with there being substances

128

Body, Mind and Self in Hume’s Critical Realism

that are both thinking and extended. It is clear that Descartes means something stronger than this. Thought and extension are not only distinct but in a sense incompatible. They are incompatible in the sense that no individual, i.e., substance, exemplifying the one can exemplify the other. And these two substances are separable in the sense that the one can exist, and be what it is, even if the other were to be (if only by God’s Omnipotence) annihilated. Descartes clearly infers the distinguishability of the individual substances in which the two attributes inhere from the logical distinctness of the attributes. And from the distinguishability of the substances he infers the separability. It is just these two inferences, from distinctness to distinguishability and from distinguishability to separability in the relevant sense, that are characteristic of nominalism. In this sense Hume is a nominalist, accepting the two principles: (P1) Whatever is distinct is distinguishable, and conversely and (P2) Whatever is distinguishable is separable, and conversely. Thus, he tells us that ...whatever objects are different are distinguishable, and...whatever objects are distinguishable are separable by the thought and the imagination. And we may here add, that these propositions are equally true in the inverse, and that whatever objects are separable are also distinguishable, and that whatever objects are distinguishable are also different (T, p. 18).

These principles (P1) and (P2) admit of a logical atomistic interpretation. On this reading, to say that one thing is separable from a second simply means that it is logically consistent to deny the one and affirm the other. Refer to these principles so understood as (LAP1) and (LAP2). But there is a second reading, as we have now seen, that of the nominalists. On this reading, ‘distinguishable’ is to be so interpreted that (P+) two objects are distinct just in case there is a predication true of the one that is false of the other; and where ‘separable’ is to be so interpreted that (P*) for an object that is separable from another it is conceivable, and therefore possible that both (P*i) it continue to be the object it is, that is, retain its identity, and (P*ii) the objects related to it do not exist. Refer to the principles understood in these nominalist terms as (NP1) and (NP2). These nominalistic principles (NP1) and (NP2) are hardly new to Hume; as Weinberg has argued, they go back at least as far as Ockham.34 It is these principles that constitute Hume’s nominalism. It is these principles that have the consequence that relations do not exist. It is these principles

129

Two: Nominalism and Acquaintance

that are inconsistent with PA, that is, with Hume’s empiricism. And here we must emphasize the difference between Hume’s logical atomism by which he establishes his thesis about causation and the stronger nominalism which he also holds, and shares with other thinkers like Descartes, as we have just seen. For the logical atomist, the distinctness in experience of properties with which we are acquainted implies that they are logically separable. But the logical atomist need not go on to make the further claim that every distinction in predicates implies a distinction in individuals, nor, of course, the claim consequent upon this that one distinct individual can exist totally as it is without any of the others. These latter two claims, in effect (NP1) and (NP2), are those that distinguish the nominalist from the logical atomist. Hume is both a nominalist and a logical atomist, and he uses both positions to argue for his account of causation. All he needs to defend that position, however, is the logical atomism established by his empiricist Principle of Acquaintance. And in fact the nominalism leads him into positions that make it difficult for him to defend his account of causation. (5) Nominalism and Relations The nominalist’s problem with relations can easily be established. Begin with the relational fact (1) Rab Assume that the relation R is asymmetrical: (2) (x)(y)(Rxy –> ~Ryx) Let us take it that a is distinguishable from all other things. In that case, from (NP1) and (NP2) it follows that a is separable from all other things. That being so, it is possible that a exist and all else cease to exist. But this in fact cannot be. For, from (1) and (2) it follows that (3) (›y)[~(y = a) & Ray] (3) is of course not a necessary truth. But the fact that it follows from (1) and (2) establishes that under those conditions, things that are distinguishable are not separable. That is, one cannot jointly accept both that the facts (1) and (2) obtain and the nominalist principles (NP1) and (NP2). Or, to put it yet another way, one cannot accept both nominalism and the existence of asymmetrical relations. It is easy to see what the problem is. By virtue of (1), the predicate (4) Rxb is true of a while by virtue of (2) it follows from this that b must lack at

130

Body, Mind and Self in Hume’s Critical Realism

least this property, and therefore be different from a according to the principle (I). For, assume that b=a Then b has the property (4): Rbb But from (2) we obtain Rbb –> ~Rbb Modus ponens immediately yields the contradiction ~Rbb So b cannot have the property (4). Thus, given the existence of asymmetrical relations, there is, as (3) asserts, at least one individual that is distinct from a by the traditional criterion of identity, (I) x = y : : (f)(fx fy) but which is not separable from a.35 To put this in terms of a concrete example, take the relation parent of. Anne is a parent of Sam. Anne has the property of being a parent; this implies, necessarily, that there is an offspring who is different from Anne. This is the simple point of (3). Anne and her child are distinct by virtue of the fact that Anne has the property of being a parent and Sam has the property of being an offspring. More specifically, Anne has the property of being the parent of Sam, while Sam has the property of being the offspring of Anne. Thus, since the being of a thing, its identity according to (I), that is, the sum of the ways in which it may be said to be, the being of Anne necessarily implies the existence of Sam while the being of Sam necessarily implies the existence of Anne. In this sense the existence of relations introduces “necessary” connections into the world. On the other hand, as we shall argue, this is not what Locke and Hume were about when they rejected necessary causal connections on the basis of an appeal to PA. The point we have just made about relations is, of course, that their existence is incompatible with the nominalistic thesis that the distinguishable are the separable. Given that the total characterization of an individual x according to (I) includes all its relational as well as its non-relational characteristics, and given the existence of asymmetrical relations, it follows that an empiricist, one at least who takes it that we are in fact acquainted with genuine relations, – such a one cannot consistently be a nominalist. Asymmetrical relations make the strongest case; their existence entails the falsity of nominalism without further premises. But any relation will serve to create problems for the nominalist. Let (1) represent any rela-

131

Two: Nominalism and Acquaintance

tional state of affairs obtaining at t1. Then, if a and b are separable according to the nominalist doctrine, the a could continue to exist at t2 without being so related to b. That would mean that (1), that is, Rab is true at t1 while ~Rab is true at t2. Thus, at t1 the predicate (4) is true of a while at t2 its negation is true of a. In that case the antecedent of (I) is fulfilled, and we must conclude that a is not identical with itself. But of course, a is identical with itself. Thus, admitting the existence of relations and maintaining the thesis that, for any object a, a = a, is inconsistent with nominalism. Or at least, it is so if one adds an additional premise. This additional premise is the doctrine that absolute time is false. Suppose that each moment is constituted by a non-relational temporal entity. In that case ‘t1’ and ‘t2’ can be construed as the names of two unanalyzable moments. If one now makes the basic sentence for one’s language relational, of the sort ‘Fxt’ indicating that x has F at moment t, then one can avoid the above conclusion. The conflict between relations and nominalism derives from the fact that it requires us to affirm a certain predicate of a at t1 and to affirm the negation of the same predicate of a at t2. To avoid the conflict with nominalism, one must so describe the situations at t1 and t2 that one does not affirm both a predicate and its negation of a. With absolute time the situation at t1 may be described as Rabt1 and that at t2 as Rabt2. What is affirmed of a at t1 is ‘Rxbt1’ while what is affirmed of a at t2 is ‘Rxbt2’ where the latter is not the negation of the former. In this way the anti-nominalist conclusion can be avoided. On the other hand, one can certainly argue on the basis of PA that there are no non-relational temporal entities. That is, if one accepts an empiricism defined by PA then this argues against construing temporal entities in the way they must be if they are to be used to save the nominalistic claim. The nominalist, however, has independent grounds for construing temporal entities absolutely rather than relationally. In fact, a nominalistic case can be made against all relations. If such a case is successful, then the problems that we have just been discussing will not arise. The nominalists stated their position in terms of an individual remaining the same even though its relations to other individuals change. This, after all, is just the notion of separability that they used. The point is that ‘remain the same’ is construed as ‘no change of non-relational properties’. If, then, the range of ‘F’ in (I) is implicitly restricted to non-relational properties, it can appear that their notion of separability is not incompatible with the definition of identity (I). But now suppose that all relational prop-

132

Body, Mind and Self in Hume’s Critical Realism

erties could somehow be defined in terms of non-relational properties. In that case, it would be legitimate to restrict the range of ‘F’ to non-relational properties. Nominalists generally attempted to reduce relations to non-relational properties. These attempts propose to analyze a state of affairs of the form ‘Rxy’ in terms of a conjunction of non-relational states of affairs of the form ‘r1x & r2y’. Call t1 and r2 the foundations of the relation R. Then the relation R can be construed as contributed by the mind of the judger as he or she judges first that x is r1 and then that y is r2. This means that the relation R will not be predicated in the strict sense of x and y. Only the foundations of R will be predicated in the strict sense of the individuals. To say that the relations in which an individual stands change now amounts to saying that the mind is no longer able to contribute R. It will be unable to do that just in case one of the foundations disappeared. Suppose that God destroyed y, and thereby r2y. Then the relation of x to y will have disappeared, but there will have been no change of properties predicable in the strict sense of x. Before God’s action only r1 was predicable in the strict sense of x, and after God’s action it is still predicable of x. Thus, the antecedent of (I) is not fulfilled; x remains self-identical after God has destroyed y. Both Locke and Hume adopt the above account of relations as analyzable into non-relational facts. For Locke, ideas of relations are all complex ideas. The basic sort of idea, whether simple or complex, are ideas the mind has of “things as they are in themselves,” and then, besides these, “there are others it gets from their comparison one with another” (Essay, II, xxv, 1). The idea of a relation is a particular sort of complex idea. This idea is formed “by bringing two ideas, whether simple or complex, together, and setting them one by another, so as to take a view of them at once, without uniting them into one” (II, xii, 1). Or, as he restates the point later, When the mind so considers one thing, that it does as it were bring it to, and set it by another, and carries its view from one to the other – this is, as the words import, relation and respect (II, xxv, 1).

He uses the example of Caius. When we consider Caius as a positive being, the mind “takes nothing into that idea but what really exists in Caius” (ibid.), e.g., the species man or the colour white.

133

Two: Nominalism and Acquaintance

But when...I give him the name whiter, I intimate some other thing:...my thought is led to something beyond Caius, and there are two things brought into consideration (ibid.).

Furthermore, ...since any idea, whether simple or complex, may be the occasion why the mind thus brings two things together, and as it were takes a view of them at once, though still considered as distinct: therefore any of our ideas may be the foundation of relation. As in the above mentioned instance,...the colour white [is] the occasion why he is said to be whiter than free-stone (ibid.).

Relation is always between two things: There must always be in relation two ideas or things, either in themselves really separate, or considered as distinct, and then a ground or occasion for their comparison (II, xxv, 6).

Relations can change without any change in the things related: The nature...of relation consists in the referring or comparing two things one to another; from which comparison one or both comes to be denominated. And if either of those things be removed, or cease to be, the relation ceases and the denomination consequent to it, though the other receive in itself no alteration at all: v.g. Caius, whom I consider today as a father, ceases to be so tomorrow, only by the death of his son, without any alteration made in himself. Nay, barely by the mind's changing the object to which it compares anything, the same thing is capable of having contrary denominations at the same time: v.g. Caius, compared to several persons, may truly be said to be older and younger, stronger and weaker, &c. (II, xxv, 5).

Locke clearly reduces relations to non-relational foundations. The one individual can exist as it is, and without a change, if the individual to which it is related ceases to exist. The relation itself is contributed by the mind: it is the act of comparison, the act by which the mind compares the two individuals in respect of the relevant non-relational foundations. Hume’s account is much the same. As with Locke, there are ideas, simple or complex, of things (events or impressions) as they are in themselves. Ideas of relation have such non-relational ideas as their foundation. Suppose we have the two ideas of being r1 and being r2. Simply conjoining these, that is, the idea of being r1 and being r2, yields a philosophical relation. If, in addition, being r1 is associated with being r2, so that the mind is

134

Body, Mind and Self in Hume’s Critical Realism

carried from the one to the other, so that, for example, the latter is called up by the former, then we have a natural relation. The word RELATION is commonly used in two senses considerably different from each other. Either for that quality, by which two ideas are connected together in the imagination, and the one naturally introduces the other...; or for that particular circumstance, in which, even upon the arbitrary union of two ideas in the fancy, we may think proper to compare them. In common language the former is always the sense, in which we use the word, relation; and 'tis only in philosophy, that we extend it to any particular subject of comparison, without a connecting principle (T, pp. 13-14).

Of all philosophical relations, resemblance is the most general. Indeed, Hume thinks it necessary to all other relations, “since no objects will admit of comparison, but what have some degree of resemblance.” However, the psychological laws that describe and explain how associations arise, deem that resemblance is always not sufficient for associating ideas: “...tho’ resemblance be necessary to all philosophical relations, it does not follow it always produces a connexion or association of ideas” (T, p. 14). Other relations are those of space and time, and, of course, of cause and effect. In this account of relations there are objective and subjective components. The objective component is the relation qua philosophical; the subjective component enters when the relation is taken naturally. This account clearly should not be taken as construing relations as mere conjunctions of non-relational states of affairs. For, there is, obviously, also the element contributed by the mind, by the subject who judges that the relation obtains. On the other hand, what this subject judges about is simply the two individuals qua having the two foundations. So it is equally obvious that while the whole complex which is the relational state of affairs may, upon this account, in some way involve the mental contribution, the objective part of the complex is constituted by the individuals qua having the foundations. Only by restricting the range of ‘F’ in (I) to non-relational predicates can the nominalists make plausible their principle (NP2) that distinguishable individuals are separable. The nominalists’ attempt to justify such a restriction is based on the claim that relations are reducible to their foundations. Moreover, if we are correct, their position can be plausible only if they hold that time is absolute. But this latter claim follows from their claim that relations are eliminable. It is, therefore, this claim about the eliminability of relations that is crucial for the nominalist position. In effect

135

Two: Nominalism and Acquaintance

(NP2) presupposes rather than establishes the non-existence of relations. And as we shall now argue, the nominalists’ other principle, (NP1) also presupposes rather than supports nominalism. This suggests, of course, that nominalism rests on some premise other than (NP1) and (NP2). It can be seen that (NP1) either simply asserts nominalism or does not entail the nominalistic conclusion.36 We must distinguish two readings that can be given to (NP1). The nominalists insist that ‘distinguishable’ means ‘non-identical’; that is the force of their insisting upon (I). But (I) applies only to individuals; it is necessary to give a corresponding notion for properties. Considering individuals, we have “distinguishablei”: Two individuals can be distinguishedi just in case there is a predication true of the one which is false of the other. This is the force of (I). But there is a second sense, “distinguishablep”: Two properties can be distinguishedp just in case that they are not coextensive.37 The point is that we have here two notions of non-identity applying to different types, in the Russellian sense of ‘type’. This means that from the occurrence of distinct predications or properties one cannot infer that the predications apply to distinct individuals. Or, at least, one cannot do so without begging the issue which the principle is supposed to establish. That is, only if it is already assumed that properties are as particular as the individuals of which they are properties is it possible to infer from two predications, i.e., from two properties, that there are two particulars distinct from each other in just the sense in which individuals are distinct from each other. Still differently, (NP1) reads “Whatever is distinct is distinguishable (and conversely)”. The ‘distinct’ applies to properties. Either it implies that the properties are distinguishablep, in which case the nominalists’ conclusion does not follow; or it implies the properties are distinguishablei, in which case the issue is begged. It is thus safe to conclude, I believe, that the nominalists’ principle (NP1) implicitly presupposes rather than supports the nominalist conclusion. This would suggest once again that implicitly there is a different argument for nominalism to which appeal is implicitly made. Nor do I think that this suggestion is false. The nominalists’ conclusion is this: all properties are non-relational, and all properties are as particular as the individuals of which they are properties. The implicit argument for this is to be found, I believe, in what may be called the Principle of Localization: everything that exists is localized in space and time.38 Properties exist: that is, there is something in individuals which make certain predicates (e. g., ‘red’) true of them and not of others (e.g., ‘green’) and this something which is there is what is called

136

Body, Mind and Self in Hume’s Critical Realism

a property. By the Principle of Localization it follows that properties are as particular as the things in which they exist.39 The same Principle of Localization demands that all relational properties be reducible to non-relational properties. If relational properties were not so reducible, then a relation would as it were span the two individuals of which it is jointly predicated. Certainly, it would not be localized in space and time in the way individuals are localized. It would, therefore, not exist. Relations exist only insofar as their foundations exist.40 We thus see how the Principle of Localization, which is presupposed by the nominalists’ principle (NP1), also forces upon the nominalists the belief that makes (NP2) plausible, namely, the belief that the range of ‘F’ in (I) may be restricted to non-relational properties. It is of course true that one does not want to reject any nominalistic analysis of relations. For some so-called relations, the nominalist account of Hume is adequate. For example, it is adequate as an account of the nomological tie of causal necessity; at least, so many hold.41 But it will not do as a general account of relations.42 For, in the first place, a formal proof can be given that dyadic relations cannot in general be reduced without contradiction to monadic predicates.43 Consider the pair of formulae (s1) (›x)(y)(Rxy) and (s2) (y)(›x)(Rxy) For any interpretation of ‘R’, these mean different things, since the second is entailed by the first, but not conversely. Now suppose that ‘Rxy’ can systematically be replaced without loss of meaning or of objective content by ‘r1x & r2y’. This yields (s1') (›x)(y)(r1x & r2y) and (s2') (y)(›x)(r1x & r2y) But these are logically equivalent to (s1") (›x)(r1x) & (y)(r2y) and (s2") (y)(r2y) & (›x)(r1x) Since conjunction is commutative, it is evident that (s1") and (s2") are logically equivalent. In that case, so are (s1') and (s2') logically equivalent. That is, each entails the other. However, (s1) and (s2) do not entail each other: (s2) does not entail (s1). Hence, statements of the sort (s1') and (s2') cannot be equivalent to statements of the forms (s1) and (s2). Relational statements

137

Two: Nominalism and Acquaintance

cannot, in short, be reduced to non-relational statements.44,45 The second point to be made about relations is, of course, that they are presented to us. To be sure, the objective relation of causal necessity is not among those that are presented – that, after all, is why the empiricist rejects the claim that there are such relations. But there are relations that are presented to one, objective relations, there in the states of affairs that are presented to one. Thus, for example, when I see a red patch that is wholly surrounded by a white patch I see that the red patch is within the white patch, where “x is within y” is a property neither of x alone nor of y alone but of x and y jointly. It is only nominalist dogma that insists that a predication cannot have two subjects. This dogma goes back to Aristotle at least,46 in his doctrine that “Everything except primary substances is either predicable of a primary substance or present in a primary substance” (Categories, Ch. 5, 2a34ff), and that “All things except primary substances are predicable of or present in them, and if primary substance did not exist, it would be impossible for any of the others to exist” (Categories, 2b4), doctrines naturally interpreted to mean that any attribute is either predicable of a substance or present in a substance, and if present in a substance then present in the substance at a time. Numerically one accident is, therefore, not in two subjects at the same time. It follows that there can be no relations. There is a second important consequence implicit in this Aristotelian position. If one property, or, as Aristotle says, an accident, cannot be in two subjects at the same time it follows immediately that numerically one property cannot be in two different individuals: properties are, in other words, as particular as the things of which they are properties. As some speak, properties are “tropes.” It is the dogma of Localization that lies behind the nominalist position on predication that Aristotle passed on to posterity. This nominalism continued through the Middle Ages – where various needs with respect to the Trinity laid further restrictions on how a philosopher could deal with relations47 – and at least so far as relations are concerned, found its most forceful defender in Leibniz, who repeatedly used the argument that an accident cannot be in two subjects.48 Locke and Hume simply unquestioningly take over the doctrine from the tradition. But the consequence of this nominalism, based on the Principle of Localization, is a doctrine about relations that must be rejected by an empiricist as incompatible with PA: we are acquainted with relations – relational facts are presented to us – and so relations must be admitted into any

138

Body, Mind and Self in Hume’s Critical Realism

empiricist ontology. It follows that the empiricist must reject the Principle, or rather, as it has been rightly called, the Dogma of Localization49 or whatever other principle the nominalist uses to defend his position. We must therefore carefully disentangle the logical atomism of Locke and Hume, which is acceptable to the empiricist, from their nominalism, which the empiricist must reject. In particular, we have argued that the Humean account of causation depends upon the logical atomism and not upon the nominalism. If we can correctly so conclude then criticisms of Hume’s position that are directed at his nominalism do not affect Hume’s logical atomistic and empiricist defence of the Humean account of causation. Hume does appeal to (P1) and (P2) in his discussion of the causal tie. But once nominalism is clearly distinguished from logical atomism, these two principles admit of two interpretations. Upon the nominalist interpretation, distinct properties imply separable substances. Upon the logical atomist interpretation they mean only that (LAP1) to be distinct properties and to be distinguishable in perception amount to the same and that (LAP2) to be given in perception as distinct and to be logically separable or independent amount to the same. While it is true that Hume appeals to (P1) and (P2), understood nominalistically as (NP1) and (NP2), he in fact needs only (LAP1) and (LAP2) to establish his position on causation. Hume is a nominalist, and he accepts (NP1) and (NP2). But as we have seen, his defence of the Humean account of causation depends on nothing stronger than (LAP1) and (LAP2). That makes his account of causation acceptable to the empiricist. But by way of his nominalism he leaves himself open to critics who wish to reject his account of causation. It is to this issue that we now turn. (6) Nominalism, Causation, Substances and Things A number of critics of the Humean account of causation suggest that Hume’s case against objective necessary connection depends upon his nominalism. Thus, Sterling P. Lamprecht seems to think that Hume was committed to the view that all causal regularities are of the form (L1) Whenever F then G = (x)(Fx –> Gx) where this form implies some sort of discontinuity between cause and effect, a discontinuity that Lamprecht finds incompatible with the perceived facts: “There is not merely a fire and then a hot oven...There is not merely an avalanche and then destruction; but there is an irresistible sweep in which the avalanche destroys whatever is in its path. There is not merely a

139

Two: Nominalism and Acquaintance

storm on the coast and then the loss of many feet of property; but there is the inevitable process in which the stormy waters eat away the soft soil.”50 Mandelbaum has made the same point: “...if I am dragging a heavy object at the end of a rope, or pushing a piece of furniture from one part of a room to another, however short a distance or however far I move it – and whether I ever get it to the spot to which I intend to move it – the cause is not some separate prior event but consists in all that is involved in the act of hauling and pushing: It is included within the boundaries of the event itself.”51 The significance of these examples, as Mandelbaum puts it, is that “they illustrate the fact that in many of our everyday uses of causal notions these apply to what occurs within what is seem as a continuous, ongoing process.”52 These examples are taken to refute Hume: “Hume’s puzzle as to how we could possibly be said to see any connection between cause and effect arose only because he started from the assumption that in the causeeffect relationship we are always dealing with what were originally seen as two separate wholes.”53 The contrast Mandelbaum draws makes clear that on his view, if Hume is correct, then cause and effect are separate in the sense of unrelated. But causal processes are, as his examples and those of Lamprecht make clear, in many cases continuous process in which cause and effect are related to each other in certain ways. Since we do perceive this relation between cause and effect Hume must be wrong. And in fact we are of course acquainted with the relations; PA requires us to include them as part of our ontology. That means that what we have for these processes are laws like (L2) (x)[Fx –> (E!y)(Gy & Rxy)] B where the relation R is assumed to have certain structural properties, such as asymmetry, transitivity, etc., to ensure the continuity between the cause and the effect. This emphasis on the continuity of causal processes can also be found in other empiricists like G. Bergmann54 and Bertrand Russell.55 The important point is that a law like (L2) is synthetic statement of fact; that is, it is not contradictory to affirm of an individual a that Fa while denying that (E!y)(Gy & Ray) (L2) therefore satisfies the Humean thesis that causal laws are not necessary truths, because the antecedent and consequent are separable in the logical sense. Contrary to Mandelbaum and Lamprecht, then, the Humean ii

. “(E!x)(Fx)” means that there is exactly one x such that this x is F.

140

Body, Mind and Self in Hume’s Critical Realism

can recognize the correctness of their examples while still defending his or her account of causation. These critics think that Hume cannot accept these examples because he rejects all relations. That excludes laws like (L2) and therefore the possibility of giving an account in Humean terms of the examples they propose. These critics have some justice in their critique. For example, we find Hume saying that “One event follows another, but we never can observe any tie between them. They seem conjoined, but never connected.”56 It is this that the critics wish to deny. Thus, we find Lamprecht saying that “Causality is basically a matter of a kind of connection between elements within single events.”57 But the conjunction that denies any sort of connectedness is the consequence of Hume’s nominalism, not his logical atomism, and all he needs to establish his point about causation against the Aristotelians and the rationalists is his logical atomism. It may well be that a failure to distinguish Hume’s logical atomism from his nominalism has misled Lamprecht and Mandelbaum into a belief that a Humean must hold that all causal laws are of the form (L1), and thereby misled them further into a rejection of the Humean account of causation and laws. W. Edwards has praised Lamprecht because he applies the empirical method to a subject-matter that has not already been distorted.58 It is sound advice, indeed, that the empiricist should not let metaphysical principles distort the view of the world given to him or her in perception: metaphysical principles should not lead the empiricist to deny what is admitted by the principle, PA, that he takes to be fundamental. That is why the empiricist must reject the nominalist rejection, based on the Principle of Localization, of relations. Hume the nominalist does begin from an ontologically distorted subjectmatter, and it is a virtue of Lamprecht that he restores relations to their rightful place in an empiricist ontology. On the other hand, it is the opposite of a virtue to include among the class of relations one that does not exist, one that is not presented and therefore deemed by PA to be inadmissible to any empiricist ontology. When Lamprecht includes in his ontology a primitive causal relation he distorts the subject-matter as much as does Hume the nominalist, and if such distortion is the opposite of empiricist virtue in Hume it is also the opposite of empiricist virtue in Lamprecht. Similar points about relations apply to another example that Mandelbaum uses to criticize Hume. He considers the case where we watch the coming of a summer thunderstorm. The breeze springs up, the air cools, the leaves rustle, and soon branches bend, then toss in the wind; there is thunder, and the rain finally comes. “There is a clear sequential order of

141

Two: Nominalism and Acquaintance

which we are aware,” Mandelbaum argues; “However, we do not experience this order as being a single linear progression from a to b to c to d. The rising storm is felt through all these effects together: The sudden breeze that cools the air also rustles the leaves, but their connection is felt and seem in the change all about us, not through the knowledge that the breeze I feel also rustles the leaves. What gives unity to what is experienced in watching a storm grow is a crescendo of change: change in the air and in the light and the sky, but above all in the increase of movements and sound as the storm gathers and breaks.”59 In this case we have a certain pattern among the things in the process and this pattern changes over time. The later pattern R2 is related to the earlier pattern R1 by a certain secondorder relation R2. In this case we will have a law that goes something like this: (L3) Whenever a group of objects are in a pattern R1 then those objects will soon be in a pattern r such that R2(R1, r) and where it is assumed that R2 has certain structural properties that will guarantee a continuous succession of patterns of the first order: there is a “crescendo of change” and “what is experienced as common to the individual elements that enter into its [the storm’s] changes is that each is included within a developing pattern, that each is passing through a transformation similar to that which is also characteristic of the others.”60 It is evident that the nominalist can think only in terms of elements, not the relations and patterns that structure these, nor the second order relations that structure these patterns into a series of similar but progressively developing patterns. That means that the nominalist cannot begin to accept that there are laws like (L3), nor, therefore, can he begin to make sense of the sorts of case that Mandelbaum has advanced. Insofar as Hume is a nominalist, then Mandelbaum has made a correct observation. But for Hume to defend the Humean account of causation, he does not need his nominalism; all that he requires is his logical atomism. And logical atomism is quite compatible with the existence of structuring relations and with the existence of laws like (L3). So, while Mandelbaum may have made a point against nominalism, he has not touched the Humean account of laws and causation. Lamprecht makes another point. We see in objects around us, he says, “active compulsion,” and this yields a “necessary relation between” events. “There is not merely an avalanche and then destruction; but there is an irresistible sweep in which the avalanche destroys whatever is in its path.”61 Now, it is no doubt true that we recognize in the situation of the

142

Body, Mind and Self in Hume’s Critical Realism

avalanche an irresistible force causing destruction; it is also true no doubt that we witness the exercise of a power of active compulsion; but does it follow from this that we observe causality as a simple relation among states of affairs? The facts being granted, must we therefore reject the Humean account of laws as false? Or are these facts compatible after all with the Humean position that there are no objective unanalyzable necessary connections? Let us see. Lamprecht notes later that The term “causality” does not designate the undisturbed development of a situation (if there be such situations), nor that part of the development of the situation which is accounted for and understood in terms of the elements that were present prior to some intrusion. Causality is essentially the interference of new elements in existing situations.62

Apply this to the case of the avalanche. We have the regular conditions, the hill, the valley, the Heidegger’s hut standing below the hill at the edge of the valley. And, of course, the absence of an avalanche. Then a moment later we have the destruction of the hut. And the presence of avalanche. We have, in other words, the following: +))))))))))))))))))))))))))))))))))))))))))))))))))))))))))), *

hut standing

avalanche

other conditions

* * time 1: present

* absent

present

* * time 2: absent *

* * *

present

present

* *

.)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))-

Given the assumption, perfectly reasonable and picked up from experience, that for any event like the destruction of a hut there is a sufficient condition for its occurrence, then a simple application of Mill’s Method of Difference to the facts before and after the intervention of the avalanche leads directly to the conclusion that the following regularity holds: (L4) Whenever the other conditions obtain, then being struck by an avalanche is sufficient for the destruction of a hut.

143

Two: Nominalism and Acquaintance

That is, from the observed facts we can directly infer from those facts that there is a causal law of which the facts are an instance. The causal explanation of the facts is discovered in the facts themselves.63 Moreover, in the same facts we do, as Lamprecht holds, discover a power that the avalanche has. It is true that for the Humean there are no unanalyzable powers that could provide necessary connections in nature, or, as we have seen Hume put it, “the distinction...betwixt power and the exercise of it, is...without foundation” (T, p. 171) and “entirely frivolous” (T, p. 311). It does not follow, however, that Lamprecht has shown that the Humean account of powers is wrong. For, as we have seen, Hume goes on to point out that “tho’ this be strictly true in a just and philosophical way of thinking, ’tis certain it is not the philosophy of our passions...” (T, p. 311). The language of “powers” has a certain place in our discourse, and must be accounted for. Hume’s proposal, we know, is that “power has always a reference to its exercise, either actual or probable, and...we consider a person as endow’d with any ability when we find from past experience, that ’tis probable, or at least possible he may exert it” (T, p. 313). Hume’s suggestion here, as we know, is that ‘P Æ Q’ represents a power just in case that there is a regularity in which it appears, in the case of the passions a regularity to the effect that For any time, a is P Æ a is Q and more generally a regularity to the effect that (x)[Fx Æ (Px Æ Qx)] where ‘F’ represents some categorical property.64 Hume is here using the Lockean criterion of significance (as Bergmann calls it), arguing that the concepts that we define are those that are useful in the sense of appearing in statements of empirical regularity. If we now take the property struck by an avalanche –> destroyed then we recognize that this becomes a power that avalanches exercise on huts by virtue of the fact that it occurs in the law (L4). In short, when we discover in the facts we observe the law that explains those facts we thereby also discover that the avalanche has the power to destroy huts when it strikes them – exactly as Lamprecht recognizes! Except that we see that in granting Lamprecht’s point we are granting nothing that is in conflict with the Humean account of laws. Turn now to Harré and Madden65 who have also criticized the Humean account of causation. As it turns out, they are misled like Mandelbaum and Lamprecht into the belief that rejecting Hume’s nominalism suffices to reject his account of causation. They hold that nominalism is the

144

Body, Mind and Self in Hume’s Critical Realism

central premise for Hume’s rejection of nomological necessity: Since [Hume] takes punctiform and atomistic sensations as epistemically basic, it would be impossible for any single impression to be the original of any relational concept, let alone the original of ‘the action of causal power’.66

The “atomism” to which these authors refer is, of course, Hume’s nominalism. And, as these authors see, if relations are to be accepted – as they should be – then the nominalism must go. But it does not follow, as they seem to suggest, that once relations are admitted causal ties become permissible. Nor does it follow that there are substances. Harré and Madden point out that Hume’s nominalism, that is, his “atomist” view of impressions, is equally a view about the world, that “...it [the world] too comes in atoms, but of course it atoms are events.”67 From this, they suggest, derives a claim they believe to be Humean, ...that the world as experienced can with equal facility be conceived as a system of things or as a flux of events, and that the latter assumption contains all that is present in the former without its allegedly unwarranted assumption of continuity.68

Now, it is true that the traditional notion of a substance as a simple entity underlying the qualities of things and enduring through change is excluded by PA. As Hume says, “We have...no idea of substance, distinct from that of a collection of particular qualities, nor have we any other meaning when we either talk or reason concerning it”, adding later that “the particular qualities, which form a substance, are commonly refer’d to an unknown something, in which they are supposed to inhere; or granting this fiction should not take place, are at least supposed to be closely and inseparably connected by the relations of contiguity and causation” (T, p. 16). In this sense, in which continuity is grounded in substances, it is true that PA excludes continuity. But a group of events can form a continuous series, without any reference to substances, provided they stand in the appropriate relations of contiguity and succession. PA does no exclude such relations, and therefore does not exclude continuity. If, as Harré and Madden suggest, ‘thing’ implies ‘continuity’, then a flux of events can be a system of things. It is not any assumption about continuity that PA deems unwarranted, nor any assumption that the world of events is also a world of things; but only the assumption that continuity must be accounted for in terms of a substance or continuant underlying the flux of events. Of course,

145

Two: Nominalism and Acquaintance

if we accept Hume's nominalism along with his logical atomism, then the continuity that relations secure will indeed be lost; but so far as the Humean argues only about causation (and that is the main concern of Harré and Madden) then he needs only Hume’s logical atomistic assumptions and can accept that the flux of events is so structured within itself that the events are in fact combined into things. Harré and Madden introduce their concern about substances because they see the latter playing a vital role in their non-Humean account of causation. Which, indeed, they must do, as in any Aristotelian account of causation: in this account the continuant is a natured substance and the nature provides the necessary tie between the events that are present in the substance. With the assumption that successive events are independent, they view the Humean as making two additional assumptions. First, It is assumed that if a material thing undergoes a sequence of changes, thus generating successive events, then these events are absolutely independent on of another, that is, that no matter what the structure of the previous succession h ad been, it is possible that the thing could take on any property whatever in the future.

And second, ...it is assumed that no matter what properties a body has, it could, for all we can know, at any time take on simultaneously any other properties whatever.69

The only exception is that a thing cannot simultaneously take on two determinates under the same determinable; e.g., a thing cannot simultaneously be red and green (all over), or any other pair of properties under the determinable characteristic of being coloured. These assumptions lead to the Humean problem of induction. The atomicity of events...ensures the sequential independence of properties. The redness of x at t1 and t2 happened to occur in that sequence but they do not belong in that sequence, since the event-ontology eliminates the notion of an enduring entity the nature of which explains why the thing exhibits a given property continuously. It is this sequential independence of properties that leads to the Humean problem of induction. If properties are independent, then indeed what reason is there for supposing that at t2 the ‘apple’ will be round instead of square?70

Clearly, the second of the above assumptions implies, Harré and Madden

146

Body, Mind and Self in Hume’s Critical Realism

hold, we cannot explain why the properties of the apple co-exist simultaneously as they do.71 Harré and Madden conclude that Since the events are instantaneous time-slices of the physical thing, the persistence of the pattern is an inexplicable and basic fact in that there is no requirement for the successive time slices to bear any resemblance to the event previously occurring at that place. The result is the accidental persistence of the pattern for which there is no rational explanation.72

The natured continuant can, Harré and Madden argue, solve the problem of induction to which Hume’s “atomism”, i.e., his nominalism, gives rise. Let us look a little more closely at the Humean ontology of events. To say of an event that “This event is (part of) a thing” is to say that there are other events related in relevant ways to this event. The relevant ways are just those relations of contiguity and succession that constitute the continuity and therefore the (Humean) identity of the thing through time. When we are presented with the event of which we are speaking, how do we know the other events are there that must be there if the event we are talking about is (part of) a thing? Primarily, we rely upon knowing the structural properties of relations. We might, for example, know that if an event is of the F-sort and is itself R-ed by some other event (i.e., is in the counter-domain, CDR, of R) then it itself R-s some other event. That is, we might know a law of the sort (L5) (x)[Fx & x , CDR: Æ (E!y)Rxy] If the fact were then presented to us that event b is F and that a is R to b then we could deduce that there is a unique event that is R-ed by b. If we also knew something like (L6) (x)(y)[Fx & Rxy: Æ Fy] we could deduce further that the unique event that is R-ed by b is also F. And from this we can deduce that this individual R-s yet another unique event. So laws of the sort (L5) and (L6) can provide us with good reasons for believing that this thing will continue to be the same sort of thing, that is, that the events succeeding the event in question will have the same property as the event in question. Thus, such laws give us good reasons for supposing that at t2 the apple, which is round, will continue to be round and not change to square. Such laws can also explain why things change their properties. Thus, if we have (L5) and that instead of (L6) we have the law (L7) (x)(z)[Fx & (›y)R’xy & Rxz : Æ Hz] and we observe that b is G and that R’bc, then we have good reason for thinking that the unique event R-ed by b (the existence of which is guaran-

147

Two: Nominalism and Acquaintance

teed by (L5)) will be H; which is to say, we will have good reasons for believing that the event in the thing which is F will be succeeded by an event that is H; which, again, is to say, we will have good reasons for believing that the thing we now perceive to be F will change to being H. In short, that events are things, that patterns of properties persist in such things, and that these things also undergo various changes – all these can be explained by laws. Harré and Madden notwithstanding, the Humean is not committed to the thesis that persistence of patterns is accidental and that there can be no rational explanation for it; for, laws, regularities, do explain such things. Of course, the explanation is not in terms of Aristotelian natures or unanalyzable powers. But why should the Humean agree that ‘good reason’, and ‘explanation’, and so on, are to be understood only in non-Humean terms? The Humean will agree that on his position the thing continues to be the same sort of thing, or even that it continues, is a matter of inductive inference and therefore not certain. And he will agree that it is always logically possible that a thing will cease to have a pattern of properties his lawbeliefs have led him to suppose would persist. That, after all, is simply part and parcel of his anti-Aristotelianism. But the Humean will not allow that he never has good reason for his beliefs about things.73 It is just that his good reasons are always fallible. This sort of reply by the Humean to Harré and Madden, insofar as it emphasizes the relations that structure events into things, is, it is clear, not available to the Humean who is also a nominalist, one who denies the existence of relations. To this extent, Harré and Madden have a point to make against the historical Hume, but their discussion does not touch one who adopts only Hume's view on causation and natured substances and rejects his nominalistic account of relations. The admission of relations has another consequence, so far as the identity of a thing or event is concerned, which we should note, since it gives a further sense to the concerns of Lamprecht and Mandelbaum that in causal processes there is a sort of interconnection of cause and effect that makes them in some way inseparable. Recall the case of Anne, who is parent of Sam, who, conversely, is offspring of Anne. As we have seen above, this example illustrates that there is a sense in which the admission of relations does introduce non-contingent connections among events. Consider the events that enter into the constitution of a thing: an event being a successor is necessarily connected to having a predecessor. One cannot conceive of an R-ed event without an R-er; that is what it means to be an R-ed event, an event in the domain of R. But it is evident that does not in any way touch the Humean case against there being necessary causal connec-

148

Body, Mind and Self in Hume’s Critical Realism

tion. But it does mean that cause and effect are necessarily implicated with each other, once we admit (non-causal) relations – as we have agreed with Lamprecht and Mandelbaum that we must admit. To see this, recall the notion of identity that nominalist uses, that is, the Russell-Leibniz sense of (I): x = y .. (f)(fx fy) If an event is the upshot of a process then its identity, in this sense, is constituted, in part, by its being the upshot of the process. Consider a process explained by (L2) (x)[Fx Æ (E!y)(Gy & Rxy)] Let the process be a and the upshot b. From the observed fact that (z1) Fa we deduce (z2) (E!y)(Gy & Ray) We have thereby explained why there is a unique event, call it $, which is G and is R-ed by a: (z3) G$ & Ra$ Observation now confirms that (z4) b=$ So we have explained why (z5) Gb & Rab that is, why b, which is G, has the property of being R-ed by a. This enters into its Russell-Leibniz identity. b could therefore not be what it is unless the process a both existed and R-ed b. We cannot explain this fire without citing this cause, the cause that enters into the very being of this fire. And of course, conversely: we cannot conceive this cause without reference to this effect, the effect that enters into the very being of the cause.74 It is quite obviously true that b would not be what b is – that is, b would not have all the properties, relational and non-relational, that b has – if some of the relational states of affairs in which b stands were not to obtain. If a were not to exist, and Rab were not to obtain, then b would not have the property of being R-ed by a: under the hypothetical conditions, b would not be what b actually is. Thus, being R-ed by its cause is essential to the very being of b. One cannot conceive b being what it is without conceiving a as existing and as R-ing b; for that would be to conceive b being, among other things, R-ed by a without conceiving a as existing and as R-ing b, i.e., it would be to conceive a contradiction. b, in short, is inseparable from a. And a is equally inseparable from b. All this may be granted. Nonetheless, the Humean can consistently maintain that a as cause is separable

149

Two: Nominalism and Acquaintance

from b as effect and that b as effect is separable from a as cause. To say that a as cause is separable from b as effect is to say that one can consistently assert both that a is what it is and that a as such did not cause b. What a is is (so far as we are here concerned) given by (z1) and the second conjunct of (z5). One can consistently assert that these hold and yet deny that a causes b or any other event that if F is R-ed by a. For, to deny the latter is to deny the law (L2). The denial of (L2) is, clearly, consistent with the affirmation of (z1) and (z5). Similarly, we can conceive a unique event that if F and is R-ed by a while not conceiving a to be its cause. For that would be simply to suppose (z2) while not supposing (L2), which is quite consistent. Similarly, we can conceive b being what it is without conceiving b as the unique event that is caused by a. For that would be to suppose (z5) while not supposing (z4), which is quite consistent. And finally, we can conceive b being what it is without conceiving b to be caused by a. For that would be to suppose (z2) while not supposing (L2), which yet once more is quite consistent. Cause and effect can be inseparable in their being yet separable as cause and effect. The nominalist, denying that there are relations, would, of course, argue the stronger thesis that cause and effect are not only separable as cause and effect but are also separable in their being. But as we know, one who adopts the Humean account of causation is not also committed to adopting Hume’s nominalism. The last point must be kept firmly in mind when reading Hume, if we are to attain any clarity at all about his thought. Consider the following passage: The effect is totally different from the cause, and consequently can never be discovered in it. Motion in the second Billiard-ball is a quite distinct event from motion in the first; nor is there anything in the one to suggest the smallest hint of the other....When I see, for instance, a Billiard-ball moving in a straight line towards another, even suppose motion in the second ball should by accident be suggested to me, as a result of their contact or impulse; may I not conceive, that a hundred different events might as well follow from that cause? May not both these balls remain at absolute rest? May not the first ball return in a straight line, or leap off from the second in any line or direction? All these suppositions are consistent and conceivable?75

Only if one is a nominalist, denying that there are relations, can one hold that the effect is separable from the cause, totally different from it. For, once one admits relations then the effect, insofar as it is related in any way to the cause, can be discovered in the latter, that is, in its being. The con-

150

Body, Mind and Self in Hume’s Critical Realism

joining of the first of the just-quoted sentences with the second indicates how Hume uses his nominalism to support his analysis of causation. If a reader now rejects that nominalism, he or she may be misled into the belief that the subsequent discussion is without support. But in fact the statements following the ellipsis stand quite on their own, and can consistently be affirmed by one who denies the nominalism of the preceding remarks. “May I not conceive,” asks Hume, “that a hundred different events might as follow from that cause?” That is, as the italicized words indicate, that the event in question, the motion of the first billiard ball, could have had other causal consequences that it did? Can I not consistently conceive that motions of billiard balls obey laws other than those of Newton? To which the answer is surely affirmative. For, contrary to what the Aristotelian holds, there is nothing about the motions we perceive that compels us, when conceiving of one, to conceive of another as necessarily consequent upon it. And here I say ‘the motions we perceive’ to emphasize that the authority to reject necessary connection derives ultimately from the empiricist’s Principle of Acquaintance. (7) Resemblance, Abstract Ideas, the Distinction of Reason, and the Simplicity of Things Mandelbaum has discussed Hume’s case against causal relations in a most illuminating way – illuminating, not because it is correct (which it isn’t), but because it raises in an illuminating way the issue of the relation between the distinguishability and separability principles (P1) and (P2) and the Principle of Acquaintance. It will also enable us to explore further Hume’s doctrine of concept formation (“abstract ideas”), and the implications of the nominalistic account of relations for the relation of resemblance that Hume uses ontologically to account for the sameness of things. Consider again the billiard-ball example to which we have referred in the preceding section. Concerning this, Mandelbaum raises two, obviously intended to be rhetorical, questions: Had Hume been considering actual impressions, rather than ideas, could he have said that what a person actually observes when one billiard-ball strikes another is a case in which “the effect is totally different from the cause”? Can it also be said that an observer clearly distinguishes and can separate the motion of one billiard-ball from that of the other at the moment at which he witnesses their impact, or can this only be done in his imagination when he recalls his earlier impression, transforming them into sets of complex ideas?76

151

Two: Nominalism and Acquaintance

The first question can be taken as challenging Hume’s denial of relations. This nominalism has the consequence that the events in a causal sequence are “entirely loose and separate,”77 and this, perceptually, is just wrong, as the billiard-ball example itself should make perfectly clear. In raising this point, Mandelbaum is quite correct. But, contrary to what he apparently believes, it is irrelevant to the Humean account of causation. The second of Mandelbaum’s questions does, however, attack not just the nominalism but also the account of causation. The argument for the Humean account of causation turns upon the claim that certain ideas are distinguishable. It infers that such ideas are separable. It then further infers that therefore the events (impressions) falling under those ideas are separable. It is this last step that Mandelbaum challenges. Mandelbaum holds that the principles (P1) Whatever is distinct is distinguishable, and conversely and (P2) Whatever is distinguishable is separable, and conversely apply to ideas. For these purposes we must, as we insisted above, clearly distinguish these principles as implied by the nominalist’s Principle of Localization, which understands them as (NP1) and (NP2), and the same principles understood in logical atomistic terms, as (LAP1) and (LAP2). Now, at times Hume does seem to limit these principles to ideas, as the italics in the following statement of the principles apparently suggest: Whatever objects are different are distinguishable, and whatever objects are distinguishable are separable by the thought and the imagination (T, p. 18; italics added).

But at other times Hume equally explicitly applies the principles to events or impressions: Every thing, that is different, is distinguishable; and every thing, that is distinguishable, may be separated....[And conversely] if...they be not different, they are not distinguishable; and if they be not distinguishable, they cannot be separated (T, p. 36).

and again: What is distinct, is distinguishable; and what is distinguishable, is separable by the thought or imagination. All perceptions are distinct. They are, therefore, dis-

152

Body, Mind and Self in Hume’s Critical Realism

tinguishable, and separable, and may be conceiv’d as separately existent, and may exist separately, without any contradiction or absurdity (T, pp. 86-7).

And this move from separable ideas to separable events (impressions) is what Mandelbaum holds is illegitimate. There are, he holds, cases where ideas are separable but impressions are not. Of these, the billiard-ball case is one. But there are cases where even Hume is prepared to say that ideas are separable but impressions are not. Mandelbaum introduces the wellknown case where Hume attempts to interpret and make sense in his own terms of what were called “distinctions of reason.” Hume puts this as follows: Before I leave this subject I shall employ the same principles [viz., those concerning abstract ideas] to explain that distinction of reason, which is so much talk’d of, and is so little understood, in the schools. Of this kind is the distinction betwixt figure and the body figur’d; the motion and the body mov’d. The difficulty of explaining this distinction arises from the principle above explain’d, that all ideas, which are different, are separable. For it follows from thence, that if the figure be different from the body, their ideas must be separable as well as distinguishable; if they be not different, their ideas can neither be separable nor distinguishable. What then is meant by a distinction of reason, since it implies neither a difference nor separation (T, pp. 24-5; his italics).

We make a distinction of reason with respect to pairs of entities when those entities are inseparable but the ideas of them are different, and therefore separable. On the other hand, Hume also suggests, where the entities are inseparable in an impression they are also inseparable in the idea of that impression: The length is inseparable from the breadth both in nature and in our minds; but this excludes not a partial consideration, and a distinction of reason, after the manner above explain’d (T, p. 43, his italics).

Hume believes that his account of the distinction of reason enables him to resolve this apparent contradiction. Hume points out that “’Tis certain that the mind wou’d never have dream’d of distinguishing a figure from the body figur’d, as being in reality neither distinguishable, nor different, nor separable; did it not observe, that even in this simplicity there might be contain’d many different resemblances and relations” (T, p. 25). Choosing impressions of colour and shape, Hume proceeds:

153

Two: Nominalism and Acquaintance

...when a globe of white marble is presented we receive only the impression of a white colour dispos’d in a certain form, nor are we able to separate and distinguish the colour from the form. But observing afterwards a globe of black marble and a cube of white, and comparing them with our former object, we find two separate resemblances, in what formerly seem'd, and really is, perfectly inseparable. After a little more practice of this kind, we begin to distinguish the figure from the colour by a distinction of reason; that is, we consider the figure and colour together, since they are in effect the same and indistinguishable, but still view them in different aspects, according to the resemblances, of which they are susceptible (T, p. 25; his italics).

About this Mandelbaum comments that “I need not proceed, for what is evident here is that Hume’s axiom, as it is applied to our impressions, does not permit us to separate shape and color; we must learn to regard them as distinct through acts of comparison with other sets of impression, that is, by transforming what were originally complex impression into ideas.”78 But what holds in this case can surely with equal justice be applied to the case of cause and effect. “Consequently, we may say that [Hume] misused his axiom when he attempted to argue on the basis of it that our impression never disclose a relation between a cause and its effect.”79 This ingenious challenge to Hume’s case about causation raises directly the question whether (P1) and (P2) apply to impressions, that is, to events given in perception. To answer it adequately will require us to examine more carefully than we have Hume’s views on ideas, and, after that, to clarify his discussion of the “distinction of reason.” At the same time we must recognize that the basis of the objection, that the colour and the shape of the globe are inseparable, and therefore not distinct, is a thoroughly odd thesis. It means that the thing in question, the event or impression, is a simple entity, with no perceptually distinguishable parts. But that is surely wrong, for the colour white and the globular shape are perceptually distinct parts of the whole with which we are presented. It is, moreover, a view that is in conflict with some of the implications of Hume’s nominalism, as we shall see. It follows that if we are to deal with Mandelbaum’s objection, we will have to probe more deeply into how Hume conceives individual things. Let us begin by examining an actual argument in which Hume explicitly deploys the principles (P1) and (P2) in his discussion of causation. This is the argument Hume uses to show that the causal maxim, that whatever has a beginning must have a cause, is neither intuitive nor demonstra-

154

Body, Mind and Self in Hume’s Critical Realism

tively certain. Hume reasons as follows: As all distinct ideas are separable from each other, and as the ideas of cause and effect are evidently distinct, ’twill be easy for us to conceive any object to be non-existent this moment and existent the next without conjoining to it the distinct idea of a cause or productive principle...The actual separation of these objects is so far possible, that it implies no contradiction or absurdity (T, pp. 789).

In the perspicuous notation of the later logical atomists, Hume is claiming that (›y)[By & ~(E!x)Pxy] is not a contradiction.80 Here, ‘By’ means “y has a beginning” and ‘Pxy’ means “x produces y.” At this point in his argument, Hume does not have to tell us more about what ‘x produces y’ means. (The passage quoted is from Treatise, Bk. I, Pt. iii, Ch. 3, and he does not tell us what ‘x produces y’ means until I, iii, 14.) All that he needs to know about the meanings in order for the argument to go through is that the concept of beginning or to be after not having been does not imply the concept of producing – which, surely, it does not. As Hume says, the concepts, or, in his terms, the ideas, are distinct, i.e., logically distinct, and so separable. Now, for our present purposes it is not the argument itself that concerns us (though it is a sound enough argument81), but rather the nature of the ideas that argument asserts to be distinct and separable. These ideas are abstract ideas. Now, how does Hume characterize an abstract idea? It is, first of all, an idea, and all ideas derive from impressions.82 Ideas are all images (cf. T, p. 20). The latter claim is, of course, one of the theses that Hume argues for as part of his research programme in introspective psychology, a programme that attempted to analyze all thought into images.83 If an impression has inseparable parts, so does the idea of that impression. All impressions are fully determinate in degree and quality, nor are such degrees and qualities separable. So all ideas derived from such impressions are fully determinate in degree and quality. Whatever abstract ideas are, then, they are ideas that are fully determinate in all respects: we have no abstract idea of, let us say, a triangle as such, for there is no idea of a triangle which is not that of a determinate triangle, equilateral or isosceles or whatever (T, p. 18-20). But, if an abstract idea is an image it is also abstract or general. What we have just established is that the abstract idea, say of a triangle, is not general because triangularity as such has somehow been detached from

155

Two: Nominalism and Acquaintance

ideas that are individual, ideas of determinate triangles. In the case of an abstract idea, “The image in the mind is only that of a particular object, tho’ the application of it in our reasoning be the same, as if it were universal.” (T, p. 20). The abstract idea of man represents all men. It cannot do this by consisting straight-forwardly of images of all possible sizes and all possible qualities. For, an infinite number of such images would be required and these would have all to be before the mind simultaneously. But the mind does not have such an infinite capacity (T, p. 18), and so Hume proposes that these other ideas are there but only dispositionally, not actually. “They are not really and in fact present to the mind, but only in power; nor do we draw them all out distinctly in the imagination, but keep ourselves in readiness to survey any of them, as we may be prompted by a present design or necessity” (T, p. 20). The ideas are brought together by custom or habit, and this habit is mediated by a general term. “A particular idea becomes general by being annex’d to a general term; that is, to a term, which from a customary conjunction has a relation to many other particular ideas, and readily recalls them in imagination” (T, p. 22). An abstract idea is a class (possibly infinite) of particular ideas each of which has come by custom to be associated with all the others, and each of which has come to be associated with a certain general term. Only one of the particular ideas in the class is actually before the mind at any one time when the general term is used; the rest are present to the mind only potentially. When the mind uses an abstract idea it has before it the general term and one, it matters not which, of the particular ideas in the class of ideas with which the general term is associated. The same particular idea can belong to several abstract ideas. This can happen when several different words are attached to it, and it is a member of several of those classes of particular ideas the association of which renders them into an abstract idea: “...the very same idea may be annext to several different words, and may be employ’d in different reasonings, without any danger of mistake” (T, p. 21). To say that there is an association or custom that links the various particular ideas and renders them abstract is to say that those ideas stand in a natural relation. This natural relation is based on a philosophical relation. This philosophical relation, that is, the objective basis of the relation, is that of resemblance. “When we have found a resemblance among several objects, that often occur to us, we apply the same name to all of them, whatever difference we may observe in the degrees of their quantity and quality, and whatever other differences may appear among them” (T, p.

156

Body, Mind and Self in Hume’s Critical Realism

20).The ideas that are customarily united to form an abstract idea all resemble, in some respect or respects, each other and the impressions from which they are derived. Ideas derived from further impressions will be as it were picked up by the custom provided those impression resemble in the relevant respects the ideas that already constitute the abstract idea. Of course, not all resemblance classes of impressions and ideas have an abstract idea corresponding to them and a general term that, in effect, names them. A resemblance class forms an abstract idea when custom, (a) unites ideas of that class together, and (b) so attaches a general term to each and all that when one of the class of ideas appears in consciousness so does the general term. It is only when the resemblance becomes natural as well as philosophical that there an abstract idea for that resemblance class. Thus, particular impressions or events fall under an abstract idea insofar as the former resemble each other in a certain respect. We must now note that resemblance turns on internal features or aspects of impressions. This is because resemblance is a relation. Since it is a relation, the nominalistic analysis of relations will apply to it no less than to other relations. This means that any fact to the effect that x resembles y, considered objectively, that is, as Hume puts it, philosophically, must be a conjunction of two facts, one to the effect that a certain aspect is present in x, and the other to the effect that a certain aspect is present in y. The aspect that is in x and the aspect that is in y will be the foundations of the relation of resemblance between x and y. But resemblance is a difficult relation. We usually speak of both exact and inexact resemblances. Fido and Rover do resemble each other, at least insofar as they are both dogs, but the resemblance is certainly not exact. And there are degrees of inexact resemblance. The various shades of red inexactly resemble each other, but the reds also inexactly resemble oranges, more at least than they inexactly resemble blues. So there is no single dyadic relation of inexact resemblance. What we need, it seems, is a tetradic relation: ‘x resembles y more than v resembles w’. But this won’t do either. The relation asserted by the statement that ‘x resembles y more than v resembles w’ can only be understood as asserting a dyadic relation of difference in degree the terms of which are two separate instances of a dyadic relation of resemblance, the first instance having x and y as its terms and the second having v and w.84 Thus, what we need to make the view work is, in the first place, a complete set of relations of exact resemblance. That is, we will need, for example, for each shade of colour a relation of exact resemblance, and then for each specific shape a relation of

157

Two: Nominalism and Acquaintance

exact resemblance, and for each tone a relation of exact resemblance, and so on. Second, these must be grouped into families, the colour resemblances, the sound resemblances, the shape resemblances, and so on. Each relation of exact resemblance must therefore have a certain property which it will share with the other relations of exact resemblance in the same family. And third, within each family there must be a second order dyadic relation (or perhaps several) that orders the relations of exact resemblance according to degree. But the exact resemblances must be analyzed in terms of nonrelational foundations. It is clear that the non-relational foundations of a specific relation of exact resemblance, say with respect to a determinate shade of colour, can be none other than the properties in the two things that are said to resemble. It is the properties of things that found the relations of exact resemblance. Moreover, the properties must also found the order among the resemblances. Each property, in other words, must have within it that which determines it to have the exact degree that it has relative to other properties of the relevant family. In fact, what founds the degree of a property must be inseparable from the property. We see Hume recognizing these points when he asserts that, “...the mind cannot form any notion of quantity or quality without forming a precise notion of degrees of each” (T, p. 18). It may very well be that we want to account for resemblance in terms of properties that found the relation. That is, it may be that the account of relations that Locke and Hume give applies correctly to resemblance as we have argued that it applies correctly to the causal relation. Indeed, that is not unreasonable.85 At the same time, however, it certainly won’t do as an account of the orders among properties. Begin with a certain quality that is less than in quantity some other quality. Since less than generates an order, this quality that we have started with will stand in the same relation to all the other qualities in the series. But all these relational facts must be so analyzed that they have foundations in the quality with which we began. Each quality will have a quantity of foundations depending upon where the quality is in the series, one for the lowest quality in the series, two for the second, three for the third, and so on.86 Two properties will then be placed at the same place in the quality order just in case that they have equal numbers of equal constituents.87 On phenomenological grounds the result must be condemned as absurd; it simply won't pass the empiricist test of PA.88 Moreover, it is ultimately inadequate, for, while order among properties is defined in terms of the order among the numbers which number the con-

158

Body, Mind and Self in Hume’s Critical Realism

stituents of these properties, the asymmetrical arithmetical relation remains unreduced to non-relations.89 So, ultimately we have to reject Hume’s doctrine on this point. For our present purposes, though, all that we need to recognize is the less controversial point that resemblances among things are founded on the properties that are present in things. Now, sensible impression A may resemble sensible impression B in one respect and C in another respect, where B and C resemble each other in neither respect. This means that there is a certain property in A and one in B by virtue of which A and B exactly resemble each other; similarly A and C will share a property; while B and C do not have any property in common. For this to be so, the things must have several properties. The example that Hume uses is the white globe. Here we have an object with two properties, that of being white and that of being globular. The problem that we are confronting is that Hume insists that these two properties are not distinct! But, on the one hand, if there is no distinction, then the impression is a simple entity, incapable of analysis. On the other hand, it does, it is clear, fall under two abstract ideas, that of white and that of globularity. Thus, an event or impression can have two different and separable abstract ideas apply to it even where the impression is simple: “’Tis certain that the mind would never have dream’d of distinguishing a figure from the body figur’d, as being in reality neither distinguishable, nor different, nor separable; did it not observe, that even in this simplicity there might be contain’d many different resemblances and relations” (T, p. 25). This presents, it seems to me, a real difficulty for Hume’s position. His nominalism requires him to found relations of resemblance on the properties of things. This in turn requires him to recognize a multiplicity of properties in things. The latter is also required by his empiricism, that is, PA. On the other hand, he also wants to hold that these objects which resemble each other in multiple ways to be simple. That means that relations of resemblance must after all be external, and not grounded in properties, contrary to the nominalistic account of relations. Thus, both his nominalism and his empiricism seem to force Hume to reject the doctrine that events and impressions are simple unanalyzable wholes. Why, then, does Hume adhere to this position? That is the question we have to deal with. But we also see that there are real difficulties in Mandelbaum using this example of the simplicity of impression and the inseparability of their properties to attack the Humean account of causation. What we have concluded about Hume’s doctrine of abstract ideas, or,

159

Two: Nominalism and Acquaintance

equivalently, his theory of concept formation, is that every abstract idea is an association grounded on a relation of resemblance. There are many problems with this theory of concept-formation. The abstract ideas with which Hume is primarily concerned are those that are based on the inseparable aspects of impressions and ideas, that is, on simple impressions and ideas. (Call these “simple abstract ideas.”) But it is clear that the discussion could, and, if the theory is to be fully worked out, must be extended to complex impressions and complex ideas.90 Even then, difficulties remain – e.g., how, really, are disjunctive concepts formed?91 For our purposes, however, these need not be pursued. These problems do not arise from the philosophically important point imposed by PA, that basic concepts (here: simple abstract ideas) are to designate only such aspects of things (here: resemblances among simple impressions) as are presented to one. The problems arise, rather, within the context of Hume’s research programme in introspective psychology. In its detail, about the role of images and so forth, Hume's theory of concept-formation is clearly part and parcel of the introspectionist programme of analyzing all thought into imagistic contents. Nor, in the end can we neglect the serious problems that Hume has with the relation of resemblance, as we shall note below. And finally, as we have insisted, Hume should include relations among the primitive or simple ideas, that is, in particular resemblances among ordered pairs of simple impressions. But none of these difficulties affect the substance of the point that we just emphasized about simple impressions falling under simple abstract ideas. In terms more appropriate to the later logical atomists, exactly the same point would be made by stating that no primitive predicates are to be admitted to the empiricist’s language unless they designate properties that are presented, where properties are simply those aspects of events by virtue of which the events resemble one another. With this rule of translation to help us, we can see that when Hume speaks of abstract ideas being distinct and separable, as in his argument about the status of the principle of causation, then what he is talking about are predicates or concepts being distinct in meaning and separable in the logical sense that one can consistently suppose one to hold of a sensible event while not supposing that the other does. Now consider once again the billiard-ball case. We have two billiard balls, say A and B. A displays one kind of motion and B another kind. Let us say that A displays motion1 and that B displays motion2. Just as the abstract idea of something beginning is distinct from and separable from the abstract idea of something being caused, so also, Hume holds (correctly),

160

Body, Mind and Self in Hume’s Critical Realism

the abstract idea of displaying motion1 is distinct and separable from the abstract idea of motion2. The separability of these ideas implies, Hume holds, that the A qua displaying motion1, is not necessarily connected to B qua displaying motion2. Mandelbaum challenges this inference of Hume as a transfer of principles properly holding only of ideas to impressions to which they do not apply. However, we now see that Mandelbaum’s challenge is mistaken, at least in the context of a metaphysics based on PA. The abstract idea of motion1 is, we just saw, basically a noticed way in which things resemble one another. Similarly, the abstract idea of motion2 is also basically a noticed way in which things resemble one another. The two abstract ideas are distinct just because they are different resemblance classes, that is, classes of (impressions and) ideas based on different resemblances, noticeable aspects of things which are noticeably different aspects. Or, in terms of our translation rule, the two abstract ideas or concepts designate properties that so far as our acquaintance with them is concerned are ontologically independent of each other. So far as the two resemblances are presented to us, we find in the motion1 way of resembling nothing that ties it necessarily to the motion2 way of resembling. The separability of the two abstract ideas is grounded in the two resemblances being noticeably different or distinct. Hume’s commitment to PA thus guarantees that the (logical) separability of two (abstract) ideas implies the nonexistence of a necessary connection between an event falling under the one and an event falling under the other – at least, no necessary connection so far as those two resemblances are concerned: there may of course be other aspects with respect to which the two events do turn out to be necessarily connected. Mandelbaum’s challenge to Hume is thus without substance. A point remains, however, to raise a doubt about the cogency of Hume’s position, and certainly always to give some credence to the sort of point Mandelbaum makes. What, it will be asked, of the inseparability of the shape and the colour of the white globe of marble that Hume used as an example when he discussed the distinction of reason? Does this inseparability not argue for the inseparability and therefore (via (P2)) the indistinguishability and therefore in turn (via (P1)) the non-distinctness of the resemblance-aspect we call “white” from the resemblance-aspect we call “globular”? But clearly the two abstract ideas are separable. At least, anyone’s view is quite plainly nonsense if these are not separable. Hence, if Hume’s account of abstract ideas is not to be nonsensical, then separability of ideas does not, after all, support the distinctness and logical separability of resemblance-aspects. In which case Mandelbaum’s objection remains. It

161

Two: Nominalism and Acquaintance

is not by accident that Mandelbaum to make his point chose Hume’s discussion of the “distinction of reason.” We must therefore return to a closer examination of Hume’s discussion of this topic. As we shall discover, perhaps not without surprise, Hume’s nominalism again enters the picture. What Hume does when he introduces the distinction of reason is use his own theory of abstract ideas to account for what others had said on the topic of the “distinction of reason.” What was the background to which he was responding? Hume himself mentions only the Port Royal Logic, that is, The Art of Thinking of Arnauld and Nicole (T, p. 43), but behind that is Descartes, and Descartes himself follows Suarez. Suarez considers two kinds of mental distinction.92 One, the distinction of reasoning reason (distictio rationis ratiocinantis), arises exclusively from reflection and the activity of the intellect, and has no foundation in things. Thus, Peter is distinguished from himself as term and subject of a relation when he is said to be the same as himself.93 The other sort of mental distinction, that of reasoned reason (distinctio rationis ratiocinatae), has a foundation in reality. Suarez tells us that “...this type of mental distinction can be understood as pre-existing in reality, prior to the discriminating operation of the mind, so as to be thought of as imposing itself, as it were, on the intellect, and to require the intellect only to recognize it, but not to constitute it. In this acceptation of the term the distinction would be called mental rather than real only because it is not so great, and in itself is not so evident, as a real distinction, and hence would need attentive inspection by the mind to discern it.”94 A little later he adds that ...[this] kind of mental distinction arises from inadequate concepts of one and the same thing. Although the same object is apprehended in each concept, the whole reality contained in the object is not adequately represented, nor is its entire essence and objective notion exhausted, by either of them....Hence such a distinction invariably has a foundation in fact, even though formally it will be said to spring from inadequate concepts of the same thing. Thus in God we do not conceive the sublimely simple virtue of God as it is in itself and according to the full range of its energy.95

Suarez goes on to argue that there are only three kinds of distinction, real, modal and mental.96 Descartes follows Suarez on these points,97 saying about the distinction of reason that it “...is between substance and some one of its attributes without which it is not possible that we should have a distinct knowledge

162

Body, Mind and Self in Hume’s Critical Realism

of, or between two such attributes of the same substance. This distinction is made manifest from the fact that we cannot have a clear and distinct idea of such a substance if we exclude from it such an attribute; or we cannot have a clear idea of the on of the two attributes if we separate from it the other.”98 This distinction of reason is, clearly, Suarez’s distinctio rationis ratiocinate, as Descartes in fact elsewhere confirms.99 Descartes notes a difference which he regards as of some importance between distinguishing and abstracting: “...in distinguishing a substance from its accidents, we must consider both one and the other, and this helps greatly in becoming acquainted with substance; whereas if instead one only separates by abstraction this substance from these accidents, i.e. if one considers it quite alone without thinking of them, that prevents one from knowing it well, because it is by its accidents that substance is manifested.”100 For the Art of Thinking (the Port Royal Logic), knowledge by abstraction is also knowledge by parts.101 One mode of knowledge by parts concerns things that possess several attributes. Between these attributes there is only a distinction of reason, but by virtue of that one can think of the one attribute without thinking of the other.102 The Logic gives this example: ...if I draw an equilateral triangle on a piece of paper and direct my attention to the place where the triangle is drawn and to all the accidental characteristics of the triangle, I shall have an idea of only that particular triangle. But if instead of considering all these particular circumstances, I consider only that what I have drawn is a figure bounded by three equal lines, then my idea will represent all other equilateral triangles. If I go even further, not stopping with the equality of lines, and consider only that what I have drawn in a figure determined by three straight lines, then my idea represents every kind of triangle. And then, if I go still further and consider only that what I have drawn is a plane surface bounded by straight lines, my idea represents all rectilinear figures. Thus, degree by degree I can ascend to an idea of extension itself. We see in these abstractions each lower degree is some particular determination of the higher degree....equilateral triangle [is] a determination of triangle; and triangle, of rectilinear figure. The higher degree being less determinate, stands for more things. Through such abstractions we move from the idea of a particular to a more general idea, then to an even more general idea, and so on.103

Elsewhere Arnauld remarks that “...the genus can be conceived without the species...Thus, I can conceive figure without conceiving any of the attributes proper to the circle.”104 Descartes agrees: “...there is always one principle property of substance which constitutes its nature and essence, and on which all others depend....we can conceive extension without figure or ac-

163

Two: Nominalism and Acquaintance

tion, and thinking without imagination or sensation....”105 Thus, there is a distinction of reason not merely between the attributes of a particular but also between genus and species. We conceive the genus without conceiving the species, and the attribute without the other attributes and without the substance. We do this by forming abstract ideas, universals. “Universals,” Descartes tells us, “arise solely from the fact that we avail ourselves of one and the same idea in order to think of all individual things which have a certain similitude; and when we comprehend under the same name all the objects represented by this idea, that name is universal.”106 For Suarez, for Descartes, and for Arnauld, a distinction of reason is a distinctio rationis ratiocinatae. But it requires, as Suarez says, “the discriminating operation of the mind.”107 The mind attends to things, and through this attending comes to recognize a distinction in things – recognize it but not constitute it, which is to say that it is a distinction with a foundation in the things. The thing in which the distinction is made may well be simple; the simple entity can nonetheless resemble other things in a variety of ways, and these resemblances are the foundation in reality of the distinction of reason among the properties of the thing. Thus, we can “...partition it [the sublimely simple virtue of God] into concepts [distinguishing His justice from His mercy]...by analogy with various virtues which we find distinct in man, but which in an ineffably eminent manner are found united in the absolutely simple virtue of God.”108 For the Port Royal Logic one attends to particular triangles and forms ideas which represent all triangles by neglecting the accidental characteristics of the particular, noting only what it has in common with other triangles. For Descartes and Arnauld, this process is one of forming an abstract idea. In major respects, Hume’s doctrine of abstract ideas agrees with all this. One forms an abstract idea only after one has occasion for noting similarities and differences. If we have no reason to classify then we won’t, and it can only be after comparisons are made that we will form abstract ideas. A white globe, if that alone were presented to us, would not lead us to form separate abstract ideas of colour and of figure. But when a black cube is also presented, then we begin to recognize resemblances, and thereby to form abstract ideas. Once we have the abstract ideas of colour and shape, we can say of the absolutely simple white globe that it is coloured and that it is shaped, distinguishing by reason, by means of our abstract ideas, two aspects of the globe, aspects which, since the globe is simple, really are inseparable in themselves. When we recognize that a thing falls under an abstract idea then,

164

Body, Mind and Self in Hume’s Critical Realism

says Descartes, we have only an inadequate understanding of it. Suarez makes the same point. In Cartesian terminology, adequate understanding of a thing is attained only when we have distinguished all its parts, that is, all the parts with respect to which abstract ideas could be formed. So an adequate understanding of a thing is possible only in terms of all abstract ideas that are or could be predicated of it. The being of a thing is given, according to the criterion (I), the Russell-Leibniz criterion of identity, in terms of all the properties truly predicable of that thing. Except, of course, we must remember that for these philosophers, all nominalists, every one of the predications is non-relational. And when the being of a thing is thus given by all the predication true of it, one has shown how it is distinguishable from all other things. Concerning this aspect of the doctrine of abstract ideas, Hume says nothing in the section of the Treatise (Bk. I, Part i, sec. 7) that deals with abstract ideas, though elsewhere (I, iii,1) he does assert that “Two objects, though perfectly resembling each other, and even appearing in the same place at different times, may be numerically different” (T, p. 69). But if Hume does not object to this part of the Cartesian doctrine of abstract ideas, there is another part that he vigorously criticizes. Descartes holds that “...all universals are simply modes of thought.”109 And the Port Royal Logic asserts the same when it asserts that “All existent things are singular.”110 Moreover, as we have seen both hold that species are separable in thought by abstraction from particulars, and genera from species. At the same time, these philosophers also hold that what is conceivable is possible, that is, that what is possible in thought is possible in the world. These three propositions are inconsistent. They constitute the core of Berkeley's critique of abstract ideas, as Weinberg has pointed out,111 and Hume also offers that critique of the “commonly inferr’d” (T, p. 18) theory. ...’tis a principle generally receiv’d in philosophy, that every thing in nature is individual, and that ’tis utterly absurd to suppose a triangle really existent, which has no precise proportion of sides and angles. If this therefore be absurd in fact and reality, it must also be absurd in idea; since nothing of which we can form a clear and distinct idea is absurd and impossible. But to form an idea of an object, and to form an idea simply is that same thing; the reference of the idea to an object being an extraneous denomination, of which in itself it bears no mark or character. Now 'tis impossible to form an idea of an object, that is possest of quantity and quality, and yet is possest of no precise degree of either; it follows, that there is an equal impossibility of forming an idea, that is not limited and confin’d in both these particulars. Abstract ideas are therefore in themselves individual, however they may become general in their representation (T,

165

Two: Nominalism and Acquaintance

pp. 19-20).

It is a virtue of Hume’s own account of abstract ideas that it does not generate this inconsistency. Upon Hume’s account, abstract ideas are to be understood in terms of resemblance classes of ideas and impressions. Clearly, the resemblance class of coloured objects will include the resemblance class of red objects. The abstract idea of a coloured object will not be separable from the abstract idea of a red object. The genus will, in other words, not be separable from the species; there will be a necessary connection between them. Similarly, every shaped entity will be the class of coloured entities, so the two ideas will be inseparable and there will be a necessary connection between being coloured and being shaped. A great many generalities commonly thought to be synthetic a priori thus turn out to be simple logical truths on Hume’s view. So does the basic Principle of Exemplification, that there are no (basic) properties that are not in some individual or other. Furthermore, since being an individual thing is the supreme genus under which some thing can fall, all species will be inseparable from that genus. Every species will have to be present in some individual or other: it will be impossible for there to be any thing but individuals. One cannot but admire the way in which Hume accounts for the necessity of such generalities as that everything shaped is coloured.112 Nonetheless, it is basically wrong. For what should be at issue is not whether every member of the resemblance class of coloured things contains every member of the resemblance class of shaped things, but whether these two ways of resembling are distinct. If they are, then they are separable, even if, as a matter of fact, everything shaped is coloured. Hume’s account goes through only if “distinct” applies only to ideas and impressions, and not to the aspects of these in terms of which they fall into resemblance classes. And this, I think, is wrong. We must, as we have seen, distinguish “distinguishablei” as applied to individuals and “distinguishablep” as applied to properties, the latter not implying the former. Nonetheless, Hume has reasons for thinking it does, reasons other than that of securing the status “necessary” for such propositions as those that everything coloured is shaped and that there are no properties that are not in an individual. But even in the absence of those other reasons, the desire to so dignify these propositions would provide a powerful motive for not applying ‘distinct’ to aspects. Let us recall the problem that has confronted us. We argued above that, for Hume, an abstract idea and a way of resembling go hand in hand.

166

Body, Mind and Self in Hume’s Critical Realism

The one implies the other, and it was this that permitted Hume quite correctly (and contrary to Mandelbaum) to apply the principles (P1) and (P2) to both ideas and impressions. The abstract idea of white is separable from the abstract idea of globular. There is therefore no necessary connection between this thing being white and this thing being globular. And to say that there is no such necessary connection is to say that the mode of resembling we call being white is distinct from the mode of resembling we call being globular. Nonetheless, Hume also holds that the white globe is absolutely simple. The white of this thing is therefore not distinct from the globularity of the thing. For, if it were it would be separable and the thing would not be absolutely simple. For this to make any sense at all requires that Hume draw a distinction between (1) the properties as such, and (2) the properties qua in the thing. The former are separable, while the latter are not. The properties as such constitute the foundations of the resemblance relations to which the abstract ideas correspond. These abstract ideas are separable, and therefore so are the properties as such. But qua in the thing the properties are not separable, but rather indistinguishable parts of the simple this. Let us for the moment take this distinction for granted, and let us call properties as such “properties1”and properties qua in a thing as “properties2.” In fact Hume has some grounds for making a distinction between properties qua resembling, properties1, and properties qua in things, properties2. After all, his nominalism requires him to hold that the properties of things are as particular as the things of which they are properties. That is, as we have said, Hume as a nominalist must construe properties as tropes. That means that the white in this and the white in that will exactly resemble each other and yet be intrinsically different. It therefore seems that Hume could hold that the abstract ideas of white and globular based on the relations of resemblance are separable while the tropes, the white in this and the globular in this, are inseparable. The first point to make is that the inseparability of the properties2, even if it be granted, does not in any way help Mandelbaum’s case against Hume. The question is that of the necessary causal connectedness of billiard ball A’s displaying motion1 with billiard ball B’s displaying motion2. In the field of one’s perceptions, the motion is continuous, the one part not separable from the other part. If this connectedness of the motion1 of A with the motion2 of B were a necessary causal connection – with all that we know that implies – then, as Aristotle recognized, the necessary truth of the generality connecting motion1 with motion2 would have to be discerni-

167

Two: Nominalism and Acquaintance

ble in the particular instance. As we put it above, we need the necessity of both the real definition (*) and the particular proposition (+) that ascribes the real essence to the individual. For, if we could not thus discern the necessity in the individual, the existence of the necessary connection will not permit one to overcome the Humean gap between sample and population. Mandelbaum’s case for necessary connections depends upon an affirmative answer to the question: is there anything about the motion1 of A that enables us to recognize that of necessity all motion1 is succeeded by motion2? If the tie between the motions is, as Mandelbaum takes it to be, something that resides in the inseparability of simple impressions into parts, then the answer is negative. For, precisely the same tie ties the two properties2 into the simple white globe, yet there is no necessity whatsoever that the two properties1 are regularly copresent. To the contrary: it is false that they are regularly copresent. So Mandelbaum cannot appeal to the Humean simplicity of impressions to argue against the Humean analysis of causation. There is a deeper point, however. The nominalistic account of properties as tropes is inconsistent with the nominalistic account of the relation of resemblance. The mode of resembling that is the basis of the abstract idea of white is different from the mode of resembling that is the basis of the abstract idea of globularity. Moreover, the mode of resembling that white is, is common to all pairs of things that exactly resemble each other in this respect. Otherwise they could not all fall under the same abstract idea. The relation of resemblance must, therefore, as Russell pointed out long ago,113 be taken to be the same for each pair of things that are said to resemble, and in this sense be a universal. This, of course, is compatible with construing properties as tropes. This and that exactly resemble by virtue of the white in this, white1, and the white in that, white2, and similarly two other things exactly resemble in the same respect by virtue of the two properties (tropes) that are in them, white3 and white4, and the resemblance in the two cases is exactly the same. But the nominalist Principle of Localization requires that this relation, objectively considered, be reduced to its foundations. So, consider this and that which are both white. The foundations of the relation of exact resemblance between these things is the property white, the trope, in this and the property white, another trope, in that. But the relation of resemblance is symmetrical: if x resembles y then y resembles x. Consider the fact that x resembles y, and let r1 found the relation in x and r2 found it in y. Since y also resembles x and since the resemblance is the same in the two facts, it follows that r1 must found that relation in y and r2 must found it in x. But of course it is clear that when we

168

Body, Mind and Self in Hume’s Critical Realism

observe x we do not perceive two properties, r1 and r2, founding the relation of exactly resembling with respect to white; there is but one property. So r1 and r2 must in fact be the same. But in that case the property in the one particular that founds the relation of resemblance must be the same as the property in the other particular that founds that relation. In other words, the nominalist account of the relation of resemblance as analyzable into foundations requires that the properties in things that found the relation be universals. The nominalist account of relations requires one to abandon the nominalist claim that properties are tropes. In the end, therefore, Hume’s nominalism gives no support for the distinction between properties1 and properties2: properties, those entities that found the relations of similarities, must be understood as universals. In that case, what of the distinction, that Hume seems to make, between properties1 and the properties2: just how do they differ? In particular, why are the former logically separable and the latter not? The mode of resemblance that white is, is different from the mode of resemblance that globularity is. The mode of resemblance that white is, a property1, is common to each member of all pairs of things that exactly resemble each other in this respect. Let us suppose that P and Q are two impressions which resemble each other in being white. We have the mode of resemblance that white is, i.e., the property1, that is in each member of the pair, and then besides this the two properties2, the mode of resemblance that white is qua in P and the mode of resemblance that white is qua in Q. But this surely is a multiplication of entities beyond necessity. When we look at the way in which P resembles Q, there is the property1 white in P and then it is supposed that there is a further entity, the property2, that is an indistinguishable part of P. However, when we mention the property2 we seem to be referring not to a further entity but rather only saying about the same entity that it is present in P. When we have a painting that is present in the gallery, there is not the painting and then a further entity, the painting qua present in the gallery; there is, rather, the painting and the fact about the painting that it is present in the gallery. Similarly, if John owns a pair of pants and a coat, we can speak of the pants qua owned by John and the coat qua owned by John, but these are not further entities and moreover the pants and the coat do not cease to be separable simply because it is true of both of them that they are owned by John. The facts of ownership do not transform them into inseparable entities. Similarly, the facts of ownership of the white and the globularity do not make those two properties somehow inseparable. I conclude that if Hume holds that two abstract ideas in his sense are logically separa-

169

Two: Nominalism and Acquaintance

ble, and that, when they are, this separability is due to their being based on different modes of resemblance, then he ought to hold that those modes qua properties of impressions and ideas are separable. But Hume does not hold this. It is therefore an error. To be sure, it is not an error that vitiates his account of causation. But since it is an error it certainly does not support the case Mandelbaum makes against Hume's view of causation. There remains the question, however, of how Hume could fall into this error of ontology. We should try to understand this. Just what is it to say that the white of the globe is separable from the globularity? Simply that it is logically possible for that same white to be combined with some other shape, say cubicity. In Hume's terms that means that we can form the idea (= image) of a white cube. We can re-combine the properties founding resemblances given in impressions to form ideas of things of which no impression has ever been had. No doubt if we are correct this is just what Hume should say. Yet this picture of how the imagination works makes it difficult indeed for him to say it. His picture seems to be similar to one in which a person takes two chairs apart, separating all the parts from each, then selecting some that belonged to the one chair, some that belonged to the other, and finally assembling these selected parts into a new chair, one that is quite unlike either of the originals. Thus, he tells us that “The imagination has the command over all its ideas, and can join, and mix, and vary them in all ways possible” (T, p. 629). And earlier he has proposed the principle “...of the liberty of the imagination to transpose and change its ideas. The fables we meet with in poems and romances put this entirely out of the question. Nature there is totally confounded, and nothing mentioned but winged horses, fiery dragons, and monstrous giants” (T, p. 10; his italics). The point is clear. If this is Hume’s dominant picture then it is clear why he would want to hold the imagination cannot separate the white of the globe to combine it with other aspects of resemblance to form the idea of a white cube. For, his picture of the imagination would leave him detaching the white quite literally, as the wing can be separated from the bird, and combined with another idea, as the idea of a wing can be joined to the idea of a horse. And that would mean that white would be existing as such as an image, quite apart from any shape. This, of course, is quite impossible – it is not possible in reality, so it is not possible in thought –, and Hume is therefore committed by his misleading picture of the imagination to the thesis that the white of the globe and its shape are quite inseparable.

170

Body, Mind and Self in Hume’s Critical Realism

I think that this point is correct. But it is not the whole story. In particular it leaves us in the rather unsatisfactory state of attributing to Hume a hidden ambiguity in ‘separable’. The accusation is that he does not notice that once he takes it to mean “logically separable” and once to mean “separable by the workings of the mechanical imagination.” Some have been content to locate such an ambiguity in Hume,114 but ‘separable’ is a term of art and it is unlikely that Hume would have been mistakenly confused in its use in the fairly simple way that I have just suggested. In this context it will pay to note another aspect of Hume’s ontological position, namely, his commitment to the idea that existence and unity are one. Thus, ...existence itself belongs only to unity, and is never applicable to number [here meaning “plurality”], but on account of the unity of which the number [“plurality”] is composed....It is therefore utterly absurd to suppose any number to exist, and yet deny the existence of unites....But the unity, which can exist alone, and whose existence is necessary to that of all number, is...perfectly indivisible, and incapable of being resolved into any lesser unity (T, pp. 30-31).

Hume was not the only one of his age to adopt this principle. Leibniz, for example, argues that “where there are only entities by aggregation, there will be no real entities. What is not truly one entity, is not truly an entity.”115 This is the old maxim that ens et unum convenuntur, which goes back at least as far as Boethius’ Commentary on Porphyry’s Isagoge: For anything that is common at one time to many can not be one; indeed, that which is common in many, particularly when one and the same thing is completely in many things at one time. Howsoever many species indeed there are, there is one genus in them all, not that the individual species share, as it were, some part of it, but each of them has at one time the whole genus. It follows from this that the whole genus, placed at one time in many individuals, can not be one; nor in fact can it happen that, since it is wholly in many at one time, it be one in number itself. But if this is so, no genus can possibly be one, from which it follows that it is absolutely nothing; for everything that is, is because it is one.116

Ockham, too, argued against the realist position on the basis of the premise: ...if such a thing as a universal thing were assumed, it would be one in number since it is not multiplied in many singulars.117

171

Two: Nominalism and Acquaintance

And Ockham ends up defending an account much like Hume’s: It should not be conceded that Socrates and Plato are alike in something or in some things; but it ought to be conceded that they are alike by means of some things, because they are alike by means of their very own selves.118

Or elsewhere: Socrates and Plato are more alike than Socrates and an ass....But it suffices that Socrates and Plato are more alike simply by means of their very own selves [and not in virtue of a common nature].119

This is hardly different from Hume’s statement that ’Tis evident, that even different simple ideas may have a similarity or resemblance to each other; nor is it necessary, that the point or circumstance of resemblance shou’d be distinct or separable from that in which they differ. Blue and green are different simple ideas, but are more resembling than blue and scarlet; tho’ their perfect simplicity excludes all possibility of separation or distinction. ’Tis the same case with particular sounds, and tastes and smells. These admit of infinite resemblances upon the general appearance and comparisons, without having any common circumstance the same...’Tis the same case with all degrees in any quality. They are all resembling, and yet the quality, in any individual, is not distinct from the degree (T, p. 637).

Hume is at least heir to a long tradition. No doubt this alone can partially account for his holding to be inseparable the various points of resemblance that are in any individual. Even so, mere tradition is not enough. There should be in that tradition a pervasive, if not persuasive, philosophical thesis, the adoption of which by Hume could have led to his mistaken views about the separability of the colour and the shape of the white globe. There is such a philosophical thesis: nominalism. Recall that in discussing (P1), the thesis that whatever is distinct is distinguishable, we had to distinguish two senses of ‘distinguishable’: we said that individuals were distinguishablei and properties were distinguishablep. Presupposing the Principle (or Dogma) of Localization, the nominalists construe ‘distinguishable’ throughout as distinguishablei, and infer that where there are distinct predications (“white is in this and white is in that”) then there are distinct individuals (the white in this is an individual distinguishablei from the individual that is the white in that). The white in the white globe and

172

Body, Mind and Self in Hume’s Critical Realism

the globularity in the globe will turn out to be distinct and distinguishable individuals on this doctrine. But the Principle of Localization will put pressure on the nominalist to deny even this. For, to be two actual individuals, the things must be located at two different places. To be separate individuals is to be spatially separate individuals. The white globe, however, is manifestly not composed of a colour that is spatially separate from the globularity. It can therefore only be that the white and the globularity of the globe are not distinguishablei nor, therefore, distinct: the globe is one. Or so the nominalist. Of course, once we distinguish “distinguishablei’ from ‘distinguishablep’, the empiricist can easily maintain, what Hume ought to have maintained, that the colour and the shape of the globe are distinguishablep but that from this it does not follow that they are distinguishablei, nor that the globe is a simple whole somehow devoid of distinguishable parts. Once one acknowledges that things (events, impressions) have properties as parts that are separable, even if one takes them to be tropes, then it is necessary to account for the unity of the thing. One must allow that the properties are with each other or are copresent. These of course are relations.120 That is, the unity of the thing is matter of its properties being related in a special way one to another. This is the solution adopted by later thinkers like Russell121 and Bergmann.122 Since what we are presented with are complexes in which properties clearly do stand unified into wholes by such a relation, this solution is, on the one hand, one that is sanctioned by the empiricist Principle of Acquaintance.123 On the other hand, this is not a solution that could be used by the nominalist, who insists that all relations must be analyzable into foundations. As a consequence, a relation provides no objective unity; a relational fact is, objectively, only a conjunction of non-relational facts. Such unity as there is, is provided by the subjective relational judgement, but this is not an objective unity in the thing.124 Upon the nominalist account of relations, then, the solution of the empiricist to the problem of the unity of the thing is not available. The Aristotelians did have a solution of sorts to this problem, a way in which they could allow a thing or substance to have present in it two properties, that is, tropes, which are different but nonetheless inseparable. Consider an object a which has, as in (+), the real essence A and suppose further that (*) is the real definition of A. Finally, suppose that a is in situation P. Since a is A it follows by (*) that a as in P will also be Q. The necessity of (*) provides a necessary connection between A, P and Q: something’s being A will necessarily tie its being P to its being Q. But (+) is also necessary. There is therefore an objective necessary connection be-

173

Two: Nominalism and Acquaintance

tween A and the two properties P and Q that are present in it. The two different properties are in fact qua in a inseparable from each other and from the thing a in which they are both present. This is, of course, the position that Hume adopted as part of his nominalism: the parts of a thing (event, impression) are inseparable from each other. But where the Aristotelians had grounds for holding that this is true, Hume’s empiricism had removed those grounds when it had eliminated objective necessary connections. At the same time, his nominalism prevented him from acknowledging the objective existence of irreducible relational facts. He could not therefore offer an alternative empiricist account of the unity of things. He was left in the untenable position in which he wanted to hold to the traditional view that things are simple wholes while accepting an empiricism that denied the traditional basis for the claim and a nominalism that denied the solution that later empiricists were to give. The moral to be drawn once again is that the nominalism in Hume’s philosophy leads him to adopt a position that creates problems for his empiricism. Those who accept the latter would do well to reject the former. At the same time we should also note once again that, even if we adopt Hume’s nominalistic view on the unity of things (events, impressions), this, contrary to Mandelbaum, does not give us grounds for rejecting the Humean account of causation. * Hume adopted both empiricism and nominalism. The former is based on the Principle of Acquaintance (PA), while the latter is based on the principles (P1) and (P2), understood as (NP1) and (NP2), and on the Principle of Localization. He used arguments from both these positions to support his views on causation. We have found that the nominalism leads to doctrines and arguments that are incompatible with the empiricism, particularly on the issue of relations. But we have also found that the Humean view of causation does not depend upon the nominalistic arguments; it can be defended on empiricist grounds alone. Moreover, since the nominalism is incompatible with empiricism, the empiricist would do best to reject the nominalism. But in any case we have also discovered that there are real problems within the nominalism that Hume adopts. It requires on the one hand that properties be construed as tropes and on the other hand that relations be so analyzed that properties must be construed as universals. One

174

Body, Mind and Self in Hume’s Critical Realism

cannot have it both ways. Moreover, problems about the unity of things that arise given Hume’s wedding of empiricism and nominalism do not arise if one rejects the nominalism. The moral of the story is: a good empiricist is not a nominalist. Once one rejects the nominalism, then there is no basis for the empiricist to reject relations. This means in particular that the empiricist can allow that there are really are relations that structure the bundle of impressions that is the non-substantialist self that alone can be admitted into the empiricist ontology. G. Rees, let us remember from the preceding chapter, in his memoir, recollected Bradley’s retort to the psychologist Bain: “Mr. Bain collects that the mind is a collection. Has he ever thought who collects Mr. Bain?” which was taken by Bradley and by Rees’ tutor H. W. B. Joseph to suffice as a refutation of the Humean views of the mind and of the self. But we now see that one can after all collect the parts of the bundle into a whole provided one has an adequate account of relations. We now see that, while Hume’s account of relations is ontologically inadequate, that is the consequence of his nominalism and not of his empiricism based on the Principle of Acquaintance (PA). Hume’s empiricism thus does not preclude his developing an adequate account of the self. But before turning to that topic of the mind and the self it will pay to see how Hume works out the bundle view of ordinary objects. After all, this is Hume’s account of body, that is, bodies in the ordinary sense of material objects, and if we are to make sense of Hume on minds then we cannot avoid discussing the mind-body relation. So the topic of the next chapter (Chapter Three), will be Hume’s account of the objects of the so-called external world. Only when we have dealt with this relatively easy case will we return (in Chapter Four) to the more difficult topic of the identity of the self. One needs relations for an adequate account of a Humean “bundle” view of the self. That makes it important to show that one can defend such an account while also accepting Hume’s empiricism, that is, an ontology based on PA. Hume’s nominalism gets in the way of such an account: the nominalism is incompatible with the needed account of relations. Our argument has been that one can reject the nominalism and the consequent denial of relations while accepting PA and providing the needed account of relations. But it is also important to defend Hume’s account of causation. For with the regularity view many of the problems that confront many discussions of the mind-body relation turn on those discussions taking for

175

Two: Nominalism and Acquaintance

granted a non-Humean account of causation. Hume was clear that if one accepted his view of causation, then there was no problem with regard to the possibility of mind affecting body and body affecting mind. For the Cartesians, the world consists of substances. These are of two kinds or essences: there are substances the essence of which is extension (or perhaps just one such substance), and there are substances the essence of which is thought. God constitutes a third kind of substance, but He or She or It is clearly a special case. Now, the Cartesians accept the argument that causation occurs only where there is a necessary connection between cause and effect. But one can discern no such necessary connection between something extended and something mental, and since one can discern no such connection, there is no such – God has not created us to be deceived on matters clear and distinct. So there can be no causal connection between mind and body or body and mind. When I will that my arm go up, there is such a movement, but the former does not cause the latter. And when a body external to our own affects our own sense organs, the animal spirits in our body are caused to effect changes in the brain, and consequent upon these changes (they occur in pituitary gland, Descartes thought) I experience a sensible impression and have the mental experience of perceiving the object that so affected my senses that I have that impression. But there is no necessary connection between mind and body, and so we cannot say that the object perceived caused the sensible impression and perceiving of that object. There are, to be sure, regularities between mental events and bodily events: in the presence of a tree in normal light and when my eyes are open then regularly I have a sensible impression of the tree and regularly I perceive the tree; and similarly, regularly when I will my arm to rise then it does rise. But regularity is not causation. That is why Descartes is unable to say that from the existence of the appropriate mental events he knows that there are bodily events that cause those events. It is his firm belief that there are such objects corresponding to the mental events, but a belief, however firm, is not guaranteed by the awareness of a necessary connection and is therefore not knowledge. So Descartes cannot demonstrate the existence of body – demonstration requires necessary connections. He does argue, however, that just as God in his veracity guarantees the truth of clear and distinct ideas, so such a God will guarantee the truth of my inescapable belief that body exists. However, Simon Foucher argued that it is our firm and inescapable belief that the objects we perceive have colours. But, if the essence of body is extension, and extension alone, then bodies cannot be coloured. Now, if God

176

Body, Mind and Self in Hume’s Critical Realism

guarantees our firm beliefs, then he guarantees our firm belief that bodies are coloured. But then Descartes has proved too much, since, he also argues, bodies cannot be coloured. At the same time, if all God does is guarantee the existence of body as an extended thing, but not our belief that it is coloured, then God’s veracity is not after all to be trusted in the case of belief, and we have no grounds for thinking that there are objects corresponding to our impressions of objects. So, if Descartes hasn’t proved too much then he has proved too little. Foucher’s critique is devastating.125 That is not, however, our present concern, though it does show the sorts of problems that arise when one adopts an ontology in which our knowledge of causes depends upon our perceiving or intuiting objective necessary connections. The problem is not that mind and body are different kinds, as the myth would have it, but with the requirement that causation involves objective necessary connections and that our knowing causal relations to obtain requires that we have rational intuitions of such connections. And as Hume makes clear, there are not such connections, and we have no such intuitions. So, given that this is our experience of the world, causation remains utterly mysterious on the Cartesian view. One removes the mystery by adopting Hume’s regularity view of causation. And in particular, it turns out that there is nothing mysterious about mind-body interaction. “They first raise a dust and then complain that they cannot see.” For the Cartesians there are regular connections between mind and body and between body and mind, but these are mere regularities, not genuinely causal. The mental event is the occasion of but not the cause of the corresponding bodily event, and the bodily event is the occasion of but not the cause of the corresponding mental event. This occasionalism satisfied no one but the Cartesians – they convinced no one but themselves, and even there, there is something about it that makes them uncomfortable: Descartes himself is relatively coy about it, never coming right out and saying that that is his position. But it is. Malbranche was more open. He does explicitly hold the occasionalist position, but holds that mental events and bodily events are caused: they are caused by God himself (or herself or itself) so acting that the regular patterns that the occasionalism requires, while not themselves causal, are patterns that really are regular, and therefore apparently causal. But the appeal to God hardly makes the position philosophically less mysterious. Why not just drop God and make the obvious move: we are not presented with any objective necessary connections and therefore causality is, it turns out, nothing more than regularity – we experience regularity between mind body and they therefore do interact

177

Two: Nominalism and Acquaintance

causally, just as commonsense demands – the mystery has disappeared. Hume is perfectly clear that his on causation does eliminate the mystery. It comes at the cost of eliminating God as a causal force in the world, but that is not at all bad. If, as Malebranche holds, I am not the real cause of my bodily motions consequent upon my mental acts, then the view leads to the “grossest impieties and absurdities” (T, p. 249); for, upon this view God is the real source the good I do, but also the source of the bad: “If nothing be active but what has apparent power, thought is in no case more active than matter; and if this inactivity must make us have recourse to a deity, the supreme being is the real cause of all our actions, bad as well as good, vicious as well as virtue.” (T, p. 249) But he is also serious: the move is philosophically indefensible. Occasionalism is adopted because we are not presented with any productive power or necessary causal connection in things of which we have an impression; but in that case, since all our ideas are either innate or derive from impressions, and since Locke has shown there are no innate ideas, then we have no idea of any necessary causal connection, and in particular “we have no idea of a being endow’d with any power, much less one endow’d with infinite power.” (T, p. 248) If the claim that billiard balls are active is nonsense, then the notion that God can supply the required activity is just as much nonsense. The notion of a necessary connection is rejected because we cannot imagine a necessary connection such that body as extended substance can affect mind which is not extended. ’Tis absurd to imagine, that motion in a circle, for instance, should be nothing but motion in a circle; while motion in another direction, as in an ellipse, shou’d also be a passion or moral reflexion: That the shocking of two globular particles should become a sensation of pain, and that the meeting of two triangular ones shou’d afford a pleasure. (T, p. 246)

We are reduced to the position that “all objects, which are found to be constantly conjoin’d, are upon that account only to be regarded as causes and effects.” (T, p. 249) It follows that “... as the constant conjunction of objects constitutes the very essence of cause and effect, matter and motion may often be regarded as the causes of thought, as far as we have any notion of that relation.” (T, p. 250) The mystery is gone: mind and body causally interact. Hume is perfectly clear this taking the mystery out of the mind-body connection is one of the consequence of accepting his regularity view of causation

178

Body, Mind and Self in Hume’s Critical Realism

This virtue of Hume’s account of causation for any sensible discussion of the mind-body relation was early seen by Scottish thinkers. Hume’s view of causation removes the mystery from it, and from the mind-body relation. It enables investigators to get on with the task of empirical research rather worry about how to decide metaphysically where causal connections really lie. Here is one example of a scientific researcher who recognized this point: William Cullen, the Scots physician who was an influential teacher of medicine at the University of Edinburgh for many years.126 In his Institutions of Medicine,127 Cullen makes clear the outlines of his physiology. The impulses of external bodies act on the ends of the nerves. These generate a process which culminates in those events in the mind which we call sensation, and, more generally, perception. These sensations in turn give rise to volition, or the willing of certain ends to be obtained though a process beginning with motions of the body. Volitions then give rise to changes in the muscles which in turn produce the relevant bodily motion (pp. 23-4). As a practising physician who is concerned about the effects of the body on the mind, e.g., in determining both human temperaments and idiosyncrasies (p. 214ff), Cullen takes it for granted that there are causal connections that go from body to mind and from mind to body, and that it is incumbent upon the physician to take note of them, and rely upon them in his diagnoses and interventions. ...[the] immaterial and thinking part of man is so connected with the material and corporeal part of him, and particularly with the nervous system, that motions excited in this give occasion to thought; and thought, however occasioned, gives occasion to new motions in the nervous system. This mutual communication of influence we assume with confidence as fact (pp. 17-18).

Interaction of mind and body is common sense. However, he goes on, “the mode of it we do not understand, nor pretend to explain...” (p. 18). To be sure, philosophers have advanced their opinions, but these all turn out to be totally irrelevant to the concerns of the physician. The system of Aristotle, which is adopted by many moderns, is called the system of the physical influx. Des Cartes has proposed another, which is known by the name of the system of occasional causes; and Leibnitz has proposed a third, which supposes the existence of a pre-established harmony. You may consult these, but I do not say a word of them, because when I have considered them as well as I can, I cannot perceive that they have the least effect or influence in explaining any thing: they do not admit of any application, either in physic or in

179

Two: Nominalism and Acquaintance

any other part of science, that I see. With regard to the mode of mutual influence, it will be allowed that as you adopt one or other of these hypotheses, it may affect religious belief, but it can have no effect in physic, so that it is not my province to consider them (pp. 18-19).

For Cullen’s purposes it suffices to note constant conjunctions: “...the soul and body are distinct substances; but they are so united, that thought is constantly connected with certain states of the body, and, on the contrary, that these states produce a certain state of thought” (pp. 20-1). Once we have recognized this, we can get on with the task of the physician: ...the mutual influence of [body and soul] being supposed and granted, physicians have differed very much with respect to the degree and extent of that influence, and therefore I am obliged to take some notice of these opinions...(p. 19)

With regard to “...these...different opinions with respect to the share which the soul has in the action of the economy [of the system of mind and body],” Cullen proposes to “give my reasons for rejecting some of these and admitting others” (p. 22). Hume’s view of causation, sceptical though it may in a way be, does permit the scientist to get on with the task of finding causes.128 Our concern here, of course, as was Cullen’s, is with the mind-body relation. But there were other researchers in other sciences, e.g., physics, who equally defended the Humean account of causation as the one appropriate to the discussion of natural laws, that is, laws in the sense in which empirical natural science speaks of laws. Among the important scientist who agreed with Cullen’s position on the utility of Hume’s views on causation was John Robison, who was elected to the professorship of natural philosophy at the University of Edinburgh in 1773. Robison is important for being among those who changed Baconianism from “a blind veneration into a rational worship”, to use the phrase of Henry Hallam,129 mainly through his articles in the Encyclopaedia Britannica. In his contribution on “Philosophy,”130 he tells the reader that “causes are no more cognoscible by our intellectual powers than colours by a man born blind: nay, whoever will be at pains to consider this matter agreeably to the received rules of logic, will find that necessary connection, or the bond of causation, can no more be the subject of philosophical discussion by man, than the ultimate nature of truth” (p. 587 a), and that we “know ... nothing of cause and effect but the conjunction of two events” (p. 593 b). In his article on “Phys-

180

Body, Mind and Self in Hume’s Critical Realism

ics,”131 he tells the reader that “the principle which connects the pairs of concomitant events, rendering the one the inseparable companion of the other, is totally unknown to us, because it is not the immediate object of our perception” (p. 649 b). His article on “Dynamics,” reprinted in his System of Mechanical Philosophy,132 he tells us that “...the powers or forces, of which we speak so much, are never the objects of our perception” (vol. i, p. 33). As for metaphysical hypotheses about entities that are “thrust in between two which we really observe to be united in nature” and that supposedly supply the necessary connections, “it is impossible that they can given any addition of knowledge” (“Philosophy,” p. 594 b). Nor is there any need for such hypotheses. As Robison explains in his article on “Corpuscular Forces,”133 once constant conjunction is accepted as the way causation is, then there is no reason to exclude action at a distance as a case of causation, nor, therefore, any need for hypotheses that try to tie closer together these phenomena in which one objects acts, regularly, at a distance from another. Separate those objects may be, and therefore separable, but for all that causally connected. ...it is most probably that our attempts to explain it [cohesion], that is, to deduce it from more simple or original principles, so as to be able to state cohesion as an effect, will be as fruitless as they have been in the case of gravity, and that it will be more prudent in us to abstain from the research, and content ourselves with collecting the laws by which the operations of this force, whatever it is, are regulated (pp. 210-11).

Attempting to go beyond observed constant conjunction is not only fruitless but contrary to any prudent pursuit of scientific knowledge. Hume’s view of causation is thus useful to researchers in science as a tool to justify their getting on with the task of empirical research and ignoring the metaphysical disputes of the past. It is therefore necessary to try, as we have tried, to separate Hume’s argument for the regularity view that is based on his nominalism from his other argument that is based on his empiricist appeal to PA. To obtain an adequate account of relations, we have argued, one must give up the nominalism. But this is to give up one of Hume’s arguments for the regularity account of causation. But, as we have also seen, giving up the nominalism does not require one to give up the regularity account of causation, the account that takes that mystery our of causation that is consequent upon the view that causation requires some sort of non-empirical but objective necessary connection. For, Hume has a second argument for the regularity

181

Two: Nominalism and Acquaintance

view, one that is based on his empiricist principles rather than on his nominalism. Hume’s empiricism enables one to have an adequate account of relations while also rejecting the Aristotelian rationalist objective necessary connections and accepting instead as the more reasonable the regularity account that Hume proposes and defends. * Our aim remains, of course, to examine and defend Hume’s “bundle” view of mind. An adequate account of relations and the regularity account of causation are needed, it has been suggested, if any defence is to be plausible. But there is another issue with which we need to deal before turning to Hume’s account of the mental. This is the fact that human minds are incarnate. One therefore, in discussing Humean minds, must talk about bodies, and we need to become clear on what exactly is Hume’s view on body. So, before taking up the discussion of Humean minds, let us first turn to Hume’s account of body. This is the topic of the next chapter. Perhaps it will not surprise one to know that we will conclude that Hume’s view is empiricist and pretty commonsensical, as Cullen and Robison suggested about the Humean account of causation. Certainly we should not be surprised that Hume’s account of body is not the sceptical position that is usually attributed to him.

Endnotes to Chapter Two

1 J. Butler, “Dissertation on Personal Identity,” pp. 280-81. 2 This view has more recently been defended by David Wiggins, Sameness and Substance. 3 In Spinoza’s terms, the “real essence” is the natura naturans and the definitional essence is the natura naturata. See F. Wilson, Socrates, Lucretius, Camus, Chapter Four. 4 R. Harré and E. H. Madden, Causal Powers. 5 W. Edwards, “Foreword” to S. P. Lamprecht, The Metaphysics of Naturalism, p. vii. 6G. Bergmann, Philosophy of Science, Ch. 1.182

182

Body, Mind and Self in Hume’s Critical Realism

7Cf. F. Wilson, “Hume’s Defence of Science,” and “Hume’s Cognitive Stoicism,” 8For a detailed discussion of this defence, see F. Wilson, “Hume’s Theory of Mental Activity,” and “Hume’s Defence of Causal Inference.” 9For a detailed discussion of Hume’s case, cf. F. Wilson, Hume’s Defence of Causal Inference. 10Cf. F. Wilson, “Dispositions: Defined or Reduced?” 11I have defended this Humean analysis of dispositions elsewhere in some detail. See “The Notion of Logical Necessity in the Later Philosophy of R. Carnap”; “Dispositions: Defined or Reduced?”; “Definition and Discovery”; “A Note on Operationism”; “Is Operationism Unjust to Temperature?”; “Dispositions Defined: Harré and Madden on Analyzing Disposition Concepts”; “Addis on Defining Disposition Concepts”; and Review of E. Prior, Dispositions. 12R. Descartes, Meditations, Haldane and Ross, vol. I, p. 152; cf. Principles of Philosophy, Bk. I, prop. 8, Haldane and Ross, vol. I, p. 221. 13Haldane and Ross, vol. II, p. 82. 14A. Arnauld and P. Nicole, The Art of Thinking, trans, J. Dickoff and P. James, p. 50. 15Descartes, Principles, I, 11, Haldane and Ross, vol. I, p 223. 16At least, this is what Descartes believes that the cogito gives him. 17Berkeley presents this attack in the “Introduction” to his Principles of Human Knowledge. Hume repeats the same critique in the Treatise, p. 18-19. For a general discussion of this argument, cf. J. Weinberg, “The Nominalism of Berkeley and Hume.” 18Descartes, Meditations, Haldane and Ross, vol. I, p. 152. 19Haldane and Ross, vol. I, p. 319. 20Descartes, Meditation VI, Haldane and Ross, vol. I, p. 190. 21Descartes to P. Gibieuf, 19 Jan. 1642, in Descartes, Oeuvres et Lettres, p. 1142. 22Cf. Weinberg, Abstraction, Relation, and Induction, p. 13ff. 23In the original, the entire passage is in italics, save for the ‘in other words.’ 24 Cf. J. Weinberg, Ockham, Descartes, Hume, p. 138.

183

Two: Nominalism and Acquaintance

25For more on this ontology and its history, see F. Wilson, “Placing Bergmann.” 26Weinberg, Abstraction, Relation, and Induction, p. 54. 27Ockham, Sententiarum, i, d. 2, 2, q. 4, D, and q. 1, D; cited in Weinberg, Abstraction, Relation and Induction, p. 45. 28Aristotle, Topics, vii, 1, 1, 152a33-36. 29T. Hobbes, Elements, Part II, Ch. 11, sec. 1, in English Works, ed. W. Molesworth ,vol. I, p. 132. 30F. Suarez, On the Various Kinds of Distinctions, trans. C. Vollert, p. 40. 31Ibid., p. 46. 32Aristotle, Topics, vii, 1, 152b34-5. 33Descartes, Meditations, Haldane and Ross, vol. I, p. 190. 34Weinberg, Abstraction, Relation, and Induction, p. 41-56. 35Cf. ibid., p. 54ff. 36Ibid., p. 56. 37This assumes the Axiom of Extensionality. If one rejects it, then one requires in the definition of ‘identity’ for properties, that for two properties to be identical they must be both true of the same individuals (co-extensive) and also every second order property true of the one must be true of the other. Extensionality implies that coextensivity implies identity in this stronger sense. 38Cf. E. A. Moody, The Logic of William of Ockham, p. 79f, p. 162f. 39Cf. G. E. Moore’s discussion of G. F. Stout’s position: “Are the Characteristics of Particular Things Universal or Particular?” 40Cf. G. Bergmann, “Inclusion, Exemplification, and Inherence in G. E. Moore.” 41For an extended defence of the Humean position, cf. F. Wilson, Explanation, Causation and Deduction and Laws and Other Worlds. 42Cf. F. Wilson, “Weinberg’s Refutation of Nominalism,” for details on some of the troubles that one runs into when one attempts to analyze relational into non-relational predicates.

184

Body, Mind and Self in Hume’s Critical Realism

43Cf. C. I. Lewis and C. H. Langford, Symbolic Logic, Second Edition, p. 387-88. 44Cf. Weinberg, Ockham, Descartes and Hume, p. 130ff. 45It should not really surprise one that relational statements cannot be reduced to nonrelational. If they could then first order logic could be reduced to monadic predicate logic. But the latter is decidable while the former is not. So relations, terms which take two (or more) individuals as their subject, cannot be reduced to terms which take only one individual as their subject. 46 Cf. Weinberg, Abstraction, Relation, and Induction, p. 61ff. 47Ibid., p. 86ff. 48Cf. Leibniz, Fifth Letter to Clarke, sec. 42; New Essays, Bk. II, ch. 12, sec. 7. 49Cf. R. Grossmann, “The Dogman of Localization.” 50S. P. Lamprecht, The Metaphysics of Naturalism, p. 131. 51M. Mandelbaum, The Anatomy of Historical Knowledge, p. 56-7. 52Ibid., p. 57. 53Ibid., p. 58. 54G. Bergmann, Philosophy of Science, Ch. 2. 55B. Russell, “On the Notion of Cause.” 56D. Hume, Enquiries concerning Human Understanding and the Principles of Morals, ed. L. A. Selby-Bigge, Second Edition, p. 74. 57Lamprecht, The Metaphysics of Naturalism, p. 139. 58W. Edwards, “Foreword” to Lamprecht, The Metaphysics of Naturalism, p. vii. 59Mandelbaum, The Anatomy of History, p. 60. 60Ibid. 61Lamprecht, The Metaphysics of Naturalism, p. 131. 62Ibid., p. 142-43. 63Cf. F. Wilson, “Hume and Ducasse on Causal Inference from a Single Experiment.”

185

Two: Nominalism and Acquaintance

64Cf. Wilson, “Dispositions: Defined or Reduced?” 65Harré and Madden, Causal Powers. For discussion of other aspects of this work, cf. F. Wilson, “Dispositions Defined: Harré and Madden on Defining Disposition Concepts.” 66Harré and Madden, Causal Powers, p. 55. 67Ibid., p. 3. 68Ibid. 69Ibid. 70Ibid., p. 110. 71Ibid. 72Ibid. 73Cf. Wilson, Hume’s Defence of Causal Inference. 74Cf. D. Davidson, “Causal Relations.” For a general discussion of Davidson’s position, cf. F. Wilson, Explanation, Causation and Deduction, Ch. 3. 75Hume, Enquiries, p. 29-30; cf. T, p. 86-7. 76M. Mandelbaum, “The Distinguishable and the Separable: A Note on Hume and Causation.” 77Hume, Enquiries, p. 74. 78Mandelbaum, “The Distinguishable and the Separable,” p. 245; his italics. 79Ibid., p. 246. 80Cf. Weinberg, Ockham, Descartes, Hume, p. 143. 81As Weinberg, ibid., ch. 10, has pointed out against the obscure criticisms of Miss Anscombe, “Hume Reconsidered,” p. 188, and of Anthony Kenney The Five Ways, p. 67. 82Concerning the infamous missing shade of blue, cf. F. Wilson, “Hume’s Fictional Continuant,” where it is pointed out that Hume’s problems here stem in part from difficulties in his doctrine of ideas and also from difficulties in his account of relations.

186

Body, Mind and Self in Hume’s Critical Realism

83For a detailed discussion of this research programme, see F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, “Some Controversies about Method in Nineteenth-Century Psychology,” and “Wordsworth and the Culture of Science.” 84Cf. P. Butchvarov, Resemblance and Identity, p. 115ff. Also F. Wilson, “Resemblance, Universals and Sorites: Comments on March on Sorting Out Sorites.” 85Cf. F. Wilson, “Effability, Ontology, and Method,” in his Acquaintance, Ontology and Knowledge. 86This should remind one of the suggestions that certain utilitarians such as Bentham made, that there can be a quantitative measure of pleasure. The historical connection is important. For a discussion of such views in psychology and ethics, see Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, Ch. 7. 87 This was Meinong’s way of formulating the notion; cf. K. Barber, Meinong’s Hume Studies: Translation and Commentary, p. 82ff. 88This empiricist appeal to PA would seem also to be valid grounds against Meinong’s account, since he seems willing to appeal to this principle; cf. ibid. It is also a sound appeal in the case of those utilitarians who suggested that a cardinal order of pleasures could be developed; cf. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, Ch. 7. 89This is a point that Russell emphasized; cf. B. Russell, Principles of Mathematics, Second Edition, p. 223. For a general discussion of the ontology of relations, see F. Wilson, “Bradley’s Account of Relations and Its Impact on Empiricism,” and “Burgersdijck, Coleridge, Bradley, Russell, Bergmann, Hochberg: Six Philosophers on the Ontology of Relations,” both in his Acquaintance, Ontology and Knowledge. 90Cf. S. Tweyman, “Hume on Separating the Inseparable,” p. 32-3. 91Some of these problems are discussed in Wilson, Psychological Analysis and the Philosophy of John Stuart Mill. 92F. Suarez, On the Various Kinds of Distinctions, p. 18. 93Ibid., p. 19. 94Ibid., p. 18. 95Ibid., p. 19. 96Ibid., p. 32

187

Two: Nominalism and Acquaintance

97Descartes, Principles, I, secs. 60-62, Haldane and Ross, vol. I, p. 243-5. 98Principles, I, sec. 62, Haldane and Ross, vol. I, p. 245. 99See Descartes, Oeuvres, ed. C. Adam and P. Tannery, vol. 4, p. 348-50. 100Descartes, Letter to Clerselier, Haldane and Ross, vol. II, p. 134. 101Arnauld and Nicole, The Art of Thinking, Part I, Ch. 5. 102Ibid., p. 49; as the translation puts it, attributes are characteristics “separable only in thought.” 103Ibid., p. 49-50; see above, endnote 9. Cf. Descartes, Principles, I, sec. 59, Haldane and Ross, vol. II, pp. 242-3. 104Arnauld, “Objections” to Descartes’ Meditations, Haldane and Ross, vol. II, p. 82. 105Descartes, Principles, I, sec. 53, Haldane and Ross, vol. I, p. 240. 106Descartes, Principles, I, sec. 59, Haldane and Ross, p. 242-3. 107.See above, footnote 94. 108Suarez, On the Various Kinds of Distinction, p. 19. 109Descartes, Principles, I, sec. 58, Haldane and Ross, vol. I, p. 242. 110Arnauld and Nicole, The Art of Thinking, p. 50. 111Weinberg, Abstraction, Relation, and Induction, p. 5ff. 112For a more recent empiricist attempt, cf. G. Bergmann, “Synthetic A Priori.” 113B. Russell, “On the Relations of Universals and Particulars.” 114 Cf. Tweyman, “Hume on Separating the Inseparable,” p. 33. 115Letter to Arnauld, 30 April 1687; see Lettres de Leibniz à Arnauld, ed. G. Lewis, p. 69. 116Boethius, Commentary on Porphyry’s Isagoge, Bk. I, sec. 10, translated in R. McKeon, ed., Selections from Mediaeval Philosophers, vol. I, p. 93. 117William of Ockham, Sententiarum, i, d. 2, q. 4, D, cited in Weinberg, Abstraction, Relation, and Induction, pp. 44-5.

188

Body, Mind and Self in Hume’s Critical Realism

118Sententiarum, i, d. 2, q. 6, EE; cited in Weinberg, p. 37. 119William of Ockham, Summa Logicae, Part I, Ch. 17; cited in Weinberg, p. 46. 120It could be argued that these are a special sort of relation, of the sort that some have called a “fundamental tie,” but this is a detail that need not concern us here. 121B. Russell, Inquiry into Meaning and Truth, p. 127. 122G. Bergmann, “Ineffability, Ontology and Method,” and “Stenius on the Tractatus.” Bergmann develops the view that the relation that provides the structural unity to a thing has the ontological status of a “fundamental tie.” 123Cf. Wilson, “Effability, Ontology and Method.” 124Such is the way to idealism, where the mind contributes the objective structure to the world. This is Kant’s move. See also G. Bergmann, “The Ontology of Edmund Husserl.” 125For an extended discussion of these aspects of Descartes’ philosophy, see F. Wilson, The Logic and Methodology of Science in Early Modern Thought, Study Seven. See also F. Wilson, The External World and Our Knowledge of It: Hume’s Critical Realism, Chapter Two. 126.For discussion of Cullen’s views, see J. Wright, “Metaphysics and Physiology: Mind, Body, and the Animal Economy in Eighteenth-Century Scotland.” 127.In vol. i of The Works of William Cullen, M.D.. 128.See also J. R. R. Christie, “Ether and the Science of Chemistry,” who also ties Cullen’s academic scepticism to that of Hume. 129.Henry Hallam, Introduction to the Literature of Europe in the 15th, 16th, and 17th Centuries, vol. ii, p. 435. 130.Encyclopaedia Britannica, Third Edition (1797), vol. xiv. 131.Vol. xiv. 132.Four volumes (Edinburgh: John Murray, 1822). 133.System of Mechanical Philosophy, vol. i.

189

Chapter Three From the Substance Tradition through Locke to Hume: Ordinary Things and Critical Realism (1) Up to Locke Hobbes’ re-statement of materialism was a crucial development in the history of British philosophy. It generated, as we have seen, many responses among the defenders of religion, who charged, with some justification, that it undermined religion and, with much less justification, that it undermined morality. But those who responded were not only the moralizers, but also the metaphysicians, those who attempted to defend the older moral view of human beings and the universe by defending the substance tradition. The revived materialism in turn revived the arguments against it, those of Plato, Aristotle and Plotinus. Among those who led in this response, and who was particularly influential, was Henry More,1 in such works as his Antidote against Atheism.2 More uses the causal argument (that Hume was later to criticize) that matter, being passive, cannot move itself; there must therefore be a spiritual being that moves it. Moreover, More went on to argue, even if matter were self-moving, or partly self-moving, it could not bring about the formation of compact masses of the sort that we observe in the stars and planets. Nor is it possible to hold that the laws of motion that govern the movements of the earth are casual or fortuitous. We must, therefore, conclude that “there is a Divine Providence that does at least approve, if not direct, all the motions of matter” (Bk. II, Ch. 1). More then goes on the infer the action of divine providence from such facts and the perpetual parallelism of the earth’s axis and from the laws of day and night, winter and summer, etc. (Bk. II, Ch. 2). This is the same sort of argument that the Newtonians Bentley and Clarke were later to develop using more sophisticated data based upon the explanations that Newtonian mechanics gave for the phenomena of the solar system. These appeals from physical

Three: Ordinary Things and Critical Realism

motion and structure are then followed by arguments for divine wisdom and benevolence based on the evidence of the adaptation of plants and animals for survival and reproduction (Bk. II, Ch. 3-11), including the adaptation of the human body (Bk. II, Ch. 12).3 Many of these same points are repeated in More’s later treatise on The Immortality of the Soul.4 The basic thrust is clear enough. Hobbes (and Descartes too) had suggested that it was possible to explain all natural phenomena in terms of inert matter in motion. Different observed patterns arise from the random motions of particles with changes occurring as the result of motions that are passed on through impulse upon contact. But, More argues (and he was not the last so to argue – consider the proponents of the “intelligent design” argument), random motions passed on by contact cannot explain the observed order in the world. One must therefore introduce other nonmaterial entities to account for the observed patterns. These spirits provide the forms that matter takes on; the phenomena of gravitation, magnetism, cohesion, etc., are due to their activities. These spiritual entities whose activities account for the observed form of the world provide the glue of causal necessity that distinguishes the observed order from the merely accidental patterns that alone could result if all that there was were particles undergoing motions and random collisions. Of some at least of these forms we have a priori knowledge which is not derived from our senses, specifically geometrical propositions (Antidote, Bk. I, Ch. 6), and also such causal principles as that every contingent event must have a cause (Bk. I, Ch. 8). At the same time, however, while we can know the forms which spirits move matter to take on, and, indeed, know some of these forms a priori, it is also true that we cannot know these active substances in themselves; in fact, we can know no substance, either active (spirit) or passive (matter) as it is in itself: For the evidencing of this Truth, there needs nothing more then a silent appeal to a mans owne Mind, if he do not find it so; and that if he take away all Aptitudes, Operations, Properties, and Modifications from a Subject, that his conception thereof vanishes into nothing, but into the Idea of a mere Undiversificated Substance; so that one Substance is not then distinguishable from another, but onely from Accidents or Modes, to which properly belongs no subsistence (Immortality, Bk. I, Ch. 2, Axiom viii; cf. axiom ix).

While we do not know substances in themselves it is nonetheless possible to infer their existence from other information that we have. Thus, we can use causal principles to infer that matter must be moved by

192

Body, Mind and Self in Hume’s Critical Realism

immaterial spirits (Bk. I, Ch. 11-12), and we can use the logical principle that incompatible attributes must inhere in distinct substances to infer distinctness of substances (Bk. I, Ch. 2, Axiom x). The latter, for example, argues for the distinctness of soul and matter (Bk. 1, Ch. 1, § 1); one relevant property is that of memory and imagination: “It is as impossible o conceive memory compatible with such a subject [matter] as it is how to write characters in the water or in the wind” (Antidote, Bk. I, Ch. 11, § 3; cf. Immortality, Bk. II, Ch. 5, § 7). Since we do not know substances in themselves, we can acquire no notion from our experience of the necessary connections which they effect among observed events. As More remarks at one point: This substance [the three-dimensional substance], which we call Matter, might as well have been penetrable as impenetrable, and yet have been Substance: But now that it does so certainly and irresistibly keep one part of it self from penetrating another, it is so, we know not why. For there is no necessary connexion discernible betwixt Substance with three dimensions, and Impenetrability (Immortality, Bk. I, Ch. 2, § 11).

As we never observe a necessary connection among the events that we know in experience, we can form no general idea of it from that source. Rather, our knowledge of the general concept of cause and effect comes from within ourselves; they are innate. Cause, effect, whole and part,...and such like [are] no material impresses from without upon the soul, but her own active conception proceeding from herself whilst she takes notice of external things (Antidote, Bk. I, Ch. 6, § 3; cf. Immortality, Bk. II, Ch. 2, § 9).

And when we join these a priori concepts with others to form propositions, we often have propositions whose truth is evident a priori; the innate ideas yield innate knowledge. This innate knowledge includes the causal and logical principles that enable us to infer the existence of spiritual substances. Even if we never know any specific form or spiritual substance, God has nonetheless provided us with the general concept of cause and with general knowledge about events and causes sufficient to enable us to infer that certain causes exist as the unexperienced necessary connections that tie together the events, the properties and attributes, that are given to us in the world of sensible experience. These patterns laid down by More were taken up and developed by

193

Three: Ordinary Things and Critical Realism

Newton. The latter tells us in the “General Scholium” to the Principia that he has “explained the phenomena of the heavens and of our sea by the power of gravity,” but he has “not yet assigned the cause of this power.”5 The role of gravity that he has succeeded in describing in the Principia has been “inferred from the phenomena, and afterwards rendered general by induction” (p. 170). On the other hand, as Newton argues in the “Queries” to the Opticks,6 since matter is in itself inert, it could not by itself have generated the forms or patterns of behaviour that we have discovered by induction (p. 177, p. 179). It follows that there must be an active principle to account for gravity, and for such facts as those of fermentation or of the cohesion of atoms into large compact masses (p. 179). These active powers are not specific, as Aristotelian occult powers were, but generic or general, and therefore an appeal to them carries an explanatory force that an appeal to occult qualities does not (p. 179). But, though it is clear that these powers are active (and therefore non-corporeal or spiritual), Newton will say no more about the forms that they can impose on matter beyond what can be discovered by induction (p. 180). Or, as he puts it in the “General Scholium,” hypotheses non fingo (p. 170). He is willing, however, to notice that mere chance could not have made all the planets move in the same direction in roughly circular orbits about the sun, and that there are similar non-chance uniformities in the structures of animals that adapt them to survival and reproduction. Since mere chance could not have created these patterns, we are entitled on the basis of ordinary causal reasoning to attribute them to the activities of a wise, powerful and benevolent deity (pp. 190-1). Thus, the lines of argument that we find in More we find revived and appearing in Newton. But whereas More emphasizes that some at least of our knowledge of the forms that active powers impose on matter is a priori, Newton emphasizes that we arrive at this knowledge a posteriori, by induction. Newton’s followers, Bentley and Clarke, in their Boyle Lectures, both invoke the argument from design as Newton did, that is, as a piece of a posteriori reasoning, but at the same time we find in them the sort of a priori reasoning that one finds in More, e.g., the argument for the existence of God as a necessary being. Now, this sort of appeal to the a priori to provide substantive explanations of matters of empirical fact makes sense only if one accepts something like a Platonic or Aristotelian account of forms or natures. But More, Bentley and Clarke all accept something like this, and so such appeals are for them perfectly legitimate. The question becomes, however, what happens to this Newtonian

194

Body, Mind and Self in Hume’s Critical Realism

philosophy if one rejects the Platonistic account of ideas and instead argues for an empiricist account. This is precisely what Locke did in his Essay concerning Human Understanding. Locke took our having various ideas to be matters of fact and proposed to use the “historical plain method” (Intro. to the Essay, sec. vii) to describe the origins and natural history of our ideas taken thus to be natural phenomena. By ‘historical’ Locke means the method of careful, accurate observation of the sequences of matters of fact.7 The sort of natural history that Locke proposed to give was precisely the sort of natural history that his friend Thomas Sydenham, for the first time since Hippocrates, was giving of diseases.8 It was an observationally based account of the patterns and regularities that characterized, for Sydenham, the development of a disease over time and, for Locke, the origins of our various ideas. In effect this inference from observation to pattern is none other than the observation followed by generalization that Newton is speaking of when he talks about “induction.” In other words, Locke proposes to take our ideas to be phenomena among phenomena that are given in experience and then to use the Newtonian method to investigate their natural causes. Here, of course, he is in direct contrast to the Aristotelian tradition, which accepted that it was legitimate to infer that thought must follow such and such a pattern given certain transcendental arguments. Where the Aristotelians allow their philosophy to impose certain patterns as those which thought must follow, phenomena be what they may, Locke, in contrast, and the empiricists who followed him, insisted that we must take even thought to be as it appears to be in experience, and to treat its causes as causes that may be found among the phenomena that we encounter in experience, or, of course, infer from that experience by means of acceptable principles. From this latter perspective, the appeal to innateness that one finds in More or Descartes is indeed vacuous, as Locke argues in Book I of the Essay. If one takes any of the principles, metaphysical or moral, that these philosophers held were innate, then a careful examination shows that they are in fact not universally assented to, and even if they are assented to by adults it is still the case that infants do not assent to them. So the question then arises as to the conditions that determine when one group assents to them and another does not, and also the question of the conditions under which children are first brought to the point of assenting to them. These are causal questions that must be answered by any natural history of our ideas. To claim that the ideas are innate is simply to refuse to answer these

195

Three: Ordinary Things and Critical Realism

questions; it is to respond to the request for a psychological theory by offering a non-theory. To insist that one develop a natural history of our ideas, however, is not yet to establish the claim that Locke also makes that all our ideas have their origins in experience, either in sensations impressed upon us by socalled external objects or the operations which the soul experiences within itself. It is the object of Book II of the Essay to argue systematically for this claim. Book IV then discusses knowledge and belief and the limits of human knowledge in the light of the genetic account of the origins of our ideas of Book II. In certain respects the discussion in Locke of substance is not remarkably different from that in More; in other respects there are radical innovations. But the latter were to find their full development only in Hume. (2) From Locke to Hume9 Locke’s discussion of real and nominal essences is of crucial importance for understanding his empiricist position on material substance. On his view, we repeatedly notice in our experience of things that certain properties are copresent with each other, e.g., in the case of gold the properties yellow, impenetrability, malleability, ductility, etc. They are, let us say, the properties A, B, and C. Since we have regularly observed them together, we know that (1) (›x)(Ax & Bx & Cx) To explain this, the Aristotelians use the principle (2) (x)(Rx. .Ax & Bx & Cx) where ‘R’ refers to the real essence of the thing in which A, B, and C are copresent. Now recall once again what the Aristotelians hold. On their view, this principle (2) is taken to be explanatory; it is therefore a substantive truth. That means that R cannot be a mere abbreviation for A&B&C as it is when (2) is taken as a nominal definition, that is, as a statement of nominal essence. For, if ‘R’ merely abbreviated the complex expression, then (2) would be a trivial linguistic truth rather than substantive. But the Aristotelian of course take ‘R’ to refer to the unanalyzable active potentiality that is the nature or form or “real essence” of the thing. Appeal to the nature or form of something is thus explanatory because that real

196

Body, Mind and Self in Hume’s Critical Realism

essence is an unanalyzable form. At the same time the appeal to the unanalyzable form R can provide a necessary connection among the sensible properties A, B, and C only if (2) is itself not a contingent truth but is, rather, necessary, true a priori. Or, to put it as the Aristotelians did, R can provide a necessary connection among the sensible properties it is supposed to explain only if it is a real definition and not just a nominal definition. The Aristotelians, we also recall, held that we know the real essences of things because we have a rational intuition of them. More, as we saw, joined the many who condemned the Aristotelians of the schools for appealing too often to our rational intuition to discover a real essence capable of explaining observed phenomena. We cannot know these essences in themselves, More held, but we can know that they exist. This knowledge that they exist is based upon inferences from certain principles that we know a priori to be true. This account is paralleled by Locke’s. Upon Locke’s account, our ideas of substances are collections of sensible ideas to which is joined the unclear idea of substance. Such a joining involves the supposition that the properties represented by the simple sensible ideas inhere in or are supported by a substance of the sort represented by the idea of substance, that is, of the sort that is unclear to us because it is represented by an inadequate idea. ...the ideas of substances are such combinations of simple ideas as are taken to represent distinct particular things subsisting by themselves; in which the supposed or confused idea of substance, such as it is, is always the first and chief. Thus if to substance be joined the simple idea of a certain dull whitish colour, with certain degrees of weight, hardness, ductility, and fusibility, we have the idea of lead; and a combination of the ideas of a certain sort of figure, with the powers of motion, thought and reasoning joined to substance, make the ordinary idea of a man (Essay, II, xii, 6).

Any person’s idea of a substance in general is an idea of “something, he knew not what”; it is something of which “they have no distinct idea...at all, and so are perfectly ignorant of it, and in the dark” (II, xxiii, 2). There are two aspects to this. On the one hand, our idea of substance is simply a complex idea which to which we attach a single name and thereby treat as a unity. “The mind being...furnished with a great number of the simple ideas, conveyed in by the senses as they are found in exterior things, or by reflection on its own operations, takes notice also that a certain number of these simple ideas go constantly together; which being presumed to belong

197

Three: Ordinary Things and Critical Realism

to one thing, and words being suited to common apprehensions, and made use of for quick dispatch, are called, so united in one subject, by one name; which, by inadvertency, we are apt afterward to talk of and consider as one simple idea, which indeed is a complication of many ideas together...” (II, xxiii, 1). At the same time, besides the easy grouping of ideas deriving from similar groups given in experience, there is the presumption that we make that there is indeed a simple subject to which the apprehended attributes jointly belong; this subject is not given by sense or in reflection, so it is not part of the complex idea of the substance as it is determined by our experience of things. The idea of the common subject is introduced rather as a presumption, or, as Locke also says, as a supposition. It is introduced, in other words, not by experience but by inference. On the basis of certain accepted principles, we infer that the substance is there even if we are not presented with it in experience: “not imagining how these simple ideas can subsist by themselves, we accustom ourselves to suppose some substratum wherein they do subsist, and from which they do result, which therefore we call substance” (II, xxiii, 1). Nor is this supposition a mere supposition. The supposition is rather a case in which even though we are not presented with a certain entity in sense experience nonetheless it is hypothesized on the basis of rational principles that this entity exists. As Locke makes clear in his discussion with Stillingfleet, we must not let ourselves be misled by his use of the term ‘supposition’: Your lordship goes on to insist upon “supposing” only, as that which gives rise to, and is included in, our idea of substance, thus resting it on mere supposition; and you yourself, if I understand your reasoning, conclude that there is substance, because it is a repugnancy to our conception of things that modes should subsist by themselves; and I conclude the same thing, because we cannot conceive how qualities should subsist by themselves (Third Letter, p. 375).

With respect to material objects in particular, we do not know their substance; for anyone, “he knows not...what the substance is of that solid thing.” Nor does he know what it is about that substance that accounts for the coherence of its parts into a whole; “Neither knows he how...the solid parts of body are united, or cohere together to make extension” (Essay, II, xxiii, 23). That which accounts for the coherence of the parts of substance into the particular order that it has is, of course, the real essence of the thing; it is this which makes lead and antinomy fusible, wood and stones not, it is this which makes lead and iron malleable, antimony and stones not, it is these which determine the “fine contrivances” of plants and

198

Body, Mind and Self in Hume’s Critical Realism

animals of species as diverse as those of sheep and cassiowaries (III, vi, 9). We sort things into species by the nominal essences that we form of them rather than their real essences (III, vi, 8). So on this point, too, Locke agrees with More that we can know that a real essence, or necessary connection, exists which accounts for the order we observe in material things, even though we are never presented with this connection in experience. Now, More justified this inference by appeal to the innate idea of cause and the proposition known a priori that contingent events must have causes. Locke has argued against innate ideas, so he must give a different account of how the inference goes. For Locke, a cause is that which makes a thing begin to be (II, xxvi, 2); a power, or rather, an active power, is the possibility of making a change (II, xxi, 2); so causation involves the exercise of an active power. Our notion of power is a simple idea (II, vii, 8; xxi, 3); powers are therefore unanalyzable properties of things. What we observe by means of our senses are changes in the properties of things. These observed changes evoke the idea of power, on the basis of the assumption that nature is regular, that is, the assumption that like causes produce like effects. The mind...observing a constant change of its ideas...; and concluding from what it has so constantly observed to have been, that the like changes will for the future be made in the same things by like agents, and by like ways, – considers in one thing the possibility of having any of its simple ideas changed, and in another the possibility of making that change; and so comes by that idea which we call power (II, xxi, 1).

This simple idea of power which inference leads us to include in the complex ideas we have of things such as fire, with its power to melt gold, is “a principle ingredient in our complex ideas of substances” (II, xxi, 3). Among the ideas that we include in the idea we have of a substance as constituting its real essence “are to be reckoned its active powers” (II, xxiii, 7). In the case of body, however, we are not presented in experience with the exercise of the power, with the necessary connection between cause and effect. Thus, for example, the observation of one billiard ball striking another and moving it by impulse, “gives us but a very obscure idea of an active power of moving in body, whilst we observe it only to transfer, but not to produce any motion” (II, xxi, 4). But if we do not get the idea of power from sense experience itself and if that idea is not innate, then from whence do we obtain it? Locke argues that there is one special

199

Three: Ordinary Things and Critical Realism

case in which we are in fact acquainted with the exercise of a power effecting a necessary connection between cause and effect. This is the case of volition. Volition is “an act of the mind knowingly exerting that dominion it takes itself to have over any part of the man, by employing it in, or withdrawing it from, any particular action” (II, xxi, 15). A volition is the exercise of the active power of the will (II, xxi, 5). In this case we witness in change not only one event being followed by another but one event being produced by another. Thus, the mind “doth...receive its idea of active power clearer from reflection on its own operations, than it doth from any external sensation” (II, xxi, 4). From here, then, in experience, rather than from innate ideas, we obtain the simple idea of an active power or necessary connection. Once we have the idea then we are in a position to impute active powers to other substances. At least, we can do so provided that we have knowledge of principles that enable us so to infer. More relied here basically upon the principle that he held was a priori true, that every contingent event must have a cause. But Locke too holds that this proposition is self-evident; it is the basis of his inference to the existence of God: ...man knows, by an intuitive certainty, that bare nothing can no more produce any real being, than it can be equal to two right angles....If, therefore, we know there is some real being,...it is an evident demonstration, that from eternity there has been something; since what was not from eternity has a beginning; and what had a beginning must be produced by something else (IV, x, 3).

It is this same principle that leads us to conclude the existence of external causes to the sense impressions which we passively receive: “It is the actual receiving of ideas from without that gives us notice of the existence of other things, and makes us know, that something doth exist at that time without us, which causes that idea in us...” (IV, xi, 2). Locke goes on to point out, however, that while we can have this knowledge that there is a cause of our sensations, this knowledge is limited in its certainty; in particular, we do not have any knowledge of the sort of real essence that is present in the thing and which determines its causal powers. ...when our sense do actually convey into our understandings any idea, we cannot but be satisfied that there doth something at that time really exist without us, which doth affect our senses, and by them give notice of itself to our apprehensive faculties, and actually produce that idea which we then perceive: and we cannot so far distrust their testimony, as to doubt that such collections of

200

Body, Mind and Self in Hume’s Critical Realism

simple ideas as we have observed by our senses to be united together, do really exist together. But this knowledge extends as far as the present testimony of our senses, employed about particular objects that do then affect them, and no further. For if I saw such a collection of simple ideas as is wont to be called man, existing together one minute since, and an now alone, I cannot be certain that the same man exists now, since there is no necessary connexion of his existence a minute since with his existence now: by a thousand ways he may cease to be, since I had the testimony of my senses for his existence. And if I cannot be certain that the man I saw last to-day is now in being, I can less be certain that he is so who hath been longer removed from my senses, and I have not seen since yesterday, or since the last year: and much less can I be certain of the existence of men that I never saw.

Note how Locke is using here the principle that if two entities are separate in our experience of them, then we have no a priori grounds to assume that the two entities are pats of the same thing. To be sure, what appears separate in experience need not be so in reality; but we do not know that reality, we do not know, that is, the substratum and the real essence that alone could provide among the perceptually separate parts of things the sort of necessary connection that could guarantee an inseparable continuity. “And, therefore,” he continues, though it be highly probable that millions of men do now exist, yet, whilst I am alone, writing this, I have not that certainty of it which we strictly call knowledge; though the great likelihood of it put me past doubt, and it be reasonable for me to do several things upon the confidence that there are men (and men also of my acquaintance, with whom I have to do) now in the world: but this is but probability, not knowledge (IV, xi, 11).

These judgements of probability, like those of demonstration, establish the agreement or disagreement of two ideas, but do so through proofs in which “the connexion is not constant and immutable..., but is, or appears for the most part to be so, and is enough to induce the mind to judge the proposition to be true or false, rather than the contrary” (IV, xv, 1). The grounds for judgements of probability are “the conformity of anything with our own knowledge, observation, and experience” and the testimony of others (IV, xv, 4). In short, More and Locke agree that our knowledge of body does not extend much beyond what we can learn about it in experience. To be sure, we can infer the existence of the substratum and of the real essence – we can know that they exist – but we do not know these things directly. These

201

Three: Ordinary Things and Critical Realism

inferences are based upon self-evidently true causal principles, according to both Locke and More. But More’s account of the origin of the ideas that appear in those principles is innatist, and is criticized by Locke, who offers a different account, finding the basic idea of active power or necessary connection in the active exercise of such a power in the case of our own volitions. There is a determined shift in emphasis from More to Locke, however. This is the shift to the empirical method of Newton. This leads Locke to introduce the “historical plain method” into the discussion of our ideas, and to the subsequent critique of innatism. But more importantly it directs one away from a concern like More’s which is attached to the nonsensible order – to substances and real essences – and leads one to a more direct concern for what experience can teach about material things – even where that experience yields judgments that are not knowledge, only probable. It is this thrust that Hume takes up and develops, criticizing both the notion of real essences and of substances as substrata, and moving even further than Locke from the substance tradition. If we recall our real essence R that explained the observable features A, B, and C of things, we should recall these latter features include the powers of things, so far as they can be known by observation. Thus, malleability is included among the characteristics defining gold. We should, therefore, think of each of these features, A for example, as short for something like P –> Q That will leave us, in the simplest case, with (3) (x)( Rx. .Px –> Qx ) as our basic model for a principle relating a real essence to the observable features for which it is supposed to account. As before, on the one hand, (3) cannot be a mere definition, for then it cannot be explanatory. ‘R’ must therefore refer to an unanalyzable potentiality that lies behind the observable pattern represented by ‘P –> Q’. And (3) must also be principle that is at once substantive, because explanatory, and necessary, for otherwise it will require the introduction of still further principles, ad infinitum, to explain the contingent connection R to P and Q. On the other hand, if (3) is a mere definition then there is absolutely no difference between attributing the disposition or power R to a thing and attributing Q to a thing when that thing is P. This is Hume’s position: “...neither man not

202

Body, Mind and Self in Hume’s Critical Realism

any other being ought ever to be thought possest of any ability, unless it be exerted and put in action” (T, p. 311). It is clear that upon this position there is, in effect, no difference between a power and its exercise. As Hume puts it, “the distinction, which we often make betwixt power and the exercise of it, is...without foundation” (T, p. 171); the distinction is, as he puts it elsewhere, “entirely frivolous” (T, p. 311). This, at least, is one way of putting the upshot of Hume’s critique of the idea of necessary connection. In denying the existence of real potentialities in things, Hume is of course disagreeing with the Aristotelian position. Now, as we have seen, that position argued that unless there were such potentialities, one would be committed to the Megarian position (at least as Aristotle described it) that one can do something only at the moment at which one is in fact doing it, i.e., that possibility and actuality are one. But Hume argues that this is not so. He takes up Locke’s point that we attribute a power to something in those cases where we have reason to suppose that if a like change were to occur again then it would be followed by a like effect (Essay, II, xxi, 1). Hume considers an example, and then proposes a criterion for distinguishing those powers that we consider real or genuine, even apart from their exercise, and those powers that are unreal. ...if we compare...two cases, that of a person, who has very strong motives of interest or safety to forbear any action, and that of another, who lies under no such obligation, we shall find, according to the philosophy explain’d in the foregoing book, that the only known difference betwixt them lies in this, that in the former case we conclude from past experience, that the person never will perform that action, and in the latter, that he possibly or probably will perform it (T, p. 312).

Thus, we attribute P –> Q to Jones as a real power which he has provided that we have reason to believe on the basis of our past experience of Jones that the regularity For any time, if Jones is P then Jones is Q More generally, ‘P –> Q’ is attributed as a real power or disposition to an individual a just in case that there is an observable categorical characteristic S of things such that (x) [Sx –> (Px –> Qx)] is a regularity that we have confirmed in experience and such that the individual a is S. Thus, solubility is a real power of this powdery substance

203

Three: Ordinary Things and Critical Realism

before me because I can tell that what is before me is sugar and I know on the basis of experience that, for any bit of sugar, if it is in water, then it dissolves. Hume puts the conclusion this way: ...power consists in the possibility or probability of any action, as discover’d by experience and the practice of the world (T, p. 313).

So Hume, while denying the distinction between a power and its exercise is nonetheless not committed to the Megarian position that one can do something only when one is actually doing it. For Hume, Jones can do Q even when he is not doing it just in case that he will do Q if he is P or, and this is equally important, that he would do Q if he were P. That is, the regularities that justify the attribution of real powers or dispositions to things must be laws rather than accidental generalities. What Hume argues, of course, is that objectively there is no difference in form between a law and an accidental generality; both are of the form “Whenever A then B.” This is his first definition of cause: “an object precedent and contiguous to another, and where all the objects resembling the former are plac'd in a like relation of priority and contiguity to those objects, that resemble the latter” (T, p. 172). What distinguishes a regularity that is causal or a law from which is merely accidental is our subjective attitude towards it: a regularity is lawful just in case that we are prepared to use it to predict and to make contrary-to-fact inferences. This is Hume’s second definition of cause: “an object precedent and contiguous to another, and so united with it in the imagination, that the idea of the one determines the mind to form the idea of the other [contrary-to-fact inference], and the impression of the one to form a more lively idea of the other [prediction]” (T, p. 172). Of course, not every case where that lawassertion attitude is adopted towards a regularity is rational; otherwise we would not be able to distinguish science and superstition. Hume spends considerable time establishing that so far as we can tell, conformity to the rules of the scientific method is more likely to lead to truth than the rules that lie behind superstition, e.g., the rule that the wish should be the father of the belief. Thus, what distinguishes those cases where the law-assertion attitude is rational is that the attitude, or habit, has been acquired through processes in which the mind has conformed to the rules of the scientific method, that is, what Hume refers to as the “rules by which to judge of causes and effects” (T, p. 173). One thing should perhaps be emphasized. Hume does not prohibit

204

Body, Mind and Self in Hume’s Critical Realism

inferences to unobserved, and perhaps unobservable entities. He explains to us how the vulgar are content to say that a watch sometimes works, and sometimes does not. What happens, so far as the vulgar are concerned, is that this is a matter of chance. But the artisan knows better. From his experience in discovering causes, the artisan can infer that when the watch stops working there is an unobserved cause for why it stops; it is, perhaps, a speck of dust. The philosopher generalizes from this experience to the more general conclusion that every event has a cause: From the observation of several parallel instances, philosophers form a maxim, that the connexion betwixt all causes and effects is equally necessary, and that its seeming uncertainty in some instances proceeds from the secret opposition of contrary causes (T, p. 132).

Of course, such inferences to unknown causes cannot arise from the repeated experience of the cause being succeeded by the effect, for, ex hypothesi, the cause is not known. It is, rather, a case where certain more general rules control the inferences that we make with less general rules (T, pp. 149-50), a case where higher level habits control the formation of lower level habits (T, pp. 137-8). As Hume puts it, “our reasonings of this kind arise not directly from the habit, but in an oblique manner” (T, p. 133). Since these reasonings fully conform to the “rules by which to judge of causes” (T, p. 149), their conclusions that certain unknown causes exist are fully rational. What Hume excludes as unreasonable are not inferences to unobserved events but only inferences to entities of which we can form no idea, and in particular inference to substances and to real essences and necessary connections. The argument against substances is the same as the argument against objective necessary connections: we are not presented with such entities in experience. We have no impression of an entity with a continued existence, i.e., a continuant (T, pp. 191-2); the only entities with which we are acquainted are “perishing” (T, p. 194). Locke agreed with this; the substance or continuant was something that we “supposed” was there, i.e., inferred was there. But since he had (he thought) reason to suppose that it was in fact there, Locke could continue to hold that ordinary perceptual objects consisted of properties inhering a substance. For Locke, then, predication represented what it did for Aristotle, the tie of inherence. But for Hume, since we have no acquaintance in experience with substances, then from his basic principle that we have no ideas without antecedent

205

Three: Ordinary Things and Critical Realism

impressions, it follows that ordinary bodies are what they appear to be, namely, collections consisting of the properties or attributes that are predicated of them. As he puts it in the Treatise, “the idea of a substance...is nothing but a collection of simple ideas...” (T, p. 16); or, as he puts it in the Abstract of the Treatise,10 “...our idea of any body, a peach, for instance, is only that of a particular taste, colour, figure, size, consistence, etc.” (p. 25). But even if there is no continuant, ordinary objects or bodies are, Hume tells us (T, p. 188), taken to have a continued existence, even when not perceived,11 and an existence distinct from the perceiver, and therefore of the perceiving of them, where he includes in the notion of ‘distinctness’ both the external position, i.e., as he soon explains (T, p. 190), their externality to, or distance from, our body, and their independence. Both independence and externality are matters of inference based upon laws (T, p. 191); impressions are therefore not only perishing (T, p. 194) but have no intrinsic sign of independence. This does not mean that they are intrinsically mental, ontologically dependent upon the mind. For, as the mind is but a bundle of perceptions, “there is no absurdity in separating any particular perceptions from the mind”; hence “we can satisfy ourselves in supposing a perception to be absent from the mind without being annihilated.” (T, p. 207) Externality is therefore a matter of relations to other impressions, including, Hume will argue, impressions that are not perceived. Equally, an impression is not intrinsically mental, but becomes such only when bundled in certain way with other impressions; “internality” is also a matter of the relations of an impression to other impressions. It is therefore reason, ordinary causal reason, that is, the imagination, that fills in the gaps so as to make our impressions into impressions of ordinary things as “continuing” objects. Note, that this is ordinary causal reason that is at work and gives us knowledge of body, which is something “we must take for granted in all our reasoning.” (T, p. 187) Importantly, this is a matter of Hume’s commitment to empiricism. The suppositions that form our complex ideas of continuing objects do not arise from (the sort of) reason that proceeds a priori from metaphysical principles (T, p. 193), since it is perfectly evident that the vulgar conceive of objects as continuing and distinct, but have not gone through metaphysical proofs of the continuing and distinct existence of body, e.g., of the kind proposed by Descartes. The source of the ideas, then, must be the imagination (T, p. 193), though we must not forget that this does not condemn them as

206

Body, Mind and Self in Hume’s Critical Realism

contrary to reason, since the imagination also includes the “understanding” as “the more general and established principles of the imagination” (T, p. 267) which are the best habits of causal inference (T, p. 150, p. 170, pp. 173-6). What is it, then, if it is not a continuant, that distinguishes those impressions “to which we attribute a continu’d existence”? What sort of reasoning is it, or rather what sort of acts of the imagination that induces us to believe in body? Now, what distinguishes those perceptions that we attribute to bodies is ...a peculiar constancy, which distinguishes them from the impressions, whose existence depends upon our perception. Those mountains, and houses, and trees, which lie at present under my eye, have always appear’d to me in the same order; and when I lose sight of them by shutting my eyes or turning my head, I soon after find them return upon me without the least alteration. My bed and table, my books and papers, present themselves in the same uniform manner, and change not upon account of any interruption in my seeing or perceiving them (T, pp. 194-5).

To be sure, the constancy is not perfect, but even though objects do change position and quality, those impressions that are reckoned external have a coherence that others do not (T, p. 195). If I do notice an alteration of an object after an interruption, e.g., the fire in my fireplace after leaving and returning to the room, I may still attribute externality that ...I am accustomed in other instances to see a like alteration produc’d in a like time, whether I am present or absent, near or remote (T, p. 195).

If constancy is not perfect, then at least there is coherence. Hume then proceeds to examine in detail how these inferences go (T, pp. 195-7).12 It is like this. We have a series of sensible particulars a1, b1, c1 with qualities F, G, H: (a) Fa1, Ra1b1, Gb1, Rb1c1, Hc1 where “R” represents that these particulars are in a continuous series. The pattern of properties is continued in other series that we observe: (b) Fa2, Ra2b2, Gb2, Rb2c2, Hc2 (c) Fa3, Ra3b3, Gb3, Rb3c3, Hc3 But we now observe a “gappy” series of two particulars a4 and c4 which have the F and H properties, but for which there is no intervening particular of sort G:

207

Three: Ordinary Things and Critical Realism

(d) Fa4 .............................. Hc4 The understanding “fills” in the “gap” by forming the idea of a particular that is G and is R-ed by a4 and R’s c4. Let us call this particular, of which we have no impression, only an idea, ß. Then the series (d), as (re)constructed by the mind is (e) Fa4, Ra4ß, Gß, Rßc4, Hc4 (a), (b) and (c) are the other instances which have accustomed me to expect a G to be in a continuous series with an F and an H; this custom or habit leads me to attribute the same continuity or coherence to the gappy pattern (d); and the result is (e) which now fits the pattern (a)-(c). Now, it would seem on the face of it that (d) falsifies the pattern that has been inferred from (a)-(c). More generally, gappy series cannot support customary inference, inference based upon mere habit, and in fact imply that those inferences very often ought to be rejected. Any degree, therefore, of regularity in our perceptions, can never be a foundation for us to infer a greater degree of regularity in some objects, which are not perceiv’d; since this supposes a contradiction, viz. a habit acquir’d by what was never present to the mind (T, p. 197).

Nonetheless, instead of permitting (d) to falsify the inference we to the contrary wish to maintain the inference and judge that our observations are at fault, that the series really is complete. We are, therefore, “involv’d in a kind of contradiction”: In order to free ourselves from this difficulty, we disguise, as much as possible, the interruption, or rather remove it entirely, by supposing that these interrupted perceptions are connected by a real existence, of which we are insensible (T, p. 199).

Thus, we do not let (d) falsify our judgment that we again have an FG-H sequence or series. We might have formed the habit, Whenever an F then a G We now have a4 is F We infer There is an x that is R-ed by a4 and which is G Notice that this is an existential claim, and such a claim is not falsifiable: failure to notice or perceive a G that is R-ed by a4 does not falsify this existential claim; failure to notice it does not imply that it does not exist.

208

Body, Mind and Self in Hume’s Critical Realism

And only if it does not exist do we have a falsification. And now, instead of letting (d) falsify the inference based on the observational data of series (a)-(c) we to the contrary maintain the inference and judge that our observations are at fault, that the series really is complete only we don’t know it, and that there is in fact a G, though we’ve not observed it, between the F and the H of (d). The idea ß of this gap-filling particular constitutes the “idea of continu’d existence” which is essential to our ideas of bodies, that is, of material objects like tables or chairs; its existence is a “supposition” (T, p. 199), that is, an existential hypothesis that has not been verified by observation. It is, however, a supposition or hypothesis which is believed (T, p. 199), and, indeed, is made worthy of belief by customary inferences derived from past experiences (T, p. 199). In fact, these inferences to unobserved sensible particulars are parallel to other such inferences to unobserved entities that fill gaps in our perceptual life; inferences of this sort arise “from the understanding, and from custom in an indirect and oblique manner” (T, p. 197). Now, Hume has earlier given a similar example of such reasoning, where the failure to observe a cause we believe to be there is not taken as grounds for rejecting the inference to its presence. A watch sometimes works, and sometimes does not. The vulgar attribute this to chance (T, p. 132). But philosophers observing, that almost in every part of nature there is contain’d a vast variety of springs and principles, which are hid, by reason of their minuteness or remoteness, find that it is at least possible the contrariety of events may not proceed from any contingency in the cause, but from the secret operation of contrary causes (T, p. 132).

Indeed, Hume goes on to remark, such a supposition or existential hypothesis about an unobserved cause can be converted into a justified belief by use of the appropriate causal inferences: This possibility is converted into certainty by further observation . . . that upon an exact scrutiny, a contrariety of effects always betrays a contrariety of causes, and proceeds from their mutual hindrance and opposition (T, p. 132).

But this is just an inference in conformity with the sixth of Hume’s “rules by which to judge of causes and effects” (T, p. 174), that is, the rules of eliminative induction that we now generally refer to as “Mill’s Methods.” This rule, the method of difference, presupposes for its logical efficacy, the

209

Three: Ordinary Things and Critical Realism

fourth of Hume’s rules, the principle of determinism that every event has a cause, and that similar causes (effects) have similar effects (causes). It is, at bottom, this rule that justifies the philosopher, in contrast to the vulgar, in attributing the non-working of the watch to an unobserved cause – that is, in believing the hypothesis that such a cause exists – and accounting for our failure to observe it in terms not of its non-existence but of our not being in a position to observe it (e.g., our eyes are closed, or it is too small to see). In such reasoning as this, we have a more general rule (the principle of determinism) correcting a less general rule (T, p. 150); and, Hume tells us, “our reasonings of this kind arise not directly from the habit, but in an oblique manner” (T, p. 133). What is the nature of the idea that we have denominated “ß”? If we take “R2” to denote the relative product of R with itself, so that R2xy means that y is R-ed by something that x R’s then, while we cannot say that the particular a4 continues through the series (d)=(e) in a way such that c4 = a4 nonetheless we do have it, and quite correctly, given Russell’s account of these things, that (*) c4 = (,x)(R2a4x) which plausibly construes c4 is the same body as a4 Technically speaking, we can use definite descriptions here in our inferences to fill in the gap in the series (d) to give us (**) ß = (,x)(Rax) only if R has certain structural properties, like symmetry, which together guarantee that that relation generates a continuous series; but Hume’s discussion implies, though he does not (could not) know it, that such structural properties are assumed. This being so, a statement like (*) is perfectly justified within the context of the gap-filing inferences that turn our series of impressions into bodies. This shows that Hume was essentially correct when he held that the idea of a body involves the idea of a given particular being connected with each succeeding member of the continuous series that constitutes that body. But his account of ordinary things, if it is to be defended, must include an adequate account of relations such as “R”. That means, as we have seen, that Hume must give up his nominalism.

210

Body, Mind and Self in Hume’s Critical Realism

An ordinary object like a table or a tree does in fact continue to exist. It has moreover a nature that is inseparable from it, that is, a nominal essence, a nature as these are understood in an empiricist philosophy. The earlier tradition had construed such entities as simple substances with simple natures, real essences. Hume has given a very different account, one that fits with his revised concept of reason. For Hume, an ordinary thing is not a simple substance but a collection, a bundle of sensible particulars. It endures, but as a process rather than a thing. It has a nature but this nature is a matter of the parts of the process or bundle being instances of a pattern. The constancy of the nature is the persistence of a pattern. This nature is grasped not by intuition, but as all causal regularities are grasped, by (fallible) inference from experience. Where Hume goes wrong is his notion that because the idea of a4 occurs in the idea that denotes ß in (**) and in the idea that denotes c4 in (*), therefore that particular, namely a4, continues in the objective series. This leads him to think that we attribute a “perfect identity” (T, p. 199) to the two different impressions at the beginning and the end of the series, and to think that we are “not apt to regard these interrupted perceptions as different, (which they really are)” (T, p. 199). The problem here is his simple inability to handle definite descriptions adequately, a problem equally evident in his famous discussion of the missing shade of blue. It is this inability which leads Hume to think that the series of perceptions that constitutes a material object contains within it a continuing particular, though one which he knows on other grounds to be “fictional,” that is, something constructed by the mind.13 Now, for Locke, there is a distinction between some observed qualities, which are real, and the rest, which are not. Those that are real are those that directly express the real essences of things. For the rest, they are the mere appearances of things in consciousness. For Hume, in contrast, all sensible qualities are equally real in the sense that none can be distinguished from any other by reference to substances and their real essences. The distinction between those observed qualities that are somehow the “real” properties of things and those which are “mere appearances” is just that some cohere as parts of those patterns that constitute material objects where others do not, and since material objects are important to us in our lives and in our capacities to communicate with others, these objects (not: substances) are singled out by convention as defining “reality.” We adopt a certain standpoint in assessing the perceptible qualities we experience because this has a certain utility in

211

Three: Ordinary Things and Critical Realism

communication. This is parallel, as Hume points out (T, p.582), to what he called the artificial virtues, those virtues that arise only because certain behaviour is generally coordinated by convention, e.g., justice. In each case we have a convention. This convention is explained by showing that it arises from self-interest in a context of constrained benevolence. In the case of justice it is an interest in establishing the property rights that are essential to peaceful living together. In the case of perception it is an interest in communication and action: we pick out from among all the patterns of sensible events that we experience those with respect to which we can most readily come to intersubjective agreement. ...corrections are common with regard to all the senses; and indeed ’twere impossible we cou’d ever make use of language, or communicate our sentiments to one another, did we not correct the momentary appearances of things, and overlook our present situation (T, p. 582).

Assessments both of the qualities of mind and character, i.e., recognition of virtues and vices, and also of the public qualities of material objects are extremely important in action both in society and in the natural world. The convention therefore becomes instituted that certain among all the qualities that we experience are to count as real. The distinction among sensible qualities of the real and the unreal in terms of coherence is thus artificial, but it is not therefore arbitrary. And, of course, once the conventions of reality have become instituted, and we talk about real colours, size, hardness, tastes, etc., then we come in turn to approve morally of conformity to these conventions for the same reason that we approve morally of the artificial virtues, namely because of their utility and because of our sense of sympathy which leads us to be concerned with the needs and concerns of others. This, then, is the account of body, of material objects, that results when one does away with the substances of the Aristotelian tradition. The simple continuant disappears; so does the nature in the sense of real essence. Thus, on the one hand, a body, an ordinary material object, becomes an ordered or patterned sequence of collections of qualities; predication does not reflect the tie of inherence but rather the relation of part to whole. The nature or form as real essence also disappears. Thus, on the other hand, to know that a presented collection of qualities is part of that sequence that we ran into yesterday or is part of a series that constitutes, say, a tree is to subsume that presented collection under certain

212

Body, Mind and Self in Hume’s Critical Realism

regularities which experience has taught us apply in these circumstances; sorting things into species is a matter of coming to know laws, general matter-of-fact truths, rather than a matter of having rational intuitions of certain special properties, the “real essences”, of things. For Aristotle, change in the first place was the coming to be and the passing away of properties in substances; for Hume it is simply a matter of the transition from an earlier stage in a series to a later. But for Aristotle change was also, in the second place, the actualization in time of a real potentiality of the substance; for Hume that actualization of potentiality is simply the transition from one member of a series to the next in accordance with a general pattern or law. The order in which changes in things took place was a matter of the form or real essence of the thing for Aristotle; for Hume it is a matter of the changes being one instance of a pattern that holds generally among all things of a similar kind. Where the substance account of body appeals to entities beyond the world of sense experience, the Humean account of body is wholly this-worldly. What is crucial for the substance account is a timeless entity outside the world of ordinary experience; what is crucial for the Humean account is a timeless pattern of things in the world of ordinary experience. One thing should be noted. We must distinguish perception from causal inference. To be sure, as Hume argued, objectively their content is the same, namely, patterns of inference which go beyond the data and infer the future and the unobserved from present and past sense experience. Thus, since, on the one hand, the two are objectively the same, the same rules may be used to evaluate both. On the other hand, subjectively perception and causal inference are very different. Both are movements of the imagination (T, p. 198), but causal inferences are more a matter of reflective thought than are our ordinary perceptual judgments, which are more a matter of instinctive response. Our ordinary experience in which we fill in the gaps between our impressions is primarily a function of memory and the inertia of thought moving in a pattern, in this case, the pattern of the object the initial part of which we have observed (T, p.198, p. 204, p. 208). Thus, causal inference and perceptual judgment are two different modes of association. Nonetheless, that they are different does not imply that causal inferences cannot show that our perceptual judgments are justified; that is, it does not follow that one cannot use what Hume has argued to be the best norms for inductive inference, namely, the rules by which to judge of causes and effects, to judge that our perceptual judgments, which are also objectively inductive inferences, are justified.

213

Three: Ordinary Things and Critical Realism

This concludes Hume’s description of what he calls the “system of the vulgar” (T, Bk. I, Pt.!V, sec. 2) It is the world of ordinary experience, the world of tables, chairs, rainbows, and straight sticks that appear bent when in water. It is a realism in which the things that we experience are as we experience them to be. (3) Hume’s Causal Inference to Critical Realism We have seen that causal inference and perceptual judgment are two different modes of association. Nonetheless, that they are different does not imply that causal inferences cannot show that our perceptual judgments are justified; that is, it does not follow that one cannot use the best norms for inductive inference, namely, the rules by which to judge of causes and effects, to judge that our perceptual judgments, which are also objectively inductive inferences, are justified. At the same time, however, causal inference might also show that various perceptual judgments are false. In such a case Hume speaks of a contrast between imagination and judgment (T, p. 148, p. 267); though, as he also holds, in a broader sense the imagination includes both solid and weak inferences (T, p. 225), with the understanding, or judgment, being those inferences of the imagination that are more solidly established, that is, can be shown to be more rationally justified (T, p. 267). Now, in point of fact, causal inferences can often correct erroneous perceptual judgments. Suppose I see white paper under yellow light. Then what I perceive is that this paper is yellow. That, though, is mistaken, and I can use my knowledge of the laws of illumination to infer that, although the paper appears yellow in perception, it is in fact white. This causal knowledge enables me to assert of the yellow impression that it is a white piece of paper. This latter non-perceptual inference enables me to correct the erroneous perceptual judgment. Note, however, that while I reject the perceptual judgment as (partially) false, I do not cease to make it. As Hume would say, the inertia of the imagination carries on, determining that I continue to make the same perceptual judgment. When causal inference is used to correct causal inferences, the latter are in general modified or abandoned (T, p. 147f); but sometimes other natural mental propensities determine us to continue to make judgments we are prepared to say are erroneous (T, p. 148f), and cases of perceptual error are of this sort. In the case of inferring that the apparently yellow paper is white, one has not strayed from the system of the vulgar. One can continue to hold

214

Body, Mind and Self in Hume’s Critical Realism

what has been called a “Selective theory” of perception. Such a theory holds that objects are bundles of sensory data, and that in perception one or more of these data is as it were selected as the one that will also the sense impression that is part of the content of one’s awareness. On such a theory, one has, in cases of error like those we have just noted, selected what is as it were a wild sensible particular, one that is not among those that are “real”, that is, selecting a wild particular rather than selecting a sensible particular that is among those the pattern has us discriminate as “real.” However, causal reasoning about perceptual objects in the system of the vulgar if pursued systematically leads to the system of the philosophers. What does the case of seeing the tree look like in the system of the philosophers? There remain the entities with which we are directly acquainted. These are the green impression and the perceptual judgment that this impression is a tree is just the way a tree is described in the system of the vulgar. But in the system of the philosophers, as in the system of the American Critical Realists, there is a distinction between impressions and objects. In this system, unlike that of the vulgar, the impression is not the tree. As in the system of the vulgar, for the philosophers the tree is a continuing entity, and (we are taking it) this continuity is the continuity of a series. But the impression is not part of this series. The series which is the continuing tree has none of the qualities of impressions. This is the second way in which the system of the Critical Realists differs from that of the New Realists. Impressions exhibit such secondary qualities as colours, tastes, etc., and none of these qualities given to us in our sense experience is ever exemplified by the entities that are the various part of the continuing entity which is the real tree. We perceptually judge of the impression that it is a tree, but it is not a tree nor even part of the real object and the judgment is false. As for the real tree, neither it nor any part of it is ever given to us, either directly in sensation or in our perceptual judgments. The relation between the real tree and the green impression is not part-whole as in the system of the vulgar but causal. In perception in the system of the philosophers, the “real” or physical tree interacts with the “real” or physical body. When the body is in such and such a state and comes to be situated appropriately to the tree, then an interaction occurs, bringing about a change in the state of the body. The sense impression is the next step in this sequence of events: it, and our awareness of it, are brought about by the change in bodily state. . . . our perceptions have no more a continu’d than an independent existence;

215

Three: Ordinary Things and Critical Realism

and indeed philosophers have so far run into this opinion, that they change their system, and distinguish, . . . betwixt perceptions and objects, of which the former are supposed to be interrupted, and perishing, and different at every return; the latter to be uninterrupted, and to preserve a continued existence and identity (T, p. 211).

This system of the philosophers involves what has been called, in contrast to a selective account of perception, a “generative theory of perception”,14 which holds that the sense impression is actually brought into being by the physiological process which precedes the sensing of it, and would have no existence otherwise. In the system of the vulgar, the laws of perception are such that a physiological process causes and explains the awareness of the impression, but the impression itself is explained by its location in the lawful pattern of sensible particulars that is the continuing material object. In the system of the philosophers, the laws of perception are such that the physiological process not only causes the awareness of the impression but also the very existence of the impression itself. Clearly, the system of the philosophers contradicts the system of the vulgar. Hume is clear that he believes he can argue to an acceptance of the generative theory of perception as it appears in the system of the philosophers (T, p. 211). He performs certain experiments, and these are designed to show that “all our perceptions are dependent on our organs” (T, p. 211). Note, first, that the dependence is a causal dependence on our organs, not an ontological dependence on our mind, i.e., substantial mind, as in Berkeley and the idealists. As we noted above, Hume makes clear that he agrees with realists of all sorts, and in particular with the Critical Realists, that the idealistic account of the data given in perception is wrong: impressions, are not ontologically dependent on minds. It does not follow that there cannot be a causal dependence of impressions upon the perception of them. The point is that the latter sort of dependence does not imply idealism or subjectivism. Note, second, that Hume’s argument is one which assumes the existence of sense organs, nerves, animal spirits, etc., that is, to put it briefly, it accepts realism. As Hume says, we arrive at the system of philosophers only by passing through the common hypothesis of unsensed objects, that is, the continuance of our interrupted perception (T, p. 211). It follows directly that Hume cannot here be trying to establish some form of subjectivism.15 Hume’s argument is from such phenomena as the appearance of double images when the eye is pressed in a certain way. But its concern is not to establish subjectivism or scepticism; rather, it is to establish something about the causal status of sensed impressions.

216

Body, Mind and Self in Hume’s Critical Realism

Moreover, Hume clearly takes it to be a sound causal inference. What, then, is the structure of the causal argument that Hume uses? To begin, we must note that it begins within the system of the vulgar. The system of the philosophers distinguishes perceptions, i.e., impressions, that exist only when perceived, and objects that have a continued existence. Now, Hume points out that, since the mind has no acquaintance with these objects, the system therefore has no primary recommendation to causal reason (T, p. 212). It does not follow that it has no appeal to reason; for, it could still have a secondary recommendation to reason. The relevant habits of causal inferences must be acquired elsewhere, and this elsewhere can only be the system of the vulgar: thus, “the philosophical system acquires all its influence on the imagination from the vulgar one” (T, p. 213).16 The point is that reasoning within the system of the vulgar will lead one outside that system to another, contrary, system. But the habits cannot proceed directly but in an indirect and oblique way, which, however, they already do in the system of the vulgar when the latter hypothesizes the existence of unsensed objects (T, p. 197): it is such indirect and oblique reasoning based on the principle that every event has a cause and the same effects have the same causes (T, p. 133) that yields a secondary recommendation to reason. Causal reasoning, like the mental inertia that yields perceptual judgments, leads us to impose uniformity on nature. In fact, it yields greater uniformity than does perceptual experience. It is this search for uniformity that leads to the system of the philosophers that contradicts the system of the vulgar as it is given to us in our ordinary perceptual judgments. Hume’s argument is from such events as double vision, perspectival variation, and so on. In this he was to be followed by the Critical Realists.17 Price has argued that from cases of this sort one cannot infer, as Hume and the Critical Realists do, the causal dependence of sense impressions on our organs. All that the evidence cited shows is that the yellow impression is dependent upon the presence of yellow light. Similarly, the existence of perspectival distortions does not establish dependence upon the sense organs but shows only that they are dependent upon spatial position. “The flat and perspectivally distorted shape which I see when I look at a distant mountain could still continue in existence – for all that has been shown – when I go away and shut my eyes. But it would only exist from a certain place, not from other places”.18 One would have a realism of perspectives and appearances of the sort once proposed by the American New Realists.

217

Three: Ordinary Things and Critical Realism

In Hume’s presentation of the experiments, the reference to perspectival variation only comes second, after the reference to double vision. Of the latter Hume says that “we do not attribute a continu’d existence to both these perceptions” (T, p. 211). Thus, this first case at least is taken by Hume to establish that the existence of some sense impressions is dependent on our sense organs and bodily processes. But Price also questions whether the example establishes even this much.19 If I am looking at a chair and press my eye-ball, then I am presented with two impressions of the chair; but if I am looking at a tree and press my eyeball, then I am presented with two impressions of the tree. Thus, double vision phenomena are causally dependent, it seems, not only on sense organs but also on the external cause. Hence, The phenomena of Double Vision has no tendency to prove that any of our sense impression are totally dependent on ‘our organs and the disposition of our nerves and animal spirits’ [Treatise, p. 211], as Hume thinks they have, and still less that all are.20

We may take this in Price’s order, and attempt to see, first, whether double vision entails any conclusion about whether some sense impressions are totally dependent on our bodily state, and then, second, to see whether it entails that all are totally dependent. As for the first, it is clear that what is crucial is the meaning of “totally dependent” (Price’s phrase). Hume cannot have ignored that when double vision occurs then, in order to explain the impressions I have, reference must be made not only to the state of my organs but also the physical objects that confront me and with which I, or my body, causally interacts in the perceptual situation. After all, even Hume specifically says, “all the objects ... become double”, referring here to the objects of perception rather than the perceptions themselves. The point simply put is that the laws that causally explain the existence of the two impressions make reference to certain physical objects as initial conditions. Now, in the system of the vulgar, the case of veridical perception of the white paper, reference to the white impression occurs in the initial conditions. The white impression is part of the physical object and what the laws of perception do is show how this object, with this part, related thus and so to me who is in such and such a bodily state, has as its effect my awareness of the white impression. But the second image in the phenomena of double vision is not included in the initial conditions. Rather, it is one of the

218

Body, Mind and Self in Hume’s Critical Realism

effects. Hume’s point is not to deny the causal relevance of external objects in producing the doubled impressions I experience. In that sense, the second image is, of course, only partially dependent on my organs. What Hume is concerned with is locating the event which the second impression is in the total causal process of perceiving. And what he does establish is that this event is not among the events that appear in the initial conditions as constituting the physical object that provides the perceptual stimulus. As he puts this point, “... we do not attribute a continu’d existence to both these perceptions ...” (T, p. 211). The existence of the second impression is not explained by its being part of the series of sensible particulars that constitute the physical object mentioned in the initial conditions. The event, which the second impression is, appears only elsewhere in the process, at the point where the awareness of the impression also occurs, as the immediate causal upshot of my bodily state. Its existence is explained not in terms of its being part of a physical object but in terms of its being the immediate effect, like the awareness of it, of my bodily state. This is all Hume means when he uses the example of double vision to establish that some perceptions are “dependent on our organs”. Price takes Hume to be arguing for a stronger thesis than he actually argues for. Price sees that this stronger thesis is indeed not entailed by Hume’s premises, and therefore, rejects Hume’s case. But he misses the point that Hume was actually trying to make, and which actually is valid. As for why Price attributes the stronger thesis to Hume, this is simply another aspect of common tendency to misconstrue Hume in these passages as arguing for some form of subjectivism. But even if the existence of some impressions is, in Hume’s sense, dependent on our bodily state, can one conclude that all are so dependent? Clearly, not without additional premises. Hume explicitly introduces a second premise: the two impressions in double vision are, he says, “both of the same nature” (T, p. 211). How are we to interpret this? I think its context makes this clear enough. The context is that of a causal argument, that of an experiment. The premise that all experimental reasoning relies upon comes immediately to mind: like events have like effects and like events have like causes, or, as Hume puts it in the “rules by which to judge of causes and effects”, “the same cause always produces the same effect, and the same effect never arises but from the same cause” (T, p. 173). The two impressions are of the same nature; they are like events. Sound causal reasoning, that is, as Hume has argued, reasoning in accord with the rules by which to judge of causes, including rule 4 that like effects (causes) have

219

Three: Ordinary Things and Critical Realism

like causes (effects), should therefore assign like causes to the two impressions that occur in double vision, and locate them at like points in the causal process of perceiving. The vulgar, or the New Realists, may be able to offer an explanation in which one of the impressions in the double image is located as part of the object that causes the process. But it cannot do so for both. Since the two images are like, and must therefore have like causes, it follows that, if one cannot be located as part of the physical cause, then neither can. But the one is located not as part of the object perceived but as the immediate upshot of our bodily state. The other impression should be given the same location in the perceptual process. So both should be so located in the process as to have as the immediate cause of their existence the state of our sense organs – though of course the mediate or distal cause is the physical object. But further, these impressions are of the same nature as all other impressions. So all impressions should be located at the same point in the perceptual process. None should be treated as part of the physical objects perceived; all should be treated as the immediate upshot of our bodily state. Hume therefore concludes that “all our perceptions are dependent [in his, not Price’s, sense] on our organs, and the disposition of our nerves and animal spirits”.21 Causal reasoning has thus led Hume to develop a theory of the perceptual process in which all impressions are causally dependent on our bodily states. The facts of variation of size in perspective, the facts of apparent colours changing according to the conditions of illumination, the fact that objects appear differently when we have jaundice or when we are dizzy, and so on, can all be easily accounted for upon this theory. As Hume says, “This opinion [i.e., his theory of perception] is confirm’d” (T, p. 211) by all these other facts about perception. From all this “we learn, that our sensible perceptions are not possest of any distinct or independent existence” (T, p. 211). That is, we have justified the hypothesis of the system of the philosophers. And, contrary to what Price and Mandelbaum and others22 have argued, Hume’s reasoning here is sound causal reasoning, at least if we presuppose that, as Hume argued, his “rules by which to judge of causes and effects” are the best rules for causal inference. And in any case, the logic of Hume’s inference is of a piece with that which was later employed by the Critical Realists in their polemic against what they called “natural realism.” Thus, for one, the facts of error, illusion, and perspectival variance were used to argue against the position

220

Body, Mind and Self in Hume’s Critical Realism

of natural realism which identifies the datum with the physical object perceived; each of these facts was used to argue that appearances are one thing, the physical object another, and that, since some are not, none of the appearances are to be construed as part of the object perceived. Thus, Sellars pére writes: When I approach a house, what I perceive changes continuously; the house grows larger and I can see details which were not at first apparent... Now I know that it took at least a year to build this year and that it has a stability which contradicts these changes... Hence, instead of saying that things change, I assert that their appearance to me changes. But how can I reconcile this assertion with my other natural belief that, in perception, things reveal themselves as they are? When is the moment of my approach or departure that the thing supplants the appearance, and the appearance the thing. Since I am aware of no such mysterious moment, I may well be sceptical of its existence... The suspicion arises, as a consequence, that the individual perceives only the appearance of the thing and never the thing itself.23

What, then, is the status of the data, or, as Sellars calls them, percepts? Sellars goes on to point out how the percept depends not only on the physical object but also the position of the observer and, indeed, his past history, that is, his education, the associations that have built up, and the inferences that he has learned to make. All this “proves beyond doubt that the percept arises not in the object but at the brain”24: in other words, working within the system of the vulgar, causal inferences with respect to situations of error, etc., one arrives at the system of the philosophers in which all sense impressions depend causally for their existence upon the state of our organs. ...Natural Realism...is forced to testify against its own possibility and to furnish the basis for an explanation of that which occurs. The result is...a compromise: things are where we judge them to be, but we do not perceive them. Instead, we perceive the percepts causally connected with them, and these percepts are spatially and temporally more directly related to the brain than to the things with which ordinarily identify them.25

It may be that these arguments offered by the Critical Realists are unsound. It may further be that in spite of their intentions, the Critical Realists end up in a sort of subjectivism and scepticism as it is claimed so often that Hume does. But that is not the issue. The point here of presenting these arguments of the Critical Realists is not to defend them,

221

Three: Ordinary Things and Critical Realism

but rather to defend Hume, by showing that his discussion of the systems of the vulgar and the philosophers is, in its intention and in the main thrust of its argument, of a piece with the intention and thrust of argument in the Critical Realists. No one reads the latter as subjectivists or as sceptics. Nor should one so read Hume.

There is, of course, a sort of philosophical system similar to the system of the philosophers. This further system does involve a kind of ontological subjectivism, and therefore a kind of scepticism. Hume, however, rejects this further system and rejects also the subjectivism and scepticism consequent upon it. What I am referring to, of course, is what Hume calls the system of “modern philosophy,” which he discusses in Treatise, Bk. I, Part iv, section iv. The difference between the system of the philosophers and that of the modern philosophy lies is clear; so are the similarities. Both rely upon an argument from sense variation and other such phenomena to the conclusion that sensible qualities are not in the world of material objects. In the system of the philosophers the conclusion is that these qualities, which we do sense, but as they do not qualify the material objects that cause them and our awareness of them, are dependent for their existence on some other entity. In the system of the philosophers, the conclusion drawn from sense variation etc., and the premise that like effects have like causes (T, p. 211; T, p. 227), is that “all our perceptions are dependent on our organs, and the disposition of our nerves and animal spirits” (T, p. 211). In contrast, in the modern philosophy concludes that the so-called secondary qualities, colour etc., are “nothing but impressions in the mind…” (T, p. 226). Note well the difference: in the one case, the impressions are dependent for their existences on the state of our “organs,” in the other case the impressions are dependent on the “mind.” In the former case, the dependence is on the state of a material object, namely, our body, and there is moreover no ontological dependence of the impressions on one’s mind, for, as Hume points out, in the system of the philosophers, given the distinguishable – separable principle, and since every impression is distinguishable and may therefore “be consider’d as separately existent” (T, p. 207), it follows that “there is no absurdity in separating any particular perception from the mind, i.e., as Hume adds, referring in this context to his bundle view of the mind (T, p. 206), there is no absurdity “in breaking off [any particular perception] all its relations, with that connected mass of perceptions, which

222

Body, Mind and Self in Hume’s Critical Realism

constitute a thinking being” (T, p. 207). In other words, in the system of the philosophers, our sense impressions are not ontologically dependent on the mind. There is therefore no subjectivism and no scepticism implied in the system of the philosophers. In the modern philosophy, that is, in effect Locke’s philosophy, there is a dependence but it is contrasted to the dependence that obtains in the critical realism of the system of the philosophers: it is a dependence on a mind, an ontological dependence and not merely a causal dependence upon the state of a material object. As Hume puts it, “nothing we can conceive is possest of a real, continu’d, and independent existence…”: in the system of the philosophers sensible impressions can be conceived apart of minds, but in the system of the modern philosophy they cannot. Minds must therefore be different in kind in the modern philosophy from what they are in the system of the philosophers. In the latter they are bundles, in the system of the modern philosophy they therefore cannot be bundles. But the alternative to a bundle is a simple entity. In the system of the modern philosophy, the mind is a simple entity, which is to say it is a substance. Whatever quality we can imagine or conceive, we cannot think of or conceive it apart from a mind. (“… [A]n idea can be like nothing but an idea…,” as Berkeley put it [Principles, sec. 90].) Every quality is ontologically dependent for its existence on a substantial mind. There are no qualities that can be had by any non-mental substance; there is therefore nothing that an external object can be. And “when we exclude these sensible qualities [as qualities that can be exemplified by an external object] there remains nothing in the universe, which has such an existence” (T, p. 232). In the system of the modern philosophy, “we annihilate all these objects [that is “external objects”], and reduce ourselves to the opinions of the most extravagant concerning them” (T, p. 228). The system Hume is defending, that is, the system of the philosophers, does not lead to the annihilation of all external objects. Therefore no extravagant scepticism is consequent upon it. The system of the modern philosophy does imply such a scepticism. The difference between the two systems consists in this, that in the system of the philosophers the mind is a bundle, in the system of the modern philosophy is not a bundle but rather a simple entity, in effect a substance. But minds as simple entities do not exist: they are excluded by Hume’s appeal to PA. There is a philosophy with sceptical consequences in the neighbourhood of Hume’s Critical Realism. But that Critical Realism does not have those sceptical consequences. One can suppose that Hume falls

223

Three: Ordinary Things and Critical Realism

into subjectivism and scepticism only if one attributed to Hume a system in which minds are simple entities. But one can do that only if one ignored the fact that Hume adopts the empiricists’ PA which excludes simple minds: one can suppose that Hume falls into subjectivism and scepticism only if one ignores that fact that Hume is an empiricist. (4) The System of the Vulgar as False, Inevitable and Reasonable Hume has now arrived at the following position. There are two systems. One is that of the vulgar, consisting of the objects that we do or can perceive through our sensible experience of the world. The other is the system of the philosophers, which we arrive at, not through perception, but by causal reasoning. Both systems generate predictions about what impressions we will and would perceive under such and such appropriate circumstances. Up to a point, these predictions agree. But they disagree about the unsensed particulars. Both systems introduce these, but that of the vulgar asserts that they are coloured, etc., qualified in just the way that sensed particulars, i.e., impressions, are qualified, whereas the system of the philosophers asserts that the unsensed particulars differ from the sensed particulars in kind in not having the sensible qualities of the latter. The idea of a particular that is not sensed is an abstract relative idea, in effect a definite description. Both systems require such ideas. In the system of the philosophers the idea of an unsensed particular is the idea of the event that causes the impression I now have. The ideas in this idea are the abstract generic idea of an event, and the (abstract) relative idea of a cause. Since we do not sense the particulars that fall under this idea, and since they lack all the specific sensible properties of (sensed) impressions, we do not form a specific idea of them, only a generic idea. (Hume makes this point repeatedly in the Treatise; cf. p. 68, p. 211, p. 216, p. 227.) The concepts of event and cause appear, of course, in the causal principle that for every event there is an event that is its cause. It is this principle that justifies26 the introduction of the definite description, ‘the event that causes this impression.’ The causal principle has been confirmed in practice through the discovery of causes.27 Its acceptance is therefore justified. In turn it justifies introducing as hypotheses unsensed particulars, in both the systems of the vulgar and that of the philosophers. Failure to sense these particulars does not falsify the causal principle, owing to the presence of the existential quantifier.28 Coherence is achieved that would otherwise be lost if we rejected the hypotheses about the unsensed particulars: for that

224

Body, Mind and Self in Hume’s Critical Realism

would involve rejecting the causal principle. And achieving such coherence is a reasonable conclusion, since we have reasons for accepting, and none for rejecting, the causal principle.29 At the point of Hume’s argument that we have reached, he gives a brief characterization of how the mind arrives at the hypothesis of objects distinct from perception. The imagination tells us, that our resembling perceptions have a continu’d and uninterrupted existence, and are not annihilated by their absence. Reflection tells us, that even our resembling perceptions are interrupted in this existence, and different from each other. The contradiction between these opinions we elude by a new fiction, which is conformable to the hypotheses both of reflection and fancy, by ascribing these contrary qualities to different existences; the interruption to perceptions, and the continuance to objects (T, p. 215).

Imagination leads us to impose order or uniformity on our perceptions. Reflection, i.e., causal reasoning, tells us our perceptions are dependent on our organs just as our awareness of them is dependent on our organs, and that our impressions are therefore annihilated by their absence. But this causal reasoning turns upon the principle of determinism (“like causes have like effects, and conversely”) and this itself is a way of imposing uniformity. Reason thus takes up the project of the imagination to impose uniformity. A feature of the theory that reason proposes to explain our perceptual processes is the hypothesizing of objects distinct from our perceptions, which objects exhibit the same sort of continuance that the imagination imposes in our perceptions. The system of the philosophers thus defends a central idea of the imagination, that nature is uniform. At the same time, however, it also contradicts the system of the vulgar, when it asserts our impressions do not have a continued existence. As Hume puts it a little later, the system of the philosophers “at once denies and establishes the vulgar supposition” (T, p. 218). But meanwhile Hume elaborates the point that they hypothesis of the philosophers satisfies the demands of both reason and the imagination. Nature is obstinate, and will not quit the field, however strongly attack’d by reason; and at the same time reason is so clear in the point, that there is no possibility of disguising her. Not being able to reconcile these two enemies, we endeavour to set ourselves at ease as much as possible, by successively granting to each what ever it demands, and by feigning a double existence, where each may find something, that has all the conditions it desires. Were we fully

225

Three: Ordinary Things and Critical Realism

convinc’d that our resembling perceptions are continu’d, and identical, and independent, we should never run into the opinion of a double existence; since we shou’d find satisfaction in our first supposition, and wou’d not look beyond. Again, we were fully convinc’d, that our perceptions are dependent, and interrupted, and different we shou’d be as little inclin’d to embrace the opinion of a double existence; since in that case we shou’d clearly perceive the error of our first supposition of a continu’d existence, and wou’d never regard it farther. ’Tis therefore from the intermediate situation of the mind, that this opinion arises, and from such an adherence to these two contrary principles, as makes us seek some pretext to justify our receiving both; which happily at least is found in the system of double existence (T, pp. 215-6).

On the one hand, nature leads us to make our perceptual judgments. In these we are presented with a world of continuing entities. Reasoning cannot destroy the tendency to impute an order and uniformity to experience. On the other hand, reasoning clearly asserts that the continuity given in perception is false; for, perceptual error is a fact. If our perceptions never deceived us about continuity, if there never was error in our perceptual judgments, then there would be no contrariety in experience. But error is present, and this provides the motive to go beyond the system of the vulgar. However, if all we had were perceptions, with no tendency to recognize a uniformity in nature, reasoning would never lead us to the hypothesis of double existence: we can arrive at the latter only through considerations rooted in reasoning that is oblique and indirect and which presupposes the uniformity of nature. There is, on the one hand, the recognized dis-continuity in our perceptions, and there is, on the other hand, the tendency to impute uniformity to nature: the mind tries to have both. And, happily, it finds a way: the system of the philosophers. In a sense, this is a pretext for having our cake and eating it, too. But in fact, it is not a mere pretext: the only sort of pretext that will do, the only one that will in the long run actually satisfy us, is one that satisfies our best causal reasoning. We must not let Hume’s somewhat bantering tone mislead us. The use of ‘pretext’ suggests that any hypothesis will do to set the mind at rest as it endeavours to reconcile its apparently contrary aims; it suggests some pretence as a reason for adopting the hypothesis, rather than a real reason. But this is misleading: the pretext is to be one for adopting an hypothesis that is satisfying to the mind, and specifically, of course, to the mind as it conforms to the standards of the best causal reason. A bit earlier Hume made the same point that he is making her, but in a way less misleading in its suggestions: “In order to set ourselves at ease...we

226

Body, Mind and Self in Hume’s Critical Realism

contrive a new hypothesis, which seems to comprehend both these principles of reason and imagination” (T, p. 215). The only sort of pretext that will serve is one that satisfies reason, i.e., causal reason. Here, as elsewhere, Hume’s somewhat bantering tone has the tendency to mislead. Another example should be considered. Hume clearly regards as sound the reasoning that leads to the conclusion that sense impressions are dependent on our sense organs. But does he regard the conclusion to which it leads to be acceptable? I think the answer to this question is clearly affirmative. At the same time, a negative answer is invited when Hume proceeds to describe the resulting system as follows: “This philosophical system”, he tells, “. . . is the monstrous offspring of two principles, which are contrary to each other, and which are mutually unable to destroy each other” (T, p. 215). The use of ‘monstrous’ here is suggestive. The term ‘monster’ may mean an imaginary animal or one of huge or extraordinary size. It may also mean a person of inhuman cruelty or wickedness. But none of these uses in appropriate. The term ‘monstrous’ might also express indignation or wondering contempt, or, equally, it may express nothing more than is expressed by ‘exceeding’ or ‘wonderful,’ as when Swift wrote to Stella about a day in which “We have a monstrous deal of snow” (Feb. 8, 1710-11). But its conjunction with ‘offspring’ indicates Hume is using it in a metaphor playing on the common usage in which it refers simply to an animal or plant deviating in one or more of its parts from the normal type, an animal that is malformed relative to the norm. The system of the philosophers is monstrous, then, in no other sense than that the two principles each have normal offspring, and the system of the philosophers, an offspring of both, diverges from the norm of each: the point is no different from the one Hume made earlier when he argued that “this philosophical hypothesis has no primary recommendation, either to reason or to the imagination” (T, p. 212), or the point he makes later when he refers to reasoning “such as that of Sect. 2 [i.e., the section we are now discussing], from the coherence of our perceptions” as reasoning not “by any principle [i.e., a habit caused directly by an observed constant conjunction], but by an irregular kind of reasoning from experience . . .” (T, p. 242). Passmore suggests that reasoning cannot be used to defend the “monstrous” hypothesis. But, it seems clear, reason can, after all, defend the hypothesis. Passmore is very likely misled into thinking that the hypothesis of the philosophical system is indefensible by nothing more than Hume’s playful use of ‘monstrous’.30 The system of the philosophers, we thus see, does conform to the

227

Three: Ordinary Things and Critical Realism

best canons of causal reason. We ought, therefore, by those canons to reject the system which contradicts that of the philosophers; we ought, in other words, to reject the system of the vulgar. We must, as one Critical Realist put it, distinguish perception and knowledge.31 In perception we have thing-experiences but do not perceive physical things; we have only knowledge about, not experience of the latter, and epistemologically “perception . . . [is] subordinate to knowledge”.32 In spite of this, however, we continue to perceive, and cannot help it, or, in Hume's terms, we cannot but affirm the system of the vulgar. The explanation for this is in terms of psychology: perceiving is an automatic process that occurs prior to, and independently of, reflective thought. In Hume’s associationist psychology, what this amounts to is that the principle or mode of association that leads us to perceive as we do, i.e., to perceive the world as described in the system of the vulgar, is a principle irreducible to causal reasoning. But now Hume carefully proceeds to point out, the philosophical system is sufficiently similar to that of the vulgar, that we can revert to the latter without serious conflict. Another advantage of this philosophical system is its similarity to the vulgar one; by which means we can humour our reason for a moment, when it becomes troublesome and solicitous; and yet upon its least negligence or inattention, can easily return to our vulgar and natural notions. Accordingly we find, that philosophers neglect not advantage, but immediately upon leaving their closet, mingle with the rest of mankind in those exploded opinions, that our perceptions are our only objects, and continue identically and uninterruptedly the same in all their interrupted appearances (T, p. 216).

Nor is this reversion to the system of the vulgar irrational, in spite of the fact that we have every good reason to believe system to be false. How can we understand this? Both systems generate predictions about what impressions we will and would perceive under such appropriate circumstances, and, up to a point, these predictions agree. The radical disagreement concerns the nature of the un-perceived particulars. Both systems introduce unsensed impressions, but that of the vulgar asserts that they are qualified in just the way sensed impressions are qualified, whereas the system of the philosophers denies that the unsensed particulars are thus qualified. What the system of the philosophers does better than that of the vulgar is introduce greater coherence among the sensed impressions. The two systems disagree in the predictions they make. They disagree of course

228

Body, Mind and Self in Hume’s Critical Realism

about the unsensed particulars, but up to a point they agree in their predictions concerning sensed impressions. Nonetheless, this latter agreement is only up to a point: beyond that, they disagree, and it is the system of the philosophers that yields the better predictions, it is the system that introduces greater coherence. Even so, within the system of the vulgar, causal reasoning can be used to correct perceptual errors, thereby introducing a greater degree of coherence than is contributed by perceptual judgments alone. The point at which disagreement appears can thus be postponed. But when the causal reasoning is generalized to all domains, and so carried through as to lead to a unified theory of science, including a science of perception, then conflict does ultimately appear and it is the system of the philosophers that must in the end be judge superior. Thus, in terms of the logic of the situation, the system of the vulgar provides knowledge which, in its specific predictions, is imperfect relative to the knowledge of the system of the philosophers, and which, at the generic level, is less unified than the knowledge provided by the system of the philosophers.33 Still, it must also be recognized that within limits it is knowledge, and if one ignores what is of practical inconsequence, the specific nature of the unsensed particulars, then for everyday purposes, for most pragmatic interests, the knowledge provided by the system of the vulgar is quite adequate. That is why philosophers revert to, and even they can get by with, the system of the vulgar when they shift from contexts where their motivation is that of idle curiosity, the realm of “pure” reason, to contexts shaped by everyday pragmatic interests. To satisfy their pragmatic interests, we need knowledge of the world, knowledge of means. Often we do not need all the knowledge science can give us; less will suffice. (In order to cure diseases, the physician does not need to know the details of the processes by which penicillin kills bacteria; the more imperfect piece of knowledge, that it does kill germs, suffices to enable the physician to achieve his or her ends.) So long as we stick with the system of the vulgar, and proceed somewhat indolently (T, p. 269) in applying the general maxims of the world, that is, avoid applying in full strictness that causal reasoning of the philosopher, then we avoid fundamental incoherence. All that happens is that we leave such phenomena as that of double vision causally unexplained; but that can hardly affect our ordinary life. Only an unusually motivated person, e.g., a philosopher (empirical scientist) motivated by idle curiosity, the passion that moves the academic philosopher (T, p. 270), a love of truth for its own sake, would pursue causal reasoning beyond the point required by ordinary pragmatic interests.

229

Three: Ordinary Things and Critical Realism

But, so motivated, and so tracing out the causal connections, one arrives at the system of the philosophers, which, like the system of the vulgar, is internally consistent, but which are the same time contradicts the system of the vulgar. The system of the philosophers shows the extent to which the system of the vulgar is and is not a true account of unperceived reality and the extent to which it will and will not yield true predictions. In particular, among its consequences is the proposition that within the realm delimited by our practical concerns the system of the vulgar can be expected to yield correct predictions. The disinterested search after truth or curiosity which motivates the academic philosopher leads us34, as we apply the best rules of causal inference, to the rejection of the system of the vulgar, and to accept that of the philosophers, yet the latter then shows that, as a matter of fact, within the area of our practical interests we are justified in relying upon the system of the vulgar to provide us with knowledge of means. The system of the philosophers itself provides the argument that within the realm of everyday interests the system of the vulgar is rationally justified.35 Such, I am proposing, is the thrust of Hume’s argument.

Price has suggested that Hume’s discussion aims to show not that the system of the vulgar has a limited rationality, but that the system is wholly inadequate because it is inconsistent.36 But the system of the vulgar is not by itself inconsistent. Nor does Hume suggest that it is. Indeed, it is the system all of us normally use: even philosophers “...immediately upon leaving their closets, mingle with the rest of mankind in those exploded opinions, that our perceptions are our only objects, and continue identically and uninterruptedly the same in all their interrupted appearances” (T, p. 216). The system of the vulgar, that is, the world as given to us in our perceptual judgments, has its roots in our imagination, and these judgments of (the mere) imagination are not reducible to causal reasoning. So long as we stick with this system, and proceed somewhat indolently (T 269) in applying the general maxims of the world, that is, avoid applying in full strictness the causal reasoning of the philosopher, then we get into no fundamental incoherence. I say ‘fundamental’ because particular perceptual judgments might disagree, as indeed they often do, and we have techniques based on causal reasoning for setting them aright, but this does not mean that there is a defect in the framework as such. But there is nothing in the ordinary flow of the world that compels or motivates the ordinary person to use the best causal reasoning to resolve his or her

230

Body, Mind and Self in Hume’s Critical Realism

quotidian misperceptions, nor anything that compels him or her to causally explain such phenomena as double vision which hardly distract one from the customary flow of one’s perceptual experience. Only an unusually motivated person, e.g., a philosopher (= empirical scientist) motivated by something like idle curiosity would pursue causal reasoning beyond the point required by ordinary pragmatic interests. Within these limits, then, the system of the vulgar is quite consistent, and, what is of present importance, Hume sees it as consistent. Price is therefore wrong to suppose that Hume aims to find an inconsistency in the system of the vulgar. The conflict that Hume discovers is in neither the system of the vulgar by itself nor the system of the philosophers by itself but between the two, one of which we must accept (that of the vulgar) and one of which one best causal reasoning tells us that we ought to accept (that of the philosophers). In effect this settles the argument against those who charge Hume with scepticism that Hume is not a subjectivist, but a Critical Realist. Moreover, Hume holds that there is only one system, that of the philosophers, that we ought rationally to accept. Some (like Richard Popkin) hold that Hume falls into a Pyrrhonism, because Hume (it is argued) holds that there is the system of the vulgar to which our human nature takes us, and the system of the philosophers to which reason takes us, that these contradict one another, and that, since there is no way of resolving the conflict, we can only conclude that there is an incoherence which implies we know nothing: as natural beings we do of course believe as we must, but those beliefs are all irrational. However, Popkin’s reading is wrong: we ought rationally to accept the system of the philosophers, not that of the vulgar. There is thus nothing contrary to reason that would force Hume into a Pyrrhonist conclusion. We cannot conclude that Hume is a sceptic with regard to the senses, at least not on the basis of the argument so far examined. But Hume’s discussion continues of course for some time, and he himself raises a variety of sceptical issues. Before we turn to these, we should explore the world of the philosophers in greater detail. (5) The World of the Philosophers Hume has established that it is reasonable to suppose that there are external physical objects which cause our impressions, and that the latter are dependent for their existence upon the state of our organs.37 He has also established that the reasoning which justifies this conclusion is not direct

231

Three: Ordinary Things and Critical Realism

cause-effect inference but the imagination inferring “from custom in an indirect and oblique manner” (T, p. 197). At this point Hume turns briefly to make a couple of points about the detailed descriptions which the system of the philosophers gives of the physical objects which it asserts to be the cause of our perceptions. These, too, he tells us, are not dependent upon direct cause-effect inferences. There are other particulars of this system [of the philosophers], wherein we may remark its dependence on the fancy in a very conspicuous manner (T, p. 216).

Now, the use of ‘fancy’ here may suggest playfulness or whim, the very contrary of reason. It may thereby tend to support the sceptical reading of Hume. It is true that the fancy may indeed, for Hume, produce false beliefs. In speaking of the miser who receives delight from his money, Hume explains that what motivates him is the pleasure of its use, not its actual use but its imagined use. “Since he cannot form any such conclusion in a way of reasoning concerning the nearer approach of the pleasure, ’tis certain he imagines it to approach nearer, whenever all external obstacles are remov’d, along with the more powerful motives of interest and danger, which oppose it” (T, p. 314; his italics). Here reason and imagination are contrasted, with the latter producing a belief which the former cannot support. Hume moreover goes on to ascribe the belief or judgment to the fancy: “we judge from an illusion of the fancy, that the pleasure is still closer and more immediate” (T, pp. 314-5; his italics). Here fancy and imagination are equated and are taken to be the source of an illusion which reason cannot support. Yet, as we know, for Hume causal reason is also at times characterized as a species of imagination, when, for example, it is contrasted to the a priori reason of mathematics (T, p. 97); so that to characterize an inference as a product of the imagination is not necessarily to condemn it as unsound or irrational from the point of view of satisfying our curiosity, that is, achieving, so far as we can, the discovering of matter-of-fact truth. Moreover, Hume has earlier, in discussing “the probability of causes” (T, I, III, xiii), attempted a psychological explanation, in terms of associationism, of how the mind comes to form the contrary hypotheses that constitute probability estimates of causes and which are the basis upon which the rules of eliminative induction go to work in order to discover which among the possible alternatives is the true cause (T, pp. 133-140). So Hume is here giving a psychological description of the mind’s working when causes are inferred

232

Body, Mind and Self in Hume’s Critical Realism

in conformity with Hume’s “rules by which to judge of causes and effects”; and that is to say that Hume is here giving an explanation of what he argues is sound causal reasoning. At the same time, Hume also ascribes the inference here to the fancy. In cases where causes are only probable, an A is followed sometimes by a B and sometimes by a C, and here Our past experience presents no determinate object; and as our belief, however faint, fixes itself on a determinate object, ’tis evident that the belief arises not merely from the transference of past to future, but from some operation of the fancy conjoin’d with it (T, p. 140; his italics).

Thus, for Hume inferences of the fancy need not be unsound; some in fact constitute the best causal reasoning there is! And when Hume ascribes certain features of the system of the philosophers to the fancy, we cannot conclude that those inferences are irrational. We must read Hume’s comments on the system of the philosophers with this caution in mind. Hume has now argued for the system of the philosophers, and at this point proceeds to make two brief comments (T, pp. 216-7) that address the question: what, in detail, do we know about the nature of the objects which the system of the philosophers asserts to be there? The two comments Hume makes are relatively brief, but he has in fact commented upon the subject throughout earlier portions of the Treatise, and the texts that we are now considering from Book I, Part iv, Sec. ii, must be looked at in the light of these earlier remarks, and also, one should add, in the light of later remarks. Perhaps the most important point to be recalled is that any idea we form of such objects must conform to Hume’s theory of ideas, or, to put it more accurately, any concept of such unperceived objects must conform to the Principle of Acquaintance (PA), that we can have no idea without an antecedent impression. It does not follow, however, that for every idea we must be acquainted with the object to which it applies. Moreover, we might have a general idea of the object without having an idea of the object in all its specificity. The general idea would be inadequate to the object, of course, but not false of the object. Thus, knowing all persons have fathers we can form the relative idea, “the father of Jones”. This is a general idea. It is also a relative idea; it refers to an individual by means of the relation in which the individual stands to another. Moreover, it is an idea which applies to exactly one individual. On the other hand, it is not an idea of that

233

Three: Ordinary Things and Critical Realism

individual as the specific person that he or she is: from this general idea of Jones’ father we can infer only that he has the properties mentioned in the idea, and from it alone can learn nothing of the other properties Jones’ father might have. To that extent, the general and relative idea we have formed of Jones’ father is abstract when compared to the concrete ideas we form of those persons with whom we are acquainted; the latter, unlike the former, are ideas which are specifically adequate to the objects. Hume has made clear earlier in the Treatise that we can form such general and relative ideas of unperceived objects, without our having any idea of what specifically those substances are like. Hume lays down the general rule that: . . . since nothing is ever present to the mind but perceptions, and since all ideas are deriv’d from something antecedently present to the mind; it follows, that ’tis impossible for us so much to conceive or form an idea of any thing specifically different from ideas and impressions. Let us fix our attention out of ourselves as much as possible: Let us chase our imagination to the heavens, or to the utmost limits of the universe; we never really advance a step beyond ourselves, nor conceive any kind of existence, but those perceptions, which have appear’d in the narrow compass. This is the universe of the imagination, nor have we any idea but what is there produc’d (T, pp. 67-8).

But to this he hastens to add a qualification: The farthest we can go towards a conception of external objects, when suppos’d specifically different from our perceptions, is to form a relative idea of them, without pretending to comprehend the related objects. Generally speaking we do not suppose them specifically different; but only attribute to them different relations, connexions and durations. But of this more fully hereafter (T, p. 68; his italics).

And then Hume gives a reference to Part iv, Sec. ii, the section we are now examining in detail. This qualification to the general rule about our ideas therefore applies to the external objects of the system of the vulgar and of the philosophers. The external objects of the vulgar are, of course, not supposed to be specifically different from our perceptions. But the external objects of the philosophers are supposed to differ specifically from our perceptions. What Hume argues is that we can form ideas of the external objects of the philosophers, but that these ideas are relative, and, as such, do not enable us to comprehend the related objects, that is (I take it), comprehend them in their own full specificity.

234

Body, Mind and Self in Hume’s Critical Realism

Hume makes the same point again later on in the Treatise, explicitly referring in a footnote to the passage just quoted: ...remember, ’tis impossible our idea of a perception, and that of an object or external existence can ever represent what are specifically different from each other. Whatever difference we may suppose betwixt them, ’tis still incomprehensible to us; and we are oblig’d either to conceive an external object merely as a relation without a relative, or to make it the very same with a perception or impression (T, p. 241).

Hume here speaks of ‘relations’ and ‘relatives’ rather than of ‘relative ideas’, though the latter clearly refers to what ‘relations’ refers to in this second passage. This is helpful, for Hume says little indeed about “relative ideas” and rather more about “relations”. We must infer from his discussion of “relations” what he means by “relative ideas”. The “relations” and “relatives” terminology of such philosophers as Locke, and Hume’s remarks fit in with the latter’s discussion. The Port Royal Logic is of little help in illuminating Hume: it gives two brief references to relations, neither of which are relevant to Hume's discussion. Watts and Crousaz have rather more to say, though neither addresses himself in detail directly to “relative ideas”. Both their accounts derive largely from Locke. Hume’s own account is complicated by his peculiar doctrine of abstraction. It suffices to recognize that the standard account of relatives, or relative ideas, found in Watts and Crousaz, and deriving from Locke, finds its place in the Humean analysis of ideas.38 The relative idea of a father thus finds a perfectly natural place within the Humean framework which thereby recognizes the possibility of one’s having knowledge by description even though knowledge by acquaintance is absent. Knowledge by description is possible by means of relative ideas. Moreover, these relative ideas conform to Hume’s basic principle PA. Of course, if we have no acquaintance with the object, if it is not actually perceived, then we will have no adequate idea of it. Nonetheless, it does not follow that we can have no idea of it: for we can form a relative idea. In particular, we have the relation of cause and effect. Using the relative idea of cause and given a certain perception, we can form the idea of the (external) cause of this (impression), where ‘this’ refers to or names the impression, in just the way we can form the less generic idea of the father of, say, Jones. Because the generic notion of cause is derived from experience, from perception, the external causes of perception will not be conceived as generically different from the objects

235

Three: Ordinary Things and Critical Realism

perceived. But given this generic sameness it does not follow that the two must be, or be conceived as, specifically the same. It is in the light of these considerations that we must interpret Hume’s two comments on particular features of the system of the philosophers. The first is this: We suppose external objects to resemble internal perception. I have already shown, that the relation of cause and effect can never afford us any just conclusion from the existence of external continu’d objects: And I shall further add, that even tho’ they cou’d afford such a conclusion, we shou’d never have any reason to infer, that our objects resemble our perceptions. That opinion, therefore, is deriv’d from nothing but the quality of the fancy above explain’d, that it borrows all its ideas from some precedent perception. We can never conceive anything but perceptions, and therefore must make everything resemble them (T, p. 216).

We can, I propose, explain this as follows. The habits of cause-effect inference arise from our regularly experiencing one sort of perceived object to be followed by perceived objects of another sort. But ex hypothesi we do not experience the external objects. This much we have seen previously, though we also recall that this conclusion is qualified so as to admit that an inference to external objects can still be justified, as Hume put it earlier, by an “inference [which] arises from the understanding, and from custom in an indirect and oblique manner” (T, p. 197), or, as Hume puts it later, “by an irregular kind of reasoning from experience”, such as reasoning “from the coherence of our perceptions” (T, p. 242). Imagination or the fancy must supplement the simple cause-effect inference habits if we are to arrive at the hypothesis of the external objects of the system of the philosophers. In the passage we are now considering Hume is adding the further point that, even if we do get to external objects, then the same qualities of simple causal inference that prevent it from justifying the existential hypothesis also prevent it from justifying any claim that the hypothesized external objects resemble our perceptions. It is therefore the imaginative processes involved in the inference from the coherence of our perceptions that give rise to the opinion that external objects resemble perceptions – even for resemblance of the most generic sort. Moreover, we cannot but think of external objects as somehow or in some respect (even if it be only the most generic) resembling our perceptions, for the imagination can form only in such ideas as conform to PA: “the fancy...borrows all its ideas from some precedent perception”. Note,

236

Body, Mind and Self in Hume’s Critical Realism

however, that such suppositions of resemblance need not be unjustified simply because they are not justified by simple cause-effect inferences: the suppositions can be justified by reasoning from the coherence of our perceptions. Moreover, we may suppose these external objects that resemble our perceptions also cause those perceptions; it does not follow that the entities, once imagination leads us to suppose they exist, cannot then also be supposed to stand in cause-effect relations with our perceptions. Hume’s second comment on the system of the philosophers is this: As we suppose our objects in general to resemble our perceptions, so we take it for granted, that every particular object resembles that perception which it causes. The relation of cause and effect determines us to join the other of resemblance: and the ideas of these existences being already united together in the fancy by the former relation, we naturally add the latter to complete the union. We have a strong propensity to complete every union by joining new relations to those which we have before observ’d betwixt any ideas, as we shall have occasion to observe presently (T, p. 217).

And here Hume adds a reference to Book I, Part iv, Section v (T, p. 237ff), which is, it is clear when we turn to the latter, a reference to his discussion of the taste of a fig. In the comment we are now examining, Hume attributes a certain tendency of the mind to conceive the external causes of our perceptions not merely as resembling in general those perceptions which they cause, but also in another way, namely, as specifically similar to the perceptions they cause. The mind does not rest content with a relative idea of those external causes but tends to form an absolute idea of them. Now, the “indirect and oblique” reasoning that leads us to ascribe existence and a resemblance in general to the external causes of perception is justified, but Hume does not say this further tendency to replace a relative by an absolute idea is justified. This may indeed be a characteristic of our thinking about the external objects which the system of the philosophers believes to exist, as Hume here suggests. But such a characteristic movement of thought may well be contrary to reason, and one reason ought to resist. Hume does not assert that this natural movement of thought is a rationally justified movement of thought. It might well be that rationally we ought to rest content with the relative idea of external objects and to resist the tendency to substitute for that an absolute idea, however natural is the tendency of thought to make that substitution. And, in fact, if we turn to his later discussion of the taste of

237

Three: Ordinary Things and Critical Realism

the fig to which he refers us, we can see clearly that he judges such a movement of thought as not reasonable. The taste of a fig can neither be said to be in only one part of the fig nor be said to be in every part (T, p. 238). Not the former, since experience convinces us every part has that taste. But if the taste is in every part then it, like colour, will have to be figured and extended, which is also contrary to experience. So the second alternative is also excluded. In order to reconcile these two options we reply upon the supposition, summarized in the scholastic dictum, totum in toto and totum in quaelibet parte (T, p. 238), that the taste exists within the body, but whole and entire in every part. This, however, does not eliminate the difficulty, only obscures it, for it “is much the same, as if we shou’d say, that a thing is in a certain place, and yet is not there” (T, p. 238). The problem arises because we attempt to locate the taste in a place, specifically in the place where the fig is: “... this absurdity proceeds from our endeavouring to bestow a place on what is utterly incapable of it ...” (T, p. 238). Why do we thus endeavour to locate the taste in the fig? To answer this we might glance at a discussion Hume has elsewhere of the example of taste,39 only in this case the taste of wine, which he is comparing to the feelings that mark aesthetic taste. About these Hume tells us that: Though it be certain that beauty and deformity, more than sweet and bitter, are not qualities in objects, but belong entirely to the sentiment, internal or external, it must be allowed, that there are certain qualities in objects which are fitted by nature to produce those particular feelings.40

Both feelings of beauty and such tastes as those of bitter and sweet are not qualities of the objects to which they are attributed, but rather are caused by qualities in those objects. The mind recognizes this causal connection but then adds to it a conjunction in place: “…’tis a quality which I shall often have occasion to remark in human nature,...that when objects are united by any relation, we have a strong propensity to add some new relation to them, in order to complete the union” (T, p. 237). What Hume’s discussion of taste makes clear is that this propensity is one that sometimes goes contrary to reason, and is therefore one that reason at times ought to resist. Now, reasoning from the coherence justifies the inference to causes of one’s sense impressions or perceptions, but only to those causes conceived as only generically similar to the impressions they cause. But to this relation of cause and effect human nature tends to add the relation of

238

Body, Mind and Self in Hume’s Critical Realism

resemblance. We will thus tend to suppose that the external object that causes our perception resembles the perception. The tendency will be not merely to suppose a generic resemblance but also a specific resemblance. And, as he has said earlier, when discussion “the effects of other relations and other habits [on belief]” (Bk. I, Pt. iii, Sec. ix), “resemblance, when conjoin’d with causation, fortifies our reasonings” (T, p. 113). The fig causes the taste; we will then suppose it is by virtue of a quality resembling the taste that the fig causes that perception is us, and moreover, we will suppose that this quality is not merely conjoined with the other qualities of the fig but that it is both everywhere within the fig and wholly in every part. To what extent is one justified in supposing an external object resembles the perception it causes? Not necessarily as far as the mind tends to think. And, in fact, only so far as is justified by reasoning from experience, that is, the “irregular reasoning from experience” based on considerations of causal coherence that we have already examined. The point is that this reasoning may well and indeed does run contrary to the human tendency to sometimes attribute a greater resemblance than reason can justify. The “irregular reasoning from experience”, i.e., from coherence, concludes with the attribution of some, at least generic, resemblance between the external object and the perception it causes. At the same time, the same sort of reasoning leads us also to conclude that each of the specific kinds of entities we perceive (tastes, smells, colours, etc.) all exist dependently upon our sense organs. The latter means that no external object can have any specific resemblance to any perception it causes. Reason must therefore resist the tendency of the mind to attribute such specific qualities to the external objects. A natural tendency of the mind leads us to move from a relative idea of external objects to an absolute idea, but reason does not justify this move, and ought to resist it. The system of the philosophers as Hume has now described and defended it, is, we recognize, of a piece with the world-view proposed later by the Critical Realists. The physical object causes or perceptions, but so far as we know, that is, so far as legitimate causal inference will take us, the physical object lacks the secondary qualities that we experience in perception. As with Critical Realism, so in Hume’s system of the philosophers “the physical world cannot be like our ideas” and the position therefore “involves a relinquishment of all attempts to picture the physical world”.41 Even so, says Hume, there is still a tendency to try so to picture the physical object, though no causal inference produces that conclusion, and though it is the opposite conclusion that legitimate causal inference

239

Three: Ordinary Things and Critical Realism

justifies, the conclusion that entities exemplifying the secondary qualities are causal dependent upon the state of our sense organs. The Critical Realists made this point, too: “we usually take it [ = the datum or percept, which exemplifies the secondary qualities] to be physical, since its qualities are what we usually mean by physical qualities, and we inevitably feel that they belong to our object”42; that is, as they would put it, we often uncritically fall into our ordinary “natural realism.” This is not, however, for them, as for Hume, the final judgment at which reason arrives. “Common sense may indeed give a snap judgment upon it and insist upon identifying the datum with the object; but there is no reason why common sense, which is merely primitive philosophy, should have the final decision”.43 So reason should resist, or at least discount when it must, the tendency of common sense and the vulgar to identify the cause of our perceptions with objects that have the secondary qualities of those perceptions. Nor do we have here one of those contradictions which those who offer a sceptical reading love to find Hume believing we fall into. The suggestion would be that the best causal reasoning leads to the conclusion that physical objects lack secondary qualities, but that there is also a natural tendency of the imagination to attribute such qualities to physical objects. There is thus a conflict between causal reason and the imagination the inevitability of which entails a scepticism from which nature alone and not reason permits an escape. This tale will not do, however. For it works only if the influence of the imagination is either sound or at least natural in the sense of inevitable. But it is not. Not only is the inference unsound according to what Hume has argued are the canons of the best causal reasoning, but there is no reason why the disciplined understanding cannot, when it pays attention, avoid the inference. It is only the inattentive and indolent mind that slips into the error of the imagination that physical objects resemble perceptions. The erroneous inference of the imagination is indeed natural but only in the sense in which a disease is natural (T, p. 226).44 Once again the sceptical reading simply is not implied by the texts. Once again we arrive at the conclusion that Hume is defending Critical Realism. This, I suggest, without being able to deal with many other relevant aspects of the text, is the main thrust of Hume’s argument in the section “Of scepticism with regard to the senses” (T, I, iv, ii). If so, then at this point Hume has arrived at a position that is more akin to that of the Critical Realists than it is to scepticism.

240

Body, Mind and Self in Hume’s Critical Realism

(6) Conclusion ...[during the reign of Charles II] there arose in England some men of superior genius who ... drew on themselves and on their native country the regard and attention of Europe. Besides Wilkins, Wren, Wallis, eminent mathematicians, Hooke, an accurate observer by microscopes, and Sydenham, the restorer of true physic; there flourished during this period a Boyle and a Newton; men who trod with cautious, and therefore the more secure steps, the only road which leads to true philosophy. Boyle improved the pneumatic engine invented by Otto Guericke, and was thereby enabled to make several new and curious experiments on the air, as well as on other bodies: His chemistry is much admired by those who are acquainted with that art: His hydrostatics contain a greater mixture of reasoning and invention with experiment than any other of his works; but his reasoning is still remote from that boldness and temerity which had led astray so many philosophers. Boyle was a great partisan of the mechanical philosophy; a theory which, by discovering some of the secrets of nature, and allowing us to imagine the rest, is so agreeable to the natural vanity and curiosity of man.45

My own sense is that a statement such as this could not have been written by someone who rejected as unworthy of belief the framework of the new mechanical philosophy of Boyle, Locke and Newton. To be sure, Hume does not accept the account dogmatically – the mechanical philosophy has led us to the discovery of “some of the secrets of nature,” for the rest it only “allows us to imagine.” But he clearly does accept it. Now, the standard reading of Hume as a sceptic does not make this possible. This reading divides Hume against himself. On the one hand, Hume is a sceptic; there are no rational foundations for our beliefs, and therefore, rationally, we ought to reject all belief, suspend judgement. On the other hand, as natural, animal beings we cannot suspend all judgment – we must believe – and so, irrationally, we believe. In particular, then, Hume’s commitment to the scientific world picture of the new mechanical philosophy must be irrational, and in no way superior, rationally, to the superstition and false philosophy that he thought and hoped it was replacing. Although this may fit the outdated picture of Hume as a person aiming, no matter the cost, at fame and literary reputation, it fits ill with the picture that we now accept of a person dedicated to the pursuit of truth in both philosophy and history. We have argued that the standard account of Hume as a sceptic redeemed only by his irrational naturalism is mistaken; he attempts, to the

241

Three: Ordinary Things and Critical Realism

contrary, to offer a rational defence of naturalism and natural science.46 In particular, we have argued that, in the famous, perhaps notorious, discussion of “scepticism with regard to the senses” Hume is not defending a sceptical view of reason but rather a form of critical realism, that is, the position that, details aside, had been developed by the mechanical philosophy of Locke, Boyle and Newton.47 The traditional, sceptical, reading gains much of its force by selectively quoting out of context, or, perhaps more charitably, by ignoring the context in which some well known Humean comments occur. Thus, it is often suggested that Hume accepts none of the philosophical theories that he examines, neither the system of the vulgar nor that of the philosophers, nor any other – this rejection of all philosophical theories is Hume’s scepticism; in favour of this one might cite the passage in which Hume asserts that “’Tis impossible upon any system to defend either our understanding or senses...” (T, p. 218). But what comes after the ellipsis? The whole sentence goes like this: ’Tis impossible upon any system to defend either our understanding or senses; and we but expose them farther when we endeavour to justify them in that manner.

Note what Hume goes on to say: we get into more and more trouble if we attempt to defend either our understanding or our senses “in that manner,” that is, in the way indicated in the first part of the sentence, by appeal to a system. This does not preclude some other sort of justification, and, indeed, the force of the passage is to suggest that another way of justifying either our understanding or our senses might well be available. One cannot support a sceptical reading of Hume by ignoring this clear indication that another sort of justification will be advanced. This other sort of justification will not be one based upon a system; it will not be anything of the sort that Descartes might have attempted. For such a justification just leads to a deeper and deeper scepticism. Not that anyone was ever genuinely a sceptic; nature provides a remedy. “Carelessness and in-attention alone can afford us any remedy” (T, p. 218). Taking scepticism seriously would lead inevitably to “melancholy and delirium,” but “nature...cures me” (T, p. 269). This is often taken as further textual justification that reason cannot escape the destruction that leads to Pyhrronnism. But once again, one must note that there is a context in which this occurs. Hume does not end here, but goes on to argue that,

242

Body, Mind and Self in Hume’s Critical Realism

while nature inclines me to accept the system of the vulgar, to acquiesce in this is a matter of “the sentiments of my spleen and indolence” (T, p. 270). To these sentiments he contrasts “curiosity” (T, p. 270), which passion he discusses in further detail in Book II of the Treatise (Bk. II, Part iii, Sec. x). This is the passion that Boyle and the other defenders of the mechanical philosophy attempted to satisfy. It is also a passion which motivates the philosopher to go beyond what satisfies the sentiments of the spleen and indolence. Thus led, the philosopher passes beyond the system of the vulgar to that of the philosophers (T, p. 271). The justification for this system is not a priori, as Descartes insisted it must be, but pragmatically, by the fact that it is a means to an end. The end is that provided by the motive of curiosity; philosophy is justified by virtue of the fact that it is, so far as we can tell, the best way to satisfy our disinterested concern to discover matter-of-fact truth.

Hume’s critical realism, then, amounts to the claim that ordinary things are really a set of objects knowledge of which derives from science. The objects in this world are not given to us in sense, nor therefore have we any experience of the properties of these objects: all that we can know is the structure of such objects and the structure of the relations among those objects. Since all we know is structure, it perforce follows that this is all we can now about the objects that produce in us our impressions and ideas and, more generally, our conscious states. Hume does not speak at any length about these objects. It is evident enough from things he says, almost in passing, in the Treatise, that he has certain beliefs that do go beyond the simple claim that there are such objects and that we can reasonably form the belief that there is some structure or other that these objects do exemplify: Hume clearly has beliefs more detailed than this about the sort of structure had by these hypothetical objects of the world of the philosophers. Not surprisingly this world as Hume sketchily describes it is much of a piece with the ideas about such a world that were common to the Newtonians, and in fact which the Newtonians, in outline at least, shared with the Cartesians: it is the world as described by the new science. Hume clearly accepts the position that our sensations and ideas have physiological causes. Included in the story are, on the one hand, external stimulations of our peripheral organs of sensation, and, on the other hand there is internally the motion of “animal spirits”:

243

Three: Ordinary Things and Critical Realism

Original impressions or impressions of sensation are such as without any antecedent perception arise in the soul, from the constitution of the body, from the animal spirits, or from the application of objects to the external organs. (T, p. 275)

Or, as he put it at another point: An object, that presses upon any of our members, meets with resistance; and that resistance, by the motion it gives to the nerves and animal spirits, conveys a certain sensation to the mind. (T, p. 230)

There are other descriptions of the hypothetical physiological causes of our mental processes and in particular the association of ideas. To be sure, Hume takes it to be clear that the explanation of these processes is in terms of the laws of association. But he also allows himself on occasions to join with others in physiological speculation, speculation which he himself characterizes as “specious and plausible.” (T, p. 60)48 He suggests that “’Twou’d have been easy to have made an imaginary dissection of the brain, and have shewn, why upon our conception of any idea, animal spirits run into all the contiguous traces, and rouze up the other ideas, that are related to it.” (T, p. 60) The genuine explanation by confirmed laws is the explanation in terms of the laws of the association of idea, but one can speculate about the underlying physiological causes. The latter is merely “imaginary” when contrasted to any real explanation in psychological terms; that is why it is “specious”; but at the same time it is not mere fantasy, and that is what makes it “plausible.” Here is the speculation about the physiological causes underlying the association of ideas. ...as the mind is endow’d with a power of exciting any idea it pleases; whenever it dispatches the spirits into that region of the brain, in which the idea is plac’d; these spirits always excite the idea, when they run precisely into the proper traces, and rummage that cell, which belongs to the idea. But as their motion is seldom direct, and naturally turns a little to the one side or the other; for this reason the animal spirits, falling into the contiguous traces, present other related ideas in lieu of that, which the mind desir’d first to survey. (T, pp. 60-61)49

There is a physiological basis to the passions also, Hume elsewhere notes: he mentions the “emotion [which] poetical enthusiasm may give to the spirits” (T, p. 630), and suggests that a passion experienced in an aesthetic context “has no other than the agreeable effect of exciting the spirits, and

244

Body, Mind and Self in Hume’s Critical Realism

rouzing the attention.” (T, p. 631) These remarks are scattered, but occur with enough regularity in the Treatise that taken together they are sufficient to make it clear that, though the specific details of the physiological theory are purely speculative and “specious,” Hume does assume that, there is, speaking generically, a physiological basis for the confirmed (and therefore not “specious”) regularities which are the laws of the association of ideas.50 It is perhaps worth noting that Hume refers to the same sort of physiological speculations in the first Enquiry, where he discusses the case of one’s volition causes one’s arm to move in some way. That action is the effect of the volition, but there is an intervening physiological process in which the direct effect of the volition is a change in one’s brain state: We learn from anatomy, that the immediate object of power in voluntary motion, is not the member itself which is moved, but certain muscles, and nerves, and animal spirits, and, perhaps [note this ‘perhaps’], something more minute and more unknown, through which motion is successively propogated [that is, the causal process which we now know to be more than matter in motion but also includes electrical impulses, various chemical reactions, and much more], ere it reach the member itself, whose motion is the immediate object of volition ... Here the mind wills a certain event: Immediately another event, unknown to ourselves, and totally different from the one intended, is produced: This event produces another, equally unknown Till, at last, through a long succession, the desired event is produced.51

Hume is careful to note that there are limitations on the power of volitions to move bodily parts: “the will [has] an influence over the tongue and fingers, not over the heart or liver.”52 There is not much detail in Hume’s account of the structures that constitute bodies in the system of the philosophers. But there are such structures. The real world – the one that is there, according to the critical realist, the one that underlies what we do experience, but which is also the world we cannot experience – this world is not devoid of form. And in principle at least, that form can be explored by science to locate its more specific details.

To understand the Humean mind we need to understand the mind-body relation, or at least Hume’s account of that relation. We now have a good idea of Hume’s account of the body side of this connection, Hume’s view

245

Three: Ordinary Things and Critical Realism

of body is evidently not that of a sceptic, but rather that of a critical realist who takes seriously the scientific theorizing of his age. Let us now, therefore, turn finally to the other side of this relation, and get on more directly with the task of becoming clear on the nature of Humean minds.

Endnotes to Chapter Three

1. For the impact of More on Newtonian philosophy, see the classic study of E. A. Burtt, The Metaphysical Foundations of Modern Science. 2. Henry More, An Antidote against Atheism (1657). 3. Book III shows how much More was still part of an earlier pre-scientific era: it deals with the evidence for witchcraft and fairies. 4. Henry More, The Immortality of the Soul (1662). 5. “General Scholium,” in The Leibniz-Clarke Correspondence, p. 169. 6. Ibid. 7.Cf. F. H. Anderson, “The Influence of Contemporary Science on Locke’s Method and Results.” 8.Cf. Thomas Sydenham, His Life and Original Writings, ed. K. Dewhurst. 9.For greater detail on the interpretation of Hume that follows, see F. Wilson, The External World and Our Knowledge of It: Hume’s Critical Realism: Exposition and Defence. 10.D. Hume, Abstract of a Treatise of Human Nature. 11.For greater detail on Hume’s account of body in the sense of ordinary things, see F. Wilson, The External World and Our Knowledge of It: Hume’s Critical Realism: An Exposition and Defence. 12.Cf. F. Wilson, “Was There a Prussian Hume?” 13.For more details on this point, see F. Wilson, “Hume’s Fictional Continuant.” 14. Price, Hume’s Theory of the External World. 15. Livingston, Hume’s Philosophy of the Common Life, p. 13, p. 14, misreads the

246

Body, Mind and Self in Hume’s Critical Realism

causal argument as one that has phenomenalism and subjectivism as its conclusion. 16.Livingston, ibid., p. 19, on the basis of passages such as this, argues that for Hume causal reason must limit itself to the popular system or system of the vulgar on pain of not being able to understand experience itself. In fact, Livingston holds, p. 25, the structure of common life is the a priori structure of philosophy. If philosophy does not so restrict itself, it ends up in Pyrrhonism, and such Pyrrhonism is the mark of any attempt to reason to transcend its own a priori structure. This reading makes sense, however, only if one construes the causal argument Hume employs as leading to phenomenalism. But if we are correct, the causal argument has no such paradoxical consequences. If this is so, then there are no grounds for reason to limit itself to the world of ordinary experience; it can after all attempt to understand the latter by appealing, on the basis of sound inference, to the causal role of entities that lie outside the world of ordinary experience. Since no paradox results, common life need place no a priori constraints on reason, so the latter can transcend its origins and lead us to the world of science that lies outside our ordinary experiences. 17.Cf. R. W. Sellars, Critical Realism, Ch. I; or J. B. Pratt, “Critical Realism and the Possibility of Knowledge”, p. 96. 18.Price, Hume’s Theory of the External World, p. 107. 19.Ibid., 108-13 20.Ibid., p. 112. 21.Livingston, Hume’s Philosophy of Common Life, p. 14, finds three systems in Hume, the popular system or that of the vulgar, a phenomenalistic system, and the system of double existence or that of the philosophers. He arrives at this by simply assuming that one stage of the causal argument has phenomenalism as its conclusion. But, as we have indicated, this assumption is erroneous. (Livingston also confuses the system of the philosophers of Section ii, with the modern philosophy of Section iv, which Hume argues does, unlike the system of the philosophers, have Berkeleian phenomenalism as its inevitable upshot; cf. Livingston, p. 18). 22.Price, Hume’s Theory of the External World, p. 120; M. Mandelbaum, Philosophy, Science and Sense Perception, p. 121, p. 123, p. 134ff; J. Passmore, Hume’s Intentions, p. 137; B. Stroud, Hume, p. 111. 23.Sellars, Critical Realism, pp. 10-11. 24.Ibid., p. 10. 25.Ibid., p. 14.

247

Three: Ordinary Things and Critical Realism

26.That is, justifies after the manner of Russell’s analysis of the use of definite descriptions as denoting, or referring, singular terms. For the logic of this, see Wilson, Laws and Other Worlds, Ch. I, Sec. 2, and Explanation, Causation and Deduction. 27.Hume makes this clear in the tale of the watch (T, p. 132), where an “artizan” or philosopher can through discovery confirm the hypothesis, justified by the causal principle, that there exists a minute cause which prevents the watch from working. 28.Hume clearly recognizes the non-falsifiability of the causal principle. Thus, in the discussion of miracles, a miracle must lack a naturalistic cause. But one cannot conclude, contrary to the causal principle, that a failure to observe a cause implies that there is no cause. As Hume puts it, where one has “extra-ordinary” events, for which no cause is discernible, such events, while not “conformable” to one’s experience, are “not contrary” to it (see Hume, Enquiries, p. 114, and p. 114n). The logic of such mixed quantificational principles is discussed in detail in Wilson, Laws and Other Worlds, Ch. I, Sec. 2; and its connection with explanation is discussed in Wilson, Explanation, Causation and Deduction. See also F. Wilson, The Logic and Methodology of Science and Pseudoscience. 29.Cf. Wilson, “Mill on the Operation of Discovering and Proving General Propositions,” and “Kuhn and Goodman: Revolutionary vs. Conservative Science.” 30.Passmore, Hume’s Intentions, p. 139. 31.Sellars, Critical Realism, p. viii, p. 20; Pratt, “Critical Realism and the Possibility of Knowledge”, p. 98. 32.Sellars, Critical Realism, p. 20. 33.For the distinction between knowledge which is more, and knowledge which is less imperfect, cf. G. Bergmann, Philosophy of Science, Ch. II; F. Wilson, Explanation, Causation and Deduction; F. Wilson, The Logic and Methodology of Science and Pseudoscience. 34.P. Árdal, in his insightful discussion of reasonableness in Hume’s philosophy, “Some Implications of the Virtue of Reasonableness in Hume’s Treatise”, proposes that Hume is not concerned about truth. But surely it is better to read Hume as we are trying to do, as concerned with the truth, as arguing to the truth of the system of the philosophers, and as also holding that the reason that leads one to this system nonetheless also justifies the reasonable man in accepting the system of the vulgar in all ordinary contexts. Cf. Wilson, “Hume’s Defence of Science.” 35.Livingston, Hume’s Philosophy of Common Life, p 25, holds that “true philosophy” can correct common life but at the same time cannot go beyond it in any radical way.

248

Body, Mind and Self in Hume’s Critical Realism

We now see that this is not Hume's view: in attempting to correct common life, reason is led inevitably and justifiably to the system of the philosophers which, in an important way, does conflict with, and refute, the system of common life. Still, we are justified in continuing, as we must, to use the latter. 36.Price, Hume’s Theory of the External World, p. 122, p. 136. 37.Some of the background physiological thinking of the Cartesians that Hume takes for granted can be found discussed in John Wright, The Sceptical Realism of David Hume. 38.Cf. D. Flage, “Locke’s Relative Ideas,” and “Hume’s Relative Ideas.” 39.Hume, “Of the Standard of Taste”, pp. 10-11. 40.Ibid., p. 11. 41.Sellars, Critical Realism, p. vi. 42.Pratt, “Critical Realism and the Possibility of Knowledge”, p. 95. 43.Ibid., p. 96. 44.For the relevance of this point elsewhere in Hume, see F. Wilson, “Hume’s Defence of Science”, and “Hume’s Fictional Continuants.” 45. D. Hume, History of England (London: T. Cadell, 1794), vol. XIII, pp. 133-134. 46. Cf. F. Wilson, Hume’s Defence of Causal Inference. 47. Cf. M. Mandelbaum, Philosophy, Science and Sense Perception. 48.I have dealt with this passage in greater detail, and Hume’s point in introducing the physiological speculations at this place in his argument, in F. Wilson, The External World and Our Knowledge of It: Hume’s Critical Realism, An Exposition and Defence, chapter one. I am there critical of certain claims by John Wright, The Sceptical Realism of David Hume. 49.See also Chapter Four, below, note 97. 50.Hume likely gets many of his speculative physiological notions from Malebranche, either directly or from Ephraim Chambers’ Cyclopedia, first edition 1728, where many entries are redolent with Cartesian and Malebranchean physiological speculations. It is throughout “specious but plausible,” but was the common way of thinking about the human mind become incarnate, how it fits into the material world. John Wright, The Sceptical Realism of David Hume, argues convincingly for the relevance of Chamber’s Cyclopedia.

249

Three: Ordinary Things and Critical Realism

51.Hume, Enquiries, p. 66. 52.Ibid., p. 65.

250

Chapter Four The Disappearance of the Simple Self: Its Problems (1) Substance and Self in Locke1 If material substance or body was handled without too much difficulty once the notion of substance and real essence was subjected to the Humean critique, the same cannot be said for mental substance. But that should not surprise us. For, after all, Plotinus had added a twist to the substance tradition at this point, a twist that was later to be emphasized by Descartes, that the unity of consciousness and of self-consciousness requires that one construe mind or self as an Aristotelian substance. Now, as we have seen, Locke points out that the notion of a person is primarily a forensic notion. It is a forensic term, appropriating actions and their merit; and so belongs only to intelligent agents, capable of a law, and happiness, and misery. This personality extends itself beyond present existence to what is past, only by consciousness, – whereby it becomes concerned and accountable; owns and imputes to itself past actions, just upon the same ground and for the same reason as it does the present. All which is founded in a concern for happiness, the unavoidable concomitant of consciousness; that which is conscious of pleasure and pain, desiring that that self should be happy (Essay, II, xxvii, 26).

A person is a substance, but more must be said. After all, there is more to the identity of things than the continuity of one and the same entity throughout the sequence of events that constitute the history of the substance. To be sure, the presence of the continuant is the criterion of identity for simple substances, but in the case of substances which have parts, that is, compound substances, the case is more complicated. It depends upon how the compounding occurs, that is, in effect on the relation by which the compound substance is constituted out the simple substances which are its parts. This means in particular that one cannot simply identify the continuity of a person with the continuity of a

Four: Disappearance of the Simple Self: Its Problems

substance; that could be done only if one established that persons were simple substances. Consider some examples of compound substances. The identity of a heap of rocks through time is constituted by the identities of its parts: take away one rock and we have a different heap (II, xxvii, 4). Locke goes on to contrast the case of plants and animals. Here the parts in themselves are not essential; one particle of matter may be replaced by another, but it is the same living animal. That being one...plant which has such an organization of parts in once coherent body, partaking of one common life, it continues to be the same plant as long as it partakes of the same life, though that life be communicated to new particles of matter vitally united to the living plant, in a like continued organization conformable to that sort of plants (II, xxvii, 5).

The same pattern of continuity may be found in certain artificial machines, such as watches. Again, even if one part is replaced, the machine does not lose its identity. What is crucial is not the particular part, but that there be a part of that sort performing the same function as that part, that is, the same function in maintaining the entity in its characteristic activity, keeping time in the case of the watch, maintaining the process of metabolism in the cases of plants and animals. At the same time, however, it should be noted that only one or a few parts can be replaced, at any one time if identity is to be preserved: replace all the parts of the watch simultaneously and one then has a new watch. What this means is that, in considering whether a sequence of events form a single identical entity, one must take into account not just the inherence relation but also other relations that define the order that we observe in the series. We should not in fact find this surprising. It amounts to the idea that a substance is not merely a continuant but also a form. That is, as Aristotle argued, a substance is not merely a continuant but a continuant of a certain specific kind, and that kind is inseparable from the continuant. If Socrates ceases to be human, he ceases to be Socrates. Locke is making the same point. He is in effect saying that, whatever one says about simple substances, for compound substances at least a substance will continue to be the same identical substance only so long as “its” real essence remains the same. And so determining whether certain events in a sequence are parts of one and the same thing, it will be necessary to make reference to the particular sort of order that is exemplified in the sequence. In the case of man, that is, in the case of persons, the relevant sort of

252

Body, Mind and Self in Hume’s Critical Realism

order is that of a living thing. We will have the same man if we have “one fitly organized body, taken in any one instant, and from thence continued, under one organization of life, in several successively fleeting particles of matter united to it...” (II, xxvii, 7). The identity of a man or person is thus of a piece with the identity of an animal. But the identity of a man or person cannot be constituted by either his soul or his rationality. It cannot be the former, since the soul is immortal, and this would therefore allow the possibility for the transmigration of souls, discussed by the Cambridge Platonists but absurd in Locke’s view, as it would entail that “men, living in distant ages, and of different tempers, may have been the same man” (ibid.). Nor can the identity of a man be constituted by his rationality, since a rational parrot, though rational, could not be counted a man (ibid.). But this idea of man or person, separated from the ideas of his or her soul and his or her rationality, makes of man or person something that is essentially corporeal, and cannot account for the fact of personality and moral responsibility. What gives man or person his or her special position that distinguishes him or her from other animals is his or her moral status as a person, something that can be held responsible for what he or she has done and rewarded or punished for that. Traditionally what this amounted to was the point that man or person is a substance that is not only animal but also rational. This is the point that Boethius made when he defined a person as “The individual substance of a rational nature” (Theological Tractates, Contra Evtychen, p. 85); and it is the point that Locke makes when he asserts that a person is “a thinking intelligent being” (II, xxvii, 11). [As Bishop Butler pointed out, “...Being and Substance in this place stand for the same thing”.2] The consciousness of the rational being as a thinking being includes not merely thought, consciousness, but also selfconsciousness; it can “consider itself as itself” (II, xxvii, 11). Locke is here accepting the Cartesian point, already argued for by Plotinus, that consciousness implies self-consciousness. We cannot, for both Locke and Descartes, think, feel, contemplate, or will anything without being conscious of, or aware of, that thinking, feeling, contemplating, or willing. As Locke says, there is a “...consciousness which is inseparable from thinking, and, as it seems to me, essential to it: it being impossible for any one to perceive without perceiving that he does perceive....Thus it is always as to our present sensations and perceptions: and by this every one is to himself that which he calls self:...”. However, as Locke hastens to add, “it [is] not...considered, in this case, whether the same self be continued in the same or divers substances” (II, xxvii, 11). But we also remember our

253

Four: Disappearance of the Simple Self: Its Problems

past actions: our consciousness here is not only conscious of itself but conscious of earlier conscious events and conscious that the earlier consciousness of them is the same as present consciousness. ...since consciousness always accompanies thinking, and it is that which makes every one to be what he calls self, and thereby distinguishes himself from all other thinking things, in this alone consists personal identity, i.e. the sameness of a rational being: and as far as this consciousness can be extended backwards to any past action or thought, so far reaches the identity of that person; it is the same self now it was then; and it is by the same self with this present one that now reflects on it, that that action was done (II, xxvii, 11).

The continuity here need not be that of a substance. In fact, we do not know whether the self is a single substance or a compound of several substances. If the latter, then of course the traditional argument for immortality based on the simplicity of the substance is rendered unsound. This argument finds its origins in Plato, in the Phaedo, is developed by Plotinus, and made its way down through history to Descartes and More. Locke uses the argument from the separateness of certain events in experience to draw the conclusion that there is nothing in what we experience of the self to guarantee that it is a single simple substance. The link between my present consciousness and my past is provided by memory. But memory is only a contingent tie; otherwise we would be unable to remember and then forget, we have false memories, we dream (II, xxvii, 13). There are gaps in our memory – we sleep, we are drunk –, and these gaps are bridged not by some necessary tie but only by memory. We could conclude that there was a single substance only if we experience such a substance actually tying the various stages of the self together. But we do not experience substances, nor do we experience necessary connections. It follows that the self may be a compound substance (ibid.). If the continuity is not that of a simple substance, it is also misleading to say that the link is merely memory. To be sure, memory is a necessary condition. That is why Locke argues as he does that when the drunkard cannot remember certain things that he has done, those things were not really done by him (or her); he (or she) was not responsible for them – though, to be sure, he (or she) was responsible for getting drunk, and if he (or she) could not be punished for what he (or she) did when drunk then he (or she) could be punished for becoming drunk. But as we saw, what is equally important is that the memory is one in which the self which was earlier conscious is the same self as that which is now

254

Body, Mind and Self in Hume’s Critical Realism

conscious; that is, in memory we recognize the earlier self as the same as the present self. Bishop Butler argued that memory does not constitute personal identity, but rather presupposes it: All attempts to define personal identity would but perplex it. Yet there is no difficulty at all in ascertaining the idea. For as upon two triangles being compared together, there arises to the mind the idea of similitude; or upon twice two and four the idea of equality; so likewise upon comparing the consciousness of oneself in any two moments, there as immediately arises to the mind the idea of personal identity....By reflecting on that which was myself twenty years ago, I discern that they are not two, but one and the same self (“Dissertation of Personal Identity,” p. 279).

Locke quite agrees that in the memory you recognize, or, if you wish, discern, that the self that was conscious then is the same as the self that is conscious now. But contrary to Butler, it does not follow from this that the identity discerned is the identity of a simple substance – no more than the discerned identity of a man, say Tom Jones, implies that that identity is one of a simple substance. Unfortunately, Locke leaves it as it is, at the discernment in memory of an identity, and does not pursue the issue further of what the nature of that judgement is. Other problems were, perhaps, more pressing. In particular, there was the claim that memory was a necessary condition for personal identity. What, then, of the actions of the drunkard that he forgets? Or the actions of a sleepwalker? Is the drunk to be punished for what he or she was never really aware of doing? Is the man or person awake to be punished for the actions of him- (or her-)self asleep? Molyneux was to argue against Locke that the two cases had to be distinguished. He appealed to the forensic criterion that Locke himself insisted upon, that a person is responsible for his actions. The sleepwalker, Molyneaux argued, is not responsible for what he or she does when he or she is asleep, and should therefore not be punished for anything that he or she might do when in that state. But drunkenness is itself a crime and so the drunkard should be punished both for it and for any other crimes that he or she commits while in that state (M to L, Works of JL,3 VII, p. 329). Locke replied that “...it is an argument against me, for if a man may be punished for any crime which he committed when drunk, whereof he is allowed not to be conscious, it overturns my hypothesis” (p. 331). But he had already argued in the Essay, that a human judicature could not allow a drinking man to plead innocent for want of remembering something,

255

Four: Disappearance of the Simple Self: Its Problems

“because the fact is roved against him, but want of consciousness cannot be proved for him” (Essay, II, xxvii, 22); human laws must punish both with “a justice that is suitable to their way of knowledge.” But, he also went on to say, ...in the Great Day, wherein the secrets of all hearts shall be laid open, it may be reasonable to think, no one shall be made to answer for what he knows nothing of; but shall receive his doom, his conscience accusing or excusing him (II, xxvii, 22).

But Molyneaux was not convinced, and re-stated his argument for Locke: drunkenness is a voluntarily induced state, and thus the drunkard is responsible for all of the consequences thereof, while sleepwalking is involuntary, and the somnambulist ought not to be held responsible for what he does (M to L, Works of JL, VII, p. 334). At this point Locke resigned and gave in to the argument of his Irish friend; he admitted that “want of consciousness ought not to be presumed in favour of the drunkard” (L to M, p. 326). But this surely is to give up his position. He has allowed that Molyneaux’s objection, if sound, strikes at the heart of his account of a person. And he then accepted the objection. The problem is, of course, that the criterion of consciousness and the criterion of responsibility do not coincide. In the traditional picture deriving from Plato, Aristotle and Plotinus, these criteria do not, of course, diverge. The self as the timeless centre of consciousness sums up within itself the past that it has already, as it were, produced, and is pregnant with the future that it will, in due course, create. The self-conscious centre of being is the reason why the events happen in its history, and, as that reason, that self-conscious centre is also responsible for those events. It does not matter that, empirically, this selfconsciousness is sometimes not conscious of itself. The position is not defended by the appeal to experience, but by transcendental arguments: the awareness of a series of events must lie outside the temporal series of events of which it is aware for otherwise it could not grasp them as a whole and all at once. But Locke has no recourse to this sort of argument: he is committed to the “historical plain method”, and what we experience are separate and separable events contingently linked in time. We are selfconscious but that self being conscious of self is an event in time, and an event that we experience in time. It is part of the series, not an entity standing apart from it. It is, moreover, part of the causal order of the

256

Body, Mind and Self in Hume’s Critical Realism

events in the world of ordinary experience. It is a product of what has happened in the past and plays an important role in the production of actions that are yet to come. Given Locke’s incapacity to go beyond the world of ordinary experience (he accepts PA as the guiding rule in ontology), there is no way that the self could play that special role that it plays in the traditional substance philosophy as the transcendental cause of the events that happen in its history. Self-consciousness no longer has that special place; it is, to repeat, an event among events, produced by events that preceded it. This is one causal process. What we must recognize is that the relations defining responsibility are also causal processes. In the traditional picture, of course, the relations of responsibility lead back to the transcendental self. But why, when one treats mind empirically, should one expect chains of responsibility to always lead back to a state of selfconsciousness? and why should one expect that a centre of selfconsciousness should always be the starting point of a relation of responsibility? As Locke said, when we look at the concept of a person we discover pretty quickly that it is a forensic concept. To put it simply, those actions for which a person is responsible are those that define what he or she is as a person; and, to put it even more simply, if not crudely, we are responsible for those actions where we use the techniques of reward and punishment to control behaviour. Why should we expect the area of behaviour that can be controlled by reward and punishment to coincide with the area of behaviour that is a product of self-conscious act? There is in fact no reason a priori to think this – other than the reasons given for the traditional substance account of the mind and of self-consciousness, but Locke rejects these. Indeed, it is the thrust of Molyneaux’s appeal to the two cases of the drunkard and the sleepwalker to suggest that the two criteria of personhood may indeed diverge. (2) The Contents of the Humean Mind Hume’s view of the world of ordinary perceptual experience is, we have argued, that of “natural realism.” More specifically, there is a sense in which he is a neutral monist: no perception (impression or idea) is intrinsically mental or material. Now as every perception is distinguishable from another, and may be consider’d as separately existent; it evidently follows, that there is no absurdity in separating any particular perception from the mind; that is, in breaking off all its relations, with that connected mass of perceptions, which constitute a

257

Four: Disappearance of the Simple Self: Its Problems

thinking being (T, p. 207).

At the same time, Hume has often been characterized, by commentators from Reid4 to Stroud,5 as an adherent of the “way of ideas,” that is, one who holds that all sense perceptions – impressions and ideas – are in the mind in the sense of being ontologically dependent upon it. Part of this is nothing more than a simple misreading coupled with a failure of philosophical intelligence. Given Hume’s non-substantialist view of the self, and his view on the separability of ideas and impressions, it is hard to see how Hume could be construed as holding that all sensible entities must be in the mind, that is, ontologically dependent on it. Nonetheless, there is enough in Hume to make the view sufficiently plausible for it to gain the currency that it has had. In particular, there are two deficiencies in Hume’s account of mind that at least suggest that the way in which they might be remedied is by the supposition that Hume holds, implicitly at least, that there is after all a substantial mind. This line of argument would, of course, tie in nicely with the standard reading of Hume as a sceptical naturalist. (i) The Intrinsically Mental Hume repeatedly uses the language of cognition when discussing ideas. Thus, in discussing causal inferences generally, Hume appeals to the logical properties of ideas: “We may draw inferences from the coherence of our perceptions, whether they be true or false; whether they represent nature justly, or be mere illusions of the senses” (T, p. 84). Again, we are said to “assent” to ideas (T, p. 94ff), another cognitive act, characterizing belief. The Treatise is stuffed with the language of 18th century logics such as Crousaz or the Port Royal Logic, as for example when Hume speaks of the “simple conception of any thing” (T, p. 94), an act which is contrasted with the acts of uniting, separating, mingling and confounding of our ideas which are cited later (T, p. 94). Again, it is the mind that conceives and comprehends (T, p. 161, p. 162); this understanding is preceded by acts of conception or having ideas (T, p. 164, p. 168). All these Hume refers to as “actions of the mind” (T, p. 177); they are functions of the “intellectual faculties of the mind” (T, p. 138). This cognitive aspect of Hume’s account of ideas has been emphasized by such commentators as John Yolton.6 But alongside this discourse, we also have the standard account of Hume’s views in which ideas are simply images, which therefore lack cognitive functions. There is no doubt that there are

258

Body, Mind and Self in Hume’s Critical Realism

both these strands in Hume. The usual response is that Hume was inconsistently trying to have the best of both worlds, in which ideas are both simply images and also are intentional entities playing a cognitive role. Customarily, the whole thing is treated as another example of Hume’s confused way of thinking. But it can be argued that Hume’s images become cognitive through the activities of a substantial mind. The traditional view that was part of the substantialist account of the world, was that images themselves were not cognitive, but came to be referred to the substances that caused them through the mind abstracting from its sense experience the nature or form of the object (substance) causing the image. On this account, the aboutness of images was parasitic upon the aboutness of thought, with the latter, and therefore also the former, deriving from the activity of the mental substance. The thought is that Hume makes images cognitive by implicitly presupposing what he explicitly denies, namely, an active mind that confers upon them the intentionality that intrinsically they lack. This was already a criticism that John Sergeant had made against Locke7; it was subsequently restated by Reid, who argued that cognitive acts of perception are simple entities, unanalyzable into sensation, and presupposing a substantial mind8; it was later repeated by T. H. Green against any empiricist account of thought9; one can find it once again restated by such writers as Gilson and Langan10; and A. and D. Hausman argue that for Hume the intrinsically mental can only be the substantial self, and that Hume eliminates this only by illegitimately collapsing the act/object distinction and construing all thought in terms of sensory contents.11 But of course, Hume does, as we have seen, speak of perceptions and ideas not simply in sensational and imagistic terms, but as cognitions; hence, such commentators conclude, Hume does implicitly acknowledge the existence of the active substantial self that confers intentionality on thought; implicitly he refutes his own scepticism about substances, and, further, about substance as the basis of personal identity. There is, however, an alternative way in which we can account for the ambiguity in Hume between ideas as images and ideas as cognitive. This alternative has the virtue of being fully compatible with empiricism. In effect, what this alternative does is ask us to take seriously the nature of introspective psychology, a notion that was used by Hume though was not fully understood by him, and, indeed, by no one until it was fully clarified by John Stuart Mill in the 19th century. The basic distinction of introspective psychology is that which Hume

259

Four: Disappearance of the Simple Self: Its Problems

draws between impressions and ideas (T, p. 1).12 The former enter the mind with “the most force and liveliness”; the latter are “faint images” of the former (T, p. 1). Impressions include both sensations, on the one hand, and our passions and emotions, on the other (T, p. 1); the latter are impressions of reflection (T, p. 7). Sensations are produced in us by unknown causes (T, p. 7), and “are such as without any antecedent perception arise in the soul, from the constitution of the body, from the animal spirits, or from the application of objects to the external organs” (T, p. 275). Our passions and emotions, in contrast, are caused by these original impressions or by the ideas of such impressions (T, p. 275), as when an impression strikes the senses, and causes us to feel, e.g., pleasure, the idea of the latter remains, and in turn causes a new impression of desire or hope; or, if the impression was of pain then there is produced an aversion or a fear (T, p. 8). Both impressions and ideas divide into simple, which “admit of no distinction nor separation” (T, p. 2), and complex, which “may be distinguished into parts” (T, p. 2). Complex ideas often correspond to complex impressions, e.g., my idea of Paris is the idea of a city which I have seen, but, equally, complex ideas can be created by the mind out of simple ideas, even when there is nothing that corresponds to that complex, e.g., the idea of a New Jerusalem, which no one has seen (T, p. 3). Simple ideas, in contrast, always have a corresponding simple impression. After the most accurate examination, of which I am capable, I venture to affirm, that the rule here holds without any exception,...that every simple idea has a simple impression, which resembles it; and every simple impression a correspondent idea (T, p. 3).

Note how Hume is here treating this – which has been called his first principle but also the copy principle13 – as a scientific hypothesis, a matterof-fact generalization in the science of psychology that is to be tested against experience. This first principle, that there is no simple idea without a corresponding impression, has other roles in Hume’s philosophy, but among those which it has is that of a psychological hypothesis within the context of psychology, i.e., science of human nature, that Hume is developing.14 Now, ideas, that is, images, appear in our minds in regular patterns; these patterns are known as associations. These associations bring together ideas which are separable, and which, indeed, are often separated by the imagination. Since the imagination can separate these ideas, Hume speaks

260

Body, Mind and Self in Hume’s Critical Realism

of a tendency for these ideas to be united, and of a “force” (T, p. 10) of an “attractive” sort (T, p. 12) which binds ideas together but which is sufficiently “gentle” (T, p. 10) that the imagination can exert a contrary force and separate the ideas that the binding force tends to associate with each other. Were ideas entirely loose and unconnected, chance alone wou’d join them; and ’tis impossible the same simple ideas should fall regularly into complex ones (as they commonly do) without some bond of union among them, some associating quality, by which one idea naturally introduces another. This uniting principle among ideas is not to be consider’d as an inseparable connexion; for that has already been excluded from the imagination: nor yet are we to conclude, that without it the mind cannot join two ideas; for nothing is more free than that faculty: but we are only to regard it as gentle force, which commonly prevails...(T, p. 10).

In speaking of associations, Hume is speaking of regularities – “Whenever and idea of an A enters the mind, then it is accompanied by the idea of a B” – but these regularities are conditioned – “Whenever conditions C obtain, then subsequent to that, whenever an idea of an A enters the mind, then it is accompanied by the idea of a B.” What Hume is saying in the passage just quoted is that when the mind has a will to, that is, as later psychologists were to put it, when it is in a certain set, then it will arrange its simple ideas in patterns of its own choosing; while, if it is not in that set, then the patterns to which its ideas conform are not imposed by the mind itself. The talk of a ‘force’ here is intended to suggest that an association has a certain strength: an association of ideas has a capacity to resist the separation by the mind of its components. At the same time, however, the force is not insuperable, the parts can be separated; that is why it is a “gentle” force. Now, as we know, Hume will propose to analyze causation in terms of regularities; there is no reason to suppose that when he here talks of “forces’ he intends anything that cannot be unpacked in terms of the language of regularities. At this point in the Treatise, however, Hume has not yet introduced that analysis, and so may as yet speak with the vulgar, using our ordinary language of causation. And in any case, Hume would claim, scientific reasoning about matters of empirical fact can, even if some uses of the ordinary language cannot, be unpacked in terms of his analysis of causation as involving, so far as objective content is concerned, nothing more than regularities. In fact, the metaphor of an attractive force with a gentle strength is obvious enough – certainly clear

261

Four: Disappearance of the Simple Self: Its Problems

enough for Hume’s purposes. It in fact marks a set of research problems for the psychologists that would come after Hume. The assumption is that there are laws which govern the appearance of ideas in our minds. Hume, like later psychologists, makes this deterministic assumption. For Hume, of course, it is simply a special application of the more general deterministic principle which Hume states as the fourth of the rules by which to judge of causes (T, pp. 173-4). This means, in particular, that we can infer, in accordance with Hume’s other rules by which to judge of causes, that there are conditions C which determine when the patterns exemplified by our ideas are imposed by our will and when they are not. We know that these conditions exist, without knowing specifically what these conditions are. Then, motivated by curiosity, or the love of truth, the task is to undertake research to discover these; this research is guided by the rules by which to judge of causes, as we have seen. Such research was in fact pursued by later psychologists. John Stuart Mill was, for example, concerned to clarify the conditions under which the mind separated what was associated15; and other psychologists were concerned to define the notion of the strength of an association, and with the causal conditions under which associations with different strengths occur.16 For Hume’s purposes, however, it sufficed to leave the situation in a state of imperfect knowledge of the laws involved.17 Hume is concerned, on the one hand, to combat the Aristotelians, and, on the other, to outline the empirical science of human nature that he is proposing as the new foundation for human knowledge. To give an outline does not require that he stop to fill in all the details. Indeed, if he were to stop to fill all the gaps, the outline itself would never be completed. For these purposes, then, it suffices to make the relevant point, using the metaphors of the causal language of the vulgar, and then move on to the next point. Others coming later, would do the science and remove the ignorance to which the metaphors point. For Hume’s purposes more essential than eliminating the metaphor of a gentle force” is indicating on what depend those patterns among our ideas which do not depend directly upon our will. What Hume suggests is that our ideas stand in the same relations as the impressions, which are their causal antecedents. The qualities, from which this association arises, and by which the mind is after this manner convey’d from one idea to another, are three, viz. RESEMBLANCE, CONTIGUITY in time or place, and CAUSE and EFFECT (T, p. 11).

262

Body, Mind and Self in Hume’s Critical Realism

The basic notion is that if (an) impression (of an) A stands in relation R to (an) impression (of a) B, and we several times experience such impressions standing in R to each other then the idea of a B will come to regularly associated with the idea of an A and in fact the idea of an A will stand in R to the idea of a B. Repeated experience of things in one of these relations brings about an association of ideas. There are two relations in particular that give rise to associations, that of resemblance and that of contiguity. Thus, if we have an impression of an A and an impression of a B and an A resembles a B, and we several times experience impressions resembling each other as this A resembles this B, then when an idea of, i.e., an image of an A appears in the mind that appearance will have associated with it the appearance of an idea of a B; in fact, the idea of a B that is associated with the idea of an A will resemble that latter just as the impression of a B that causes the idea of an A resembles the impression of an A that causes the idea of an A. Association based on resemblance provides Hume with the theory necessary to provide, as we have seen, a non-traditional account of “abstract ideas.” Similarly, if several simple impressions regularly appear as copresent with each other, then the ideas derived from these simple impressions also regularly appear as copresent in the mind. It is this which gives rise to our ideas of substance. Again, if one sort of impression is regularly followed in experience by one of another sort, then the idea of the former is regularly followed in the mind by an idea of the latter. In fact, we know, Hume will propose in Part iii of Book I of the Treatise to reduce the case of association due to cause and effect to association due to contiguity. (Later psychologists would also attempt to reduce association by resemblance to association by contiguity, but this latter reduction remained controversial, unlike that of cause and effect, which most psychologists were prepared to accept.18) There are two kinds of association by contiguity, that in which the ideas are successive, as in the case of causation, and that in which the ideas are simultaneous, as in the case of our ideas of substance. Since all ideas derive from impressions, this holds in particular for that of substances. This means that our idea of a substance must derive from an impression of sense or be obtained by reflection. But impressions of reflection are our passions and emotions, which are not impressions of substances. Our idea of substance must therefore derive from impressions of sense. “We have therefore no idea of substance, distinct from that of a collection of particular qualities, nor have we any other meaning when we either talk or reason concerning it” (T, p. 16). The idea of a substance is thus a collection

263

Four: Disappearance of the Simple Self: Its Problems

of particular qualities “united by the imagination” and “supposed to be closely and inseparably connected by the relations of contiguity and causation” (T, p. 16). And so our idea of gold consists of the ideas of yellow, weight, malleableness, fusibility, etc. Where we have a complex idea, whether generated by successive or by synchronous association, there the mind can distinguish the parts of the complex, or, as it was also said, can analyze the complex idea into its parts. Upon this view, the analysis of an idea consists of distinguishing its parts, that is, the simple ideas that compose it. One has, first, a description of the idea as a whole; thus, we have the idea of gold. Such a description we may call the phenomenological description, to follow a later way of talking.19 We then re-describe the idea by listing its distinguishable parts; thus, we have, in the case of gold, the parts which are the ideas of a yellow colour, weight, etc. Such a description we may call the analytical description. Two points are clear. One. There is a set of rules such that, from the analytical description one can infer the phenomenological description and conversely. This is implied in the part/whole model that clearly lies behind the simple/complex distinction. While these “rules” evidently have a normative aspect to the extent that the model incorporates, as it obviously does, the notion of a definition, in contrast the purposes of empirical psychology, these “rules” must be understood as matter of fact regularities. Introspective analysis aims to discover a complete set of unanalyzable parts or “elements” and a set of “rules” such that the latter uniquely determine a mental event on the basis of its elements. Two. Given the learning theory that simple ideas have resembling impressions as their causal antecedents, and that the relations among the antecedent impressions are replicated in the relations among ideas, then what appears in the analytical description is a list not only of the parts but also a description of the genetic antecedents of the complex idea. Introspective analysis, in uncovering the “elements” out of which a mental event is composed, also, and thereby, yields a causal account of the origins in our impressions of that phenomenon. The programme of introspective psychology thus characterized has a theory of learning, namely, associationism, which guides research, and a method, that of introspective analysis. The theory asserts that there are mental phenomena for which there are parts. These parts are the genetic antecedents of the phenomenon analyzed. The wholes are produced by the genetic antecedents, or parts, coming to be associated with each other.20

264

Body, Mind and Self in Hume’s Critical Realism

This is Hume’s psychology. There are two part/whole models which are at work. One is the model of definition. This is the model that is at work in his discussion, at which we above looked, of the complex idea of a substance. The other model is spatial. This is at work in Hume’s discussion of space, where he suggests extensions can be analyzed into sets of coloured parts (T, pp. 32-4). But a further point must be made – and this is the crucial point against those who hold that Hume implicitly, and contrary to his own claims, must introduce mental substances in order to account for the intentionality of thought. The programme of introspective psychology demands neither of these models. For introspective psychology, parts are by definition what analysis discovers. There is no reason a priori to suppose that the analytical parts are either logical parts or spatial parts, though, of course, neither is this excluded. The logical parts of a concept are wholly in that concept, as unmarried and male are in the concept of bachelor. The logical parts of a concept are wholly in the concept because they define that concept. Similarly, the spatial parts are wholly in the spatial entity of which whey are parts, as the squares on a chess-board are wholly in that space defined by the boundary of the board. But the analytical parts need not be wholly present in the phenomenon which is analyzed. The analytical parts are discovered upon analysis; for them so to be discovered, it is not necessary that they be wholly present in the phenomenon prior to analysis; all that is necessary is that they be present in the phenomenon dispositionally. We have seen that John Stuart Mill adopts this position. Hume is on to a similar point when he argues that the genetic antecedents of an abstract idea “are not really and in fact present to the mind, but only in power,,,” (T, p. 20). It follows that the phenomenon can be a simple whole, without actual parts, but rather parts that are only dispositionally present to be discovered by, or, perhaps better, to be recovered by, analysis. Since the analytical parts are the genetic antecedents, what has happened in such cases is that a process of association working on the parts that analysis has recovered has produced a mental phenomenon, which, as s simple whole, distinct from those parts, has properties not present in those genetic antecedents. We can put the same point in a slightly different way. If D1 is the phenomenological description and D2 is the analytical description, then the logical or definitional model, and the spatial model, too, applies to the relations between D1 and D2. Hume tends to think of this relation as either definitional or spatial, but so long as he does this, he cannot have the whole

265

Four: Disappearance of the Simple Self: Its Problems

D1 greater than the sum of the parts D2. But once we give up the logical model for conceptualizing the D1-D2 relation, then we are free to recognize that an idea need not be like its parts, i.e., that an idea as a cognitive entity which carries intentionality need not be an image. Once free of the constraints of the definitional model, we can hold that an imageless thought can be analyzable into images and can be derived by association – or, more generally, learning – from the impressions from which those images derive. This point was recognized explicitly only late in the history of introspective psychology, by John Stuart Mill, when he emphasized that the “parts” which analysis recovers are not real or “integrant” parts but only “metaphysical” parts.21 But others, including Hume, were at least dimly aware of the point. Thus, at one point Hume tells us that “many of our ideas are so obscure, that 'tis almost impossible even for the mind, which forms them, to tell exactly their nature and composition” (T, p. 33). Here Hume is evidently thinking of the idea as a mental phenomenon which does not explicitly contain as real parts those simple ideas or images which are derived from sense impressions and which, through association, produce that phenomenon; and the idea has, moreover, properties that distinguish it as a simple whole from those “parts” that produced it. If one takes an idea to be a mental phenomenon, and parts to be analytical parts, then an idea can be simple whole, distinct from its parts, and certainly not merely the sum of its parts. On the other hand, if one takes an idea to be a concept, and parts to be definitional parts – where, following both Locke and Aristotle, concepts are to be defined by a conjunction of terms, as man is rational and animal or bachelor is male and unmarried – then an idea will be complex whole, not distinct from, but merely the sum, or conjunction, of its parts. The distinction was to be made clearly by Horne Tooke, in his Diversions of Purley, when he insisted that while the psychologist may analyze ideas, one does not define ideas; rather, what one defines are words.22 In Hume, however, the two are never kept clearly separate. The consequence is the confusion which is so often evident in his psychology. For, on the one hand, there will be a tendency to treat a mental phenomenon as a simple whole, distinct from its parts, and, on the other hand, a contrary tendency to treat the same phenomenon as a complex whole, not distinct from, but exhausted by, and merely the sum of its parts. Thus, although officially for Hume an idea is an image, we now see that it will perhaps not consistently be treated as such. An idea as a mental phenomenon will be analyzable into images, that is, sensory contents,

266

Body, Mind and Self in Hume’s Critical Realism

which are in turn derived from impressions. Moreover, the idea as a mental phenomenon will be the product of an association of those images that introspective analysis recovers. But it itself can be a simple whole, with qualities not present in the images that analysis yields. However, because of the unclarity in Hume’s thought about the nature of psychological analysis, we are liable to find him sometimes treating ideas as nonimagistic simple wholes and at other times treating them as nothing more than images. We in fact find this in Hume’s analysis of belief. What, Hume asks, is the difference between incredulity and belief (T, p. 95)? He considers the case of a “proposition you advance” to which you do but another person “does not assent” (T, p. 95). The difference is not a matter of having “conceiv’d the object” (T, p. 95) in different ways; the principle which accounts for the difference “plainly makes no addition to out precedent ideas” (T, p. 96). Rather, the principle “can only change the manner of our conceiving them” (T, p. 96; Hume’s italics). In the “Appendix” which appeared in volume two of the original Treatise and which added corrections, changes, and comments to the previously published Book I, Hume speaks of this difference as a difference in “feeling” (T, p. 631, p. 636). Notice, first, that a belief is an idea which has an additional property. This additional property falls into the quality category of feeling; it is the feeling of assent. Notice, second, that an idea is characterized as a conceiving which has a certain object; my idea of Paris is a mental phenomenon which has, however, a non-mental entity, namely Paris, as its object. As Hume puts it elsewhere, “to form the idea of an object, and to form an idea simply is the same thing” (T, p. 20). This feature of thought has been referred to, following Brentano, as its “intentionality.”23 Now, whatever intentionality is, it is not to be cashed out as the resemblance of an image to an impression24; otherwise, since resemblance is symmetrical, the impression would have the idea as its object as equally as the idea has the impression as its objects: my idea of Paris is about Paris; Paris does not have my mental state as its (intentional) object. R. J. Butler has emphasized, quite correctly, this point, that Hume recognizes the intentionality of thought.25 The point is that intentionality is a property of ideas which mere images do not have. Ideas, in other words, are distinct from images. Yet equally, according to Hume, they are to be analyzed into images. “Our ideas,” he says, “are copy’d from our impressions, and represent

267

Four: Disappearance of the Simple Self: Its Problems

them in all their parts”; and “an opinion, therefore, or belief may be most accurately defin’d, A LIVELY IDEA RELATED TO OR ASSOCIATED WITH A PRESENT IMPRESSION” (T, p. 96). Ideas are now images, and the relations in which they stand to impressions is not that of intentionality but of resemblance and causation. What distinguishes a belief now is not a certain quality of feeling but rather its liveliness, or “degree of force and vivacity” (T, p. 97). An idea acquires the degree of force that makes it a belief by virtue of its causal relation, or association, to our impressions. “Reason can never satisfy us that the existence of any one object does ever imply that of another,” that is, reason alone can never establish one matter of fact on the basis of knowledge of an independent, or separable, fact; hence, “when we pass from the impression of one [object] to the idea or belief of another, we are not determin'd by reason but by custom or principle of association” (T, p. 97). This force Hume characterizes in terms appropriate to images; such ideas are “more strong, firm and varied, than the loose reveries of a castlebuilder” (T, p. 97). At the same time, this force is also a causal quality; the ideas of poetry and fictions have their own “vigour of conception,” and as a consequence the force they have “causes the idea to feel very different from the eternal establish’d persuasions founded on memory or custom” (T, pp. 631-2). Although Hume speaks of a belief as an image and indeed as nothing more than an image of a special sort, we now see how a more just reading can be given, one which takes into account his other notion that ideas are somehow distinct from images, and distinct from the images which are their causes. A belief, upon Hume's account, is a non-imagistic idea to which a certain quality of feeling is attached. This idea is the product of an association of images derived from sense impressions. Introspective analysis of the non-imagistic idea recovers this complex image which is its genetic antecedent. This complex image has a certain quality of vividness which the antecedent impression has caused it to have. This quality of vividness has the causal effect of evoking the attitude of assent which attaches to the non-imagistic idea and which transforms the latter from a mere conceiving into a belief. It is worth noting that Hume allows that the attitude of assent which is evoked is a function of the antecedent impression, but may also be a function of the mind's own causal interference, through the influence of general rules that we have learned: “...reflection on general rules keeps from augmenting our belief upon every encrease of the force and vivacity

268

Body, Mind and Self in Hume’s Critical Realism

of our ideas....’Tis thus the understanding corrects the appearances of the senses...” (T, p. 632). Although the mind cannot avoid assenting to ideas, it can nonetheless control the degree to which it assents by so disciplining itself that assent is proportioned to what is by the general “rules by which to judge of causes and effects” determined to be reasonable. Once the blurs involved in the notion of introspective analysis are removed, then it is clear that Hume’s account of belief is not only consistent, but even has a certain plausibility. It of course presupposes in its details the associationist account of learning. But this theory, too, is not implausible, at least as an initial stab at a scientific approach to the process of leaning.26 For our purposes perhaps what is most important is the separation of the rules of definition, and, more generally, the rules of logic, from the “rules” that describe the linkage of ideas to the elements into which they are analyzed – that is, the distinction that Horne Tooke was getting at when he insisted that it is words, not ideas, which are defined. What this means, of course, is that the logical analysis of concepts is not the same as the psychological analysis of ideas; the former is an inquiry into their logic, while the latter is an inquiry into how we acquired the capacity to use them. What this distinction does is go part way at least towards justifying the device used by many interpreters of Hume in the twentieth century of treating questions of logic by means of the notion of an empiricist’s language, in the sense of the logical positivists. This, to be sure, is not the whole story of the relation between thought and language, as we see in Hume's doctrine of abstract ideas; but it is a good part of the story. To make the separation does require one, however, to be clear – clearer, indeed, than Hume himself was – on the nature of introspective psychology. There are, unfortunately, a number of commentators who have not been as clear as required, and as a consequence have directed unfair criticisms at Hume's account of belief. Thus, Passmore notes that Hume tends to think of impressions, or sensory impressions at least, as perceptions of the objective realm. But impressions are merely the same as ideas only more forceful or vivid. In that case vivid images of the imagination ought to count, as they surely cannot, as perceptions of the objective realm. They cannot be so counted not only because they are not such perceptions but also because, as Hume says, “every one of himself will readily perceive the difference betwixt feelings [impressions] and thinking [ideas]” (T, p. 1). Passmore suggests that all this is simple confusion on Hume’s part.27 In this he is followed by Bennett.28 Bennett

269

Four: Disappearance of the Simple Self: Its Problems

also puts the supposed muddle in another way. Hume holds that thinking consists in having ideas; but, Bennett argues, quite correctly, if an idea as an image is in effect nothing more than a “faint impression” then he should not hold that having ideas is eo ipso thinking: for, not every imaging is a case of thinking. However, Hume at the same time calls ideas the “faint images of [impressions] in thinking and reasoning” (T, p. 1), which implies that any having of ideas is to count as thinking. So Hume has to distinguish thinking from the sensory processes of having impressions and images if he is to do justice to human thought, as indeed Hume himself sees when he insists that we can all “readily perceive the difference betwixt feeling and thinking.” But, as Bennett suggests, since every case of thinking is one of having images, Hume cannot draw this distinction.29 But to dismiss it as all muddle, as Passmore and Bennett both do, is to go too far, and it is certainly not to be fair. Every philosopher ought to be given a fair reading, and if we do find that he is muddled or confused we have some obligation to search for a way to explain how he or she fell into that confusion and how it does, after all, involve a certain amount of sense. This latter is what we have attempted to do. What gets Hume into his problem is his failure to keep clear the distinction between the phenomenological description of a mental phenomenon and the analytical description. Passmore and Bennett point out that this gets Hume into trouble. But once the distinction is clearly grasped, one understands how Hume got into this problem, and also that there is in this confusion nothing that seriously challenges the central empiricist thrust of Hume’s philosophy. Passmore and Bennett are insisting that belief and thought are to be identified by their phenomenological qualities. That is fair enough; Hume would not disagree. They then infer from this that they cannot be accounted for in terms of images; these critics then simply dismiss the analytical descriptions as false. But this amounts to holding that no process of analysis could ever reveal the genetic antecedents of our ideas and beliefs. That is, it amounts to arguing that since belief and thought are simple there can be no scientific account that locates their causal origins in simpler sensory processes. It is to argue, in effect, that the learning theory of associationist psychology is in principle just wrong. Hume gets into his muddle, in other words, by attempting to construct a scientific human psychology. He has his insights but gets into trouble by trying to put science into where it doesn't belong. This is not the first time that this view has infected philosophy; one finds it also in Kames and Reid, where it is tied, quite properly, to a defence of innatism.30 But it is also a view that one

270

Body, Mind and Self in Hume’s Critical Realism

must reject once one has a clear view of the nature of introspective psychology. Passmore and Bennett have not taken the trouble to try to understand the science of psychology that Hume was attempting to found. We may conclude, then, that Hume’s account of psychological analysis enables him to hold, even if he is not entirely clear about it, that ideas are, on the one hand, cognitive, or intentional, entities and also that, on the other hand, they can be analyzed into images. Moreover, he can do this in a way that is entirely compatible with PA and empiricism. Contrary to Hume’s critics, then, there is no need for him to covertly introduce mental substances in order to allow a place in his ontology for intentionality, that is, the intrinsically mental. Nor does this involve, contrary to other critics, a collapse of the act-object distinction.

Allowing Hume the sort of leeway we are suggesting for the notion of an idea, that it mean an image in some contexts and an intentional entity in others, permits a more plausible reading of the Treatise than we would otherwise have. Thus, consider Hume’s remark that “as our ideas are images of our impressions, so we can form secondary ideas, which are images of the primary; as appears from this very reasoning concerning them” (T, p. 6). He is talking about reflective thought, and more generally of our awareness of our awarenesses. There may well be, for some such awarenesses, images of images, but it is more reasonable simply to take Hume to be discussing intentional entities, awarenesses of things and awarenesses of awarenesses. In what follows, we shall in this way be reasonable. This permits us to think of conscious states in the following way. Suppose we are thinking about some fact P, that is, that our conscious state is one of thinking about P. There is, then, the event of thinking about P. This is an event which has two properties. It has in the first place the property of being directed at or about or intending P; and it has in the second place the property of being a thinking. Again, if we are morally approving of Q, then that is an event which has the property, first, of being a moral approving, and the property, second, of being about Q. In saying these mental events are of or about othr states of affairs, we are recognizing the fact that they are intentional31; this intentionality is an intrinsic feature of these events. Now, when we are thinking about P or morally approving of Q, we are aware of those states; they are present in consciousness. We need to

271

Four: Disappearance of the Simple Self: Its Problems

recognize that the thinking about P or the morally approving of Q is itself the object of an awareness, an awareness of that thinking about P or the morally approving of Q; it is this awareness which makes those events states of consciousness. Thus, mental events are states of consciousness by virtue of being the objects intended by an act of awareness; it is that awareness which constitutes the state of consciousness. That act of awareness that constitutes the state of consciousness is, of course, not itself the object of a further act of awareness. Although it is generally true that whenever we are aware of something, we are aware of that awareness, this could not be true of the awareness that constitutes the conscious state, To suppose otherwise invites an infinite regress, one that is vicious. So the awareness that constitutes events as events in a state of consciousness is not itself the object of an act of awareness. But having said that, it is also necessary to say that the awareness constituting a conscious state can become the object of a further act of awareness. This happens when we have become conscious of being conscious. We are walking down the street, seeing all the various things we are seeing, then of a sudden we catch ourselves seeing: our attention is not directed towards the objects on the street but on the fact of consciousness that we are seeing those things. This is what happened to Descartes. In his Meditation One, he thinks about objects which are external to consciousness, sensible objects, arithmetical entities, cognitive demons, and so on. Then at the start of Meditation Two, his attention shifts, he suddenly catches himself thinking about those thinkings: cogito, he notes. And then he notes that also sum: I am. There is the thinking and there is the awareness which makes that thinking an entity in consciousness: the I is that awareness which constitutes the conscious state as one which contains the thinking. Of all this more later. We should note here, however, how this fits with Hume’s Critical Realism. Consider the world of ordinary experience, the system of the vulgar. Material objects are, in this system, patterns of sensory contents, both perceived and unperceived. Among these material objects is one that is, for me, special, namely, my body. We are conscious through our senses of these material objects. We see a green leaf. When that is so, there is present in consciousness a sensory impression of a green shape. In the system of the vulgar, this sense impression which is present in one’s consciousness is part of the ordered bundle that is the green leaf. It is the green leaf, considered as a thing that causes one to be conscious of the sensory quality green: that quality is present in my consciousness but is

272

Body, Mind and Self in Hume’s Critical Realism

part of the leaf, and not merely an event that occurs in consciousness. Further, of course, as we have said, among the entities of which one is conscious are one’s mental states. Mental states, whether cognitive (e.g., believings, supposings, doubtings, etc.), conative (e.g., willings, desirings, etc.) or affective (e.g., taking pride in), are characterized by their intentionality. These mental states are contents of conscious states that also have among their contents the state of my body. The material identity of my body provides an identity among my conscious states. My mind is the series of conscious states that are thus identified. At any moment, the centre of consciousness that constitutes my mind at that moment is the act of awareness that constitutes the conscious state at that moment. In one way, the parts of a conscious state are the contents of that state. Parts in this sense will include mental states, that is, acts of awareness of various sorts, and non-mental states, e.g., impressions and images. Taken in this way, the act of awareness that constitutes the conscious state is not among its parts, since it is not among the contents of that state. Things are much the same in the system of the philosophers, save that in this system the sense impression of the leaf is not objectively there as characterizing the object that causes my awareness of it. The material object which is the external or distal cause of my sensing something green is not itself green; the green impression is not part of the external cause, as it is supposed to be in the system of the vulgar, but depends for its existence upon the state of my organs produced by the external cause. The external object causes me to be aware of something green, but also is the cause of the sense impression. That impression is among the contents of my conscious state but cannot be said to have an existence apart from that conscious state. So, when I perceive the green leaf, there is present in my conscious state the perceiving of a green leaf and also a sense impression which is green and is of a certain shape. And further, there is in the conscious state, as in the system of the vulgar, an awareness, cognitive and sensory, of my body from the inside. As in the system of the vulgar, my mind is a bundle of conscious states; this bundle is a sequence brought together and ordered by these conscious states all being states of my body, the body which I am aware of and perceive from the inside. This, at least, is how things are in outline. More must be said, and later will be said. (ii) Neither Mental Nor Material

273

Four: Disappearance of the Simple Self: Its Problems

For Hume sensations – impressions and ideas – are not intrinsically mental; as we have seen, on his account of these things, there is no logical impossibility in a sense impression existing apart from the mind. Hume is thus no idealist, no subjectivist. Indeed, he holds that it is precisely the substantial mind that generates both idealism and thereby “the most extravagant scepticism” (T, p. 228), and that the way to avoid scepticism is to abandon mental substances, which are in any case excluded by the empiricist’s PA.. But in this context we should consider the argument of the Hausmans32 that Hume for all this is nonetheless a crypto-idealist, in that he must against his own principles have mental substances. Part of their argument is that Hume needs mental substances because he collapses the act-object distinction, and in order to capture the “aboutness” of thought, Hume must implicitly invoke a substantial mind. Collapsing the act-object distinction has the consequence that perceptions – impressions as well as ideas (images) – must be in the perceiving mind, that is, in the substantial mind, and therefore ontologically dependent upon it. On their view, then, Hume, implicitly at least, is committed to the very sort of idealism that he rejected as the source of scepticism. But we have just seen that this argument is not successful, since Hume, though to be sure not clearly, can account for the “aboutness” of thought without requiring an implicit appeal to mental substances. But there is a second argument that the Hausmans use to suggest that Hume, contrary to his own intentions, is committted to substantial minds, and to idealism. Begin with the basic framework of the system of the vulgar. That is, begin with material objects of the ordinary sort – tables, chairs, stones, trees, dogs, etc. – and the properties – colours, shapes, etc.– that are predicated of these objects. Material objects of course endure through time. The properties of things are determinate – this shade of colour, or that shade, this specific shape or that; but these determinate properties fall into families – those of the determinable properties, colour, shape, etc. It is part of the framework that material objects exemplify one and only one determinable property for each determinable – e.g., one specific colour (in a particular area); that is, a material object cannot have predicated of it at the same time and in the same respect two different determinates under the same determinable. Now, material objects do, in the appropriate circumstances, have some properties over time. (Such constancy, as well as any changes, will of course be explainable in terms of the regularities or patterns that define the material object to be of the kind that it is.) But

274

Body, Mind and Self in Hume’s Critical Realism

objects often appear to have properties that they do not have: the white paper appears red (under red lighting); the material object continues to be white while in sensation it appears to be red. It is contrary to the system that the material object, the piece of paper, can have both the properties white and red, since that would imply the object could have predicated of it two determinate properties under the same determinable. The property red which is given in sense cannot be predicated of the material object which appears to be red but which is really white. Now make the anti-Platonist assumption that there are no properties that exist unexemplified, that is, that there are no properties that are not predicated of some subject. Material properties, like colours and shapes, are, of course, predicated of material objects, while mental properties, e.g., beliefs, volitions, passions, are predicated of minds. The problem is the property red that is given in sense but is not predicated of any material object. Such sensory contents are not predicated of material objects. But the anti-Platonist assumption requires that the content be predicated of some object. The only alternative is the mental substance. Hence, such properties must be in the mind that perceives it. Variations on this theme, in particular, the notion that predication is always of a substance, led Berkeley into his idealism.33 Hume, too, appeals to the fact of perceptual relativity, and the premise that if (what is true) some sensations are relative, then what holds of one must hold of all. From these he concludes that all sense impressions are dependent (causally) for their existence on the state of one’s sense organs and are therefore not in the material objects that are the distal causes of their existence. But, the Hausmans suggest, if they are not in material objects, then they must be in the mind.34 That is, given the argument from perceptual relativity, all sense properties must be in the mind. But if Hume is a neutral monist, then the distinction between the mental and the material lies in the different relations that sensory contents (impressions and ideas) bear to each other. We have, however, just concluded that all sensory contents are in the mind; hence, there are no relations available to distinguish those contents given in sense which are material from those that are mental: they are all mental. Hume must therefore have implicitly another way of characterizing the mental, and this can only be a mental substance.35 So we see once again that Hume is committed to substantial minds and idealism. The problem with this argument is that it moves, illegitimately, back and forth between the system of the vulgar, in which not all sense contents are “in” the mind, and that of the philosopher, in which are sense contents

275

Four: Disappearance of the Simple Self: Its Problems

are “in” the mind. We must, moreover, recall that Hume does insist that mental states have an intrinsically cognitive feature, and therefore does have a criterion for the mental, namely, intentionality, though, to be sure, as we have just seen, his views on the nature of psychological analysis obscure this fact. Stick for the moment to the system of the vulgar. Material objects are, in this system, patterns of sensory contents, both perceived and unperceived. Among these material objects is one that is, for me, special, namely, my body. Furthermore, among the entities of which one is conscious are one’s mental states. Mental states, whether cognitive, conative or affective, are characterized by their intentionality.36 These mental states are contents of conscious states that also have among their contents the state of my body. The material identity of my body provides an identity among my conscious states. My mind is the series of conscious states that are thus identified. At any moment, the centre of consciousness that constitutes my mind at that moment is the act of awareness that constitutes the conscious state at that moment. In one way, the parts of a conscious state are the contents of that state. Parts in this sense will include mental states, that is, acts of awareness of various sorts, and non-mental states, e.g., impressions and images. Taken in this way, the act of awareness that constitutes the conscious state is not among its parts, since it is not among the contents of that state. We can also ask, however, what are the mental qualities of a conscious state. These would be the mental acts that form the conscious state, that is, those acts that are among the contents of the conscious state and the act of awareness that constitute the conscious state. Thus, consciousness itself (i.e., the act of awareness) and the mental contents then in that consciousness are the mental qualities of the conscious state. We may therefore also say that the mind at a moment has as its parts at a moment the mental qualities of the conscious state that is the mind at that moment. As for me, this is, as Hume said, constituted by “the qualities of our mind and body, that is, self,” (T, p. 303); the person which one is, is, in other words, the complex consisting of one’s mind and one’s body – I am my mind as incarnate in my body and my body as animated by my mind. But of this, more later. Return to the argument of the Hausmans. What, then, they ask, of the sensory contents of which we are conscious when we perceive nonveridically? What of the red that I sense when I perceive a piece of white paper under red light? It is not part of the complex that is the material object that we call the paper; it cannot be predicated of it. Nor can it be

276

Body, Mind and Self in Hume’s Critical Realism

predicated of my mind; for then my mind would be red. In the system of the vulgar, it hovers uneasily between the mental and the material, existing but present in neither.37 Where material objects and minds are substances, as in Descartes and Locke, then the dangling property, unable to exist in the material substance, inevitably falls into the mind, given the anti-Platonist assumption. But does not the same anti-Platonist assumption lead in the case of Hume to a similar conclusion? For, after all, the anti-Platonist assumption holds that properties must be predicated of subjects; it does not insist that they be predicated of substances. That is, the reasons for making the anti-Platonist assumption apply equally whether one analyzes objects as substances or as bundles of properties (qualities). So, as the Hausmans suggest, the dangling properties should end up in the mind as much in Hume as in Descartes. But here the argument of the Hausmans breaks down. As it turns out, is there is no reason why Hume should not accept this. The point is that, contrary to the Hausmans, this does not lead Hume back into any sort of implicit idealism. The question is, what does it mean to say of something that it is “in the mind”? For the substance philosophy, the answer is clear: to be in the mind is to inhere in a mental substance. But for Hume, as it turns out, there are six answers to this question. In the first instance, as we have seen, the mind is a certain complex of conscious states. To be one of these conscious states that are identified with each other is to be in the mind. In the second place, to be in the mind is to be a mental act that is part of one of the conscious states that is part of the mind. In the third place, to be in the mind is to be private to that mind. This third point deserves comment. For the substance tradition, assuming nominalism is taken as central to it, this criterion is going to coincide with the inherence relation, and therefore with predication. For, given the nominalism, one property cannot be in two substances; that is, properties are unique to the substances of which they are properties. Since that of which one is directly aware is a property of the mind, it follows that that of which one is aware is present to one and only one mind, i.e., is private to it. If, moreover, direct awareness is the only kind of mental awareness that one allows in one’s ontology, as in Berkeley, then one arrives not only at idealism but at a threatening solipsism.38 But if one allows, as we have seen Hume do, that there are cognitive states other than direct awareness, then there is no reason to hold

277

Four: Disappearance of the Simple Self: Its Problems

that the mind cannot perceive public objects that other minds also perceive: the intentionality of acts of perception, and of other cognitive acts, permits several minds to perceive, and, more generally, to think of (or, for that matter, to desire or to take pride in) one and the same object. Certainly, this is so in the system of the vulgar: material objects are perceived by any person appropriately situated. Furthermore, even with respect to direct awareness, that is, the awareness that constitutes a conscious state, there is no ontological reason for Hume, when he rejects substances, to suppose that one and the same entity cannot be in two minds, that is, two complexes. On the other hand, privacy is a fact, and this provides a nonontological reason for thinking that some entities at least are not publicly perceivable. And given Hume’s suggestion that the identity of one’s mind is determined by the conscious states in that mind all belong to one and the same, that is, one’s own, body, it turns out that the parts of one mind are never parts of another mind, in none of the three senses of ‘part’ that we have just noted. Thus, in these senses, being part of the mind and being private coincide. But there is more that is private than the parts of the mind. Thus, when I am aware of my body “from the inside” I am aware of qualities of that body of which no one else is aware. As we shall argue, it is precisely this fact that enables one to clarify the notion of privacy. These qualities of the body that I alone perceive are private but they are neither conscious states nor mental components of conscious states. So they are not parts of the mind in the two senses we have noted. However, I alone am conscious of those qualities. In that sense they, too, are mine and mine alone; they are in my consciousness as they are in no one else’s consciousness. In that sense, they are part of me as the public qualities of my body, and the public qualities of other bodies, are not. Thus, as we suggested, we have in privacy a third sense of what it is to be in the mind. Besides the three meanings already given of the notion of being in the mind, there is, in the fourth place, the notion that something is in the mind if it depends for its existence upon the state of our sense organs, or, more generally, on the state of our nervous system. This fourth sense of being in the mind deserves comment also. Clearly, in this sense of being in the mind, all our conscious states, and all the mental acts that they contain, are in the mind. Private states of our body, e.g., the pain in my tooth when I have a toothache, clearly depend for their existence on the state of the body; in that sense they are, by this fourth criterion, in the mind. Beyond the private entities, there are

278

Body, Mind and Self in Hume’s Critical Realism

in addition the properties that one is presented with when one misperceives a material object, e.g., the property red that is perceived when one views white paper under a red light. Such a property is not a property of the material object; it does not exist as a property of the object independently of its being perceived. The existence of such a non-material property depends, in other words, upon the state of the perceiver. It follows that, upon the fourth meaning of what it is to be in the mind, such properties are in the mind that perceives them. Note that they are not private: anyone similarly situated would perceive the same, that is, misperceive the same. So they are not, in that sense, in the mind; they are not mental in the way that, say, pains are mental. Next, a fifth sense of being in the mind: something is in the mind if it exists only when it is among the contents of a conscious state. Mental acts like percieving, desiring, or taking pride in, seem to be mental in this sense. Note, however, that the act of awareness that constitutes a conscious state is not among the contents of that state, and is therefore in this sense not mental: if you wish, the centre that is the source of consciousness lies outside consciousness. (Recall, however, that nothing prevents that act of awareness from becoming, through a shift of attention, the content of a further conscious state.) In addition, many would argue, with some plausibility, that private entities like pains exist only when one is conscious of them; that would mean that they too are in the mind in this fifth sense that we are now considering. The same considerations suggest that, since non-material properties, like private entities such as pains, depend for their existence upon the state of the nervous system, these properties exist only so long as one is conscious of them. Material properties, that is, properties of material objects, exist as the contents of conscious states; they so exist when the object is perceived as having that property. However, at the same time, the property exists materially as a property of the object. But when we misperceive the object as having a certain property, the property that we are presented with is non-material – it does not exist as a property of the object perceived –; it therefore exists only as one of the contents of the conscious state. In this way, the non-material properties that are presented when we mispercieve are, in the fifth sense we are now considering, mental. We now see that, once Hume removes himself from the tyrrany of the substantial mind, the dialectic of the notion of being “in the mind” becomes complicated. The properties of which we are aware when we misperceive are indeed “danglers,” as the Hausmans say. Further, these

279

Four: Disappearance of the Simple Self: Its Problems

properties end up quite correctly being characterized as “in the mind.” But, contrary to what the Hausmans go on to claim, this in no way implies an implicit commitment on Hume’s part to the existence of a substantial mind. This brings us to the sixth sense of being “in the mind”: something is in the mind just in case that it is predicated of the mind. Clearly, in the first instance what one predicates of the mind are the mental acts that are the contents of one’s conscious states, e.g., (+) I feel proud that I then did so-and-so. which predicates of oneself the mental act of taking pride in so-and-so which is at present among the contents of my conscious state. That is, one predicates of the mind entities that are mental in our second sense of mental. (In the second instance, one predicates of the mind dispositions to have contents of this sort.) As Chisholm has pointed out, in sentences like (+) “the verb is used to refer to a certain type of undergoing. The undergoing is what traditionally has been called being in a conscious state, or being in a sentient state. And the adjective, in such sentences as [(+), that is, ‘proud’], is used to specify further the type of undergoing to which the verb refers. The adjective could be said to function, therefore, as an adverb.”39 Chisholm suggests that from this grammatical point it follows that the feeling of pride is not an entity in one’s conscious state (ibid.) But of course that simply does not follow; what we have just shown is that this grammatical point is quite compatible with the mental act of taking pride in something being an entity that is in, i.e., among the contents of, one's conscious state. If we take it that it is mental acts that are the contents of conscious states that are predicated of minds, then of course sensory contents will not be predicated of minds. There is no fear, then, of the sort expressed by the Hausmans,40 that the dangling property that exists when we misperceive a material object will end up being predicated of the mind. Rather, the relevant mental predication will be something like I am sensing red The verb, once again, is used to refer to a certain type of undergoing. The term ‘red’ is here functioning grammatically as an adverb making more specific the type of undergoing that is predicated of the subject. The adverb is telling one how the subject, I, is sensing. But we should note, again contrary to Chisholm,41 that this grammatical point does not imply that the property red is not among the contents of my conscious state. At the same time, however, since the property that is presented when we misperceive is not a property of a material object and exists, rather, only among the

280

Body, Mind and Self in Hume’s Critical Realism

contents of my conscious state, there is a sense in which it can be taken as modifying my consciousness alone and not a material object. But to say that it, in this sense, modifies my consciousness is not to say that it is a property of consciousness in the way that the property of which I am aware when I don’t mispercieve is a property that modifies the material object: the notion of modification is different in the two cases. So the system of the vulgar. Let us now shift to the system of the philosophers. The issue here is this, what happens in this system to the material properties of things? What Hume argues, as we know, is that all the properties that we sense are dependent for their existence upon the state of our sense organs. They therefore exist only as the contents of our conscious states. That makes them mental in the fifth of the senses that we have distinguished. In the system of the philosophers no sensory content is a material property of an object. Every sense impression is mental, a modification of one’s state of consciousness. All sense impressions have the status that non-material properties have in the system of the vulgar; that is, in the system of the philosophers all sense impressions play the role that such impressions play in misperception in the system of the vulgar. In short, in the system of the philosophers all sensory awareness is misperception and when it attributes a sensed property to a material object it projects a modification of consciousness misleadingly onto the objective cause of the sense perception. We have in fact seen, in the preceding chapter, that none of this should be puzzling. Yet it has of course bothered considerably those who have not read Hume closely enough, or who wrongly suppose that it is to fall into scepticism if one denies the objective existence of the material objects that are given in perception. In particular, there is nothing puzzling about every sense impression being “in the mind.” The Hausmans suggest that there is something problematic in this.42 We now see that they are wrong. It becomes problematic only if one moves back and forth between the system of the vulgar, where it is not true that every sensory content is nothing but a modification of consciousness, to the system of the philosophers, where every one of them is only that. Once we make the relevant distinctions we see that it is perfectly possible for Hume to hold that sensory properties are both modifications of consciousness and also danglers that are neither mental nor material.

281

Four: Disappearance of the Simple Self: Its Problems

We also see that none of it is particularly puzzling. In the system of the philosophers, there are material objects. These are not perceived, nor are their properties like the sensory qualities that we do actually experience. Our body is one of these material objects. The distal object affects the sense organs of our body. These organs in turn cause to exist both the sensory contents of which we are aware and the consciousness of them. These causal relations are understood in Humean terms, as matter of fact regularities. They relate not only the physical processes but also the mental processes, including our sensory impressions, both among themselves and to the physical processes. Shalom has written of all this in a way that makes it rather more mysterious than it is: ...I am using the concept of “internalization” to refer to the very process of the constitution of subjectivities by means of what we, one of the products of this same process, later see and interpret as a “physical process per se.” In other words, from this standpoint “green” exists in the physical universe in the sense that it is a process of inorganic reality that is occurring as such, and that can be internalized by the appropriate physical structure involving sense organs, nervous system, and all the bodily complexities belonging to the human subject. The expression of this internalization is what we call “the experience of green.”43

All this is no doubt true when it is interpreted in a Humean way. But in that case, why speak of “internalization”? Why suggest that the process is one in which the (non-perceived) objective qualities are somehow taken up by the mind which brings them into itself through a process that somehow transforms them from the way they are objectively to the sensory qualities that are given in consciousness? What does this add to our Humean understanding of the relevant processes? Why should we use this language of mystification to obscure the basically perfectly pellucid processes that account for our sensory awarenesses? One can, I guess, think of reasons, but none of them are very convincing: generally they beg the question in favour of an Aristotelian, and non-Humean account of causation.

I conclude that Hume is not, contrary to the critics we have examined, implicitly committed to the doctrine of a substantial self. But some have argued that consciousness is something that inevitably eludes explanation. We must look at this position. As it turns out, within the context of a Humean ontology, there is no reason to suppose that consciousness cannot

282

Body, Mind and Self in Hume’s Critical Realism

be incorporated into a science of human being. (3) Explaining Consciousness The position one takes on the nature of scientific explanation is conditioned in part by the metaphysical framework within which one situates explanation. It is this context that I would like to explore. I shall do so using the particular example of the mind-body relation. Explaining this relationship has often been presented as something terribly difficult or ultimately impossible: the connection of mind to body is an ultimate mystery, beyond our capacity to grasp. At least that is so if one is not clear on how such judgments are shaped by ontological principles. At the same time, much of the explanatory mystery disappears if one correctly situates one’s account of explanation in an empiricist metaphysics. The example with which I shall start is that of Colin McGinn, who has asked,44 “How is it possible for conscious states to depend upon brain states?” He continues, “How could the aggregation of millions of individually insentient neurons generate subjective awareness?” (p. 349). He does allow that consciousness is not to be eliminated, and that it has correlated with it certain brain states. “If we are not to be eliminativists about consciousness,” he says, “then some theory must exist which accounts for the psychophysical correlations we observe” (p. 353). But McGinn suggests that the question he posed cannot be answered: by the very fabric of our being, by the very structure of our cognitive faculties, we are not in a position to answer it. “The problem arises,” he suggests, “because we are cut off by our very constitution from achieving a conception of that natural property of the brain (or of consciousness) that accounts for the psychophysical link” (p. 350). McGinn defines the notion of what it is for a mind to be “cognitively closed”. A mind M is “cognitively closed with respect to a property P (or a theory T)” just in case that “the concept-forming procedures at M’s disposal cannot extend to a grasp of P (or an understanding of T)” (p. 350). McGinn then takes ‘P’ to refer the property of the brain state that is causally responsible for the correlated conscious state. We have direct access as it were to one term of the brain state/conscious state pair, namely, the conscious state. But we have no such access to the other end of the relation; the inner awareness of the conscious state does not involve any sort of similar awareness of the correlated brain state. “We have direct cognitive access to one term of the mind-brain correlation, but we do not have such access to the nature of the

283

Four: Disappearance of the Simple Self: Its Problems

link. Introspection does not present conscious states as depending upon the brain in some intelligible way” (p 354). We have by inner awareness a knowledge of the properties of conscious states. But none of these properties is presented as having an intrinsic connection to the properties defining the brain state upon which the conscious properties are dependent. As for the properties of brain states, these are presented not in inner awareness but in perception. Now, with regard to perception, we can say that the objects of perception are localized in space: “the senses are geared to representing a spatial world; they essentially present things in space with spatially defined properties” (p. 357). However, these properties too lack the requisite connection: “it is precisely such properties that seem inherently incapable of resolving the mind-body problem: we cannot link consciousness to the brain in virtue of spatial properties of the brain” (p. 357). We have two faculties, inner awareness and perception. The former presents conscious states. But the properties of these states are not presented as depending upon the brain. As for perception, this presents brain states but like all perceptual objects these states are states in space. But properties of spatial objects are not presented as linked to the properties of conscious states. So, through neither of our faculties are we presented with P, the property of brain states upon which conscious states are causally dependent. But we cannot form concepts of properties unless we are aware of them. In particular, “you cannot form concepts of conscious properties unless you yourself instantiate those properties” (p. 355). Thus, you cannot form the concept of a property of a conscious state unless that property is presented in a conscious state. But the same is true of properties of perceptual objects: unless such a property is presented to one in an object of which one is perceptually aware, then one cannot form the concept of such a property. It follows that we cannot form a concept of the property P, since it is given neither in perception nor in inner awareness. If, contrary to what the facts seem to be, we could grasp this property then equally we could grasp a theory that would explain the dependence of mind upon body. Indeed, “it seems ... that grasp of the theory that explains [...]experiences would confer a grasp of the nature of those experiences. ... How could we grasp the nature of [...]experiences without grasping the character of those experiences” (p. 355). On McGinn’s view, there is a connection between the properties that characterize the generator and the properties that characterize the generated: in grasping the former one ipso facto grasps the latter. How else are we to interpret the notion of one set of properties “conferring”

284

Body, Mind and Self in Hume’s Critical Realism

knowledge of the intrinsic character of another set of properties? Upon McGinn’s account, then, there is, there in the properties, – objectively there in the properties –, a connection that so links the brain states to the conscious states that through it alone one can infer the latter from the former. Evidently this connection, this property P that does the connecting, is an objective necessary connection. But we do not know such a connection: it is presented to us neither in outer perception nor in inner awareness. It follows from all this that the mind is, in McGinn’s terms, cognitively closed with respect to property P. Finally, it follows still further that the mind is cognitively closed also with respect to any theory about the mind-body relation since any such theory to be adequate would describe the causal relations between mind and body and that would require us to have a concept of the property P which is the property of brain states which ensures that the latter generate conscious states. The mind-body relation is therefore inevitably a mystery to us, and the problem of the mind-body relation is one that our cognitive faculties ensure we cannot solve. This is an intriguing argument. McGinn presents it almost as a piece of commonsense, or at least scientific commonsense. A closer examination suggests, however, that it is hardly that. Thus, McGinn holds, for example, to the principle of determinism, that for every event there is causal law which explains it, and in particular holds that this applies to mental events. McGinn even holds that it is a piece of this commonsense that every mental event has as its cause some state of the brain. At the same time, however, he also holds that the connection between brain states and conscious states is unknowable. How, then, could he have any evidence that there is in fact a causal connection between the brain and our conscious states? After all, he has never had, nor, if his argument is correct, could he ever have had, any instances that could confirm that there is such a causal connection. So he doesn’t really know if the causal principle actually holds between brain states and conscious states. Again, he apparently begins with a perfectly reasonable appeal to the empiricist’s Principle of Acquaintance (PA).45 For he argues that what is presented to consciousness is as it is, and neither not something else nor anything that in itself points beyond itself to some other property or being. That is the basis, one presumes, for his rejection of eliminativism with regard to conscious states. And he states, as we have seen, that “you cannot form concepts of conscious properties unless you yourself instantiate those properties.” Yet at the same time he holds that there is some sort of

285

Four: Disappearance of the Simple Self: Its Problems

connection between brain states and conscious states with which we are not acquainted and, indeed, with which, given our cognitive faculties as he views them, we could not be acquainted: as he puts it, the mind is cognitively closed to it. Yet here he does not consistently apply the empiricist’s PA, and deny that such properties should be admitted into one’s ontology. The apparently simple argument thus becomes more complex, with a display of metaphysical principles which are, however, simply not applied consistently. It is this interplay of metaphysical principles and accounts of explanation that we need here to explore. We may begin by noting that McGinn’s appeal to PA contrasts with other approaches to the mental. Thus, for example, Donald Davidson offers a very different characterization of the mental. An event is mental just in case that it is described by a verb which is mental such as hoping, believing, perceiving, etc., where such verbs are “characterized by the fact that they sometimes feature in sentences with subjects that refer to persons, and are completed by embedded sentences in which the usual rules of substitution appear to break down.”46 McGinn, like Hume, deploys an empiricist framework (though perhaps less consistently than Hume), where Davidson deploys a linguistic framework. There is another important difference. For Davidson, events are distinguished by their causal relations47: if two events stand in the same causal relations, then they are the same event. Properties do not serve to distinguish events. An event may have different descriptions but be the same event. Descriptions apparently do not apply to events by virtue of the properties present in those events. In fact, events, while they stand in causal relations, do not seem to have properties.48 This is part of what Davidson seems to mean when he rejects the correspondence account of truth.49 But it is on the basis of his notion that events are not defined by their properties, that he argues that therefore mental events can be identical with physical events. McGinn in contrast clearly allows that one must take account not only of events but of the properties that are present in them. The argument for this latter position is based on the empiricist’s PA: events are presented to us as having properties, e.g., the property of being a conscious state. or, in the case of physical events, various spatial properties, and since these properties or kinds are presented they are to be admitted into one’s ontology. Not that properties are totally neglected in Davidson’s system. A casual relation holds between events provided that there is a strict lawlike

286

Body, Mind and Self in Hume’s Critical Realism

generalization that holds between a pair of descriptions that are true of the event, where “Lawlike statements are general statements that support counterfactual and subjunctive claims, and are supported by their instances.”50 But, Davidson argues, while every mental event, insofar as it is identical to a physical event, has a cause, this cause is another physical event. There are in fact strict laws only for events under physical descriptions. There are not the same sort of strict laws for events under mental descriptions. Hence the view at which Davidson arrives he characterizes as “anomalous monism,” ‘monism’ because it holds that every mental event is identical to a physical event, but ‘anomalous’ because it holds there are no strict correlations that link events under their mental descriptions to events under their physical descriptions. The argument for the absence of physical-mental correlations is based in the first instance on the fact that we do not have such correlations. This, though, could simply be due to gaps in our knowledge of relevant factors.51 Davidson argues that it goes deeper than that. A gappy theory may be homonomic or it may be heteronomic. If the gappy generalization is the former then it “points to the form and vocabulary of the finished law”; in contrast, where the filling in of the gappy generalization requires “shifting to a different vocabulary” the theory is heteronomic. We are told that Confidence that a statement is homonomic, correctible within its own conceptual domain, demands that it draw its concepts from a theory with strong constitutive elements.52

The interpretation of behaviour requires the observer to advert to the beliefs and attitudes of the person whose behaviour is being interpreted. As we try to understand the behaviour we make the assumption that the beliefs, preferences, intentions, actions, etc., hang together rationally: “the constitutive ideal of rationality partly controls each phase in the evolution of what must be an evolving theory.”53 It is not merely a matter of finding other variables that link parts of a process, to thereby increase our understanding of the process, as in the case of physical theory; it is, rather, that there is a guiding standard that directs us how to achieve greater understanding, and this does not simply require the discovery of more variables: it requires where necessary instead a re-interpretation of the given material. Davidson’s point is satisfactory as far as it goes: this is indeed what goes on in much of everyday attempts to understand others. But we also

287

Four: Disappearance of the Simple Self: Its Problems

want to develop theories – theories of learning – which describe how people come to conform their thoughts to the principles of rationality, and how, too, they sometimes fail so to conform them. Once we pass beyond verbal and other intentional behaviour to the factors that determine how language is learned, we have passed beyond simply the principles of rationality and are demanding to know the factors that determine how these norms come to be internalized. The laws of learning connect the mental, even in Davidson’s broad sense, to the physical. Once we introduce the role of learning into the picture of human behaviour, then there seems to be little reason to suppose that the laws connecting the mental and the physical are heteronomic. Certainly, this is the position that McGinn adopts, the position in which there are strict correlations, whether we know them or not, between the physical – our brain states – and the mental – our conscious states. It seems a reasonable position. McGinn’s issue is not whether there is such a correlation, but whether it is a case of causation. McGinn asks, as we have seen, “How could the aggregation of millions of individually insentient neurons generate subjective awareness?” (emphasis added). McGinn is hardly the first to appeal to this principle. Thus, already Zeno of Citium, the founder of the Stoic school in antiquity, had made it a central point of the philosophy of his school that “Nothing which is devoid of sensation can contain anything which possesses sensation.”54 Zeno went on: “Now some parts of the universe possess sensation; therefore the universe is not devoid of sensation” (ib.). Zeno was not the last to suggest such a solution to the problem of the causal connection between body and mind, the causal connection through which body generates or causes mind. Thus, Thomas Aquinas argues that “the generator is like the generated,” and as a consequence sensitive creatures derive their being from creatures with sensitive souls or forms. Now the more powerful an agent, the greater scope its action has: for instance, the hotter a body, the greater the distance to which its heat carries. Therefore bodies not endowed with life, which are the lowest in the order of nature, generate their like, not through some medium, but by themselves; thus fire by itself generates fire. But living bodies, as being more powerful, act so as to generate their like, both without and with a medium. Without a medium – in the work of nutrition, in which flesh generates flesh: with a medium – in the act of generation, because the semen of the animal or plant derives a certain active force from the soul of the generator, just as the instrument derives a certain motive power from the principal agent. And as it matters not whether we say

288

Body, Mind and Self in Hume’s Critical Realism

that something is moved by the instrument or by the principal agent, so neither does it matter whether we say that the soul of the generated is caused by the soul of the generator, or by some seminal power derived therefrom.55

This line of argument continued into the early modern period. And so we find Samuel Clarke arguing in his Boyle Lectures argued that “Perception or intelligence is a distinct quality or perfection, and not a mere effect of composition of unintelligent figure and motion.” And, If perception or intelligence, be any real distinct quality, or perfection; and not a mere effect or composition of unintelligent figure and motion; then beings endued with perception or consciousness, can never possibly have arisen purely out of that which it self had not such quality as perception or consciousness; because nothing can ever give rise to another any perfection, which it hath not either actually in it self, or at least in a higher degree.56

And intelligence or consciousness is distinct. As Clarke puts it, “Intelligence is not figure, and consciousness is not motion. For whatever can arise from, or be compounded of any things; is still only those very things, of which it is compounded.”57 McGinn makes the same point: “Consciousness does not seem made up out of smaller spatial processes, yet perception of the brain seems limited to revealing such processes” (McGinn, p. 357). Conscious states are generically different from brain states. As McGinn puts it, ... a certain principle of homogeneity operates in our introduction of theoretical concepts on the basis of observation....consciousness itself could not be introduced simply on the basis of what we observe about the brain and its physical effects. If our data, arrived at by perception of the brain, do not include anything that brings in conscious states, then the theoretical properties we need to explain these data will not include conscious states either (McGinn, p. 358).

At the point of Clarke’s discussion that we have noted, he has already inferred that there is a necessary being as the cause of the world. The issue is whether this being could be the material world itself. Clarke argues that, as the necessary being must, among other things, cause not only cogitative substances but also sensitive substances, this being must be intelligent and therefore cannot be matter. “It must needs be,” Clarke holds, following Zeno and Aquinas, “that, in the order of causes and effects, the cause must always be more excellent than the effect...”58 The self-existent being, whose existence has already been proven, must therefore contain intelligence, and could as a result not be material.

289

Four: Disappearance of the Simple Self: Its Problems

...an unintelligent being, ’tis evident, cannot be endued with all the perfections of all things in the world; because intelligence is one of those perfections. All things therefore cannot arise from an unintelligent origin: And consequently the self-existent being, must of necessity be intelligent (ib.).

Richard Bentley had earlier argued much the same point in his Boyle Lectures. He argued that “no action and passion of the animal spirits, one particle upon another, can create any sense and perception.”59 This is because “in our conception of any quantity of body there is nothing but figure, and site, and a capacity of motion” (p. 38); from this homogeneity of matter it follows that “sense and perception can never be the product of any kind of matter and motion...” (p.46). It follows that sense and perception “must necessarily proceed from some incorporeal substance within us” (ib.). He then draws much the same conclusion as Clarke: ...since we have shewn that there is an incorporeal substance within us, whence did that proceed, and how came it into being? It did not exist from all eternity; that is too absurd to be supposed; nor could it come out of nothing into being without an efficient cause. Something, therefore, must have created our souls out of nothing; and that something (since nothing can give more than it has) must itself have all the perfections that it hath given to them (p. 47).

These inferences all depend upon much the same causal principle that McGinn invokes, that the cause generates the effect. But McGinn rejects these inferences which “invoke supernatural entities or divine interventions”60; they invoke entities which are not “naturalistic,” that is, one presumes, entities that violate the empiricist’s principle PA. In contrast to Zeno, Aquinas, Clarke and Bentley, McGinn rests content with the simple conclusion that the relation between mind and body must remain mysterious to us: “The approach I favour is naturalistic but not constructive. I do not believe that we can ever specify what it is about the brain that is responsible for consciousness, but I am sure that whatever it is it is not inherently miraculous.”61 There is another point on which McGinn seems to be in disagreement with those who earlier accepted the causal principle that denies that material objects such as brain states, so far as we know these, cannot generate conscious states. Here is how Bentley lays out his view of the conscious states with which he is concerned. I shall lay it down as self-evident, that there is something in our composition

290

Body, Mind and Self in Hume’s Critical Realism

that thinks and apprehends, and reflects and deliberates; that determines and doubts, consents and denies; that wills, and demurs, and resolves, and chooses, and rejects; that receives various sensations and impressions from external objects, and produces voluntary motions of several parts of our bodies.62

Bentley’s appeal is to the empiricist’s PA, and is much the same as McGinn’s similar appeal. But the point to be noted in Bentley’s discussion is his insistence that conscious states as well as being caused are also causes. Specifically, they are causes of changes in the states of bodies, the bodies of which they are conscious states. This is most clear in the case of willing and demurring, but if we allow that every conscious state has a corresponding brain state then the conscious states that bring about such activities as inferring and resolving will also effect changes in bodily states, namely, they will bring about those various brain states that correspond to these various processes. But if McGinn is concerned with these casual relations, he does not so indicate. It seems beyond his concern that not only do bodily states generate conscious states but conscious states generate bodily states. Clarke did allow that there are states of consciousness that are not caused by material things, brain states or what not, but which are nonetheless correlated with such states. These are the sensible impressions that we have of things. “Colours, sounds, tastes, and the like,” he tells us, “are by no means effects arising from mere figure and motion...”. Contrasting the picture of the world of the new science with the world of common sense, he argues that such sensible impressions are not in material objects: “there being nothing in the bodies themselves, the objects of the senses, that has any manner of similitude to any of these qualities.” Since these qualities are not contained in these entities, they cannot be generated by them; those objects are indeed regularly connected with them but do not have them as effects. Rather, they are simply events that are occasioned in consciousness by the material objects that appear to be their causes: “they are plainly thoughts or modifications of the mind it self, which is an intelligent being; and are not properly caused, but only occasioned, by the impressions of figure and motion.”63 Locke was to agree with Clarke on this point. Locke argues, to use McGinn’s way of speaking, that we cannot grasp how it is that body, entities extended and moving in space, can generate these impressions. ’Tis evident that the bulk, figure, and motion of several Bodies about us, produce in us several Sensations, as of Colours, Sounds, Tastes, Smells,

291

Four: Disappearance of the Simple Self: Its Problems

Pleasure and Pain, etc. These mechanical Affections of Bodies, having no affinity at all with those Ideas, they produce in us, (there being no conceivable connexion between any impulse of any sort of Body, and any perception of a Colour, or Smell, which we find in our Minds) we can have no distinct knowledge of such Operations beyond our Experience; and can reason no otherwise about them, than as effects produced by the appointment of an infinitely Wise Agent, which perfectly surpasses our Comprehensions....64

Properties are perceived to be just as they are, in themselves; to know them as they are we need not know any of the relations in which they stand to other entities. ... the immediate Perception of the agreement or disagreement of Identity being founded in the Mind’s having distinct Ideas ... affords us as many self-evident Propositions, as we have distinct Ideas. Every one that has any Knowledge at all, has as the Foundation of it, various and distinct Ideas: And it is the first Act of the Mind (without which it can never be capable of any Knowledge) to know every one of its Ideas by itself, and distinguish it from others. Every one finds in himself, that he knows the Ideas he has; that he knows also, when any one is in his Understanding, and what it is; and that when more than one are there, he knows them distinctly and unconfusedly one from another (Essay, Bk. IV, Ch. 8, sec. 2).

Locke’s appeal to an empiricist’s Principle of Acquaintance is clear.65 The point for us is, of course, that once one accepts that properties are simply as they are presented, then, as McGinn has stated, any eliminativist attempt to somehow expel or eradicate properties of conscious states in favour of the properties of extended objects must be rejected. An empiricist metaphysics based on PA immediately implies that conscious states are what they are, that they are what they are presented as, and are not anything else. For Locke what this meant was clear. We are not presented with necessary connections that link the properties of impressions of sense with the primary qualities of the objects that cause those impressions. There are therefore no necessary connections between the two sorts of event. In terms with which he would have understood, the connection is occasional rather real; in our terms, there is correlation rather than causation. What Locke is arguing on the basis of the empiricist’s PA is that in these cases the properties of things, as they are presented to us, are such that all that we can know is correlation. Locke’s scepticism about knowing more than correlations extended beyond the ideas of secondary qualities, these sensible impressions, to

292

Body, Mind and Self in Hume’s Critical Realism

many of the correlations that the new science had established among bodies, gravity for example. As he put it at one point, It is evident, that by mere Matter and Motion, none of the great Phaenomena of Nature can be resolved, to instance but in that common one of Gravity, which I think impossible to be explained by any natural Operation of Matter, or any other Law of Motion, but the positive Will of a Superior Being, so ordering it.66

From whence then do we obtain the idea of causation that enables us to judge that there cases of correlation and regularity are not cases of causation? Locke argues that it is from the will that we obtain our idea of a necessary connection, or, as he puts, our idea of an active power. He proposes “to consider here by the way, whether the mind doth not receive its idea of active power clearer from reflection on its own operations, than it doth from any external sensation” (Locke, Essay, Bk. II, Ch. 21, sec. 4). He argues that it is not through sensation that we obtain the idea of active power: here events are, as he argues, separable, and all that one can obtain is regularity, the passing on of motion (in the case of billiard balls) rather than the initiation of motion. when by impulse [one billiard ball] sets another ball in motion that lay in its way, it only communicates the motion it had received from another, and loses in itself so much as the other received: Which gives us but a very obscure idea of an active power of moving in body, whilst we observe it only to transfer, but not produce any motion (ib.).

It is from inner awareness that we obtain our idea of an active power. Specifically, we obtain it from our experience of the action of the will in volitions that cause bodily action. The idea of the beginning of motion we have only from reflection on what passes in ourselves, where we find by experience, that barely by willing it, barely by a thought of the mind, we can move the parts of our bodies, which were before at rest (ib.).

Hume was later to argue, however, that the case of the will is no different from the other cases: in both, the effect is separable from the cause, there is nothing in the one that objectively necessitates the other. Some have asserted that we feel an energy or power in our own mind; and that, having in this manner acquired the idea of power, we transfer that quality to

293

Four: Disappearance of the Simple Self: Its Problems

matter, where we are not able immediately to discover it. The motions of our body, and the thoughts and sentiments of our mind (say they) obey the will; nor do we seek any further to acquire a just notion of force or power. But to convince us how fallacious this reasoning is, we need only consider, that the will being here considered as a cause has no more a discoverable connexion with its effects than any material cause has with its proper effect....The effect is there distinguishable and separable from the cause, and could be foreseen without the experience of their constant conjunction. We have command over our mind to a certain degree, but beyond that lose all empire over it: and it is evidently impossible to fix any precise bounds to our authority, where we consult not experience. In short, the actions of the mind are, in this respect, the same with those of matter. We perceive only their constant conjunction; nor can we ever reason beyond it. No internal impression has an apparent energy, more than external objects have. Since, therefore, matter is confessed by philosophers to operate by an unknown force, we should in vain hope to attain an idea of force by consulting our own minds. (T, p. 162).

But Hume has already argued this case with regard to what Bentley and Clarke would have taken as standard examples of mechanical causation among physical objects, the case of one billiard ball striking another, as he puts it “the communication of motion, which I see result at present from the shock of two billiard balls...” (T, p. 164). We have here a case of cause and effect: the first billiard ball strikes the second, and the motion of the first is communicated to the second. Now, Hume argues that cause and effect are logically distinct. The appeal is to the empiricist’s PA. In order to find the idea of causation we must turn to the impressions from which that idea derives: “we reject at once all the vulgar definitions which philosophers have given of power and efficacy; and instead of searching for the idea in these definitions, must look for it in the impressions from which it is originally derived. If it be a compound idea, it must arise from compound impressions. If simple, from simple impressions” (T, p. 157). Consequently, ... as all distinct ideas are separable from each other, and as the ideas of cause and effect are evidently distinct, it will be easy for us to conceive any object to be non-existent this moment, and existent the next, without conjoining to it the distinct idea of a cause or productive principle. The separation therefore of the idea of a cause from that of a beginning of existence, is plainly possible for the imagination; and consequently the actual separation of these objects is so far possible, that it implies no contradiction nor absurdity... (T, p. 82).

Just as Locke could appeal to the empiricist’s PA to argue that there were

294

Body, Mind and Self in Hume’s Critical Realism

no objective necessary connections that linked colours, or coloured impressions, to their physical causes, so there are no objective necessary connections that link the parts of a physical process – not any such connections that link a mental process or state such as willing or intending to either a physical or mental upshot. This leads to Hume’s two definitions of “cause”: “If we define a cause to be an object precedent and contiguous to another, and where all the objects resembling the former are placed in a like relation of priority and contiguity to those objects that resemble the latter; we may easily conceive that there is no absolute nor metaphysical necessity, that every beginning of existence should be attended with such an object. If we define a cause to be, an object precedent and contiguous to another, and so united with it in the imagination, that the idea of the one determines the mind to form the idea of the other, and the impression of the one to form a more lively idea of the other; we shall make still less difficulty of assenting to this opinion” (T, p. 172). The first definition captures the point that objectively, in the cause and effect, there is nothing to causation but regularity. In a sense, then, for Hume all connections are occasional in the sense in which Clarke spoke of such connections as occasional. At the same time, however, Hume recognizes that there is a point to the claim that causal connections are necessary. There is the point that Davidson made, that lawlike generalizations or regularities support contrary to fact conditionals. This point Hume captures in his second definition. According to it, those regularities are lawlike which we are prepared to use to predict (“the impression of the one leads us to form a more lively idea of the other”) and to support contrary to fact or subjunctive conditionals (“the idea of the one determines the mind to form the idea of the other”). Hume’s definitions do not, of course, distinguish science from superstition. The latter like the former treats certain generalizations as lawlike, but unlike the former such treatment is unreasonable. Hume proceeds to distinguish those regularities which it is reasonable to treat as lawlike from those which are not.67 Those that are rational or reasonable are those that satisfy the condition that we have arrived at them through a process that conforms to the rules of the new experimental science, what Hume calls the “rules by which to judge of causes and effects.”68 A causal explanation now consists in the subsumption of the events to be explained under a causal regularity. Earlier thinkers such as Locke and Clarke distinguished occasional “causes” from genuine causes, where the latter provide a connection between cause and effect where the former

295

Four: Disappearance of the Simple Self: Its Problems

did not. In that sense, mere regularities did not provide explanations; the latter were available only when one could discover real causes, with objective necessary ties. But with Locke’s appeal to PA as carried out consistently by Hume, the older notion of real causes as distinguished from occasional causes disappeared. There is nothing, objectively speaking, to causation beyond regularity. Which is not to say, as we have seen, that there is no distinction to be drawn between mere regularities which do not provide good explanations and causal regularities which do: the point is that this “necessity” which is the mark of the causal is not there objectively in the cause and the effect but is there only in our thought about cause and effect. The point that I want to make is that once one accepts the empiricist’s PA, and deploys it after the fashion of Hume, then the problem raised by McGinn straight off disappears. As we saw, he asks the question, “How could the aggregation of millions of individually insentient neurons generate subjective awareness?” where this is meant to demand something – a cause or a generator – over and above mere correlation. Indeed, he tells us in the essay that if we were to know this causal property of brain states in the case of bats then “We would be in possession of the same kind of understanding we would have of our own experiences if we have a correct psychophysical theory of them.” This theory is such that a “grasp of the theory that explains [bat]-experiences would confer a grasp of the nature of those experiences.” The relevant concept for bat-experiences he abbreviates as “B,” and clearly indicates that if we are to understand such a concept then we must somehow be acquainted with its referent. The property of the brain that causally generates the experiences must somehow so connected to the property B as to contain the latter within it, that is, contain it in such a way that in grasping the former property we ipso facto grasp the latter. How else, he asks, rhetorically one presumes, “could we grasp the nature of B-experiences without grasping the character of those experiences?” (McGinn, p. 355). This connection which ensures that if the one property is presented then so is the character of the other is a connection between the two properties, and is objectively there in the properties. But McGinn also both accepts PA and accepts that we are not presented in either consciousness, inner awareness, or perception, outer awareness, with such a causal property, such an objective necessary connection, one that transforms merely correlated events into events one of which generates, by virtue of containing, the other. If we take Hume’s

296

Body, Mind and Self in Hume’s Critical Realism

empiricist argument seriously, then there is no reason to claim that there is any such causal property or generator, no such objective necessary connection. There is therefore no problem facing McGinn. The answer to the question about how could brain states generate conscious states lies precisely in the regularity that correlates them: that regularity explains the connections when the two events are subsumed under it. Of course, on this point Davidson is correct, the correlations that we have, the regularities, are gappy. But as McGinn recognizes, we have reason enough to think that the gaps can in principle at least be filled. And when they are, then those regularities will provide everything we need to know in order to understand how it is that brain states can cause or generate conscious states, or, for that matter, for conscious states to cause or generate brain states and from them various molar forms of behaviour. McGinn creates the problem that he finds by not consistently following through with his own empiricist framework. He insists that there is some objective causal property that makes for generation over and above regularity. But then, given his empiricist principles, he cannot find a property. So he proclaims an insolvable mystery. Hume himself considered a similar case, that of the Cartesians. We have established it as a principle, that as all ideas are derived from impressions, or some precedent perceptions, it is impossible we can have any idea of power and efficacy, unless some instances can be produced, wherein this power is perceived to exert itself. Now, as these instances can never be discovered in body, the Cartesians, proceeding upon their principle of innate ideas, have had recourse to a Supreme Spirit or Deity, whom they consider as the only active being in the universe, and as the immediate cause of every alteration in matter. But the principle of innate ideas being allowed to be false, it follows, that the supposition of a Deity can serve us in no stead, in accounting for that idea of agency, which we search for in vain in all the objects which are presented to our senses, or which we are internally conscious of in our own minds. For if every idea be derived from an impression, the idea of a Deity proceeds from the same origin; and if no impression, either of sensation or reflection, implies any force or efficacy, it is equally impossible to discover or even imagine any such active principle in the Deity. Since these philosophers, therefore, have concluded that matter cannot be endowed with any efficacious principle, because it is impossible to discover in it such a principle, the same course of reasoning should determine them to exclude it from the Supreme Being (T, p. 160).

If we find, as McGinn does, that we cannot form an idea or concept of the supposed causal generating power, then he ought to have no recourse to

297

Four: Disappearance of the Simple Self: Its Problems

such a power to create for us a mystery: if you can’t have recourse to it then you can’t have recourse to it to create a mystery. And without it, there is no mystery. “We know,” he says, “that life evolved from inorganic matter, so we expect there to be some explanation of this process. We cannot plausibly take the arrival of life as a primitive brute fact, nor can we accept that arose by some form of miraculous emergence” (McGinn, p. 353). But of course it is not miraculous. To the contrary, given the regularities that connect mind and body, consciousness and brains, then this is precisely what one would expect. There is another aspect to McGinn’s argument that is deserving of comment. He notes that “Crudely, you cannot form concepts of conscious properties unless you yourself instantiate those properties” (McGinn, p. 355). He uses as examples the person born blind who cannot grasp the concept of a visual experience of red, and “human beings [who] cannot conceive of the echolocatory experience of bats” (ib.). The former is a standard empiricist example. The latter is not, and in that regard is more interesting. As McGinn sees it, “one’s form of subjectivity restricts one’s concepts of subjectivity” (p. 356). McGinn’s point is that if we could understand how brain states generate conscious states then we could form the concept of what is going on in the bat’s consciousness. He refers to the bat experiences as “B”, and the causal power that generates them from the bat’s brain “P1”. He then argues that “By grasping P1 it would be perfectly intelligible to us how the bat’s brain generates B-experiences; we would have an explanatory theory of the causal nexus in question.” He then goes on to make the point that we have already noted, that “it seems to follow that grasp of the theory that explains B-experiences would confer a grasp of the nature of those experiences; for how could we understand that theory without understanding the concept B that occurs in it? How could we grasp the nature of B-experiences without grasping the character of those experiences?” (p. 355) But of course, we do not grasp the nature or character of B-experiences: our concepts of consciousness are limited to our own subjectivity. It follows that we do not have a conceptual grasp of the causal power P1. Nor equally do we have a grasp of the causal power that enables brain states to generate our own conscious states. The fact that we do not grasp the character of B-experiences shows that not only are we not aware of them subjectively as it were, as the bat is aware of them, but that we are also not aware of the causal power that is supposed to contain them necessarily and thereby generate them. Now, it is true that we grasp no such causal property. It is also true that we are not

298

Body, Mind and Self in Hume’s Critical Realism

subjectively aware of B-experiences, as is the bat. But do those experiences, the bat’s experiences, really systematically elude us? The example is not McGinn’s; it starts with T. Nagel’s essay on “What Is It Like to Be a Bat?”69 Nagel argues, like McGinn and like Bentley much earlier, that the character of conscious states cannot be eliminated: it is given to us in experience as it is. “Without some idea...,” he points out, “of what the subjective character of experience is, we cannot know what is required of physicalist theory” (Nagel, p. 167). Within this context, established by an appeal to PA, Nagel asks about the conscious experience of a bat. He argues that we cannot in fact form a concept of the experiences of a bat: we do not have those experiences nor experiences of those types, that is, those specific sorts; we do not, for example, have a way of effectively localizing sounds in the way that bats do – colours we can localize with considerable accuracy, but not sounds. We seem to be restricted to “certain general types of mental state” (p. 170) – Nagel suggests appetite and perception might be among these. The examples sound reasonable: bats certain seem to eat in the way that other creatures eat, including ourselves, and we are moved by the appetite of hunger. Similarly, we locate objects perceptually, in our case by means of sight by means of which we localize objects in three-dimensional space; bats, too, locate objects in three-dimensional space, though they do it with sounds and the sense of hearing. But our description will lack “richness of detail” (ib.); it will lack the specificity that we have in our own case as human beings. Our knowledge of the states of the bat’s consciousness will be limited to generic features. For McGinn the point is that the facts of the bat’s consciousness will be inaccessible to us cognitively. The point is no doubt to be accepted so far, at least, as concerns the specific detail. Generic features are something else, however. Moreover, at the same time we can clearly recognize that the relevant facts are indeed there: a bat’s consciousness lies behind its behaviour just a human consciousness lies behind our behaviour. It is just that because we have our consciousness and not the bat’s, we can form concepts of how we experience the world but not how the bat experiences the world. In this way, “We can be compelled to recognize the existence of such facts without being able to state or comprehend them” (p. 171). But is it so easy? Or rather, is it really all that difficult to achieve some sort of understanding of a bat’s consciousness? Now, it is certainly true that we cannot experience the world as the bat experiences the world. Or, to use a somewhat older example, if we

299

Four: Disappearance of the Simple Self: Its Problems

have not been a member of the French resistance, then we cannot have had the experiences that were had by someone who was a member. Al Gazzali pointed out that one can understand what it is to be drunk without oneself being drunk: “…the physician, although not drunk, knows well in what drunkenness consists, and what are its predisposing conditions,”70 From the fact that we have not felt the feeling, it does not follow that we cannot come to a scientific understanding of these conscious states. All that we have to recognize is that such an understanding does not require that we have had the experiences in question, or, to put it a bit differently, we can achieve a scientific understanding, in principle at least, of both bats and fighters in the French resistance and drunken persons without having to have had the experiences, either shared or imagined, of those creatures. What we have to recognize is that there are ways of referring to characteristics of conscious states other than by referring to them directly by names. Specifically, we can refer to these states by means of definite descriptions. The logic of these was first given by Bertrand Russell. Consider the relation “Fxy”: x is a father of y. We have good evidence that For every y, there is exactly one x such that x is a father of y This permits us to introduce the definite description or rather function: the x who is a father of y or, in symbols, (,x)(Fxy) If a is some individual, then we can form from this function the definite description (,x)(Fxa): “the father of a”. We can now refer to this person by means of this definite description even though we do not know specifically who he is. Such a concept was known in the philosophical tradition of Locke and Hume as a “relative term.” Locke spoke of such terms in this way: ...it suffices, for the knowing the precise idea the relative term stands for, to have a clear conception of that which is the foundation of the relation: Which may be done without having a perfect and clear idea of the thing it is attributed to. Thus having the notion, that one laid the egg out of which the other was hatched, I have a clear idea of the relation of dam and chick, between the two cassiowaries in St. James’s Park; though perhaps I have but a very obscure and imperfect idea of those birds themselves. (Locke, Essay, II, 25, viii).

And Hume, as we have seen, used the same logical point to provide us with

300

Body, Mind and Self in Hume’s Critical Realism

references to objects external to the world of sense impressions. The furthest we can go towards a conception of external objects, when supposed specifically different from our perceptions, is to form a relative idea of them, without pretending to comprehend the related objects (T, p. 68).

By means of such definite descriptions, we can refer to objects and properties of objects with which we are not acquainted. In this way we can form concepts of those objects, though it will be, as Locke says, an “obscure and imperfect” idea, referring at best only generically and leaving out specific detail.71 Let us suppose that we have reason to believe that mental states are correlated with brain and behavioural states. If C1 is a conscious state and H1 the bodily or behavioural state with which it is correlated, then we will have something like (*) (x)(C1x if and only if H1x) We may not know C1 specifically, however, as in the case of the bat (or the drunk or Jean-Paul Sartre). We may only know that the state that is correlated with H1 is a unique species of a certain genus, C, “consciousness”. Given this generic feature, there is, in McGinn’s terminology, a homogeneity between the conscious states of the bat and conscious states of humans, including oneself. Since one is acquainted with this kind of state, specifically in the case of oneself, it follows that this concept “C” satisfies the empiricist’s PA. It follows that the term or concept can be admitted into the vocabulary of the empiricist’s language, and that the kind can be admitted into the empiricist’s ontology. Where we do not know, as in the case of the bat, specifically what is correlated with brain states, then perforce we do not know a specifically detailed law of correlation like (*). We do have reason, however, to suppose that there is correlation between the brain states of the bat and its conscious states in all the specific detail. We know that whenever the bat is in the appropriate brain state then there is a conscious state, that is, that there is a state the properties of which fall under the genus conscious state. In that case, what we know is that (**) (›!f)[Cf & (x)(fx if and only if H1x)] But the uniqueness condition enables us to introduce the definite description to refer to the property in question: (,f)[Cf & (x)(fx if and only if H1x)] We can now use this definite description to refer to the relevant characteristic or property of conscious states. It constitutes a concept of a

301

Four: Disappearance of the Simple Self: Its Problems

property or characteristic with which we are not acquainted. This concept that we have of that property or characteristic is, to be sure, “obscure and imperfect”. We have only a generic description of the characteristic; we have no concept of what it is specifically. But even an “obscure and imperfect” concept is better than nothing, and more than either McGinn or Nagel suggest that we have. As Locke and Hume clearly understood, the empiricist has in his or her logical basket the tools that are needed to refer to things that have never been presented to us or to their characteristics that have never been presented to us. It is true, of course, that if we know C1 then we also know that C1 = (,f)[Cf & (x)(fx if and only if H1x)] But we don’t know C1, and so all that we have is the definite description. Let us abbreviate the definite description this way: C* = Df (,f)[Cf & (x)(fx if and only if H1x)] Since C1 = C* we have (l) (x)(C*x if and only if H1x) which, when the definite description is expanded according to the usual rules, is logically equivalent to (**). Let us now suppose that we observe j in state H1. We have H1j This permits us to use the law (l) to deduce that C*j Having thus inferred the cause from the effect, we can now explain that effect in deductive-nomological fashion (x)(C*x if and only if H1x) C*j _______________________ So, H1j Explanations of this sort, that is, ex post facto explanations based on gappy laws, are common where there is gappy knowledge.72 Thus, for example we infer the presence of the virus that causes the flu from the disease symptoms we observe in the ill person. Observing the symptoms we infer the presence of “the flu bug.” Having thus arrived at the cause of the symptoms, we now turn around and use that cause to explain the symptoms as the effect of its being present. Of course, such ex post facto explanations based on gappy laws are not good for making predictions: we cannot identify the cause prior to the occurrence of the effect. We cannot predict

302

Body, Mind and Self in Hume’s Critical Realism

that the flu symptoms will occur because we have not yet identified which virus it is that is, that is, is identical to, the flu bug. But it does not follow that we cannot offer the gappy explanation ex post facto. In the case of the flu bug, of course, there is always the possibility of actually identifying the cause. In the case of the bat’s states of consciousness, those are private to the bat; they therefore cannot be identified independently of their effects. Having no independent access to the initial conditions C*, we cannot use the gappy law (l) to predict the presence of the bodily states H1. At the same time, however, we are supposing with McGinn and Davidson that every bodily event has a physical cause. That means that we can predict H1 on the basis of the causal regularity (ll) (x)(if H0x then H1x) If we then have the initial conditions H0k, then we will be able to predict that k is H1. Indeed, since (l) and (ll) jointly imply that (lll) (x)(if H0x then C*x) we can use the initial condition that k is H0 to predict, and to explain – gappily – that k is in the conscious state C*. And if H1 is by physical law always followed by H2, that is, if we also have the regularity (L) (x)(if H1x then H2x) then we can also infer via the correlation of conscious states and bodily states that (LL) (x)(if C*x then H2x) so that from the knowledge that k is in the conscious state C* we can predict and explain – gappily – why k comes to be in the behavioural state H2. But if we want a gapless explanation of the brain event H1k in the bat then we use the purely physical laws (LL) and (L), which describe and explain a purely physical process. The same sort of explanation is available in the case of human beings, except that in principle explanations can be gapless, since we know from our own case what the specific features of conscious states are. We do need to refer to the conscious states of others by means of definite descriptions, but the descriptions can be based on specific predicates rather than the generic ones that appear in laws like (l) = (**) or like (LL). In our case, the correlating laws that we know are like (*), specific in their reference to the properties or characteristics of conscious states rather than merely generic as in the case of the bat. To be sure, that the laws are like (*) is a matter of principle rather than practice, but as we argued against

303

Four: Disappearance of the Simple Self: Its Problems

Davidson there is no need to think that the existing gaps in our knowledge of the mind-brain correlations cannot in principle be filled in. This account of the mind-body connection makes two assumptions. There is, in the first place, the assumption, which McGinn makes and which we have argued that Davidson is wrong to criticize, that there are gapless correlations between conscious states and bodily states. And it assumes, in the second place, along with McGinn and this time also with Davidson, that the physical causal process is in fact closed and gapless.73 It is this second assumption that raises certain issues. It suggests that mind has no causal impact on body. In fact, as we have seen, McGinn seems to accept this position. For he raises the issue of how bodily events can generate conscious states, but he does not consider the corresponding issue of how conscious states can generate bodily states. The physical world of which the body is a part seems to be closed to the impact of the mental. But this conflicts with the commonsense view, and the view of such earlier thinkers as Clarke, Bentley, Locke and Hume, that if physical events can bring about changes in consciousness, equally the conscious events can bring about changes in the physical. The view that the mental cannot bring about changes in the physical is often called “epiphenomenalism,” the theory once described by C. D. Broad as maintaining the one-sided action of body on mind.74 But such a characterization presupposes that mind is as it were there to be affected. However, on the view of McGinn this is not the case. Consciousness is rather a series of events which are generated or created by the physical events. But with this qualification, Broad’s idea is clear enough, and McGinn equally clearly opts for that view of the mind-body relation. T. H. Huxley argued for the same view as McGinn, dealing with frogs rather than bats. Huxley writes that “if a frog, in his natural state, possesses anything corresponding with what we call volition, there is no reason to think that it is anything but a concomitant of the molecular changes in the brain which for part of the series involved in the formation of motion.”75 But he applies the same results to human beings. “It seems to me that in men, as in brutes, there is no proof that any state of consciousness is the cause of change in the motion of the matter of the organism” (p. 244) As for the soul, it “stands related to the body as the bell of a clock to the works, and consciousness answers to the sound which the bell gives out when it is struck” (p. 242). There is a correlation between brain states and conscious states, but the causal relation goes in only one direction, from body to mind: consciousness is generated by the brain but does not affect

304

Body, Mind and Self in Hume’s Critical Realism

it. This view of the correlation between brain and consciousness was not the view of David Hartley, whose theory was, according to E. G. Boring, “the first thorough theory of psychophysical (really psychophysiological) parallelism.”76 Hartley’s book, Observations on Man,77 is the first book that a present day scientific psychologist would recognize as a text in psychology – at least he or she would so regard it until they arrived at volume II, which treats of man’s relation to God in associationist terms.78 But volume I is a thorough treatment of human psychology in a way that presupposes a systematic correlation between brain states and conscious states. Hartley argues that we must ascribe the “Performance of Sensation to Vibrations excited in the medullary Substance...” (Hartley, p. 33). He goes on to deny that this is to hold that “Matter can be endued with the Power of Sensation.” Rather, he takes the action to be a piece of commonsense, “established by the common Consent of Physicians and Philosophers...”. He agrees with Locke that we are as ignorant of the cause of gravity as we are of sensation: “...we are intirely at a Loss to determine, in which Mechanical Way each Atom contributes to the Gravity of the whole Mass....” It is indeed “impossible ... to discover in what Way Vibrations cause, or are connected with Sensations, or Ideas” since “Vibrations be of a corporeal, Sensations and Ideas of a mental Nature” (pp. 33-34), but given that there is the same problem with corporeal things themselves and their actions, it is therefore hardly an objection to his own theory of the connections between mind and body that we cannot see how body can affect mind. Like McGinn, Hartley sees here a mystery in discovering the generating causes of conscious states. But he is arguing that for his purposes the commonsense of regularity suffices. As part of his psychological theory, Hartley gives an account of “Motions which occur every Day in common Life, and which follow the Idea called the Will...” (p. 103). Details aside, the account is in terms of associations, and throughout the conscious states are taken to be correlated with brain states (“vibrations in the medullary substance”). The point here is that Hartley clearly takes it also to be part of commonsense that mind affects body: the will, an idea and therefore mental, produces motions in the bodily substance. Although somewhat earlier thinkers such as Bentley, Clarke and Locke held that it is perfectly pellucid that mind has a power over body where it is not clear how body has a power over mind, Hartley, having established that he will proceed simply in terms of commonsense regularity, takes such regularity for granted here also, in the action of mind

305

Four: Disappearance of the Simple Self: Its Problems

on body. So long as one does not try to go into metaphysics that goes beyond empirical commonsense, then there are problems neither with regard to the action of body on mind nor with regard to the action of mind on body. It was Hume of course, who saw this metaphysical point most clearly: through an appeal to PA, he exorcised the various powers and objective necessary connections that make it seem that body cannot affect mind or that mind cannot affect body. Adopt instead the account of causation and of causal necessity that he proposes and the mystery disappears. What it is important to note is that with the parallelistic hypothesis proposed by Hartley it is possible to hold two apparently incompatible theses, that the physical world is causally closed and that mind affects body.79 Given the correlation between body and mind, it is possible to hold, as McGinn and Donaldson aim to hold, that physical world is a closed causal process: no non-physical variables are necessary if we to explain causally physical events. So in this sense, body and mind do not interact. But as part of the parallelistic hypothesis, it is also possible to hold that there are regularities that link mind to body – laws on the model of our (*) or (LL) – , and if there are such regularities, then there are casual relations that lead from mind to body, from conscious states like willing to bodily actions. So in this sense, body and mind do interact.80 The odd position of epiphenomenalism of the sort defended by Huxley and apparently by McGinn, the doctrine that there is a one-way action of body on mind, is simply not reasonable once one accepts the Humean account of explanation as subsumption under matter-of-fact regularities. We can also see that another sort of objection is out of place. This is raised by H. Feigl,81 and is used as an argument for why we want to eliminate mental events in favour of physical events. The objection is that if we do not do this then we will be left with laws that irreducibly connect the mental to the physical. If we insist, as Feigl not unreasonably thinks we should insist, and as Davidson and McGinn also suppose, that the physical system is causally closed, then we must treat mental states as connected parallelistically to physical states. These correlations are then described by Feigl as “nomological danglers” (p. 15), that somehow exist but actually do nothing (p. 61). Since no one wants nomological danglers, we had best attempt to reduce the mental to the physical in some way that eliminates the former in favour of the latter. Eliminating the mental, we have no nomological danglers. But of course, if we accept PA, then there is no

306

Body, Mind and Self in Hume’s Critical Realism

question of eliminating the mental. At the same time, however, once we accept the Humean account of causation together with the parallelistic hypothesis, then there is no need to accept the notion that mental states merely dangle, doing nothing. To the contrary, one can quite clearly hold that mind affects body: given the regularity account of causation, the causation is simply not one-way. My willing it regularly brings it about that my arm rises. That simply is what it means to say that the willing causes the action. So willings do not just dangle, they do do something, they bring about actions and bodily changes. And one can make similar points about our beliefs, our preferences, our intentions, and so on: they all have effects on the body, shaping the actions that they bring about – they do not just dangle. It is true that if the properties of consciousness are irreducible to the physical, if, that is, eliminativism is to be rejected, as McGinn rejects it by appeal to PA, then the laws that relate mental states to physical states will not be deducible from the laws of physics alone. For, they will contain terms that are not contained in the premises. But to have laws that are not derivable from physical theory in the way in which the laws of chemistry are, in principle at least, derivable from the laws of physics, is a bad thing, to some like Feigl anyway (Feigl, p. 105). Nor, for the same reason, can evolutionary theory as based on purely physico-chemical considerations alone yield the laws that connect mind and body (p. 108). But whoever thought such a theory should exist? To be sure, speculation has been rife, almost as rife now as it was in an earlier age when such speculations were part of the stuff of the new world picture. At one time it was thought one could derive everything from a single law, now it is a single theory. Descartes argued that everything could be deduced from a knowledge of the laws of impact. But Locke, as we saw, disagreed: “It is evident, that by mere Matter and Motion, none of the great Phaenomena of Nature can be resolved, to instance but in that common one of Gravity, which I think impossible to be explained by any natural Operation of Matter, or any other Law of Motion, but the positive Will of a Superior Being, so ordering it.”82 But this is not to say that we cannot through physics provide ourselves with understanding other motions of objects. Locke cites Newton, who “has shewn, how far Mathematicks, applied to some Parts of Nature, may, upon Principles that Matters of Fact justifie, carry us in the knowledge of some, as I may so call them, Particular Provinces of the Incomprehensible Universe.”83 Newton’s book, he goes on, the Philosophiae naturalis principia Mathematica,

307

Four: Disappearance of the Simple Self: Its Problems

, ... will deserve to be read, and give no small light and pleasure to those, who will to understand the Motions, Properties, and Operations of the great Masses of Matter, in this our Solar System, will but carefully mind his Conclusions, which may be depended on as Propositions well proved.84

In effect what has happened is this. Physics is defined in a certain way by the Cartesians and the corpuscularians. Gravity turns out not to be deducible from physical theories so defined. It is therefore deemed mysterious. But it does turn out to be a matter of regularities among ordinary things; it does give us the same sort of understanding as those regularities that are deemed to be among the facts of physics. One therefore comes to re-define the science of physics so as to include these laws even though they are not deducible from the laws previously held to define the essence of matter. This is what we should do with regard to the laws that link consciousness and matter. They cannot be deduced from the laws of physics. But they yield the same sort of understanding of how things are and how things operate. Let us therefore admit them to the body of scientific laws. Let us simply re-define science so as to include these laws. John Stuart Mill, as in so many cases, commented astutely on this issue. He noted that in the earlier history of science there had been many projects such as that of Descartes which aimed to deduce everything from a single law or theory. He observed that “Projects of this kind were universal in the infancy of philosophy, any speculations which held out a less brilliant prospect being in those early times deemed not worth pursuing.”85 But as he also goes on, “It is therefore useful to remark that the ultimate Laws of Nature cannot possibly be less numerous than the distinguishable sensations or other feelings of our nature....” (Mill, System of Logic, III, xiv, 2). The properties of things are there in nature; we distinguish them, but the distinctions are there to be, if not discovered then at least noticed; they are the way the world is. The empiricist’s PA requires us to take these properties and the distinctions among them as the way the world is. It may be that, like Feigl or Descartes or McGinn, we would really like to obtain a theory which would show how the physical objects of the world could “generate” consciousness, that is, a theory which would enable us to deduce from a knowledge of the laws for purely physical objects the laws for consciousness and the laws that relate consciousness to brain states. But as Locke and Hume argued, the qualities of the world are given to us as logically or, if you wish, ontologically separable from one another and

308

Body, Mind and Self in Hume’s Critical Realism

separable from the properties that physics attributes to the fundamental parts of matter. A longing for a theory in physics from which everything can be deduced is in fact just wishful thinking: the world is such that not everything is physical. The empiricist’s PA forces us to recognize this, as indeed McGinn recognizes when he insists on the basis of PA that the eliminativist programme for consciousness must be unsuccessful.86 And once we establish on the basis of PA that these properties and the distinctions amongst them are basic, then we have to accept that the laws which relate these properties to physical properties are not deducible from laws that mention only the latter. To call such properties “danglers” just because they cannot be deduced from a purely physical theory and to imply that they therefore have no real claim to reality is not really an argument against parallelism. Another version of epiphenomenalism that has recently been developed is the doctrine that mental properties “supervene” on physical properties. On this view, mental properties are somehow dependent on or determined by physical properties on which they supervene but are not caused by them. Formally speaking, the supervenience of a family A of properties on another family B can be understood in this way: necessarily, for any property F in A, if any object x has F, then there exists a property G in B such that x has G, and necessarily anything having G has F.87 If F and G are thus related, then F is said to be supervenient on G, while G is the supervenient base for F. From the perspective of Hume’s empiricist account of causal relations, what we have here is a necessary regularity going from G to F. The necessity is of course not an objective connection, according to the Humean. It is rather the felt necessity that comes from our being prepared to use the regularity for prediction and for supporting contrary to fact assertions. It is for all that a case where the necessity distinguishes it from a mere or accidental regularity. The reason that this relation is proposed for describing the mindbody relation is the so-called problem of multiple realizability of mental states relative to the supposedly more basic physical states.88 The notion is that the mental state F is “realized” in some cases by G in B and in other cases by G’ in B. This means that there cannot be a neat one-one relation between mental states and brains states, nor, therefore, a neat causal relation between them. It is argued that this cannot be a case of causal necessity. As Jaegwon Kim has put it, “there seem to be mental states which are ‘nomologically incommensurable’ with respect to

309

Four: Disappearance of the Simple Self: Its Problems

neurophysiological or, more generally, physical properties; there appear to be mental states which do not nomologically correlate with physical states.”89 It is not worth arguing over the term ‘causal’. There may be some reason for holding that only a one-one relation from mind to body is a case of causation where a one-many relation from mind to body is a case of supervenience. It does mean that if we are trying to eliminate the supervening property in favour of its base, then the property is eliminated not into a single basic property but into a disjunction of such properties. But given the empiricist’s PA, we are not going to be able to eliminate it, so this is not a problem. It does not seem to be a problem anyway. Why not reduce a simple property to a disjunctive property? I can think of no reason for not so doing, except for some metaphysical bias against disjunctive, or, as they have also been called, determinable properties. Being coloured is a perfectly good property. For an object to be coloured means for the object to be red or orange or yellow or ... , where each of the disjuncts itself has the (second order) property of being a colour. To say that x is coloured is to say that there is a property f such that f is a colour and x is f: x is coloured = Df (›f)(f is a colour & x is f) So long the second order or generic property of being a colour is, as it should be, admitted on the basis of PA into one’s ontology, then there seems to be no problem in allowing being coloured to be a property among properties, not to be sure a simple property, but for all that a property. Certainly, it is a property in the sense that it can be predicated of objects, and in the sense that it can enter into regularities that can reasonably be treated as lawlike, e.g., “Whatever is coloured is extended.” The definition of supervenience tells us that where we have a base property G with F supervening upon it, then it is true that anything having G necessarily has F. This means that there is a regularity to the effect that whatever is G is F. This is the FIRST regularity we should note. We are also told that whenever we have F present in an object x then there is a property g such that g is of the generic sort B and g is in x. We have here a SECOND regularity. It is also necessary and therefore lawlike. This regularity links F on the one hand to, on the other hand, a determinable property: (›g)(g is in B & x is g) We might call this property B* Each of the G’s that is in B and is a base property for F is then a specific or

310

Body, Mind and Self in Hume’s Critical Realism

determinate instance of the determinable property B*; it is an instance of this determinable property B* because it is a species under the genus B which defines the determinable property. The point is that each G which is a specific instance of B* is a sufficient condition for F. This is the first regularity that we noticed. If each such G is sufficient, their disjunction is also sufficient. But the second regularity that we noticed states that if F is present then at least one of the G’s must be present. This means that F is sufficient for the disjunction of the relevant G’s. It follows that where we have the supervenience relation we have a regularity which is one-one between the supervening property and the determinable property which is the disjunction of the base properties. The regularity is one which is necessary, that is, which supports predictions and contrary to fact assertions. That makes it lawlike, even if for whatever reason you choose not to refer to it as “causal.” Supervenience, then, in the context of the empiricist’s PA and the Humean account of causation, amounts to a re-statement of the parallelistic hypothesis. Moreover, it does not establish epiphenomenalism, the oneway dependence of mind upon body. This is easily seen. Since the regularity of supervenience is lawlike we can assert that if the state of consciousness F were absent then so would any of the physical base properties be absent. And the absence of all of the base properties would mean that the usual events consequent upon the presence of those properties would also be absent. The physical world would be causally very different if the mental events were to be different. Which is indeed so: minds make a difference. The world is very different if I choose to insult my colleague what it would be if I choose not to; it is less different but nonetheless different if I will to raise my arm rather than to scratch my nose. Is that not precisely what Bentley was insisting upon when he insisted that mental or conscious states are not only caused but are also causes, and in particular causes of bodily changes? And is this not to say that the way the physical world is depends upon the way my conscious states are? that, in other words, the physical depends upon the mental? We may therefore conclude, I think, that the doctrine that minds are supervenient upon physical processes collapses into parallelism, once one accepts the empiricist’s PA and its consequent Humean account of causation. And parallelism is all that one needs in order to preserve the scientific commonsense, on the one hand, that the physical world is closed, and, on the other hand, the piece of commonsense that minds make a difference to what happens in the physical world.

311

Four: Disappearance of the Simple Self: Its Problems

Epiphenomenalism and related positions that hold that mind is somehow causally dependent upon body but not conversely are simply contrary to commonsense. But there is no reason to reject that commonsense. Nor does there seem to be any great mystery about how mind acts on body, let alone how body acts on mind. I find there to be no mystery either in the fact that I can raise my arm or in the fact that when someone pinches me I feel a pain. It is the philosophers who fly in the face of commonsense and who create a mystery out of the mind-body relation or try to say the mental is without its effects in the world. At the same time, it is also a piece, if not of commonsense then of scientific commonsense, that physical events have physical causes. What I have argued is that many of the apparent problems of the mind-body relation disappear once one opts into an empiricist ontology defined by the Principle of Acquaintance and adopts on that basis a Humean account of causation. McGinn’s mystery dissolves into commonsense. As Berkeley once put a similar point, they first raise a dust and then complain that they cannot see. The main point, however, is that the concept of explanation and its implications for science depends upon the metaphysical background in which it is embedded. In particular, an empiricist metaphysics will do much to clear up some of the mystery that is sometimes said to surround explanation, especially explanation in the human sciences. (4) Privacy and Other Minds We have examined several senses of ‘mental’. But before we return to our main line of argument, there is yet another sense of ‘mental’ which deserves our attention. This is the notion of “privacy”: the mental is the private. Exactly what is this notion of privacy? Does it raise any problems for the Humean view of minds? Hume thinks not – he does not deal with that problem. But he does deal with a related issue. That is the problem of other minds. This is the issue of how, given the privacy of what is mental, we know, or even talk about, what is going on in other minds. It is to this issue that we now turn. Privacy is often proposed as a criterion of the mental: what is in one’s mind is what is private to oneself. Even if this is not the only criterion of the mental – there is, for example, intentionality –, and even if what is mental in the sense of private does not coincide with what is mental on

312

Body, Mind and Self in Hume’s Critical Realism

some other criterion – non-intentional entities such as pains and pleasures are nonetheless mental in the sense of being private90 –, it is still true that the fact that certain states are private is one that is immensely important in the philosophy of mind. Certainly, it gives rise to what many have thought to be one of the central problems of the philosophy of mind, namely the problem of other minds. If minds were not private there would be no question about our knowledge of the other person’s thoughts, at least there would be no more question here than there is a question about his or her bodily motions. But, of course, while bodily motions are public, thoughts are private. How, then, do we gain access to the latter? The answer that empiricists have given to this question is, naturally, that it is by inference. As Hume put it, No passion of another discovers itself immediately to mind. We are only sensible of its causes or effects. From these we infer the passion: And consequently these give rise to our sympathy (T, p. 570).

In more detail, for example, if I notice a person in water crying out and struggling to not drown, I immediately infer, or rather conclude more or less immeidately that the other person is afraid for him- or herself, i.e., is experiencing the passion of fear which has as its object that person’s drowning, then that vivid idea or belief is transformed by the mechanism of sympathy into a passion which I now experience, namely, the passion of fear which has as its object the other person’s drowning. I become fearful for his or her safety, and in appropriate circumstances that passion moves me to act to rescue that other person from drowning. As Hume puts it, When I see the effects of passion in the voice and gesture of any person, my mind immediately passes from these effects to their causes, and forms such a lively idea of the passion, as is presently converted into the passion itself. In like manner, when I perceive the causes of any emotion, my mind is convey’d to the effects, and is actuated with a like emotion. Were I present at any of the more terrible operations of surgery, ’tis certain, that even before it begun, the preparation of the instruments, the laying of the bandages in order, the heating of the irons, with all the signs of anxiety and concern in the patients and assistants, wou’d have a great effect upon my mind, and excite the strongest sentiments of pity and terror (T, p. 370).

So we know other minds by inference. But this only postpones the difficulty: what sort of inference? Here Hume is not of much help, though it is clear that he thinks of the reasoning in terms of causes and effects:

313

Four: Disappearance of the Simple Self: Its Problems

whatever inferences they are, they are based on laws or regularities which we have learned through experience hold in experience. John Stuart Mill was more specific about the nature of the inference. He urged that it was one based on analogy: there is a resemblance or analogy between my behaviour and that of the other person, and, since same effects have the same causes, I therefore infer that his or her behaviour is caused as mine is, by certain mental states. As Mill points out, it is certainly possible upon the Humean or bundle view of minds that Mill adopts to conceive minds other than my own: ...[what] I ... admit if I receive this theory, is that other people’s Selves are also but a series of feelings, like my own. Though my Mind, as I am capable of conceiving it, be nothing but the succession of my feelings, there is nothing in that doctrine to prevent my conceiving, and believing, that there are other successions of feelings besides those of which I am conscious, and that these are as real as my own. The belief is completely consistent with the metaphysical theory.91

What, then, are the grounds that support the belief? What is the nature of my inference? Mill proposes much the same grounds as Hume: causes and effects. By what evidence do I know, or by what considerations am I led to believe, that there exist other sentient creatures; that the walking and speaking figures which I see and hear, have sensations and thoughts, or in other words, posses Minds? The most strenuous Intuitionist does not include this among the things that I know by direct intuition. I conclude it from certain things, which my experience of my own states of feeling proves to me to be marks of it. These marks are of two kinds, antecedent and subsequent; the previous conditions requisite for feeling, and the effects or consequences of it. I conclude that other human beings have feelings like me, because, first, they have bodies like me, which I know, in my own case, to be the antecedent condition of feelings; and because, secondly, they exhibit the acts, and other outward signs, which in my own case I know by experience to be caused by feelings. I am conscious in myself of a series of facts connected by an uniform sequence, of which the beginning is modifications of my body, the middle is feelings, the end is outward demeanour. In the case of other human beings I have the evidence of my senses for the first and last links of the series, but not for the intermediate link. I find, however, that the sequence between the first and last is as regular and constant in those other cases as it is in mind. In my own case I know that the first link produces the last through the intermediate link, and could not produce it without. Experience, therefore, obliges me to conclude that there must be an intermediate link; which must either be the same in others as in myself, or a

314

Body, Mind and Self in Hume’s Critical Realism

different one: I must either believe them to be alive, or to be automatons: and by believing them to be alive, that is, by supposing the link to be of the same nature as in the case of which I have experience, and which is in all other respects similar, I bring other human beings, as phaenomena, under the same generalizations which I know by experience to be the true theory of my own existence. And in doing so I conform to the legitimate rules of experimental enquiry. The process is exactly parallel to that by which Newton proved that the force which keeps the planets in their orbits is identical with that by which an apple falls to the ground. It was not incumbent on Newton to prove the impossibility of its being any other force; he was thought to have made out his point when he had simply shown, that no other force need be supposed. We know the existence of other beings by generalization from the knowledge of our own: the generalization merely postulates that what experience shows to be a mark of the existence of something within the sphere of our consciousness, may be concluded to be a mark of the same thing beyond that sphere.92

This makes clear the sort of inference that Mill – and Hume – has in mind. But it does not deal with the fact of privacy. And this fact is often cited as invalidating the notion that I can know other minds through an inference based on a law which is an inductive generalization from my own case. Thus, Plantinga93 considers a generalization like the following for inferring a mental cause from an action or bodily effect: (@) Whenever a person exhibits pain behaviour then there is a pain which causes that behaviour. This certainly seems to be roughly the sort of thing that Mill is suggesting. But Plantinga argues that this cannot form the basis of our knowledge of other minds: “I observe that B a body other than my own is displaying pain behaviour, and no matter how intensely I concentrate, no matter how carefully I canvass my feelings, my attempt to feel a pain in B is futile. I feel no pain there. But does not this state of affairs provide me with a disconfirming instance [of the generalization (@)] ... should I not take my failure to observe pain in B to provide me with a counter-instance to the generalization that every case of pain behaviour is accompanied by pain in the body displaying it.”94 If Plantinga is correct, one can start to generate a scepticism with respect to our knowledge of other minds.95 However, as we shall proceed to argue, this is wrong: if one properly understands the inference to other minds, then, contrary to Plantinga, one cannot generate a scepticism with respect to other minds in the way he believes he can. Specifically, I shall argue, what one needs to prevent the sceptical conclusion of Plantinga is a careful understanding of the fact of privacy: what is it for a mental state to belong to me?

315

Four: Disappearance of the Simple Self: Its Problems

In fact, as it turns out, if we can become clear on this point, not only will we have become clear about the nature of other minds, but at the same time we shall have clarified certain concepts that turn out, as we shall see in the next chapter, to be useful in dealing with the problem of personal identity. (i) Parallelism: Hartley, Hume and Something of Its History: Why It Is Believable Let us begin by trying to become a little more clear on what the supposed causal relation is between mind and body. Hume and Mill speak simply of environmental or distal causes and various kinds of actions as behavioural effects, in contrast to the proximate causes, the changes in the sense organs, and the micro-behavioural effects, that is, the specific bodily motions rather than the actions or macro-behavioural effects. Yet both allow that there is physiology that intervenes between the distal cause and the mental state as effect, and between the mental state as cause and the action as effect. The standard view of psychologists came to be the connection between mental states and bodily states could be best understood in terms of a parallelistic connection between mental states and states of the brain.96 The received psychological theory of the second quarter of the 18th century was associationism, first stated by John Gay,97 though it was clearly implicit in much of what Locke says. And it is certainly there in Hume’s Treatise, but this work is as much philosophical as psychological, and no contemporary psychologist would identify it as a book in psychology. It was David Hartley who, in his Observations on Man,98 first put the theory in a context that a 20th century psychologist would recognize a psychological.99 It was also Hartley who first clearly stated the parallelistic hypothesis as an account of the causal process through which mind and body interact.100 Again one must say that something like parallelism is there in the Treatise. But it appears only in passing and is not fully explicit. One can pick it up from various bits scattered about the Treatise. Thus, Hume notes “that motion ... actually is, the cause of thought and perception.” (T, p. 248) Hume states explicitly that all our sensations have bodily states as their causes: “all our perceptions [sensations] are dependent upon our organs, and the disposition of our nerves and animal spirits.” (T, p. 211)

316

Body, Mind and Self in Hume’s Critical Realism

This general claim is confirmed in particular by our sensations of smell and taste: “The nerves of the nose and palate are so dispos’d, as in certain circumstances to convey such peculiar sensations to the mind.” (T, p. 287) Affective sensations have bodily causes. Thus, Hume mentions “the pains and pleasures, that arise from the application of objects to our bodies, as by cutting of our flesh with steel, and such like.” (T, p. 192); and also, in another context, points out that “Let a man heat one hand and cool the other; the same water will at the same time, seem both hot and cold, according to the disposition of the different organs.” (T, p. 372) To be sure, Hume does at one point say that sensations “arise in the soul originally, from unknown causes,” (T, p. 7) but hastens to qualify this remark by indicating that we do know something about their causes – they are physiological – but that locating those causes is not his task as a psychologist, philosopher and moral theorist: he comments that “the examination of our sensations belongs more to anatomists and natural philosophers than to moral.” (T, p. 8) The same point is made again later in the Treatise, this time a little more accurately, emphasizing that these sensations are without antecedent psychological causes but are not without causes, it is just that these causes are physiological, and so are not within the purview of the moral philosopher: “Original impressions or impressions of sensation are such as without any antecedent perception arise in the soul” (T, p. 275) and the reason that they so arise can be found in physiology, so “as these [the perceptions that arise without any antecedent perceptions] depend upon natural and physical causes, the examination of them wou’d lead me too far,” Hume says, “from my present subject, into the sciences of anatomy and natural philosophy.” (T, pp. 275-76) So much for impressions: what about ideas? These too seem to have physiological causes. Thus, we are told how our imagination, and, one presumes, our ideas, can be disordered by “any extraordinary ferment of the blood and spirits.” and by “chimera[s] of the brain.” (T, p. 123) We have already noted101 Hume’s “specious but plausible” physiological account of how logical connections among our ideas sometimes slide into illogical connections: ...as the mind is endow’d with a power of exciting any idea it pleases; whenever it dispatches the spirits into that region of the brain, in which the idea is plac’d; these spirits always excite the idea, when they run precisely into the proper traces, and rummage that cell, which belongs to the idea. But as their motion is seldom direct, and naturally turns a little to the one side or the other; for this reason the animal spirits, falling into the contiguous traces, present other related

317

Four: Disappearance of the Simple Self: Its Problems

ideas in lieu of that, which the mind desir’d first to survey. (T, pp. 60-61)

Hume is here assuming that corresponding to mental processes involving the ideas and the ways in which they come to be associated are physiological processes in the brain. Elsewhere he makes the point just as clearly, when he states that “Bodily pains and pleasures ... arise originally in the soul, or in the body, which ever you please to call it.” (T, p. 276) Again, “everyone may perceive, that the different dispositions of his body change his thought and sentiments” (T, p. 248); and we “find by experience” that “thought and motion” are “constantly united,” and thus we “may certainly conclude, that motion ... actually is the cause of thought and perception.” (T, p. 248) None of this is conclusive evidence that Hume is like Hartley in clearly accepting parallelism – though it does make more than clear that Hume rejects any form of epiphenomenalism,102 as, indeed, would make sense given his regularity account of causality. But there is enough evidence here that is most reasonably to be read as a commitment to the sort of mind-body interaction that fits with the parallelist position. But for what Hume was about he did not need anything more than the rather vague treatment of the physiology he assumes underlies the psychological processes in which were his central interest.103 Hartley was much more a straightforward psychological theorist, who eschewed the philosophy and moral theory with which Hume was concerned – at least until one comes to the rather odd theology one finds in second volume of the Observations on Man. Hartley notes the work of John Gay and Locke, but takes the associationist doctrine, works it out in great detail (most of it speculative), and then weds it to Newtonian physiological theorizing.104 In the usual fashion, ideas are images derived from sensations, and complex ideas arise by patterns of association based on relations like those of resemblance and regularity of coexistence and succession. Hartley’s hypothesis is that corresponding to these mental changes are bodily changes, and corresponding to bodily changes are mental changes. As he puts it, “The white medullary substance of the brain is also the immediate instrument, by which ideas are presented to the mind: or, in other words, whatever changes are made in this substance, corresponding changes are made in our ideas, and vice versa” (p. 8). The ideas coalesce in the case of complex associations to form the ideas that do not resemble their genetic antecedents (p.75); among these are the associations involved with words that create those ideas called

318

Body, Mind and Self in Hume’s Critical Realism

propositions (p. 270). The same general pattern, relating motions in the brain and mental states one to the other, is applied equally by Hartley to the phenomena of the will (Ch. I, section iii, Ch. II). Propositions are of course mental states of the sort we call cognitive; while volitions are conative, as passions are affective. All have that property of intentionality: the “aboutness” that is characteristic of the mental – (though not of the private). These acts – our beliefs, desires, intentions, etc., as occurrent mental phenomena – are among the contents of our conscious states. But Hartley’s general hypothesis, which he works out with considerable ingenuity, is that the brain/mind relation for all of these is the same as it is for much simpler sensations and ideas: “whatever changes are made in this [the white medullary] substance, corresponding changes are made in our ideas, and vice versa.” There is, in other words, a parallelism between mind and body. This is clearly Hartley’s position, where Hume’s view on this issue is much less clear. This is as far as Hartley goes, however. He generally assumes a simple one-one relation between mental states and bodily states. The compound bodily state A+B+C+D corresponds to the compound mental state a+b+c+d (p. 74). Actually one would expect the relation to be more complicated than this. The same sort of mental state can on different occasions give rise to different behaviour, that is, different bodily states, depending upon different past histories. These histories leave different traces in the brain, and these different traces, all of which have the same mental state parallel to them, will generate the different sorts of behaviour. The parallelism is, in other words, one-many: mental states of sort MS correspond to brain states of the disjunctive sort BS1 v BS2 v BS3 All this talk of brain states remains largely hypothetical, of course. Hartley never did identify precisely what those brain states were that were parallel to specific mental states. Rather, he simply identified the brain states by means of definite descriptions: the brain state that is lawfully parallel to mental state MS We can similarly identify brain states by reference to behaviour: the brain state that is the lawful cause of this

319

Four: Disappearance of the Simple Self: Its Problems

behaviour This is what those materialists who call themselves “functionalists” do. Hartley, like the functionalists, does not point to known laws, known neurophysics, but rather makes a general assumption that there are laws relating brains states to mental states or to behaviour and takes it that this claim justifies the assumption they make that we can successfully, in Russell’s sense, form definite descriptions that refer to the parallel and causally antecedent brain states without our actually being able to identify them.105 Later writers like Priestley, who edited Hartley but left the physiological speculations out,106 insisted quite correctly that psychology could stand on its own as a science and did not need the help of physiological hypotheses to get on with its task of explaining the workings of the human mind.107 But physiology was there and could in the end not be ignored, as everyone knew. It was Spencer who (re-)introduced considerations of biology and physiology into the science of psychology.108 And John Stuart Mill commends Bain for the way in which he supplemented the traditional associationism with considerations drawn from physiology.109 The question for the psychologist was, what evidence was there that the general parallelistic hypothesis was true? Well, there was sufficient to make it fairly reasonable to let Descartes almost take it for granted. La Mettrie certainly assumed it; but perhaps in his case it was simply materialist dogma. In any case, by the end of the 19th century the evidence was overwhelming.110 What argued against the view that mind interacted in material causal processes is the fact that there is exceptionally strong evidence that material processes form a closed system. This is a direct consequence of the law of conservation of energy, but had been a fairly explicit assumption of the new science since Descartes. At the same time, parallelism permitted one to hold that behaviour is systematically related to our mental states. Thus, William B. Carpenter expressed the parallelistic position111 by asking us to consider “that mortal contest which [fable tells us] was once carried on by two knights respecting the material of a shield which they saw from opposite sides, the one maintaining it to be made of gold, the other silver, and each proving to be in the right as regarded the half seen by himself” (1855, p. 547). The “materialists” and “spiritualists” are like these two knights, each getting half the picture right but ignoring the other half, and ignoring too what he elsewhere112 refers to as the “intimacy of that

320

Body, Mind and Self in Hume’s Critical Realism

nexus between Mental and Bodily activity” (1890, p. 696) There is a moral, he thinks, that is to be drawn. ...the moral of this fable, as regards our present enquiry, is, that as the entire shield was really made-up of a gold-half and a silver-half which joined each other at midway, so the Mind and the Brain, notwithstanding those differences in properties which place them in different philosophical categories, are so intimately blended in their actions, that more valuable information is to be gained by seeking for it at the points of contact, than can be obtained by the prosecution of those older methods of research, in which Mind has been studied by Metaphysicians altogether without reference to its material instruments, whilst the Brain has been dissected by Anatomists and analyzed by Chemists, as if they expected to map-out the course of Thought, or to weigh or measure the intensity of Emotion (1855, p. 547)

At the same time, Carpenter clearly distinguishes parallelism from epiphenomenalism, accepting the former and rejecting the latter. Epiphenomenalism had been proposed by T. H. Huxley who argued that we are nothing but “parts of the great series of causes and effects which, in unbroken continuity, composes that which is, and has been, and shall be – the sum of existence.”113 On this view, Carpenter suggests, people are automata, with no duty, no obligations, no responsibility (1890, p. xlvi). But this position is absurd, so Carpenter rejects the epiphenomenalism of Huxley (1890, p. xlviii). What he objects to is the materialist claim that “the highest elevation of Man’s psychical nature is to be attained by due attention to all the conditions which favour his physical development,” where this is taken to imply that man’s “fancied power of self-direction [is] altogether a delusion” and that “the notions of duty or responsibility have no real foundation, Man's character being formed for him, and not by him” (1855, p. 548). For it is in fact evident that this position is false; man “really possesses a self-determining power, which can rise above all the promptings to external suggestion, and can, to a certain extent, mould external circumstances to its own requirements, instead of being completely subjugated by them” (p. 549). This is clear from “the direct testimony of Consciousness, in regard not only to the existence of this Volitional power, but also to the Self-determination of the Ego in the exercise of it” (1890, p. xxiv). It was Carpenter’s – and Hartley’s – parallelism rather than Huxley’s epiphenomenalism that was to become the working hypothesis of the practising psychologist and physiologist. 114 The real issue is not parallelism, but materialism: are there any grounds for identifying mental states with bodily states? The answer is:

321

Four: Disappearance of the Simple Self: Its Problems

undoubtedly not. Materialism is implausible because there seem to be no ways in which the properties of mental states can be eliminated in favour of properties of brain states. My occurrent believing may have properties that I am not aware of; it may, for example, be at a place, even though it is apparently only in time, and so after all fulfil at least that condition of materiality. But it is hard to argue away the properties my mental states are perceived to have: when my tooth aches,115 I am aware of the pain at a certain place in my body, and this property, localized there, is not just one of the physiological properties of the tooth, nor any other physiological property. So we end up defending not so much materialism as some sort of doctrine of emergence, not of mental things, but of mental properties. This is the thesis of supervenience. Some take it to be a version of materialism, and if they wish so to speak, why should we object? But we should also recognize that this position is but the latest version of the parallelism of Hartley and of Carpenter. The laws of supervenience are but the laws of parallelism re-labelled in a way that pleases a conscience that for some reason wishes to be a materialist but at the same time notes the difficulties of taking that position seriously. (ii) Parallelism and Interactionism Four points are central. The first is the desire to admit minds into one’s ontology; one is acquainted in our inner awareness with these states and they undeniably exist distinct from bodily states, and, in particular physiological states of the brain. The second is the generally accepted extrapolation from the gappy knowledge of current physiology that in a complete physiology every brain event would “follow recognized physical laws,”116 and that no brain event involves a “departure from physical law”117 so that “for explaining recent advances in the brain physiology is, in principle, complete”:118; mental events are therefore not needed in explaining events in the brain nor in explaining events brought about by events in the brain. So it seems that mental events, while they do exist are not relevant to the explanation of human behaviour – minds may be nice, but they are irrelevant in any science of human nature. But third, it is sheer commonsense that mental states do affect bodily states – I will my arm to go up (mental event), and it goes up (bodily event consequent upon the mental event). In other words, in spite of the fact that physiological processes are closed to any non-physiological causes, mental event do make a difference to what happens in my brain and thereby in my actions

322

Body, Mind and Self in Hume’s Critical Realism

and my behaviour. Campbell has put the point this way: Brain activity is the crucial bodily factor in both learning of our environment (perception), and in living out our lives in that environment (action). The links of the senses to the brain, and the brain to muscles show every sign of being physicochemical relations. So if spirit [mind] acts on matter in men [persons], it acts through the brain. Yet the evidence suggests that nothing happens in a man [person] save what conforms to physical and chemical law.119

It would seem, then, that we are caught up in a real contradiction, Mental states do exist and are distinct from brain and behavioural states, but we have to say, on the one hand, that these states of consciousness don’t interact with any physiological variables nor make any causal difference to any physiological process – only physiological variables are relevant to such processes – , while we also have to say, on the other hand, that our conscious states, what’s in our mind, does make a causal difference to what happens in our brain and to how we act and behave. Commonsense, for each of us our own experience, tells us that there are mental states and that these interact with our bodily processes, but science (which is after all but the long arm of commonsense) tells us that, while those mental states may exist, they don’t interact. Parallelism shows how this apparent paradox, this apparent contradiction, is only apparent: there is no real problem. Or, more accurately, parallelism solves the problem, as Cullen clearly saw, provided one also accepts the Humean account of causation as regularity. This is the fourth of the points we need to include in our list of those that generate the paradox: one needs to accept that mind and body do causally affect one another. In contrast, one does have a real mind-body problem if one sets the game up so that the nature of causal connections precludes mind acting on body or conversely. The Cartesians did this. So do others. Thus, Keith Campbell has suggested that an essential ingredient of ordinary causation is that it involve a “mechanism”, and the “links between matter and spirit” have no such mechanism. The reason lies in the impalpable character of spirit. We have no idea how such a mechanism could “get a grip” on a spirit. The mechanism by which changes in the brain can “get a grip” on the heart and change its rate of beating is intelligible; but not how the brain can get a grip on a spiritual mind, to give it fear.

323

Four: Disappearance of the Simple Self: Its Problems

Indeed, he goes on to suggest further that there could be no laws that describe the mind-body interaction. ... because fear resists description in terms of place or dimension, we have no “fear” equations linking brain and spirit, nor seem likely to have any.

From which one of course concludes that “matter-spirit interaction is impossible.”120 Now, of course mind is “impalpable.” One cannot punch a thought nor weigh a pain: but, then, one cannot punch a rainbow or weigh a shadow. But that does not prevent there being laws describing and explaining rainbows and shadows. As for “getting a grip,” one can get a grip on oneself with a pair of pliers, one can pinch a fold of flesh, for example. Getting such a grip can cause pain, but one cannot similarly use the pliers to get a grip on the pain. One can similarly get a grip on a spigot with the pliers and turn it to open a spout, causing a stream of water, or perhaps a glass of beer, to be released, but one cannot get a grip on that stream of fluid that is released: nonetheless, the gripping of the spigot and turning it caused the stream to be released. “Getting a grip” is just a bad metaphor: there are all sorts of causal connections that do not involve any sort of mechanism in which the cause brings about an effect by gripping something and wringing from it the effect. And of course fear cannot be measured. And of course it has no place or dimension. Neither does the feeling of tipsiness, but one is still quite comfortable with saying that the feeling was caused by drinking three martinis. Lacking position or dimension does not preclude something from being caused. Neither does the fact that something is “impalpable” – as if the only sort of causation that one can admit is punching or gripping or groping. Perhaps we should also note, with Hume, that not all mental events lack spatiality – on this point, the Cartesian is simply wrong: “there are impressions and ideas really extended.” (T, p. 240) Hume considers the argument that “different shocks, and variations, and mixtures are the only changes, of which matter is susceptible, and as these never afford us any idea of thought or perception, ’tis concluded to be impossible, that thought can ever be caus’d by matter.” (T, p. 247) After all, is it not inconceivable “that the shocking of two globular particles shou’d become a sensation of pain” or “that the meeting of two triangular ones shou’d afford a pleasure,” (T, p. 246) How could it be that “motion in a circle ... shou’d be nothing but merely motion in a circle; while motion in

324

Body, Mind and Self in Hume’s Critical Realism

another direction, as in an ellipse, should also be a passion or moral reflexion.” (T, p. 246) As the Cartesian argues, “there appear[s] no manner of connexion betwixt motion or thought.” (T, p. 247) However, this is no criterion that can establish the impossibility of any causal relation. For the absence of connection, the absence, that is, of any objective necessary connection, is true everywhere of causation. As he puts it at one point, so far as concerns logical or ontological possibility, “Any thing may produce anything. Creation, annihilation, motion, reason, volition: all these may arise from one another, or from any other object we can imagine.” (T, p. 173) There is nothing impossible about matter acting on mind and mind acting on matter. There is no reason to suppose that they are necessarily connected. So, causal connection just is constant conjunction: “we are never sensible of any connexion betwixt causes and effects, and that ’tis only by our experience of their constant conjunction, we can arrive at any knowledge of this connection.” (T, p. 247) Causation is simply not objective necessary connection, it is certainly not a necessary connection in which the cause brings about its effect by “coming to grips” with it – or by “generating” it. To search for such a connection is merely foolish: settle instead for the only possible thing that causation could, logically or ontologically, be, namely constant conjunction. Objective necessary connections are not there in things, and to search for them is to search for the impossible. One avoids those problems, as Cullen argued, if we adopt the Humean view, that causation is regularity, constant conjunction. With that, there is no problem in holding that mind and body interact. This is the fourth of the propositons we noted above. But with this in place, while avoid some problems, are still confronted by the problem generated by the other propositions one through three, the problem of the conflict between the commonsense that mind and matter do interact and the scientific commonsense that physiological processes are closed and that interaction is therefore not the case. The claim is that if one accepts the parallelistic hypothesis then the apparent conflict disappears. Let us see. Suppose there is an environmental state near our person Jones, say a nice looking green apple; call this state ES1. Jones is hungry; that is a mental state, call it MS1. He perceives the apple; that is mental state MS2. Straightway he wills that he pick up the apple in order to eat it (MS3). The action of his arm going out to pick up the apple immediately ensues; this behavioural state AS1. Here we have a process that goes from the

325

Four: Disappearance of the Simple Self: Its Problems

environment to the mind and issues another non-mental state, a piece of behaviour. That’s sketch enough of such a process, anyway. We have this process (mp1) ES1 –> MS1 –> MS2 –> MS3 –> AS1 We are committed to parallelism. So we have brain state BS1 parallel mental state MS1, BS2 parallel to MS2, and BS3 parallel to MS3. MS1 ; MS2 ; MS3 (c1) | | | BS1 BS2 BS3 We therefore have the physiological process: (bp1) ES1 –> BS1 –> BS2 –> BS3 –> AS1 This physiological process is closed, and in particular, no mental events are relevant. Yet, of course, the mental events are there, and so is the mental process. More importantly, the mental process produces the behavioural event AS1; that behavioural event, that action, is caused by the mental event, the volition. And earlier, the mental event MS1 has been caused by the environmental event ES1. So the non-mental event, a physical event, has caused an event in the mental process. Thus. body affects mind and mind affects body. In this respect, there is interaction between the mental and the physical. All the same, however, that same behavioural event, the action AS1, is caused by the brain state BS3. AS1 is the product of BS3 and BS3 is itself the product of a purely physical process to which no mental event is relevant, In this sense, the physical process is closed and there is no interaction., that is, body does not affect mind, body affects only body, and mind does not affect body. Given the parallelistic connections, one can have both the scientific commonsense that physical processes are closed to all but physical events, and also the ordinary commonsense that body does affect mind and mind does affect body. But is it not the case that if the physical process is closed to mental events, then the latter just irrelevant to what happens in the natural world? Mental events are epiphenomena and mind has no place in nature. Or, if it has, then it is only insofar as bodily events cause the mental events; certainly, the mental event is causally inefficacious with regard to bodily process, since the latter processes are completely accounted for by processes which are purely physical. But now suppose that we have the apple (ES1), followed by perception (MS1), followed by a feeling of being full, of being not hungry (MS2*). One then decides (MS3*) to turn away, and this results in one so doing (AS1*).. We have this process

326

Body, Mind and Self in Hume’s Critical Realism

(mp2) ES1 –> MS1 –> MS2* –> MS3* –> AS1* We are committed to parallelism. So we have brain state BS1 parallel mental state MS1, as before, but now we have BS2* parallel to MS2*, and BS3* parallel to MS3*. MS1 ; MS2* ; MS3* (c2) | | | BS1 BS2* BS3* We therefore have the physiological process: (bp2) ES1 –> BS1 –> BS2* –> BS3* –> AS1* The important point is the two parallelistic laws of (c1) and (c2): MS2 BS2 MS2* BS2* The lawful connection is one of “if and only if”.121 It follows with regard to (mp1) that if the mental state (MS2) were different then the corresponding brain state would be different, and so would the physiological process (bp1).. Thus, if instead of (MP2) there were (MP2*), then the brain state would also be different; it would be (BP2*) rather than (BP2). And if that were so then, contrary to what we have supposed really happened, namely (bp1), we would have (bp2), just as we would have (mp2) instead of (mp1). Thus, it is clear that which mental state one is in “makes a difference” to what occurs in the physiological process. It is simply wrong to suggest that “Parallelists may hold that neither matter nor spirit affects the other, or that matter can affect spirit but not vice versa.”122 Provided one accepts the Humean account of causation, then, given the parallelistic regularities, one can quite reasonably claim that body affects mind, and that mind affects body, and further that bodily processes are in themselves causally complete, closed to causal influences that are other than material. If one accepts parallelism then, provided one is a Humean with regard to causation, then body affects mind, and mind affects body, in the sense that what bodily state one is in “makes a difference” to what mental state one is in, and equally what mental state one is in “makes a difference” to what brain state one is in, and therefore also “makes a difference” to what actions and behaviour ensure. It has been suggested that on a view such as this one is committed to a sort of “double causation” which, however, is a doctrine which is “incoherent”. For, Either the material and the spiritual were both causes of the effect (both made a difference), in which case each was only part of the cause and the physical causes are not complete, or one was idling and was not effective although it

327

Four: Disappearance of the Simple Self: Its Problems

might have been. In that case the idling member is not a cause at all. If physiology is complete, then to introduce spiritual causes alongside physical causes and have one or the other idling all the time, is idle indeed.123

However, consider (mp1) and (bp1). Here we have (MS3) as sufficient for (AS1) and also (BS3) as sufficient for (AS1). We have two statements of sufficient conditions: (x) P –> R (y) Q –> R If we have these regularities, then we also have the regularity (z) (P & Q) –> R again stating a sufficient condition. Why should one hold that if the latter (z) obtains while, say, (x) also obtains, then in fact the sufficient condition Q in (y) is merely “idling”? It is true that, if (x) obtains, then so automatically does (z): (x) logically entails (z). Thus, if (x) obtains while (y) does not then (z) also obtains, then one could indeed say that were Q absent then R would nonetheless occur, for the presence of P would, by (x) be, by itself, sufficient of the presence of R. In that case, that is, if (x) alone were true, then P alone would be sufficient for R, and though P & A would also be sufficient, the factor Q would indeed be idling, not required for the presence of R. But that is compatible with both (x) and (y) being true. In that case, neither factor P nor factor Q is idling; each is sufficient and jointly they are sufficient. The objection that on the parallelistic hypothesis, one side or the other, mind or matter, must be idling as a cause for human action, is itself incoherent. One can therefore accept both the scientific commonsense that physiological processes are closed to outside, e.g., mental influence, and also that our actions and behaviour are things for which we are, through our decisions, which are mental events, morally responsible. Parallelism enables one to be both a good citizen and also scientifically respectable. (iii) Parallelism: Further Exposition In fact, the science of psychology has, though it sometimes did not know it, been committed to parallelism. It was Gustav Bergmann who most forcefully argued for this understanding of psychology in the second third of the twentieth century.124 Bergmann’s position has been elaborated by Addis.125 We can conveniently begin with this formulation. Addis considers the propositions (P1) John is now irritated

328

Body, Mind and Self in Hume’s Critical Realism

John is now irritable (P2) John is an irritable person. (P3) He assumes that to be now irritated is to be in a certain mental state. Let us call this state “I.” If we let “a” name John, and “to” stand for “now,” then, following Addis,126 we can transcribe (P1) as I(a, to) (P4) To say that John is now irritable, that is, to assert (P2), is to ascribe a disposition to John; it is to say that John is such that if the appropriate stimulus, say “SI,” is present then John will feel irritated. The relevant predicate may be defined by DI(x, t) =df SI(x, t) –> I(x, t) (df1) This permits us to transcribe (P2) as DI(a, to) (P5) = SI(a, to) –> I(a, to) As we have seen, for the Humean, one will assert (P5) as a disposition statement just in case that there is a law that enables one to infer that (P5) obtains. Such a law is available in (P3), which asserts that John is an irritable person, i.e., that he has at all times the disposition to become irritated, and which may be transcribed as (P6) (t) [SI(a, t) –> I(a, t)] From the law (P6) one can infer (P5), thus justifying the predication of DI as a disposition holding of a at to. Now let “N” represent the neurological state that is parallel to I. The relevant parallelistic law is: (A) (x)(t) [I(x, t) N(x, t)] Clearly, by virtue of this law, the stimulus SI that evokes the mental state I also evokes the parallel neurological state N. Besides the mental state and the neurological or brain state, there is also the typical behaviour that irritable persons often display. Call this behaviour “IB”. When a person is irritated, he or she does not always display this sort of behaviour; this happens only in the presence of some further stimulus, call it “SB”. Of course, that stimulus will not always evoke the behaviour characteristic of being irritated; it will do so only if the person is in fact irritated. That is, we have the following sort of law relating mental states to behaviour: (B) (x)(t) {SB(x, t) –> [I(x, t) IB(x, t)]} From this and (A) we can infer a similar law relating the neurological state parallel to I to the behavioural stimulus and response: (C) (x)(t) {SB(x, t) –> [N(x, t) IB(x, t)]}

329

Four: Disappearance of the Simple Self: Its Problems

From this we can infer directly that (D) (x)(t) {N(x, t) –> [SB(x, t) –> IB(x, t)]} The law (D) renders the following concept significant: DB(x, t) = Df SB(x, t) –> IB(x, t) (df2) Since it is significant, it can be predicated dispositionally of individuals; DB represents the tendency to overt display of the behaviour which we associate with irritated persons. Since, by (P6), the stimulus SI evokes the mental state I of being irritated, it follows from (A) and (D) that (E) (x)(t) {I(x, t) –> [SB(x, t) –> IB(x, t)]} or, what is the same, given (df2), (E') (x)(t) {I(x, t) –> DB(x, t)} For obvious reasons we can say that the disposition DB “expresses” the mental state I. We should also note that the stimulus SI that evokes the mental state I also evokes the disposition DB; that is, we have, following from (P6), the law (F) (t) {SI(a, t) –> [SB(a, t) –> IB(a, t)]} or, what is the same, (F') (x)(t) {SI(a, t) –> DB(a, t)} Now let us suppose that we have the observations (O) SB(a, to) & IB(a, to) This enables us to infer, via (B), that a is I. That is, the law relating behavioural or bodily states to mental states enables us to infer from observed behaviour in the appropriate context that the appropriate mental state is present, in this case that the person exhibiting the behaviour in the appropriate stimulus context is in fact irritated. In short, the laws relating mind and body enable us to infer mental states from observed behaviour. At the same time we can of course infer as well, via the parallelistic law (A), that the neurological state N is also present. Or, what amounts to the same, we can infer the presence of the neurological state from the behaviour in the stimulus context from the law (C) relating behaviour to brain states. From (O) it clearly follows that SB(a, to) –> IB(a, to) (P7) But it does not follow that this is a dispositional predication. This latter sort of predication is justified, at least so far as a Humean is concerned, as we have seen, only if there is a law that enables us to deduce (P7). However, the observations (O) enable us, via the law (B), to infer the presence of I in a at to. But from the presence of I it follows, via the law (D), that (P7). Thus, given the laws relating mind and body, it follows that

330

Body, Mind and Self in Hume’s Critical Realism

upon the observations (O) we can assert (P7) dispositionally. A series of comments will serve to clarify some points. (i) SI is the environmental stimulus that evokes the mental state I, together with its parallel neurological state N and behavioural expression DB = SB –> IB; SB is the stimulus that evokes the overt behaviour IB. SB and IB are different, of course, simply because even if one is in a mental state, it need not be “revealed” in overt behaviour. What corresponds to the mental state is the tendency to display certain kinds of overt behaviour (IB) under certain conditions (SB). This tendency DB = SB –> IB and the mental state (I) are both evoked by the same stimulus (SI). This is established by the laws (P6) and (F). (ii) Several behavioural tendencies may be associated lawfully as the behavioural correlates of the mental state I. We might have the further behavioural tendency defined by “SB* –> IB*.” This would be a second disposition expressing the mental state I; there would therefore be a law like (B), say (B') (x)(t) {SB*(x, t) –> [I(x, t) IB*(x, t)]} We will also be able to infer a law of the form (E') (x)(t) {I(x, t) –> [SB*(x, t) –> IB*(x, t)]} which will render the concept “SB* –> IB*” significant, and enable us to treat it like “SB –> IB”, and predicate it dispositionally of objects. Finally, we should note that from (B) and (B') we can infer a law of the following sort, which associates the two dispositions with each other by asserting that if both stimuli are present then the one response occurs if and only if the other does also: (G) (x)(t) {[SB(a, t) & SB*(x, t)] –> [IB(x, t) IB*(x, t)]}. The laws (B), (B'), etc., entail that the one mental state has several different ways in which it is expressed in behaviour; these laws ensure that observing the actualization of any one of these ways in which the mental state is expressed will imply the presence of the mental state. The law (G) entails that in effect these ways of expressing the mental state are equivalent to one another; it entails that, if one observes the actualization of any one of them, then one can infer the presence of all. Indeed, if we know that (P7) is true on the basis of observations (O), then one will be able, via (G), to infer the presence of other dispositions that are equivalent to it; and, having inferred the presence of these, one will be able to reverse

331

Four: Disappearance of the Simple Self: Its Problems

the inference to deduce the presence of the original disposition, thus justifying one's asserting (P7) dispositionally. One might say that the “same concept” has “several criteria of application.”127 In any case, observing one such tendency being actualized, or being able to infer the presence of one such tendency, will permit one, via the laws like (G), to infer the presence of the other behavioural dispositions that express the mental state I. (iii) Several of the behavioural correlates will be dispositions acquired by virtue of the person having undergone certain learning processes. That means that the laws which correlate the various dispositions will not be of the form (G) so much as historical in Bergmann’s sense.128 The laws of the sort (G) will hold only if the appropriate learning process has occurred. As Bergmann would put it,129 the laws of the sort (G) which would be imperfect relative to less imperfect laws of learning theory will yield the conditions under which the more imperfect laws obtain. (iv) We may know the truth of (P7) in one of two ways. If we know (O), that SB(a, to) and IB(a,to) both obtain, then we know (P7) is true on the basis of the observation; and, as we have seen, the laws relating mind and body justify this as a dispositional predication. We thus know (P7) to be true through the observation of its actualization. However, if, in the absence of the evoking stimulus SB, the overt behaviour IB is not displayed by a at to, then (P7) may still be true and assertible dispositionally. In that case, neither its truth, nor, perforce, its assertibility as a disposition, can be known on the basis of an observation of its actualization. We may nonetheless infer its truth using laws and other observed facts as initial conditions. E.g., observation might yield the truth of SI(a, to), in which case we could use (F) to infer the truth of (P7). The same law would justify asserting (P7) dispositionally.130 In this case, we know a cause for the mental state and knowing also the laws that relate mind and body, we can infer the presence of the disposition which is the behavioural expression of that mental state: given the laws of mind and body, the same stimulus evokes both the mental state and its expression. Or, e.g., we might observe the actualization of some other disposition that expresses the mental state. The law (G) will enable us to infer the truth of (P7). Since we have observed the actualization of a disposition expressing I, we can use the law (E) to deduce (P7), and therefore assert it dispositionally. In short: (P7) can be asserted dispositionally; when it is, it expresses the mental state I. (P7) can be asserted dispositionally on the basis of either

332

Body, Mind and Self in Hume’s Critical Realism

(") observation of its actualization, or ($) inference by means of laws and observation of other particular facts. The latter may be based upon either ($1) observation of causes that evoke it together with the mental state that it expresses, or ($2) observation of the actualization of another disposition expressing the same mental state. (v) The behaviouristic psychologist will claim that “DB (= SB --> IB)” means what “I” means. This is literally false. Yet there is a sense to what she says. (") The laws (B) and (E) establish that whenever I is predicated of an individual then DB can be predicated dispositionally of that individual, and when DB can be predicated dispositionally of the individual then I can be predicated of the individual. Thus, every statement of individual fact involving I can be replaced by a dispositional statement about behaviour. ($) With respect to statements of causes, any statement of causes of the presence of I, e.g., (P6), can be replaced according to (E), by another statement of causes of the presence of DB. (() Statements of effects are more complicated. Suppose we have a law to the effect (x) I –> E that certain events are the effects of the mental state I. We cannot replace ‘I’ in (x) by ‘DB’ since the resulting generality (SB –> IB) –> E need not be falsified by an individual that falsifies (x); any individual that falsified (x) but which was never tested by being made to be SB would be of this sort. This is one way in which one accepts the point of Molière, Locke and Hume that dispositions as such do not explain. However, if DB can be predicated dispositionally of an individual, then, as we saw in comment (iv) above, we can predicate I of that individual. Now, from an explanation of I, say (y) SI –> I we can deduce a law that explains E: (z) SI –> E Hence, if we can predicate DB dispositionally of an individual then any explanation of behaviour E based on the mental state I can be replaced by an explanation of the same behaviour that does not mention the mental state. Of course, the explanation in terms of (z) will be more gappy, in Mackie’s sense,131 or, what is the same, more imperfect, in Bergmann’s sense, than one that uses the laws (y) and (x). (z) alone fails to notice intermediate stages that (x) and (y) include. In order to eliminate these gaps without referring to mental states, one must refer instead to the neurological states that are parallel to the mental states. Assuming that N is

333

Four: Disappearance of the Simple Self: Its Problems

parallel to I, then we would have to replace the pair of laws (y) and (x) (SI –> I) & (I –> E) by the pair (SI –> N) & (N –> E) The result would be an explanation less imperfect than the purely behavioural explanation based on (z), and, if N were a disjunctive predicate, as realistically it must be, then the explanation is also potentially less imperfect than the mentalistic explanation. In general, then, the behaviourist’s replacement of mentalistic concepts by behaviourally defined disposition concepts does not affect the explanation and prediction of behaviour: what can be explained using mentalistic language can be explained using behaviouristic language. However, the purely behavioural explanations will be more imperfect than the mentalistic explanations unless one refers to the (non-behavioural) neurological states that are parallel to the mental states. In general, then, we can conclude that we can accept, when suitably qualified, the behaviourist’s claim that mentalistic concepts can be replaced by behaviourally defined disposition concepts without affecting the explanation and prediction of human behaviour, provided that we accept the parallelistic hypothesis.

We may conclude, I think, that parallelism, and in particular the schematization presented by Addis, is a sound picture of the mind-body relation.

But, having said this, we must now note that Addis’ schemata run into a serious difficulty. This is the privacy problem raised by Plantinga. We know by observation that (P7) obtains. We now use (B) to infer (P4), i.e., (P1) obtains. We do not now observe that (P4) obtains. For, after all, mental states are private. We predict (P4) obtains, yet cannot observe any individual facts verifying (P4). Where we use a law to predict a fact obtains and cannot observe that fact, we usually take this as evidence falsifying or at least disconfirming the law in question. We should therefore conclude from the privacy of mental states that the parallelistic hypothesis is false. This is the difficulty with the parallelism that has been advanced by Plantinga and others, as we have seen.132 Addis has not adequately dealt with the notion of privacy. Nor has Bergmann. Nor have Mill and Hume –

334

Body, Mind and Self in Hume’s Critical Realism

though both the latter implicitly recognize the solution. What we shall now argue is that the fact of privacy, far from providing counter-evidence to the parallelistic hypothesis, is exactly what one would expect if parallelism is true, and that therefore the fact of privacy to this extent confirms that hypothesis. In order to become clear on this matter, we shall have to examine the idea that mental states “belong to” a body. In clarifying this notion we shall, in passing, be able to clarify certain philosophically problematic ideas that certain other philosophies have recently advanced. I have in mind certain things Wittgenstein has said about pain and Strawson’s curious notion of a “P-predicate.” What I shall suggest is that if the parallelistic hypothesis is accepted, then these problematic ideas can be explicated and shown to have a sound, commonsensical core. (iv) Ownership of Mental States Since Wittgenstein talked about pain, we may as well begin with this also. When one says of Jones, sitting in that chair, that he is in pain, one is asserting that several things are true. One is asserting, first, that there is a certain conscious states that belongs to Jones and that this state is a consciousness of a phenomenal pain. And one is asserting or at least implying, second, something about Jones’ bodily state. One is asserting or implying that there is in Jones a tendency to behave in a certain way, in one of the ways that fall under the general rubric of “pain behaviour.” Thus, Jones may actually be manifesting the tendency; he may, for example, be grimacing, holding his cheek and pointing to an abscessed tooth. Jones may also be saying “I am in pain.” Wittgenstein133 explored this connection between primitive pain behaviour like grimacing and the tendency to utter “I am in pain.” He noted how the tendency to display the linguistic behaviour becomes associated with the tendency to display the more primitive behaviour. Thus, he noted how certain dispositions come to be associated with each other (see remark [iv] above) and how the relevant laws are historical (see [v]). In any case, the bodily state which we attribute to Jones when we say of him that he is in pain need not involve only overt behaviour. Also attributed to Jones will be certain dispositions or tendencies to behave in certain ways.134 Indeed, in the case of malingering, only the dispositions or tendencies will be attributed to him; the overt behaviour will not be present. In a case of something like malingering, if we know that Jones exemplifies the tendencies in question, that knowledge

335

Four: Disappearance of the Simple Self: Its Problems

will be something inferred from other known facts about him (see [iii]). With respect to this bodily state which is attributed to Jones, I have said that when we assert that Jones is in pain, we are “asserting or implying” that this bodily state obtains. We are at least implying the presence of that bodily state, in the following sense of “imply.” When we make an assertion, we “imply” that we have evidence that it is true. Now when we assert that Jones is in pain, we are at least attributing to him a certain conscious state, namely, a consciousness of pain. But we who are making the assertion about Jones do not observe this conscious state. It is, after all, private. Its presence must, therefore, be inferred from Jones’ bodily state. So Jones’ bodily state at least provides evidence for the assertion we are making about Jones. And insofar as it does provide that evidence, when we make the assertion, we are “implying” that the evidence obtains. But other philosophers seem to have held that besides implying, in this sense, the presence of the bodily state, we are also asserting its presence. I have in mind Wittgenstein and Strawson. Because of this disagreement, I have at that point expressed myself with the cautious “assert or imply.” But shortly I shall attempt to explicate the notions of these two philosophers, and if the explications I propose are adequate, then the difference between “assert” and “imply” is one that makes no difference. The physical characteristics which fall under the concept of pain behaviour will, for the most part, be defined. Many of the behaviour patterns will be acquired. One such acquired tendency toward a specific piece of pain behaviour has already been mentioned. It is that defined, roughly, by: (1) If x is asked at t “are you in pain?” then x says at t “I am in pain.” But there are other kinds of behaviour falling under the concept of pain behaviour. We all know many of them (even if we can’t spell them out). The precise details of the concept can be left to the psychologists; these details will be decided by what the laws (regularities) are in the relevant area.135 Such questions of significance are scientific and can be answered only by the future development of psychology. But we need not wait. It is the schema which interests us, not the details. So I shall brush such issues aside and stick with the simple, pre-scientific notion of “pain behaviour” and simply use (1) as an appropriate illustration. Nor is this the only part of our discussion which will be schematic, even crude. Neither phenomenological detail nor scientific detail will be

336

Body, Mind and Self in Hume’s Critical Realism

given. If I were doing either phenomenology or science, the present essay could be criticized for omitting such detail. But the purpose of the present discussion is neither phenomenology nor science, but rather philosophical illumination. Are the schemata too crude for this purpose? I think not. But, anyway, it cannot be judged until I have actually put them to use. The proof of any pudding is in its eating. Let us therefore turn to the conscious state attributed to Jones when we say of him that he is in pain. The model of conscious states is the one which we developed above, following Hume. Here again, though, we will be schematic with respect to phenomenological detail. Wherever there is a conscious state, there is a mental act. This act is one of experiencing, the (“mere”) experiencing which is there when one lives through or enjoys a conscious state. The experiencing is of or, as one says, intends the content of the conscious state. The experiencing which must be there if there is to be a conscious state is never itself part of the content of that conscious state.136 This does not mean (to make again a point we have made earlier) that the experiencing cannot be the content of another conscious state. Of course it can. All that it requires is, as the psychologists speak, a shift of set. But if such a shift of set occurs, then there is a further experiencing.137 This further experiencing makes the situation one in which one lives through the consciousness of pain. In the situation there are two acts, each of which has the character of being an experiencing. The one experiencing intends a phenomenal pain. The second experiencing intends the first. This second experiencing accounts for the fact that consciousness of pain is itself the content of a conscious state. We shall more to say about this account of conscious states in the next chapter, where we shall fit it more exactly into the concept that we have of persons, or, perhaps, into the various things that philosophers have claimed about our concept of a person. But for present purposes, we have said enough about the nature of conscious states. Return, therefore, to Jones’ case. He is in a conscious state. What constitutes this conscious state is an act of experiencing. This experiencing intends a phenomenal pain; i.e., the phenomenal pain is part of the content of the conscious state. This phenomenal pain is a fact. That is, it is an individual with a certain character, viz., the character of being a phenomenal pain. This same character can be exemplified by several individuals. Moreover, the individual that exemplifies this character is not the individual which is the experiencing that makes the phenomenal pain

337

Four: Disappearance of the Simple Self: Its Problems

the content of a conscious state. These two points can be established simultaneously by considering a single conscious state which has as its content two instances of the same (determinate) kind of phenomenal pain, e.g., two toothaches. However, the phenomenal pain is only part of the content of the conscious state. Another part of the content is a perceiving. Here I shall use “perceiving” generically.138 One may have affective and conative perceivings as well as cognitive perceivings. Thus, in willing one’s arm to go up, the bodily movement is present to one in the volition, or, in other words, the bodily movement is the intention of the volition; one perceives the bodily movement conationally. Besides being of a phenomenal pain, the experiencing, which makes the situation one in which there is a conscious state, is also of a perceiving. This perceiving is itself a mental act. It intends the physical fact that Jones’ bodily state is one of exemplifying the various tendencies to display pain behaviour. When we assert of Jones that he is in pain, we are asserting of him at least that there is a certain conscious state which belongs to Jones and that this conscious state has a certain content; it is a consciousness of pain. I am now suggesting that this conscious state has an additional content. I am suggesting that when Jones is conscious of the phenomenal pain, he is also conscious of the bodily state the exemplification of which provides the evidence on the basis of which we (who are not Jones) infer the presence of the conscious state which we assert of Jones. It is this usually ignored awareness of his own bodily state which I shall suggest is crucial for what we are about: the explications and arguments I shall offer depend upon taking it seriously. Thus when Jones is aware of the phenomenal pain, the experiencing which makes the situation one in which there is a conscious state is also an experiencing of a perceiving, namely, a perceiving of Jones’ own bodily state, the behavioural state which is parallelistically correlated to that conscious state. Now this physical fact of Jones being in that bodily state is perceived, so to speak, from a particular point of view – “from the inside,” as it were. This is a metaphor, one which summarizes a large number of phenomenological facts. By sketching what these relevant facts are, we can unpack this metaphor. Once it is unpacked, we will continue to use it. One perceives one’s own body. One also perceives the bodies of others. Jones has perceivings of Jones’ bodily state and perceivings of Smith’s bodily state. One perceives one’s own body in bits and pieces, so

338

Body, Mind and Self in Hume’s Critical Realism

to speak. The backs of others can be perceived without devices. To perceive one’s own back, one needs a mirror. One can “look down” one’s own nose, but not that of another. When eyeglasses are being worn, one’s own are, quite literally, seen from the inside; those of others, from the outside. Mach describes this fact as follows: ... I lie upon my sofa. If I close my right eye, the [following] picture ... is presented to my left eye. In a frame formed by the ridge of my eyebrow, by my nose, and by my moustache, appears a part of my body, so far as visible, with its environment. My body differs from other human bodies – beyond the fact that every intense motor idea is immediately expressed by a movement of it, and that, if it is touched, more striking changes are determined than if other bodies are touched – by the circumstance, that it is only seen piecemeal, and, especially, is seen without a head.139

Mach also notes in this passage conative and affective mental states in addition to the cognitive. These too are important in the phenomenological description of oneself. Thus, conatively one is aware of other objects as things to be manipulated; they are tools to be used. But one’s own body is not an instrument to be manipulated. Or at least not to be manipulated in the same sense. It rather is manipulated – “from the inside.” Sartre describes this phenomenological fact in these terms: . . . while each instrument refers to another instrument and this to another, all end up by indicating an instrument which stands as the key for all. . . . The first term is present everywhere but it is only indicated. I do not apprehend my hand in the act of writing, but only the pen which is writing; this means that I use my pen in order to form letters but not my hand in order to use the pen. I am not in relation to my hand in the same utilizing attitude as I am relation to my pen; I am my hand.140

We may also get at the point of the metaphor by noting that one’s pains are located at a point in one’s body.141 The difference between how one becomes aware of this location oneself, and how others become aware of it142 indicates how the metaphor is to be unpacked. These differences between how one’s own body is perceived and how others’ are perceived we may summarize with Köhler as follows: Phenomenally, the bodily “self” is not a physical entity outside immediate experience as is the physical organism; it is, rather, a percept of which we are aware, enriched by changing moods, attitudes, efforts and activities.... In this world which is that of naive everyday life, certain parts, events and properties

339

Four: Disappearance of the Simple Self: Its Problems

belong – phenomenally – to the “self,” others belong to “objects” or, more generally speaking, to the phenomenal environment of the “self.” The former have the character of “subjectivity” which, in this sense, is only another name for the fact that they appear and are counted as parts or states of the “self”; the latter have, in most cases, the character of “objectivity.”143

These phenomenological facts suffice, it seems to me, to unpack the metaphor that one perceives one’s own body “from the inside.” It is perhaps worth noting that perceiving “from the inside” and perceiving “from the outside” are not, of course, two different ways of perceiving. The mental acts are of the same species, viz., perceptual.144 It is with respect to their objects or intentions that acts differ; “inside” and “outside” refer to different objects or intentions of acts of perceiving. Suppose Jones perceives a bodily state of his own. Suppose that Smith perceives the same bodily state (Jones’) from the outside. However, though the acts differ with respect to their objects or intentions, they are in one sense still about the same thing (or object – in a different sense of “object”) so far as the same bodily state (viz., Jones’) is part of the intention of each act.145 Whether one has a perceptual act with this object or intention or one with some other object or intention is determined by the laws of nature which relate conscious states to physical states. Among the relevant physical states are (often) those physical states which the perceptual acts intend (cf. f4 in [D]). Thus, Smith’s body interacts with that of Jones’. If Jones’ body is antecedently in the appropriate state, then at the end of the process Jones’ brain will be in a certain state that it would not have been in if Smith’s body had not interacted as it did (cf. SI). Then, parallelistically connected with this brain state (cf. [A]) is a conscious state which has as part of its content the perceiving of Smith’s body, more exactly, a perceiving from the outside of a certain state of Smith’s body. But also the conscious state that is thus parallelistically connected to this state of Jones’ brain has as part of its content a perceiving from the inside of Jones’ body – Jones is perceiving Smith’s body from the outside and his own body from the inside, and these perceivings are both among the contents of Jones’ conscious state.. When Jones perceives his own body, it is the physical state of his own body which is the cause of Jones’ perceiving his own body. Jones’ body being in a certain state causes his brain to be in a certain state. Parallelistically connected to this brain state is the conscious state with the perceiving from the inside as part of its content. It is worth noting that

340

Body, Mind and Self in Hume’s Critical Realism

there are in that content both the perceving as an intentional event but also non-intentional entities – sensory impressions, some like impressions of Smith’s body in being coloured but differently located, the brown of the eyebrows but seen as framing one’s field of vision, but also other impressions which are wholly private, such as those impressions one has of the positions of one’s limbs, and also sensations of pain and pleasure, of hunger and thirst, the feeling that one is going to sneeze, and so on and so on. Such is the scientific account of perception. Again I shall be schematic. I shall ignore the scientific account of how one at times perceives veridically, as one says, and at other times non-veridically. I shall assume for the sake of simplicity that all perceptions are veridical. Further, I shall ignore the brain (that is, micro-) states, I shall assume that when Jones perceives some macro-state of Jones’ body from the inside, the parallelistic connection holds between the physical macro-state and the conscious state. We are considering Jones, sitting in that chair. He is in pain. There is in the body a tendency or disposition to display pain behaviour. There is a certain conscious state. This state exists by virtue of an experiencing. This experiencing intends the content of the conscious state. This content consists of (at least) two facts. One is a phenomenal pain. The other is a perceiving from the inside that the bodily state of Jones involves a tendency to display pain behaviour. (Recall that we are using “perceiving” generically to cover in addition to cognitive also conative and affective awareness of one’s bodily states. Clearly, on phenomenological grounds, we are aware of bodily states which involve tendencies to behaviour most often conatively and affectively rather than cognitively.) The physical state, the experiencing of the phenomenal pain, the act of perceiving, and the latter’s object are connected by the laws which give the parallelistic connection between mind and body. Schematically, the relevant parallelistic laws are: (2) If x is a body, then, whenever x is presented with the appropriate stimulus, then x displays pain behaviour just in case that there is an experiencing of both a phenomenal pain and a perceiving from the inside that x exemplifies a tendency to display pain behaviour. (3) If x is a body, then, whenever there is an experiencing of both a phenomenal pain and a

341

Four: Disappearance of the Simple Self: Its Problems

perceiving from the inside that x exemplifies a tendency to pain behaviour, x exemplifies a tendency to display pain behaviour. Here the tendencies to display behaviour are the sort (1). (2) and (3) are, of course, literally false. We have been through this before. Both (2) and (3) leave out many relevant variables. As one says, these laws hold only under “normal conditions.” Or, in other terms, they are gappy, pieces of imperfect knowledge. Many of the other relevant factors we could all fill in. All of the relevant factors, none of us could fill in. These factors are all, however, a matter of scientific detail. For our purposes we may ignore them; the schema suffices. Again, the proof of the pudding is in the eating. What (2) and (3) embody is the special lawful relationship that one’s conscious states have a certain special material object, that object which is one's own body. Mill recognized how crucial this relationship was to the understanding of our inferences to other minds. Whatever sensation I have, I at once refer it to one of the permanent groups of possibilities of sensation which I call material objects. But among these groups I find there is one (my own body) which is not composed, like the rest, of a mixed multitude of sensations and possibilities of sensation, but is also connected, in a peculiar manner, with all my sensations. Not only is this special group always present as an antecedent condition of every sensation I have, but the other groups are enabled to convert their respective possibilities of sensation into actual sensations, by means of some previous change in that particular one. I look about me, and though there is only one group (or body) which is connected with all my sensations in this peculiar manner, I observe that there is a great multitude of other bodies, closely resembling in their sensible properties (in the sensations compounding them as groups) this particular one, but whose modifications do not call up, as those of my own body do, a world of sensations in my consciousness. Since they do not do so in my consciousness, I infer that they do it out of my consciousness, and that to each of them belongs a world of consciousness of its own, to which it stands in the same relation in which what I call my body stands to mine.146

The crucial point, then, is that the laws that connect mind and body connect conscious states to bodies that own them. It is this that constitutes the significant difference between (2) and (3), on the one hand, which correctly describe the connection, and Addis’ (B) or Plantinga’s (@), on the other, which do not. (2) and (3) mention the awareness of the body which is not mentioned in (B) or (@). Addis’ I corresponds only to the awareness of the phenomenal pain; there is nothing in mentioned in the law

342

Body, Mind and Self in Hume’s Critical Realism

(B) that corresponds to the awareness of the bodily state. Similarly, there is nothing in Plantinga’s (@) that corresponds to the awareness of the bodily state. It is this difference between the phenomenologically inaccurate (B) and (@), on the one hand, and, on the other hand, the phenomenologically more accurate (2) and (3) that enable the latter to answer the privacy objection to parallelism. Suppose there is no conscious state of the kind Jones has. This premise, together with the true generality (2), entails the Jones’ body does not exemplify a tendency to manifest pain behaviour. Conversely, suppose Jones’ body does not exemplify a tendency to manifest pain behaviour. Then this, together with (3), entails that there is no conscious state of the sort Jones has. I.e., our knowledge of laws (2) and (3) justifies our asserting the subjunctive conditionals: (4) If Jones’ body were not to display pain behaviour under the appropriate stimulus, then there would be no conscious state of the appropriate sort and (5) If there were no conscious state of the appropriate sort, then Jones' body would not exemplify a tendency to display pain behaviour. (4) and (5) show the way the conscious state and the bodily state “belong to” each other.147 This point is worth stating more accurately. One defines (6) “A conscious state c belongs to a body x” is short for “c contains a perceiving from the inside of a state of x.” Then one obtains from (2) the true generality (7) If x is a body, then, whenever x is presented with the appropriate stimulus, then x displays pain behaviour, just in case that there is a conscious state belonging to x which contains both a phenomenal pain and a perceiving from the inside that x exemplifies a tendency to display pain behaviour. Similarly one defines (8) “A body x belongs to a conscious state c” is short for “c contains a perceiving from the inside of a state of x”148 and deduces from (3) the true generality

343

Four: Disappearance of the Simple Self: Its Problems

(9)

If x is body, then, whenever there is a conscious state containing both a phenomenal pain and a perceiving from the inside that x exemplifies a tendency to display pain behaviour, the body belonging to the conscious state exemplifies a tendency to display pain behaviour. Three comments are relevant. One. It is a conscious state which belongs to a body; it is a body which belongs to a conscious state.149 But that which both makes the conscious state a conscious state and which unifies the contents into the contents of this conscious state is the experiencing which intends those contents. The conscious state belongs to the body by virtue of there being among its contents a perceiving of the body from the inside, but that perceiving is among the contents of this unitary conscious state by virtue of the experiencing or direct awareness which makes the state a unified conscious state. Thus, it is this experiencing or direct awareness that determines what belongs to the body, and what the body belongs to is determined by this experiencing.150 Two. The two concepts of “belonging to” are significant. That is, they occur in true synthetic generalities, in particular, (7) and (9).151 These facts of significance show that the concepts defined by (6) and (8) are not “mere” arbitrary stipulations. They are stipulations, to be sure. But the concepts defined by these stipulations are significant. And whether or not a concept is significant is a matter of fact, not stipulation. Three. The truth of (7) and (9) depends on the truth (2) and (3). If the latter were not true or, at least, if some other synthetic generalities of a closely similar nature were not true, then the former would also be false. This means that if the parallelistic connections did not hold, definitions (6) and (8) would be logically possible put pointless. They would be pointless insofar as the concepts they define would then not be significant. That we all systematically draw a distinction between my body and other objects, between that physical object which belongs to me and other physical objects, is perfectly clear. What is problematic is the analysis this fact requires. The analysis must reveal not only the ground of the distinction but also the basis for our being able to draw it systematically. The analysis just proposed does both these things. It begins with the phenomenological fact that provides the ground for the distinction. We are able to draw the distinction systematically because the relevant notion of “belonging to” is rendered significant by the laws which parallelistically

344

Body, Mind and Self in Hume’s Critical Realism

connect conscious states to bodily states. Or, at least (and this is the only point I wish to make), if parallelism is true (which it seems to be), then we can account for our being able to draw the distinction systematically. The fact that parallelism can account for the commonsense fact that we do draw this distinction systematically lends, I suggest, support to that hypothesis. I believe, moreover, that the notion of “belonging to” just defined provides the tools necessary for the defender of parallelism to meet successfully the mentioned argument that our not being aware of the pains of others provides reason to believe that parallelism is false. Before turning to this issue, however, I want to deploy the already explicated notion of “belonging to” in the explication of several other philosophically problematic notions. The possibility of these further explications strengthens the case I am making for parallelism. One can define (10) “A body x is in pain” is short for “x exemplifies a tendency to display pain behaviour and there is a conscious state which contains both a phenomenal pain and a perceiving from the inside that x exemplifies a tendency to display pain behaviour.” Then from (2) and (3) respectively, one can deduce the true synthetic generalities (11) If x is a body, then, whenever x is presented with the appropriate stimulus, then x is in pain just in case that there is an experiencing of both a phenomenal pain and a perceiving from the inside that x exemplifies a tendency to display pain behaviour. and (12) If x is a body, then, whenever there is a conscious state which contains both a phenomenal pain and a perceiving from the inside that x exemplifies a tendency to display pain behaviour, x is in pain. Four comments are relevant. One. The concept defined by (10) is, I suggest, an adequate (for our purposes) representation of our “ordinary” concept of pain, or, perhaps more accurately, what some (e.g., Strawson152) have argued is our “ordinary” concept of pain. When we say, using this “ordinary” concept, of Jones sitting in that chair, that “Jones is in pain,” we are attributing to Jones’ body roughly the property to which the definiendum of (10) refers.

345

Four: Disappearance of the Simple Self: Its Problems

(Thus, the cautious “assert or imply” used above is unnecessary.) Two. The concept defined by (10), i.e., our “ordinary” concept of pain, is a significant concept. This is shown by the true statements of law, (11) and (12), in which the concept occurs. Three. The truth of (11) and (12) depends on the truth of (2) and (3). For (11) and (2), (12) and (3) are logically equivalent pairs. Thus, if the parallelistic connection established by (2) and (3) did not exist, the concept defined by (10) would not be significant and would for that reason be useless. To make the same point differently: if the conscious state which was parallel to the bodily state which we say exemplifies a tendency to manifest pain behaviour, if that conscious state had a green sensation as its content rather than a phenomenal pain, then the second conjunct of the definiens of (10) would mention green sensations rather than a phenomenal pain. For only in that way would a significant concept be then defined. The parallelistic connection, being what it AS A MATTER OF FACT is, accounts for why we have the concept of pain that we do have rather than some other concept.153 Four. The definition (10) of our “ordinary” concept of pain shows that there is indeed a “conceptual connection” between the concept of pain, on the one hand, and, on the other, both the tendency to display pain behaviour and phenomenal pain. We have thus explicated the views of certain philosophers.154 But the fact that this “conceptual connection” exists does not exclude the existence of lawful connections among the concepts.155 These lawful connections are (2), (3), (11), and (12). Indeed, not only is the existence of a “conceptual connection” compatible with the existence of these lawful connections, but it is dependent upon it. It is dependent upon it in the sense that if these lawful connections did not exist, our “ordinary” concept of pain would not be significant, and would therefore be useless.156 Suppose Jones, sitting in that chair, says at to. (13) I am now in pain. What Jones has said is what would be said by somebody saying (14) At to Jones is in pain. The “pain” which occurs in (14) is that defined by (10). That means both that Jones’ body exemplifies a tendency (which, since Jones utters [13], is in fact actualized at to) to manifest pain behaviour at to, and that there is at to a conscious state having as its content both a phenomenal pain and a perceiving from the inside that Jones’ body exemplifies a tendency to pain

346

Body, Mind and Self in Hume’s Critical Realism

behaviour at to. Moreover, the conscious state belongs to the body of Jones at to, and the bodily state of Jones at to belongs to the conscious state, where the senses of “belonging to” are those defined by (6) and (8). Similarly, if Jones says at to (15) I am now aware that I am in pain then what he is saying could be said by (16) At to Jones is aware that Jones is in pain. I.e., what Jones is saying with (15) is that there is a conscious state at to which belongs to Jones, and that this conscious state has as its content the fact (set of facts) that Jones is in pain. With both (13) and (15) we see the following. If what the sentence says is true, then there is a conscious state, one which belongs to the body which utters the sentence. Conversely, the bodily state of the utterer belongs to the conscious state, i.e., to the experiencing which makes that state a conscious state. The point can be generalized. If one says (17) I am now five feet seven inches tall she is attributing no mental state to herself. In such a case, then, the “I” attaches the consciousness of the speaker to the body of the speaker (since the conscious state belongs to this body) and also attaches the body of the speaker to the consciousness of the speaker (since the bodily state belongs to it)157 – even though nothing is actually said by the speaker that attributes to her any sort of conscious state. This explicates Wittgenstein’s otherwise obscure comment that “There are two different cases in the use of the word ‘I’ (or ‘my’) which I might call ‘the use as object’ and ‘the use as subject.’” As an example of the first kind of use, he suggests “I have grown six inches”; of the second kind, “I have a toothache.”158 That such explications are possible is a (partial) justification of my belief that the concepts of “belonging to” defined by (6) and (8) explicate the relevant philosophically problematic notions. Another “conceptual connection” is worth noting. Recall that part of the relevant tendency to manifest pain behaviour is the disposition “I am in pain” (cf. [1], above). Thus, “pain” and “saying ‘I am in pain’”159 are related definitionally. I.e., there is a “conceptual connection” between being in pain and saying “I am in pain.” But again this definitionally established relation exists because the definitions define significant concepts. The definitions would be pointless if there were no parallelistic laws connecting phenomenal pain and tendencies to manifest pain behaviour, and if there were no laws associating the disposition to say “I

347

Four: Disappearance of the Simple Self: Its Problems

am in pain” with tendencies to other kinds of pain behaviour. I have thus returned to a point several times emphasized. What is of crucial importance are not the “conceptual connections” but rather the laws (cf. [2] and [3]), which establish the parallelistic connection between mental (conscious states) states and bodily states; for it is these laws which make the concepts we have significant, and therefore useful. Assuming parallelism (and certain phenomenological facts), one can account for our drawing the distinctions we draw, and our drawing them systematically, and one can explicate other problematic notions. This point is one of two which I wish to make in this section. I now turn to the other. (v) Privacy Parallelism is a possible and, indeed, a plausible account of the mind-body relation. The statement of it is philosophically unproblematic; so at least we have been arguing.160 If it is true, its truth is a matter of empirical fact. It is, I believe, true. All evidence, both that from systematic psychology161 and that from our pre-scientific judgments as instances of the proverbial “man [that is, more correctly, person] in the street,”162 is in favour of parallelism; no evidence is against it – except the fact of privacy. Let us now turn to this problem.

The set of facts relevant to this problem consists of those facts associated with our knowing “our own minds” but not “other minds,” that is, the fact that we know through inner awareness our own conscious states but do not know, through either inner awareness or outer awareness, the conscious states of others. Having analyzed what it is for a conscious state to “belong to” someone, we are now in a position to deal with the privacy issue. Once again, consider our guy Jones. He is conscious of the content of his own mental states. He perceives – from the inside – his own bodily states. He perceives – from the outside – the bodily states of Smith. He neither perceives nor is conscious of the contents of Smith’s mental states. Rather, he infers the existence of Smith’s mental states from the bodily states of Smith. This much is generally acceptable to all. Let us assume it. The usual area of disagreement concerns the nature of the inference.163 People are acquainted with individual facts. A person is acquainted with an individual fact if there is a mental act, either an experiencing or a perceiving, which has that fact as its object. A generalization is about all

348

Body, Mind and Self in Hume’s Critical Realism

individual facts. Therefore, it is about the individual fact with which one is acquainted. But it is about all the others also. That means one never “knows for sure” that any generalization is true. Nonetheless, one can obtain evidence for and against. Such evidence may not be conclusive, but it is evidence nonetheless. This evidence is provided primarily by the individual facts with which one is acquainted. One’s belief that a generalization is true more or less justified depending on what this evidence is. Suppose one believe the generalization to be true that (18) For every shark there is a pilot fish which accompanies it.164 If one perceives, becomes acquainted with, the fact which verifies (19) This is a shark then one can deduce from (19) and (18) that (20) There is a pilot fish accompanying this is true. If one perceives a pilot fish with the shark, one has perceived a fact which verifies (20). If one makes successful predictions in this fashion, then one’s belief that (18) is true is more highly confirmed than before. Now consider the generalization (2). Suppose Jones believes that it is 165 true. If Jones now perceives that he himself exemplifies a tendency to manifest pain behaviour, then he can deduce from this and (2) that there is a conscious state with certain contents which belongs to Jones (i.e., to himself). The contents are a phenomenal pain and a perceiving from the inside that Jones exemplifies a tendency to manifest pain behaviour. There is an experiencing which belongs to Jones and which is of these contents. So Jones is acquainted with the contents. As for the experiencing which (2) predicts, a shift of set, to give a consciousness of the experiencing, suffices to verify its existence. Thus each time Jones himself is in pain, Jones’ belief that the generalization (2) is true becomes more highly confirmed. However, consider the case of Smith. Jones perceives (from the outside) that Smith exemplifies a tendency to manifest pain behaviour. Jones might perceive the tendency being actualized, e.g., by Smith saying “I am in pain.” Such a perception will verify (21) Smith exemplifies a tendency to display pain behaviour. From (21) and (2) Jones can now deduce that (22) There is an experiencing of both a phenomenal pain and a perceiving from the inside that Smith exemplifies a tendency to display pain behaviour is true. If the generalization (2) is true, then the notion of “belonging to” defined by (6) is significant. Hence Jones can deduce that this conscious

349

Four: Disappearance of the Simple Self: Its Problems

state, whose existence he has inferred, belongs to Smith.166 Jones never becomes acquainted with the facts which verify (22). Call this fact (Priv). (Priv) is sheer commonsense: it states the fact that mental states are private. It is (Priv) that Plantinga and others have suggested provides disconfirming evidence for Jones’ belief that (2) is true. For if in some cases one looked all about a shark and failed to find a pilot fish, one would conclude that it was very probable that (20) was false. And if (19) were true, and (20) probably false, then one would have evidence showing the generalization (18) was also probably false. Similarly, if – as everyone agrees – Jones never becomes acquainted with the facts which verify (22), then Jones has thereby acquired evidence which strongly disconfirms the belief that (2) is true.167 It is concluded that parallelism is probably false and that, however it may be that Jones can reasonably infer that Smith has certain conscious states, it is certainly not by using parallelistic laws as major premises. Or so the argument goes. But we can now see that this is in fact a bad argument against parallelism. It does not recognize an important difference between that law (2) and the law (18). The law (18) does not mention those facts, the acts of acquaintance, by virtue of which the other facts (which it [the law (18)] leads one to believe exist) become objects of acquaintance. Or, more briefly, the law (18) does not mention how the individual facts about which it talks are known. In contrast, the law (2) not only leads to predictions about certain facts, but also predicts how these facts are known and to whom belong the acts of acquaintance by which they are known. This is the crucial difference.168 (Addis’ [B] is like [18] rather than [2], which is why his schema is unable to adequately account for the fact of privacy.) (2) leads to the prediction that certain facts exist. Among the facts are the facts (acts of acquaintance, either experiencings or perceivings) by virtue of which other facts are known. (2) also predicts to whom these acts of acquaintance belong. The prediction is, first, that a phenomenal pain exists and that this is known to exist by an act of experiencing; and, second, that the act by which it is known to exist belongs to Smith. It is not predicted that the act of acquaintance will belong to Jones. If Jones infers using (2), then he should expect not to be acquainted with the facts predicted to exist. That is, he should expect there to be no act which belongs to him and which has as its contents the facts which make (22) true. Thus, far from the fact (Priv) providing evidence which disconfirms the belief that (2) is true, (Priv) is exactly what one would be led to believe if (2) is true.

350

Body, Mind and Self in Hume’s Critical Realism

In short, the fact that I am not acquainted with your conscious states is evidence that parallelism is true. This is the second point which I wish to make in this section. A final three comments will conclude. One. Suppose (Priv) were not so; suppose that Jones did in fact become acquainted with the facts which verify (22). That is, suppose there was an act of acquaintance which belongs to Jones and which had the facts verifying (22) as its object. This act of acquaintance belongs to Jones. So it intends a perceiving from the inside of a state of Jones’ body. But it also, we have just said, intends the facts which verify (22). So it also intends a perceiving from the inside of a state of Smith’s body. However, this would conflict with something else we have good reason to believe, namely, that a person perceives one and only one body from the inside. Mill saw this point clearly, expressing it in this way: “I am aware, by experience, of a group of Permanent Possibilities of Sensation which I call my body, and which my experience shows to be an universal condition of every part of my thread of consciousness. I am also aware of a great number of other groups, resembling the one that I call my body, but which have no connexion, such as that has, with the remainder of my thread of consciousness.”169 The fact that this belief, that I am aware from the inside of but one body, is well confirmed in turn supports (Priv). Thus, it indirectly supports the belief that (2) is true. Two. (2) can be confirmed directly by acquaintance with the facts the existence of which it predicts only in case those facts belong to oneself. This does not mean that (2) is a generalization “from one instance,” as Ryle, for one, has suggested.170 The generalization (2) associates the displaying of pain behaviour with the existence of a conscious state of a certain kind. Pain behaviour is observed being displayed several times. Each time it is accompanied by the appropriate conscious state. So it is not a generalization from one instance but from several, and indeed quite a few.171 This point is clear, by the way, in Mill’s statement, quoted above, describing our inference to other minds: when I infer the existence of other minds, “I bring other human beings, as phaenomena, under the same generalizations which I know by experience to be the true theory of my own existence.”172 Three. My use of (2) to infer your pain does not lead to predicating my pain of you. To be sure, all I ever experience are my own phenomenal pains. But the character of being a phenomenal pain which is presented in such an experiencing is not intrinsically (in and of itself) mine. Nor is it

351

Four: Disappearance of the Simple Self: Its Problems

intrinsically yours. It is just what it is! It is mine only by virtue of being exemplified in a fact which is intended by an experiencing which is mine. When I say of you that your conscious state contains a phenomenal pain, the character is the same as the one presented to me in my own conscious states.173 Except that when I predicate it of you, I assert that it is yours – extrinsically – rather than – extrinsically – mine.174 The conclusion that I would have to predicate my pain of you could be arrived at, it seems to me, only if one made two confusions. The first is that between a fact and a character in a fact. (The phenomenal pain which I experience is a fact; the character of being a phenomenal pain is in this fact; but the same character can be in several facts.) [One would make this confusion, of course, if one were a nominalist!.] The second confusion is that between a property a thing has by virtue of standing in a relation to something else, and a property intrinsic to a thing. (The phenomenal pain is mine by virtue of its being intended by an experiencing which is mine; the phenomenal pain is a phenomenal pain by virtue of a character which is to be found in the thing, that is, the pain, which I experience.)175 [One could make this second confusion if one were tempted to adopt a nominalist account of relations.] This second confusion could be strengthened by yet another: If one confused the act (the experiencing which makes the phenomenal pain mine) with its object or intention (the phenomenal pain), one could mistake the extrinsic property of being mine for an intrinsic one like being a phenomenal pain.176,177

We have now looked at a number of senses of ‘mental’, including both the notion of intentionality, and the notion of privacy. We may conclude, I think, that these raise no problems for the Humean or bundle view of the self. We must now return to our main line of argument. This concerns possible problems that develop when one renounces the substantialist account of minds and turns instead to the bundle view of the empiricist. (5) The Problem of the Self in Hume Contrary to what some have suggested,178 Hume does not deny that we have an idea of the self. “Ourself,” he says, “is always intimately present to us...” (T, p. 320, p. 339, p. 427, p. 354). It is of central importance to the passions of pride and humility, according to Hume: “when self enters not into the consideration, there is no room either for pride or humility” (T, p.

352

Body, Mind and Self in Hume’s Critical Realism

277). This is not to deny that the self is a bundle of perceptions; thus, when Hume explains the exact role of the self with respect to pride, he tell us that “tho’ that connected succession of perceptions, which we call self, be always the object of the object of these two passions, ’tis impossible it can be their CAUSE...” (T, p. 277). Indeed, the fact that the self is simply a bundle of perceptions is itself of importance in the explanation of certain mental phenomena. Ourself, independent of the perception of every other object, is in reality nothing: For which reason we must turn our view to external objects; and ’tis natural for us to consider with most attention such as lie contiguous to us, or resemble us. But when self is the object of a passion, ’tis not natural to quit the consideration of it, till the passion be exhausted...(T, p. 341).

What Hume does deny is that we have an idea of the self as a simple substance. A person is “a bundle or collection of different perceptions which succeed one another with an inconceivable rapidity and are in perpetual flux and movement” (T, p. 252). The problem is that of finding the principle that unites these into a person, and not merely into a person but into a person with the relatively stable character that we need in order to carry on with the practical projects in our lives and which is presupposed by our ordinary moral principles. The central text for Hume on personal identity is, of course, Treatise, Book I, Part iv, Section vi, which bears that title. In the end, however, as we shall see, we cannot limit ourselves to this text; as so often in Hume, a discussion that begins in Book I of the Treatise finds its completion in Books II and III. Hume signals this in Book I when he tells us that ...we must distinguish betwixt personal identity, as it regards our thought or imagination, and as it regards our passions or the concern we take in ourselves. The first is our present subject...(T, p. 253).

But at the end of the discussion in I, iv, vi we are not given a clear account of personal identity – not as, for example, we are given clear accounts of causation and of the identity of body. One reason for this, as we shall see in the present section, is that Hume could not in his own terms address the problem that had been raised long ago by Plotinus. A second reason for it, however, as we shall see in the next Chapter, is that the relevant relation among the parts of the self that generates the idea of our self cannot be understood apart from our passions and our system of morality, topics that

353

Four: Disappearance of the Simple Self: Its Problems

do not get discussed until Books II and III respectively of the Treatise. The issue of personal identity cannot be resolved without turning to our passions and the concern we take in ourselves. But this should surely not surprise us. After all, Locke had already established that the notion of a person is a forensic notion, and to say this is to say that, as the Treatise is written, personal identity cannot be fully understood until we complete Books II and III. We must, however, begin where Hume begins, with the discussion of personal identity in Book I. It is here that Hume first establishes, on the basis of the Principle of Acquaintance, that is, in effect, Locke’s “historical plain method,” that just as we have no idea of body as a simple substance, so we have no idea of the self as a simple substance. The bundle view follows. But as in the case of body, there is continuity in the complexity, and the presence of a relation that generates this continuity. Like Locke, Hume points out that things are often counted the same even though their parts change. A church falls down and is replaced; it is nonetheless the same church (T, p. 258). It is the same church because it serves the same end, or, as he says, stands in the same relation to the people of the parish. Similarly, the parts of plants and animals change – the small tree grows into the mighty oak, the infant into the adult – but when they are added or replaced the parts continue to function jointly to maintain the organism. What we have here is “a sympathy of parts to their common end, and [we] suppose that they bear to each other, the reciprocal relation of cause and effect in all their relations and operations” (T, p. 257). This sort of reasoning, which has worked so well in the case of plants and animals, “must be continued” (T, p. 259), Hume tells us, in the case of personal identity. Here too we have a succession of objects so related as to form a continuous series. The continuity we transform into an identity when we “imagine something unknown and mysterious connecting the parts besides their relation” (T, p. 254). This identity is a fictitious one, the continuing entity a fiction, creating no real bond among them (T, p. 259). Now, Hume is here careful not to say that the relation that generates the continuity is the same as that which holds for organisms or parish churches. He says only that the case appears to be analogous for persons. In particular, no doubt, there will be the sympathy of the parts serving a common end. But that need not be the whole of it. It does involve resemblance and causation, he tells us (T, p. 260), but that is not to go very far. In particular, memory has a role to play. Through memory we recall

354

Body, Mind and Self in Hume’s Critical Realism

past contiguities in space and time, and this produces the association of ideas that constitutes the causal judgment. The mind then applies this to the remembered events, uniting them into a series. That is, memory is now that which locates the events that the mind unites into the series; memory discovers identity “by shewing us the relation of cause and effect among our different perceptions” (T, p. 262). Memory is also relevant in the case of resemblance. The memory recalls a past object, which resembles a present one, and this creates by association the abstract idea that links them; the mind then as it were reverses itself and applies the abstract idea to the remembered entity. (For Hume, one will recall, an abstract idea consists in a word being associated with the members of a class of particulars that resemble each other in a certain respect.) In this way, then, for both resemblance and causation, memory both produces and discovers personal identity (T, p. 262). There is nothing parallel to this in the case of animals. In the case of animals, it is our thought that creates the feeling that there is a tie of necessity; and it is our thought that creates what Hume thinks of as the fictitious continuing particular. The mechanism that creates the feeling is external to the animal. In the case of the mind, however, the mechanism is internal to the entity whose parts are tied together. The memories that we have both produce and discover the identity but are at the same time part of the series to which the identity is attributed. It is this which distinguishes the case of personal identity from that of the identity of animals. What we need, therefore, is a better grasp of what precisely is the relation that generates the continuity of mind. Recall how Hume sees these things. The relation, considered philosophically, is an association of ideas. This association is produced by some “quality” in objects; this quality is the relation considered naturally (T, pp. 13-14). Thus, in the case of causation, to judge that two things are related as cause and effect, there must be an association established between our ideas of the cause and the effect. The quality that establishes this association is a constant conjunction among things. Thus, to ask for a better account of the relation which produces the identity is to ask what “quality” of things produces the relevant association (T, p. 255). What we do know is that resemblance plays a role (T, p. 260), that causation plays a role (T, p. 261), that memory plays a role (T, p. 261-2), and also that memory cannot be a necessary condition, since Hume contends, as Molyneaux did, that causation extends our judgments of identity through gaps in our memories, even when we are not drunk – do you really remember what you were doing at precisely this

355

Four: Disappearance of the Simple Self: Its Problems

time last week? (T, p. 262) – and to events prior to any memories – the infant is the same person as the adult (T, p. 257). But to say these things is not yet to give a fully developed theory. Hume puts it this way in the “Appendix”: ...no connexions amongst distinct existences are ever discoverable by the human understanding. We only feel a connexion or determination of the thought to pass from one object to another. It follows, therefore, that the thought alone finds personally identity, when reflecting on the train of past perceptions, that compose a mind, the ideas of them are felt to be connected together, and naturally introduce each other. However extraordinary this conclusion may seem, it need not surpize us. Most philosophers seem inclin’d to think, that personal identity arises from consciousness; and consciousness in nothing but a reflected thought or perception. The present philosophy, therefore, has so far a promising aspect. But all my hopes vanish, when I come to explain the principles, that unite our successive perceptions in our thought or consciousness. I cannot discover any theory, which gives me satisfaction on this head (T, pp. 635-6).

Consciousness is reflected thought or perception; it is, in other words, the perception of thought. But a perception is an impression. So consciousness must be an impression that is of and is produced by our thoughts. Consciousness of personal identity must therefore be an impression of our self qua unified; that is, since the self is a series of events, it must be an impression of the events of this series as a unified whole. But the only impressions of which we are at a moment conscious are the impressions which are then present to us, i.e., not those in the past (or the future), and the only impression of unity that is present is the feeling of union that is present by virtue of the association among the ideas of the events in the series that constitutes the self. This feeling is produced by a quality of the perceptions in the series that forms the self, but is separate from the events that produced it. Hence it is not part of the self whose unity it constitutes. Yet surely the self’s awareness of itself is part of the self. This is the problem. Hume goes on to say how he could solve these problems if he had either real essences, that is, objective necessary connections, or a substratum as a continuant. In either case, we would have an impression in experience of the tie. This impression would at once be a perception of the unity that constitutes the self and at the same time be unified into the self by the tie of which it was an impression. But there is no such impression (T 636), and so the problem, for Hume, remains unsolved.

356

Body, Mind and Self in Hume’s Critical Realism

In fact there are two problems for Hume. The first is this: what is the quality of the events in the series that forms the self that produces the impression of unity? The second is this: how can the impression of unity that is produced by that quality represent the whole of which it is one part? The second problem is, of course, the problem that was raised by Plotinus. On the one hand, if the felt connection is established by an association of ideas, then the complexity of the latter, including its temporal progression, cannot be a perception, which is simple. On the other hand, does not the simplicity of the perception imply the simplicity of what is perceived? How can a simple perception be the product of a process of association and embody within it the series of events from the past and their continuation into the future? As John Stuart Mill put it, If...we speak of the Mind as a series of feelings, we are obliged to complete the statement by calling it a series of feelings which is aware of itself as past and future; as we are reduced to the alternative of believing that the Mind, or Ego, is something different from any series of feelings, or possibilities of them, or of accepting the paradox, that something which ex hypothesi is but a series of feelings, can be aware of itself as a series.179

If we take for granted the model of psychological analysis that Hume tended to accept and which he had inherited from Locke, then this problem cannot be solved. The notion of psychological analysis is that the analysis of a mental state yields the genetic antecedents of that state. Thus, the analysis of the impression of unity of a self will, upon analysis, yield the genetic antecedents of that impression; or, in other words, analysis of the impression of the unity of the self will yield the various parts of the series that constitute the self. But, upon the Lockean model of analysis, the psychological analysis of a mental state is the same thing as a logical analysis of a concept into the parts that define it. The parts into which an idea or impression is analyzed are upon this model real parts, really and wholly present in the thing analyzed. Thus, upon the Lockean model of analysis, the impression of unity must be a whole that has within it all the various parts that constitute the self; it cannot, in other words, be simple, but must contain within itself the complexity of the series which produced it. Clearly, one cannot within this Lockean account of analysis reply to the Plotinian problem in a way that avoids the substantialist account that Plotinus proposed as a solution, and, indeed, as Hume himself saw would provide a solution. The solution of the problem actually consists in the rejection of the

357

Four: Disappearance of the Simple Self: Its Problems

Lockean model of analysis. This is what John Stuart Mill did.180 As he explained it, quite correctly, psychological analysis does not locate the real parts of a mental state. Rather, what analysis does is recover through association elements that are normally only potentially or dispositionally present; that is, it recovers parts that are actually present only when they appear as a consequence of undertaking the analytic task. These “parts” that are thus recovered are the genetic antecedents of the phenomenologically simple phenomenon that their association produced.181 This enables Mill to say straight out that a simple perception can “contain” a complexity of parts, to say that there is no mystery here, only a fact. Mill makes the relevant point this way: The real stumbling block is perhaps not in any theory of the fact, but in the fact itself. The true incomprehensibility perhaps is, that something which has ceased, or is not yet in existence, can still be, in a manner, present: that a series of feelings, the infinitely greater part of which is past or future, can be gathered up, as it were, into a single present conception, accompanied by a belief of reality. I think, by far the wisest thing we can do, is accept the inexplicable fact, without any theory of how it takes place; and when we are obliged to speak in terms which assume a theory, to them without a reservation as to their meaning.182

To be sure, some have rejected Mill’s point. One such was T. H. Green,183 who resisted Mill’s claim that a part within a whole could represent the whole. “To be conscious of it [one’s personal history],” he tells us, “we must unite its several stages as related to each other in the way of succession; and to do that we must ourselves be, and distinguish ourselves as being, out of that succession.” It is only through our holding ourselves aloof, so to speak, from the manifold affections of sense, as constant throughout their variety, that they can be presented to us as a connected series, and thus move to seek the conditions of connection between them (p. 92).

But Green’s argument for this is not so much to deny Mill’s point as it is to insist that the relational structure of the self presupposes a self that is outside time. But for this argument to go, Green presupposes a particular account of relations, that of the nominalist. According to Green, both selves and material objects have as their basic constituents entities given in sense experience, or, as he calls them, feelings. On this point he is in agreement will Mill and with Hume. But,

358

Body, Mind and Self in Hume’s Critical Realism

...feelings are facts; but they are facts only so far as determined by relations, which exist only for a thinking consciousness ad otherwise could not exist (p. 53).

Relations do not exist objectively among the sensory phenomena; sensations are simply unrelated atoms. This is the nominalist picture of reality, and Green recognizes what we have argued, that this will not do. But instead of recognizing that relations are among the objective constituents of the world, as Mill and Russell were to insist, Green instead holds to the idealist position that they are contributed by the mind that knows the phenomena. The relata in the world of facts are feelings (sensations), but the relations are mental. ...understanding, as the unifying principle which is the source of relations, acts formatively upon feelings as upon a material given to it from an opposite source called ‘things-in-themselves,’ rendering them into one system of phenomena called ‘nature,’ which is the sole object of experience, and to which all judgments as to matters of fact relate (p. 57).

The world of structured facts, which is the only world that we know is thus dependent for its existence upon the mind that knows it only in the act of constructing it: “... so far as we feel without thinking, no world of phenomena exists for us. The suspension of thought in us means also the suspension of fact or reality for us” (pp. 54-5). Mere feeling, then, as a matter unformed by thought, has no place in the world of facts, in the cosmos of possible experience....Feeling and thought are inseparable and mutually dependent in the consciousness for which the world of experience exists, inseparable and mutually dependent in the constitution of the facts which form the object of that consciousness (p. 55).

Indeed, mere feeling itself is an abstraction from a structure of relations, and does not as such exist as a fact: ...the distinction of the merely feeling consciousness is just this, that what it is really it is not consciously – that the relations by which it is really determined [and which are contributed by a more encompassing consciousness] do not exist for it, but for the thinking consciousness on which it and they alike depend for being what they are (p. 54).

The consciousness which is the source of the relational structure of the

359

Four: Disappearance of the Simple Self: Its Problems

world itself stands outside this structure. But it does so as a substance which at once actively creates that structure and contains implicitly the whole within itself. As for any substance, its character as substance depends on that relation of appearances to each other in a single order which renders them changes. It is not that first there is a substance, and that then certain changes of it ensure. The substance is the implication of the changes, and has no existence otherwise (p. 61).

There is, perhaps, a certain point to Green’s argument, provided that one accepts the nominalist account of relations which is his starting point. For, that account in effect leaves the world a heap of atoms, without any structure, and one way in which to provide that structure is to have it supplied by the knowing mind. But this idealist solution to the problem of structure that is created by the nominalist account of relations is not the only move that is possible. An alternative is to reject the nominalist account of relations. This alternative does not occur to Green. But once it is available, there is no reason why we have to accept his account of the consciousness that a self has of itself. In other words, once we reject the nominalist account of relations, there is no reason for rejecting Mill’s claim that a part of the whole that is the self can represent that whole. We may take for granted, then, that a simple impression can indeed represent the complexity of the self. Why, indeed, should that be puzzling? Taken in this way, there is no further problem of the sort that Plotinus raised; this solves the second of the two problems we noted above. This returns us to the first of the two problems, that of finding an adequate account, an adequate sketch at least, of the associational or at least psychological processes that generate the simple perception of the unity of self, that is, to discover, as Hume put it, the “quality” of the parts of the self that yields the association of ideas that is the perception of the unity of the self. This is the topic which we take up in the next chapter.

360

Body, Mind and Self in Hume’s Critical Realism

Endnotes to Chapter Four

1.For an excellent discussion of some aspects of this Chapter, see Henry E. Allison, “Locke on Personal Identity: A Re-Examination.” 2.Joseph Butler, “Dissertation of Personal Identity,” p. 281. 3.John Locke, Works, vol. VII, 1794. 4.T. Reid, Inquiry into the Human Mind, and Essays on the Intellectual Powers of Man. 5.Stroud, Hume, Ch. II. 6. Yolton, “Hume’s Ideas.” 7.John Sergeant, Solid Philosophy Asserted against the Fancies of the Ideists, 1697. 8.Reid, Essays on the Intellectual Powers of Man, p. 445. Cf. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill. 9.T. H. Green, Prolegomena to Ethics: “Feeling [sensation] and thought are inseparable and mutually dependent in the consciousness for which the world of experience exists...” (p. 55). 10.É. Gilson and T. Langan, Modern Philosophy: Descartes to Kant. The difficulty with the empiricists, these authors assert, is that they have forgotten “the balanced realism of a St. Thomas” (p. 195). More specifically, for Hume there is “a single subjective source” (p. 252) for all our perceptions. The basis of perception is Lockean sensations, but these require synthesis to become perceptions; the faculty of synthesis is the imagination which thus becomes the “central cognitive faculty” (p. 254). But what guides the imagination in its synthesizing work? If nothing, according to Gilson and Langan, then scepticism results. To avoid this we need either Thomistic realism or Kantian or Husserlian transcendentalism (p. 254), that is, a substantial active mind that can inform the imagination with the objective entity that is perceived. But this of course just begs the question. Of course, if the imagination is totally unguided then there is no reason to suppose that it will synthesize sensations into veridical perceptions. But why can there not be mechanisms in the mind that ensure conformity of thought to objective reality? Why cannot one hold, as Hume does, that through the processes of association the objective patterns of the world themselves guide the formation of the patterns of the imagination and therefore of perception? Why will this not suffice to ensure an objective validity to one’s perceptions? Gilson and Langan do not answer; there is no answer. Again, why not hold, as Hume also does, that the mind guided by the passion of idle curiosity, the love of truth, can set itself to conform its patterns of thought to the patterns of objective reality? Why will

361

Four: Disappearance of the Simple Self: Its Problems

this not suffice to ensure an objective validity to one’s mental representations? Again, Gilson and Langan have no answer; again, there is no answer. To be sure, neither of these mechanisms for securing the conformity of perception and thought to reality can guarantee truth; they are both fallible. But fallibility is not the same as scepticism. However, whether Gilson and Langan recognize this simple point is clearly an open question; but I suspect they don’t. That means that they notice falliblism, illegitimately infer scepticism, and conclude that Hume needs something more than the two suggested mechanisms to combat scepticism and to account for perception and thought, something like a substantial mind. Understood this way, it is evident that their case simply begs the question against Hume. But then begging the question against the empiricists is an old tactic that they do not shun. 11.A. and D. Hausman, “Hume’s Use of Illicit Substances,” p. 18ff. 12.For a detailed discussion of the history and nature of introspective psychology, with special reference to John Stuart Mill’s contribution, see F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill. 13.For further discussion of this principle, see F. Wilson, The External World and Our Knowledge of It: Hume’s Critical Realism, An Exposition and Defence, chapter six. 14.Some suppose that the so-called “missing shade of blue” constitutes a counterexample to Hume’s principle; I have discussed, and dismissed, that notion in F. Wilson, “Hume’s Fictional Continuant.” 15.J. S. Mill, Note 108 to Chapter XI of James Mill, Analysis of the Phenomena of the Human Mind, Second Edition, vol. I, p. 402ff. 16.Cf. E. G. Boring, A History of Experimental Psychology, Second Edition, p. 224. 17.For a detailed discussion of this notion of imperfect or gappy knowledge, cf. G. Bergmann, Philosophy of Science, Chapter Two; F. Wilson, Explanation, Causation and Deduction; and F. Wilson, The Logic and Methodology of Science and Pseudoscience. 18.Cf. Boring, A History of Experimental Psychology, pp. 228-9. 19.Cf. F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill. Also G. Bergmann, “The Problem of Relations in Classical Psychology,” in his Metaphysics of Logical Positivism. 20.For more on this theory and the connected research programme, see F. Wilson, “Some Controversies about Method in Nineteenth- Century Psychology.” 21.Cf. J. S. Mill, Examination of Sir William Hamilton’s Philosophy, p. 259.

362

Body, Mind and Self in Hume’s Critical Realism

22.John Horne Tooke, The Diversions of Purley (London, 1798; reprinted by the Scolar Press, Menston, England, 1968), p. 47ff. 23.It was introduced by Franz Brentano, Psychologie vom Empirischen Standpunkt (Leipzig, 1874). 24.Cf. G. Bergmann, “Intentionality.” 25.R. J. Butler, “Hume’s Impressions.” The same point is argued in John W. Yolton, “Hume’s Ideas.” 26.Cf. E. G. Boring, A History of Experimental Psychology, Chapter 12; and Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, Chapter Eight. 27.J. Passmore, Hume’s Intentions, p. 96. 28.J. Bennett, Locke, Berkeley, Hume, p. 224. 29.Ibid., p. 224. 30.Cf. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, where the views of Kames and Reid as critics of associationism are discussed. These philosophers argued from the phenomenological simplicity of thoughts, e.g., perceivings, to their being unlearned or innate, precisely the sort of inference that we find in Passmore and Bennett. John Stuart Mill showed once and for all that this is a fallacious inference, and that the phenomenological simplicity of a mental state is compatible with its having come to be through learning, and, more specifically, through a process of association. This in fact was the major contribution of the younger Mill to the development of psychology. For details, see ibid. See also F. Wilson, “Wordsworth and the Culture of Science.” 31.We shall hereafter ignore the qualification that Hume does not securely grasp this point by virtue of his flawed understanding of the nature of psychological analysis. 32.A. and D. Hausman, “Hume’s Use of Illicit Substances.” 33.Cummins, “Perceptual Relativity and Ideas in the Mind.” But in criticism of this reading, see Pappas, “Ideas, Mind and Body”. A. Hausman has replied to Pappas and other critics of the inherence interpretation; see his “Adhering to Inherence.” See also F. Wilson, “On the Hausmans’ ‘New Approach to Berkeley’s Ideal Reality’.” 34.Hausmans, p. 14f. 35.Hausmans, p. 21ff.

363

Four: Disappearance of the Simple Self: Its Problems

36.We should once again note that we can ignore the qualification that Hume does not securely grasp this point by virtue of his flawed understanding of the nature of psychological analysis. 37.Hausmans, p. 28. 38.Compare the following exchange: W. H. Hay, “Berkeley’s Argument from Nominalism”; E. B. Allaire, “Berkeley’s Realism”; R. J. van Iten, “Berkeley’s Alleged Solipsism”; A. Hausman, “Solipsism and Berkeley’s Alleged Realism.” 39.Chisholm, “On the Observability of the Self,” p. 16. 40.Hausmans, p. 27. 41.Chisholm, pp. 16-17. 42.Hausmans, p. 27. 43.Shalom, The Body/Mind Conceptual Framework & the Problem of Personal Identity, p. 445. 44.Colin McGinn, “Can We Solve the Mind-Body Problem?” 45.Cf. F. Wilson, “Acquaintance, Ontology and Knowledge’” 46.D. Davidson, “Mental Events,” p. 83. 47.Cf. D. Davidson, “Causal Relations” and “The Individuation of Events.” 48.For a discussion and critical evaluation of Davidson on events, causal relations and explanation, see F. Wilson, Explanation, Causation and Deduction, p. 296ff. 49.D. Davidson, “The Myth of the Subjective,” pp. 165-66. For criticism of this sort of nominalism, see the decisive essays by H. Hochberg, “Mapping, Meaning and Metaphysics,” and “Nominalism, General Terms, and Predication.” 50.Davidson, “Mental Events,” p. 82. 51.The term ‘gappy’ derives from J. L. Mackie; see his “Causes and Conditions.” The idea, however, derives from G. Bergmann’s notion of “imperfect knowledge”; see Bergmann, Philosophy of Science, Ch. II. For a detailed discussion of the role of gappy or imperfect laws in explanation, see F. Wilson, Explanation, Causation and Deduction. See also F. Wilson, The Logic and Methodology of Science and Pseudoscience.. 52.Davidson, “Mental Events,” p. 94.

364

Body, Mind and Self in Hume’s Critical Realism

53.Davidson, “Mental Events,” p. 98. 54.Cicero, The Nature of the Gods, II.22, p. 55. 55.Thomas Aquinas, Summa Theologica, First Part, Q118, A1. 56.Samuel Clarke, Discourse concerning the Being and Attributes of God, vol. 1, p. 50. 57.Ibid., p. 52. 58.Clarke, p. 48. 59.Richard Bentley, Sermons Preached at Boyle’s Lecture, vol. III, p. 36. 60.McGinn, p. 350. 61.Ibid. 62.Bentley, pp. 34-5. 63.Clarke, p. 51. 64.J. Locke, Essay concerning Human Understanding, Bk. IV, ch. iii, sec. 28,; see also IV, vi, 10. 65.Cf. F. Wilson, “Moore’s Refutation of Idealism,” in his collected essays on ontology, Acquaintance, Ontology and Knowledge. 66.John Locke, Some Thoughts concerning Education, sec. 192; see also sec. 246. 67.For a discussion of Hume on causation, see F. Wilson, Hume’s Defence of Causal Inference. For a defence of the Humean view, see F. Wilson, Laws and Other Worlds. 68.These rules occur in the Treatise, Bk. I, Pt. ii, sec. xv. For a discussion of the background to these rules, see F. Wilson, Logic and the Methodology of Science in Early Modern Thought: Seven Studies, Study Five, “‘Rules by which to Judge of Causes’ before Hume.” 69.T. Nagel, “What Is It Like to Be a Bat?” 70

Quoted in William James, The Varieties of Religious Experience, p. 438.

71.For discussion of these notions, see D. Flage, “Locke’s Relative Ideas,” and “Hume’s Relative Ideas.”

365

Four: Disappearance of the Simple Self: Its Problems

72.For a detailed logical analysis of gappy explanations, emphasizing how important it is to recognize such explanations in defending the deductive-nomological model of explanation, see F. Wilson, Explanation, Causation and Deduction. For a discussion of their role in one of the particular sciences, see F. Wilson, Empiricism and Darwin’s Science. See also F. Wilson, The Logic and Methodology of Science and Pseudoscience. 73.In this sense, it is a mark of incompleteness that psychology continues to use mentalistic language in its literal sense. On this point, see M. Brodbeck, “Mental and Physical: Identity vs. Sameness.” 74.C. D. Broad, Mind and Its Place in Nature, p. 118. 75.T. H. Huxley, “On the Hypothesis that Animals Are Automata, and Its History,” p. 240. 76.E. G. Boring, Sensation and Perception in the History of Experimental Psychology, p. 85. 77.David Hartley, Observations on Man. 78.For a discussion of these other features of Hartley’s philosophy, see F. Wilson, “The Ultimate Unifying Principle of Coleridge’s Metaphysics of Relations and Our Knowledge of Them.” 79.Parrallelism has received an extended defence in L. Addis, The Logic of Society. See also his “Behaviorism and the Philosophy of the Act.” 80.Cf. F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, pp. 299-311. 81.H. Feigl, The “Mental” and the “Physical.” For a discussion of Feigl’s views on the identity of mental events and physical events, see the important essay of M. Brodbeck, “Mental and Physical: Identity vs. Sameness.” 82.Locke, Some Thoughts concerning Education, p. 246. 83.Ibid., p. 248. 84.Ibid., p. 249. 85.John Stuart Mill, System of Logic, eighth edition, Bk. III, Ch. xiv, sec. 1. 86.As May Brodbeck once put it, even if we establish an identity between physical events and mental events, it will still be true that we cannot establish that they are the same: the properties of spatially extended physical events are not the same as the properties of mental events. See M. Brodbeck, “Mental and Physical: Identity vs.

366

Body, Mind and Self in Hume’s Critical Realism

Sameness.” 87.See Jaegwon Kim, “Concepts of Supervenience.” Kim’s discussion more or less takes for granted his account of events; for criticism of the latter, see F. Wilson, Explanation, Causation and Deduction, p. 318ff. 88.For this argument, see H. Putnam, “The Nature of Mental States.” Part of Putnam’s argument is based on an unreasonable account of meaning; I have discussed aspects of the latter in F. Wilson, “The Notion of Analyticity in Carnap’s Later Philosophy,” p. 176ff. 89.Jaegwon Kim, “Supervenience and Nomological Incommensurables,” p.150. 90.Cf. Addis, “Pains and Other Secondary Mental Entities.” 91.J. S. Mill, Examination of Sir William Hamilton’s Philosophy, p. 190. 92.Ibid., p. 191. 93.See note 44, above. 94. Ibid., p. 444. See also Wollheim, “Privacy,” especially p. 88. 95.Plantinga plays such a scepticism off rather neatly in his God and Other Minds against a similar scepticism with respect to the existence of a god. 96.See Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, Ch. 8. 97.Gay, John, A Dissertation concerning the Fundamental Principle and immediate Criterion of Virtue, 1731. 98.David Hartley, Observations on Man (Bath, 1749). 99.Hartley refers to Gay’s work; see his Observations on Man, p. v. 100.Cf. E. G. Boring, A History of Experimental Psychology. 101.See Chapter Three, above, note 49. 102.See note 109, below. 103.John Bricke similarly concludes that Hume is best read as adopting a form of parallelism; see his discussion in his important study, Hume’s Philosophy of Mind, pp. 34-35. 104.Hartley quotes Newton, Observations on Man, p. 9.

367

Four: Disappearance of the Simple Self: Its Problems

105.For a discussion of the role of definite descriptions in this sense in science, see F. Wilson, Empiricism and Darwin’s Science; also F. Wilson, “A Note on Hempel on the Logic of Reduction.” 106.And also the theology! 107. Hartley’s Theory of the Human Mind, ed. Joseph Priestley. 108.See H. Spencer, Principles of Psychology. See also F. Wilson, “Herbert Spencer.” 109. Mill, Review of Bain 110.See F. Wilson, Psychological Analysis and the Philosophy of John Stuart Mill, Ch. 8. 111.In his Principles of Human Physiology (London, 1855). 112.Principles of Mental Physiology (London, 1890). 113.T. H. Huxley, “On the Hypothesis that Animals Are Autonmata and Its History,” p. 577. 114

And so, see Hughlings Jackson: “…although active nervous states are not psychical states, there is parallelism, and what we have to do in a medical inquiry is to discover the anatomy and physiology of the various nervous states that go along with various psychical states…” (“On Epilepsies and on the After-Effects of Epileptic Discharges,” originally in West Riding Asylum Medical Reports, vol. vi, 1876, reprinted in Hughlings Jackson, Selected Writings, vol. I, p. 139). Or again: Jackson points out that the doctrine that he adopts holds “that (a) states of consciousness or synonymously states of mind) are utterly different from nervous states of the highest centres; (b) the two things occur together, for every mental state there being a correlative nervous state; (c) although the two things occur in parallelism, there is no interference of the one with the other. Hence we do not say that psychical states are functions of the brain (highest centres), but simply that they occur during the functioning of the brain. Thus in the case of visual perception, arbitrarily simplifying the process, there is an unbroken physical circuit, complete reflex action, from sensory periphery ultimately though highest centres, back to the periphery. The visual image, a purely mental state, occurs in parallelism with – arises during (not from) – the activities of the two highest links of this purely physical chain (sensori-motor elements of the highest centres) – so to speak, it ‘stands outside’ these links.” (“Remarks on Evolution and Dissolution of the Nervous System,” originally in the Journal of Medical Science, April 1887, reprinted in Hughlings Jackson, Selected Writings, vol. II, p. 84) 115.Compare the discussion of the views of W. Sellars (Sellars fils) in the Appendix to

368

Body, Mind and Self in Hume’s Critical Realism

F. Wilson, “Acquaintance, Ontology and Knowledge,” in his collection, Acquaintance, Ontology and Knowledge. 116.Keith Campbell, Body and Mind, p.51. 117.Campbell, Body and Mind, p. 18. 118.Campbell, Body and Mind, p. 52. 119.Body and Mind, p. 39. 120.Body and Mind, p. 38. 121.We are here making the simplifying assumption that the connection is one-one from mind to body; the more realistic case, where the connection is one-many from mind to body, is only slightly more complicated. But the relevant point holds for both the simplified and complicated cases, so we can do what we want using only the former. 122.Campbell, Body and Mind, p. 55. 123.Body and Mind, p. 52. 124.Cf. Bergmann, “The Contribution of John B. Watson.” 125.L. Addis, “Ryle’s Ontology of Mind,” passim, and especially pp. 44-49. See also M. Brodbeck, “Meaning and Action”; and D. Sievert, “Historical Cross-Section Laws and Austin’s Scepticism in ‘Other Minds’.” 126.”Ryle’s Ontology of Mind,” pp. 28-31. 127.Cf. “Ryle’s Ontology of Mind,” p. 31. 128.Cf. Bergmann, Philosophy of Science, p. 124 ff.; and Sievert, “Historical CrossSection laws and Austin’s Scepticism in ‘Other Minds’.” 129.Bergmann, Philosophy of Science, pp 115-124. Cf. M. Brodbeck, “Explanation, Prediction, and ‘Imperfect Knowledge’.” For a particular reference to the mind-body connection, cf. Sievert, “Historical Cross-Section Laws and Austin’s Scepticism in ‘Other Minds’.” 130.Cf. Wilson, “Dispositions: Defined or Reduced?” 131.J. L. Mackie, “Causes and Conditions.” 132.See note 44, above.

369

Four: Disappearance of the Simple Self: Its Problems

133.L. Wittgenstein, Philosophical Investigations, part 1, sec. 244. 134.This, of course, is the truth behind much of what Ryle says (The Concept of Mind, p. 46, p. 51, p. 199, and passim). From the fact that many of these dispositions are present, it does not follow that there is not something actual which is also present (as Ryle himself seems at times to admit – see his comments on “visual imagery,” p. 27). Cf. Addis, “Ryle’s Ontology of Mind,” pp. 6ff., 26ff. 135.Cf. Wilson, “Definition and Discovery,” sec. 3. 136.The analysis of a conscious state is that of Bergmann. See Bergmann, “Acts.” (It is perhaps needless to add that the acceptance of Bergmann’s analysis of the conscious state does not require that one accept his analysis of the act.) See also E. Husserl, “Phenomenology.” Husserl writes (p. 124) that “The ‘I’ and ‘we,’ which we apprehend, presuppose a hidden ‘I’ and ‘we’ to whom they are ‘present’.” (Of course, Husserl’s later idealism does not follow from this alone. For a discussion of Husserl’s move from realism to idealism, see Bergmann, “The Ontology of Edmund Husserl.” See also R. Grossmann, The Structure of Mind, chap. 1. Brentano’s analysis of the conscious state is different; he proposed that each perceptual act is both an awareness of something else and an awareness of itself; cf. F. Brentano, Psychologie vom empirischen Standpunkt, vol. 1, pp. 196-199. On phenomenological grounds, I cannot but agree with Bergmann and Husserl that this analysis does not work. At the same time, however, I do not believe that this difference between Brentano and Bergmann and Husserl makes any difference to what I want to say; everything that I express in terms of the Bergmann-Husserl analysis can, it seems to me, be translated into the Brentano analysis. For this reason I shall continue to speak in the Bergmann-Husserl fashion. 137.Not that every awareness requires a further awareness; but that every awareness can be the object of a further awareness. This cuts off any potential regress (cf. L. Addis, “Ryle’s Ontology of Mind,” pp. 58-59). 138.This point should be kept in mind. 139.E. Mach, The Analysis of Sensations, pp. 18-19. Cf. also Ryle, The Concept of Mind, p. 37; and Jean-Paul Sartre, Being and Nothingness, p. 329. 140.Being and Nothingness, p. 323. Of course, to commend Sartre for his phenomenological description is not to commend the philosophical conclusions which he (quite illegitimately) draws from these descriptions. 141.As the case of the amputee who feels a pain in his lost limb shows, one can be in error here. As in cognitive perceivings (e.g., seeings), affective and conative perceiving can also be erroneous. But we have ways of finding out when we are in

370

Body, Mind and Self in Hume’s Critical Realism

error. 142.Thus, some now speak of “non-observational knowledge” of one’s bodily movements (cf. G. E. M. Anscombe, Intention, p. 69. 143.W. Köhler, The Place of Value in a World of Facts, p. 69. 144.This, of course, is why the term “non-observational knowledge” (see note 142) is a poor way to speak about perceiving from the inside. 145.This is analogous to the case of one person perceiving the same coin from two different angles. There will be two acts of perceiving. These acts will differ with respect to their objects or intention. One has as its object or intention the coin from above, and the other the coin from the side. But the two acts still are about the same thing (or object – in the other sense) insofar as the same physical thing or object (the coin) is part of the object of intention of each act. Compare the distinction drawn in Grossmann (The Structure of Mind, p. 121) between the objects of acts and the ultimate objects of acts. 146.Mill, Examination of Sir William Hamilton’s Philosophy, pp. 191-2. 147.Cf. “Ryle’s Ontology of Mind,” pp. 54-56. 148.This makes (6) and (8) define synonymous terms. “Belonging to” is a symmetric relation. A sentence using (6) will state the same facts as a sentence using (8). The difference will not be with respect to the facts stated but with respect to emphasis: Using (6) one will be emphasizing the idea of being related to a body, while using (8) one will be emphasizing the idea of being related to a conscious state. (6) and (8) are like the active and passive voices in verbs. Compare “He hit the ball” and “The ball was hit by him.” The difference lies in the emphasis one places; in the case of the former, it is on the hitter, while in the case of the latter, it is on the object hit. (8) can be rewritten in the passive voice to read (8*): “a body x belongs to a conscious state c” is short for “a state of x is the object of a perceiving from the inside which is part of the content of c.” However, (8*) adds noting to (8) besides the verbosity of the passive. Let us therefore stick with (8). 149.I mention only Jones’ body and Jones’ conscious states. I do not mention Jones the person. Jones the person is simply the temporally and lawfully ordered sequence of bodily states which are parallelistically connected with these, in particular, memory states. Of the person, of course, more later. 150.Smith’s body is the object of Jones’ perceiving, which is intended by the direct awareness which belongs to Jones. In this extended sense, Smith’s body may be said to belong to Jones. Cf. G.E. Moore, “The Nature and Reality of Objects of Perception.” 151.It will be recalled that this is Locke’s criterion for a good definition.

371

Four: Disappearance of the Simple Self: Its Problems

152.Cf. P. F. Strawson, “Persons,” p. 390 ff., and esp. p. 395. 153.Gathering several factually coextensive concepts together into a single concept coextensive with the rest is a common pattern, one to be found both in science and in everyday life. Cf. F. Wilson, “Definition and Discovery,” and “Is Operationism Unjust to Temperature?” 154.I have in mind Strawson’s notion of a “P-predicate,” to be found in his essay “Persons,” in Gustavson, ed., Essays in Philosophical Psychology, p. 391ff. Compare N. Malcolm, “Knowledge of Other Minds,” in Gustavson, ed., Essays in Philosophical Psychology, pp. 372-373. 155.This invalid inference is made in Malcolm, “Knowledge of Other Minds,” in Gustavson, ed., Essays in Philosophical Psychology, p. 366. 156.I believe this to have been seen by Wittgenstein. See his Philosophical Investigations, part 1, secs. 242, 350-351; and part 2, sec. xii. 157.Of course, as the phrases now go, in the first case the speaker “implies” or makes it to be understood that there is a certain conscious state belonging to him. However, in the second case the existence of a conscious state is not only “implied” by the speaker’s saying something, but also what the speaker says states that such a conscious state exists. 158.L. Wittgenstein, The Blue and the Brown Books, p. 66. 159.Note the word “saying.” 160.However, Passmore for one denies this. See J. Passmore, Philosophical Reasoning, p. 55f. But Addis (“Ryle’s Ontology of Mind,” p. 54f.) has effectively met Passmore’s objections. 161.Cf. Bergmann, “The Contribution of John B. Watson,” pp. 265-276. 162.Cf. Moore, “The Nature and Reality of Objects of Perception,” and especially J. Austin, “Other Minds,” in A. Flew, ed., Logic and Language, 2nd series, p. 82ff. Specifically, Austin argues that in some ordinary sense of “know,” we (often) know with a certainty sufficient to avoid scepticism, what is going on “in” “other minds.” For this to be defensible, some form of parallelism must hold between mind and body (cf. Addis, “Ryle’s Ontology of Mind,” pp. 47-49). Austin does have hesitations, however. For an excellent discussion of these, see Sievert, “Historical Cross-Section Laws and Austin’s Scepticism in ‘Other Minds’.” 163.I ignore those, who like Watson and Ryle, heroically embrace metaphysical behaviourism. Concerning Watson, see Bergmann, “The Contribution of John B. Watson”; concerning Ryle, see Addis, “Ryle’s Ontology of Mind.” It is often

372

Body, Mind and Self in Hume’s Critical Realism

suggested, on the basis of Malcolm’s “Knowledge of Other Minds,” that he is a metaphysical behaviourist. However, as he himself maintains, and as seems perfectly evident from the paper, he is no metaphysical behaviourist; that is, he maintains that there is indeed an inference involved in any claim about other minds. He disagrees with someone like Hume or J.S. Mill (and with the position we are defending), not with respect to whether there is an inference, but with respect to the nature of this inference. John B. Watson and Ryle, on the other hand, maintain there is not an inference because there is nothing mental beyond behaviour and tendencies to display behaviour. One must say, therefore, of Malcolm that he is disputing, not the parallelistic hypothesis per se, but rather the claim that the parallelistic connections are empirical generalizations; he wants to hold, to the contrary, that these connections are somehow “non-contingent.” 164.How the belief is acquired, i.e., what causes one to have the belief, is not at issue. Nor are philosophers capable of settling this issue. The question is one of learning theory. 165.Again, how the belief is acquired is not important. Cf. Moore, “The Nature and Reality of Objects of Perception.” 166.Note that his is not an argument from analogy. It is a deductive argument, one premise of which is an inductive generalization. The analogy lies in the behaviour which form the initial conditions: you and I both say “I am in pain,” your behaviour here is analogous to mine. 167.As we have noted, this is the suggestion of Plantinga. 168.Cf. Moore, “The Nature and Reality of Objects of Perception,” p. 82ff. 169.Examination of Sir William Hamilton’s Philosophy, p. 205n. 170.Thus Ryle (The Concept of Mind, p 53) writes that “even if a person did enjoy a privileged illumination in the ascription of mental-conduct concepts to his own performances, his supposed analogical [sic] argument to the mental processes of others would be completely fallacious.” To establish this, he continues with the following analogy: “If someone has inspected a number of railway-signals and signal-boxes, he can then in a new case make a good probable inference from observed signalmovements to unobserved lever movements. But if he had examined only one signalbox and knew nothing about the standardization-methods of large corporations, his inference would be pitiably weak, for it would be a wide generalization based on a single instance.” 171.We now see where Ryle’s analogy goes wrong. It should read, “If someone has inspected a number of railway-signals and signal boxes on the LNER, he or she can then in the case of another railway, e.g., the LMS, make a good probable inference from observed signal-movements to unobserved lever-movements.” The person

373

Four: Disappearance of the Simple Self: Its Problems

examines a great many signals, etc., on one railway and, on observing similar things on another railway, he infers to the unobserved workings within the signal box. Or, in other words, it is not a generalization from one instance. The same point has been made in A. J. Ayer, The Problem of Knowledge, p. 219. 172.Mill, Examination of Sir William Hamilton’s Philosophy, p. 191. 173.Cf. Wittgenstein, Philosophical Investigations, part I, sec. 253: “In so far as it makes sense to say that my pain is the same as his, it is also possible for us both to have the same pain.” 174.Malcolm (“Knowledge of Other Minds,” p. 372f.) asserts that in such a case, unless the connection were logical there would be no “criterion” for its being the “same (kind of) pain” which I attribute to myself. But this is simply wrong. The pain behaviour and the tendency to such pain behaviour which you exemplify provides a criterion. To hold it is a criterion, it suffices to point out that the tendency is lawfully related to the conscious states for which it is a criterion. and not only is it lawfully related, but I have every reason to believe the law is true, and no reason to believe it is false. 175.It can be cogently argued that these confusions of a nominalistic nature are at the bottom of Ryle’s rejection of the existence of minds. See Addis, “Ryle’s Ontology of Mind.” 176.Cf. Moore, “The Nature and Reality of Objects of Perception.” 177.If I understand Malcolm, “The Privacy of Experience,” in this later paper on this topic, then he altered the views he had held when he wrote “Knowledge of Other Minds” and came to defend the same position the present study is defending. But perhaps I have understood neither of his papers. However, I am unable to make sense of either of them unless I read them in the ways indicated. 178.Cf. T. Penelhum, “Hume on Personal Identity,” and “Self-Identity and SelfRegard.” 179.J. S. Mill, An Examination of Sir William Hamilton’s Philosophy, p. 194. 180.For details, see Wilson, Psychological Analysis and the Philosophy of John Stuart Mill. 181.Ibid., Ch. 4. 182.Ibid. 183.T. H. Green, Prolegomena to Ethics.

374

Chapter Five Hume’s Positive Account of the Self (1) Mind and Body Contrary to the views of many philosophers, from Plato and Aristotle to Descartes, Hume argues “the true idea of the human mind, is to consider it as a system of different perceptions, or different existences, held together by the relation of cause and effect.” (T, p. 261)1 It is the causal relations that bind the parts of the bundle into a coherent whole that give us “a present concern for our past or future pains and pleasures.” (T, p. 261) This is the “bundle” account of the self; the self is a compound entity. He makes his point with a striking analogy. The mind is a kind of theatre, where several perceptions make their appearance; pass, re-pass, glide away, and mingle in an infinite varierty of postures and situations. There is properly no simplicity in at one time, nor identity in difference. (T, p. 253)

This he really takes to be, it would seem, the view of ordinary persons. Thus, in the Dialogues concerning Natural Religion, the view is attributed not only to Philo, who represents Hume, the thoughtful natural philosopher, but also to Cleanthes, who represents the liberal or broad church point of view, and Demea, who represents the dogmatic enthusiast. Thus, Demea points out that Our thought is fluctuating, uncertain, fleeting, successive, and compounded; and were we to remove these circumstances, we absolutely annihilate its essence ... 2

Hume later describes Cleanthes’ view in this way: What is the soul of man? A composition of various faculties, passions, sentiments; united, indeed, into one person, but still different from each other. When it reasons, the ideas, which are parts of his discourse, arrange themselves in a certain order; which is not preserved entire for a moment, but immediately

Five: Hume’s Positive Account

gives place to another arrangement. New opinions, new passions, new affections, new feelings arise, which continually diversify the mental scene, and produce in it the greatest variety, and most rapid succession imaginable.3

Cleanthes does not demur from this description of his opinion, and later adds that A mind, whose acts and sentiments and ideas are not distinct and successive; one, that is wholly simple, and totally immutable; is a mind that has no thought, no reason, no will, no sentiment, no love, no hatred; or,, in a word, is no mind at all.4

Nor does Philo disagree, though he has the pleasure of pointing out that “You are honouring with the appellation of atheist all the sound, orthodox divines almost, who have treated of this subject ...”5 So Hume takes the bundle view to be commonsense, and the substantialist alternative is merely the dogmatic opinion of “some metaphysicians.” (T, p. 252) Some have argued that Hume, in barely stating the bundle view, had done enough: the bundle account is commonsense, and his worries about identity and identity in difference are misplaced. Thus, for example, G. Vesey has suggested that once the standard of substantial identity is abandoned, there is no remaining problem with personal identity. Hume generates for himself the so-called problem of personal identity, on Vesey’s view, because, while he rejects the substantialist account of mind, he nonetheless insists upon substantial identity as the standard that must be met. The correct move, Vesey holds, is not only to reject substances but to reject substance as the standard criterion of identity. Once we do that then no problem remains. To say that there is nothing essential about personal identity, or that the self is not a substance, is to suggest that there is something essential about other things, or that other things are substances. To say that there is no just standard by which we can decide any dispute about identity is, surely, wrong. We can decide ordinary disputes – in the way in which we do.6

Hume would not really disagree with this. He too argues, as we have seen, that the problem with much philosophy has been the substance tradition, and that once substances are abandoned many of the problems disappear. Hume would also like to think along with Vesey that our ordinary concept of personal identity is all in order. No doubt, however, he would recognize what Vesey ignores, that much of our ordinary discourse is infected by the

376

Body, Mind and Self in Hume’s Critical Realism

Christian tradition with its view of the self as in effect a substance. But even if we agree that we can get along quite well in ordinary circumstances with our ordinary concept of a person, that is, the bundle concept, it does not follow that all the interesting questions about personal identity have been answered. For it remains to spell out what are those ordinary procedures for settling ordinary disputes, and to spell out what are the criteria that we ordinarily use in contexts where personal identity is an issue. In Hume's terms, we have still to elaborate an account of the origin of this idea of personal identity that we use. And in still more specifically Humean terms, we shall need an associationist account of the origins of this idea. Vesey stops just at the point where a serious philosopher ought not to stop, just at that point where the negative task has ended and a positive task must begin. He stops just at that point where a positive account of personal identity needs to be developed. “What we call a mind,” Hume argues, “is nothing but a heap or collection of different perceptions, united together by certain relations ....” But we also attribute an identity to the mind: it is a heap or collection which is also “suppos’d, tho’ falsely, to be endow’d with a perfect simplicity and identity.” (T, p. 207) Such a succession “answers evidently to our notion of diversity, [but] it can only be by mistake we ascribe to it an identity; and as the relation of parts, which leads us into this mistake, is really nothing but a quality, which produces an association of ideas, and an easy transition of the imagination from one to another, it can only be from the resemblance, which this act of the mind bears to that, by which we contemplate one continu’d object, that the error arises.” (T, p. 255) The point here is not, we should recognize once again, the error of supposing that in the bundle which the mind is, there is a simple continuant which endures one and the same through change. It is rather to draw attention to the point that the parts of the bundle form a connected whole by virtue of a “quality, which produces an association of ideas.” It is this set of associated ideas that constitutes our complex idea of the self and of personal identity. The issue is, what precisely is this quality that produces this association? To answer this question, at least to sketch an answer, is to give Hume’s positive view of the self. Vesey is simply wrong to suppose that, having disposed of the substantial self, Hume has completed his task, and need not address this further question. It is, of course, to this question that we are now turning. To retrieve Hume’s positive account we must proceed to Books II and III of the Treatise. Hume does not devote a further section specifically

377

Five: Hume’s Positive Account

to the topic. But he does say more than enough about the idea of the self that he insists we do have for us to be able to discern an adequate Humean theory of the “quality” or relation among the parts of the self that accounts for the association of ideas of those parts into one's idea of oneself. Locke was sufficiently part of the Christian tradition that, while he gave up the notion of a substantial self, he nonetheless still had to allow for future rewards and punishments in the hereafter. When he came to analyse the concept of a person, he had to leave a place for this. The concept of a person could not, therefore, have as part of itself the concept of a man. The latter was a material object, and Locke had to allow the possibility of the separation of the person from the man. This led to his memory criterion for being a person, with the defects that Molyneaux pointed out. Hume is not so bound to the Christian tradition, and so, as we shall see, he is prepared to allow that our body enters into the idea of our self. We discover through experience a distinction between our own body and those of others, and this identity of body enters into the identity of self. Since the body is relatively stable, that will in turn give a certain stability to the self. This is part of Hume’s answer to critics of Locke like Butler and Clarke, who held that all stability disappeared when the substantial self was abandoned. But this is not all that Hume has to say. There is also the issue of the sufficient stability of personality for the practical purposes we have in this life. Here too Hume attempts to show how the distinction between our own self and those of others arises within the world of experience, and more specifically how the stable character of the responsible person is developed through interaction with others, in the family and in devil society. And of course, when Hume develops these aspects of the issue of personal identity, he is pursuing lines of thought first introduced by Locke when he insisted that the concept of a person is a forensic notion. Hume tells us that “the idea, or rather impression, of ourselves is always intimately present with us, and our consciousness gives us so lively a conception of our own person that ’tis not possible to imagine that anything can in this particular go beyond it” (T, p. 317). The first thing to note is that this idea or impression of oneself is present in our consciousness; “ourself is always intimately present to us” (T, p. 320; italics added). In other words, our conscious states each contain this idea or impression of oneself. Hume similarly speaks elsewhere of the “self or that individual person of whose actions and sentiments each of us is intimately conscious” (T, p. 286). Our conscious states thus include not only the idea or impression of oneself but also one’s sentiments and impressions of

378

Body, Mind and Self in Hume’s Critical Realism

one’s actions. But these ideas, impressions and, among the latter, sentiments that are in one’s conscious state are “intimately present” to one. This implies a distinction between those contents and the awareness of them. When I take pride in something I have done, there are on the one hand the ideas and impressions that are “intimately present” to me, and, on the other hand, the “me” to which they are “intimately present.” We therefore have the model of what it is to be a conscious state that we have earlier proposed. There are, first, certain contents. Second, these contents are the contents of a conscious state by virtue of being the objects of an awareness.7 Two further points should be noted. First, among the contents of the conscious state will be various non-intentional entities, e.g., sensations, impressions, images, etc., and including feelings of pleasure and pain, but there will often also be various mental acts, perceptual (e.g., seeing), cognitive (e.g., believing, doubting), conative (e.g., willing, desiring, morally disapproving), and affective (e.g., loving, taking pride in). Thus, in the intention of the awareness that constitutes the conscious state will be other awarenesses, other intentional states. The point to be made is that there is absolutely no contradiction in saying that there are two mental acts and that the one act is the object of the other. Nor, as we have also emphasized earlier, does the fact that we are conscious of the consciousness of A require that we be in turn conscious of the consciousness of the consciousness of A. There is of course nothing that prevents one, in another shift of attention, from becoming conscious of the awareness of the perceiving. Capaldi has argued that Hume’s point about the unobservability of the self is that this centre of consciousness is not itself among the contents of the conscious state that it constitutes.8 He infers from this that it “can never be discursively characterized” (ibid.). But this does not follow: Capaldi gives us no reason for supposing that the awareness that is the centre of consciousness cannot itself become the object of an act of awareness and thereby become capable of discursive characterization. Moreover, neither does it follow that if we are to talk about the self as the centre and unity of consciousness, as Capaldi rightly insists that we must (p. 280), then we must abandon the discursive language of science in favour of some other (non-rational) mode of discourse in which the self is revealed to us (ibid.).9 In other words, we can avoid, as we reasonably want to do, the inference from the fact that there is a unifying centre of consciousness to the conclusion that this centre, the self, somehow lies outside the web of

379

Five: Hume’s Positive Account

that which can be explained scientifically. This conclusion is avoided by noticing that the act of awareness that is the centre of consciousness, while it is not perceived when in one’s present conscious state, can itself become the object of a further act of awareness. It is no longer the now-beforeconsciousness act of awareness which is the centre of consciousness, but it is the act of awareness that was the centre of consciousness, the very same act. These facts, that it is knowable, and at times known, enables one to place this act, the centre of consciousness, among the facts and events that are to be explained scientifically. But if the act of awareness that constitutes one’s conscious state can become the object of another act of awareness, it is also true that there is nothing that requires this further act of awareness. Ryle once argued to the contrary.10 However, Russell was surely correct in holding that consciousness of something is an event of which we may become conscious and that there is no infinite regress implied in this.11 And so too is Mill correct in his claim, also made by Sir William Hamilton, that, so long as one looks upon awareness as an event among events rather than as a simple immaterial substance then there is nothing impossible in holding that one can consciously attend to consciousness. As Descartes discovered in the cogito, we do sometimes shift our attention away from the external objects of perception to consciousness, thought itself. Mill, too, made the point that the issue whether one can attend to processes in which one is attending to other things is an empirical matter: is it in fact possible to do this? Comte had suggested one cannot, but, Mill asserts, “...M. Comte might be referred to experience...for proof that the mind can not only be conscious of, but attend to, more than one, and even a considerable number, of impressions at once.”12 Mill also refers to the writings of Cardaillac and Sir William Hamilton.13 The issue, as Hamilton sees it, is the simple empirical one of whether one can attend to only one thing at a time or several; and if several, then there is no reason in principle, whatever practical difficulties there are, to suppose that the mind cannot attend to its own states. But of course we can be conscious of several things at once – up to six suggests Hamilton –, and so the issue of principle is settled,14 and Comte is wrong. Cardaillac makes a similar point. He agrees with Comte that to attend to our mental acts and processes is difficult: “...lorsque nous nous observons nous-mêmes, l’action des facultés se devise; une portion se porte sur la manière dont le reste s’emploie, nous nous regardons regarder .... Cette division de nos facultés en affablit l’énergie, et l'observation devient plus difficile; nous regardons

380

Body, Mind and Self in Hume’s Critical Realism

mal ce que nous nous regardons regarder.”15 But not impossible. In fact, he argues, we do it regularly, whenever we make relational judgements, since we cannot be conscious of a relation between two things unless we are attending to both of them. ...pour connaître les rapports et rendre plus distinct le sentiment que nous en éprouvons, il faut nous porter simultanément sur les deux objets entre lesquels existent les rapports, et par use double attention, rendre plus distinctes les deux sensations, afin que le sentiment-rapport, qui résult de leur simultanéité, devienne à son tour plus distinct (vol. II, p. 79).

It is thus possible to attend simultaneously to two things; in fact, “nous faisons continuellement cette opération” (ibid.). Thus, for Hamilton and Cardaillac, and, following them, Mill, the issue is not whether attending to one’s inner states can be done – as a matter of experience it can! – but whether we can develop an ontology which can accommodate this fact.16 Second, given Mill’s point about awareness, that a simple act can represent, or have in its intention, a complex bundle of events, including other awarenesses, we can treat the awareness by virtue of which various entities are among the contents of our conscious state itself as a simple mental act. Every conscious state thus exists as conscious by virtue of an act of awareness which intends the contents of the conscious state. Clearly, this awareness that constitutes the conscious state will be the “centre of consciousness” to which so many philosophers, from Plotinus through Descartes to Shalom, have directed our attention. The crucial point is that, while the Humean can agree that there is always such a centre of consciousness, he or she does not have to accept the Plotinian inference that this centre is a simple substance. To the contrary, the Humean will argue that this awareness is a fact among the facts, a part among the parts, in the “bundle” that makes the self. Consider, for example, the mental state that one expresses when one uses a sentence like (+) I feel proud that I then did so-and-so. in the normal way. The conscious state that is expressed when (+) is used in a speech act in the normal way has among its contents the feeling of pride. This is the sort of thing that Hume would, of course, refer to as an impression. This is, in Hume’s terms, accompanied by an idea, which means, in our way of speaking, this feeling of pride has a certain object, which is, in our example (+), my past action. This past action becomes the

381

Five: Hume’s Positive Account

object of my feeling of pride by virtue of the latter being tied to a certain idea, namely the idea of my past action. This idea, though of course not the action itself, is also among the contents of my conscious state; it too is “intimately present” to me. (+), as a sentence, describes the very same conscious state that is described by (#) David Hume [now] feels proud that he then did so-and-so. But a speech act in which (#) is used need not express a conscious state of David Hume; its use may express, for example, a conscious state of James Boswell, namely, a conscious state of Boswell’s that has as its content the belief that the state of affairs described by (#) actually exists. Both the first ‘I’ in (+) and ‘David Hume’ in (#) refer to, or designate, the same person; that is implied when one says that (+) and (#) both describe the same state of affairs. The difference lies in the fact that (+) is so used that it expresses the conscious state of the person to whom the subject term refers while (#) does not. This means that in the characteristic use of the word ‘I’ as a subject term not only does it refer to the person speaking but it also expresses the awareness that is constitutive of the conscious state of that person. In this sense, although the referent is the same, the linguistic roles of the two terms differ, and, in that sense, so does their meaning.17 This is the point that Wittgenstein was getting at when he remarked that To say ‘I have pain’ is no more a statement about a particular person than moaning is. ‘But surely the word “I” in the mouth of a man refers to the man who says it; it points to himself; and very often a man who says it actually points to himself with his finger.’ But it was quite superfluous to point to himself. He might just as well only have raised his hand. It would be wrong to say that when someone points to the sun with his hand, he is pointing both to the sun and himself because it is he who points; on the other hand, he may by pointing attract attention both to the sun and to himself. The word ‘I’ does not mean the same as ‘L.W.’ even if I am L.W., nor does it mean the same as the expression ‘the person who is now speaking’. But that doesn't mean: that ‘L.W.’ and ‘I’ mean different things. All it means is that these words are different instruments in our language.18

What the Humean holds is that, when we thus use the word ‘I’ to express the awareness that constitutes the conscious state, that does not imply that this awareness must be an act of a simple substantial self. This fallacious inference has been made by several writers. Thus, Chisholm,19 commenting on Hume, asserts that “He [Hume] said, referring to himself,

382

Body, Mind and Self in Hume’s Critical Realism

that he found nothing but impressions” (p. 11). What Hume found...was not merely the particular perceptions, but also the fact that he found those perceptions as well as the fact that he failed to find certain other things. And these are findings with respect to himself (pp. 11-12).

From this Chisholm concludes that the traditional substance account of the self is true. We are therefore acquainted, he claims, with the self (p. 12, p. 21). This self is not the same as any one of the multiplicity of states or attributes that it exemplifies (p. 20); it is in itself without parts, a simple. It moreover is the same in all states of consciousness: Hume’s notion is “that the self is a substance persisting through time” and, according to Chisholm but contrary to Hume, “one may derive the idea of such a self from any impression whatever” (p. 12). But Chisholm’s conclusion does not follow. It is true that, when Hume found no impression of the self as a simple substance, he expressed this conscious state using a sentence beginning with ‘I’. It is also true that the awareness thus expressed was not among the contents of the conscious state. But, as Mill emphasized, there is no reason to suppose that this awareness that constitutes the conscious state is not one among the many parts that go to make up the bundle that is the self. Chisholm simply makes, without any justification at all, the leap from his correct point about ‘I’ to a human soul. Shalom offers us rather more.20 He, too, criticizes Hume, telling us of the “inconsistency of Hume’s ‘infamous phrase’ in which he asserted that he was unable to find in himself anything corresponding to ‘himself,’ thereby denying what he was, in fact, asserting by virtue of the very sentence in which he was denying it” (p. 414). But all Hume was doing was expressing his conscious state. It is, moreover, true that this is “an ‘I’ which continues to think” (p. 412). But what does not follow is that the awareness that constitutes what the use of ‘I’ at one moment expresses is literally the same as the awareness that constitutes what the use of ‘I’ at a later moment expresses. No one disagrees that they are in some sense identical – not even Hume. The point is that this identity is not, contrary to what Chisholm and Shalom think, that of a simple substance. Shalom improves upon Chisholm, however, and shows himself thereby to be more sensitive to the philosophical problems, by providing a reason for thinking that the awareness that constitutes the conscious state is a simple substance. For, as we have seen (Chapter One), he holds with Plotinus that there can be no intelligible account of the unity of the self unless we take for granted that the self as a structured entity is created by the activity of a

383

Five: Hume’s Positive Account

simple soul or transcendental substance, or, as Shalom prefers, subjectivity. We have by now disposed of this Plotinian argument in principle. It remains, however, to try to specify in detail those relations that constitute the unity of the self. Not surprisingly, the account of the identity of bodies yields a starting point. Begin with the commonplace difference between an action and a mere movement: the former is, while the latter is not, the result of a conscious intention. This means that the sentences (+) and (#) both ascribe to David Hume a certain conscious state that has preceded the one that has the feeling of pride among its contents, namely, the conscious state of intending or willing to do so and so. Call this earlier conscious state a. But when (+) and (#) ascribe a feeling of pride to David Hume, they ascribe a conscious state that is later than a. Call this later conscious state d. There are, let us assume, other conscious states that intervene between a and d; suppose that these states are b and c. There is a certain “quality” or relation that connects these conscious states a, b, c, d, uniting them into the states of a single self. Call this relation R.21 (+) and (#) ascribe these two conscious states to the same person, David Hume; they identify them as states of the same person. The pattern here is the same as we had in the case of body: we can reasonably construe d is the same person as a as d = (,x)(R3ax) because we also have both b = (,x)(Rax) and c = (,x)(R2ax) Hume concludes from the presence of the idea of a running throughout these that our idea of a person as an enduring entity contains the idea of a simple entity that continues throughout the series. But the Principle of Acquaintance excludes any such entity, that is, any simple entity. The idea of that which unites the bundle into a person is therefore, according to Hume, “ficticious.” Thus, “what we call a mind,” Hume suggests, “is nothing but a heap or collection of different perceptions, united together by certain relations, and supposed, tho’ falsely, to be endowed with a perfect simplicity and identity” (T, p. 207). The pattern here is the same as in the case of body, and we may also conclude here that we can accept the main thrust of Hume’s position without accepting that he needs to include in the idea of the self any “ficticious” simple particular that is supposed to endure

384

Body, Mind and Self in Hume’s Critical Realism

throughout the series. For what we are about now, the question with which we are concerned is simply this: what is the relation R? Now, Hume speaks of “that connected succession of perceptions which we call self” (T, p. 277). These perceptions include the perceptions of our body. Indeed, they are part of the definition of oneself, according to Hume, for he speaks of “the qualities of our mind and body, that is, self,” and elsewhere of the “self or that individual person of whose actions and sentiments each of us is intimately conscious” (T, p. 303, p. 286; italics added). In experience we distinguish a variety of bodies, including the bodies of persons. But among the bodies of persons that I can distinguish within experience there is one to which I have a unique relation; this is my body. As we have seen John Stuart Mill put it, “I am aware, by experience, of a group of Permanent Possibilities of Sensation which I call my body, and which my experience shows to be an universal condition of every part of my thread of consciousness. I am also aware of a great number of other groups, resembling the one that I call my body, but which have no connexion, such as that has, with the remainder of my thread of consciousness.”22 Others have made the same point about our standing in a special relation to a unique body. Thus, A. J. Ayer is “inclined to think that a person’s ownership of states of consciousness consists in their standing in a special causal relation to the body by which he is identified.”23 But while Ayer is undoubtedly correct in agreeing with Mill about the role of a unique body, it is not sufficient to characterize the relation as causal. The “special causal relation” that Ayer attempts to invoke must, he says, be such that “any individual experience is limited to one and only one human body” (p. 118). But it cannot be the relation of simple causal necessity, since “every one of my experiences is dependent upon the existence of bodies other than my own. This follows simply from the fact that I must have had ancestors” (p. 119). Ayer attempts to answer this difficulty by distinguishing “mediate” and “immediate” causal necessity. Ayer’s ancestors’ bodies are only mediately necessary for the occurrence of his conscious states. He argues that only the “internal states” of bodies are sufficiently immediate as causes to do the job; “the most that we can hope to maintain is that an experience belongs to a given person in virtue of the fact that some state of that person’s body is a necessary condition of its occurrence” (p. 119). Thus, “certain experiences would be mine in virtue of the fact that such things as the condition of my brain and nerves were

385

Five: Hume’s Positive Account

immediately necessary for their occurrence, and that they did not stand in precisely this relation to any other body but my own” (p. 123). This by itself will not do, however. For it is true of every one of the contents of our sense experience that it depends for its existence upon the state of our sense organs. Hume made the case in this way, as we saw: When we press one eye with a finger, we immediately perceive all the objects to become double, and one half of them to be remov’d from their common and natural position. But as we do not attribute a continu’d existence to both these perceptions, and as they are both of the same nature, we clearly perceive, that all our perceptions are dependent on our organs, and the disposition of our nerves and animal spirits (T, pp. 210-11).

This is not to deny that there are objects that are independent of our bodily states. The argument, to the contrary, aims, as we have seen, to establish what has been called critical realism, which is the thesis that, while the world of sense experience all depends for its existence upon the state of our body, nonetheless there also exists the world of objects independent of us that science attempts to describe and which cause the changes in our sense organs that in turn cause us to have the sense experiences that we have. The point is that this argument (if successful, as Hume thinks it is) establishes that everything that is given to us in sense experience must, upon Ayer’s criterion, be described as “mine.” And within the critical realist position, there may be a real point to saying that.24 John Stuart Mill’s point, however, is that within experience we draw the distinction between my body and the bodies of other persons. The distinction between self and non-self is one that is made within the world of sense experience; it is not a distinction between the world of sense experience and that which is external to that world. Ayer’s approach, therefore, will not do. Another suggestion that also introduces a connection to one’s body has been made by Ian Gallie and J. R. Jones.25 People have visual experiences: things come into, and go out of, their field of vision. They also have bodily experiences of various kinds, some short-lived, but others that persist over time. At any time when we experience something as it comes into our field of vision, we will also be aware of our bodily state. The visual field and a “somatic field” will go hand in hand. If someone says that (@) I am seeing a blue object there is an “I” of which the seeing is predicated and to which the blue object is related. This “I,” upon the proposal of Gallie and Jones, is the

386

Body, Mind and Self in Hume’s Critical Realism

somatic field, that is, the bodily state of the person making the assertion, insofar as he or she is aware of it. As Jones puts it, “The somatic field in question is the subjective referent to which the visual datum is related when its occurrence is reported in the sentence ‘I am seeing such and such a coloured patch’” (p. 60). However, this cannot be the whole story of the “I,” that is, the self, since it omits the fact that the self is a centre of consciousness. The self is defined in the first instance by a conscious state, not by a somatic field. This is the point of Plotinus, and it was the point that Mill was able to account for when he made awareness constitutive of a conscious state a simple entity, where simple is here used to contrast to bundle. As we put it above, the ‘I’ refers to a person, and when sentences like (+) and (@) are used, the ‘I’ expresses the awareness that is constitutive of the conscious state. At the same time, however, the point is surely correct that among the contents of every conscious state there is an awareness of one’s own bodily state. This is the point that we ourselves emphasized in the previous chapter. It is the point that Mill was getting at when he said that “experience shows” my body “to be a universal condition of every part of my thread of consciousness.” We are often have as parts of the contents of our conscious states awarenesses of the bodies of others, but an awareness of our own body is among the contents of all our conscious states. But this awareness of our own body is unique. As we have argued earlier, all other bodies we experience from the outside; our own body we experience from the inside. When I see something, I experience part of my own body, the eyebrows, the bridge of my nose, and so on, in a way that I never experience the same part of another person. Properties like pains and pleasures are located at places within my body – thus, my toothache is located in my tooth, the pain from the cut in my big toe is there in the big toe; these properties of parts of my body located within my skin are experienced by no one else but me. Conversely, there is one and only one body such that I experience properties like these that are located at places inside the body. Again, I can move my own body simply by willing it; this I can do with no other body. Talk of a “somatic field” is correct enough as far as it goes, for we certainly are aware – from the inside – of the state of our body. But that is only a small part of the story. We are also aware of our own body in a special way in perception, in action, and in our emotions. As we have argued, it is a fact of experience that we are aware of one and only one body from the inside and that among the contents of

387

Five: Hume’s Positive Account

every conscious state is an awareness of this body from the inside. As we have also argued, we can use this special sort of experience to define the symmetrical relation of “belonging to,” that is, to define what it is for a conscious state to belong to a body and what it is for a body to belong to a conscious state. A conscious state belongs to a body just in case that among the contents of the conscious state is an awareness of that body from the inside; conversely, a body belongs to a conscious state just in case that among the contents of the conscious state is an awareness of that body from the inside. Thus, if I assert that (+) I feel proud that I then did so-and-so then the first ‘I’ expresses an awareness that is constitutive of a conscious state. That conscious state has among its contents an awareness from the inside of a particular body. That body belongs to me, that is, the “I”; that body is mine. But (+) also describes an earlier conscious state. This too is mine. Why? Because among the contents of that conscious state was an awareness from the inside of the very same body that I am now aware of from the inside when I assert (+). That earlier conscious state belonged to the very same body that is (now) mine. The earlier conscious state and the later conscious state are related to one another through the fact that they belong to the same unique body. As we put it above, the earlier conscious state was a and the later one d, and the two are the same because d = (,x)(R3ax). We phrased the issue of personal identity as the issue of the relation R: precisely what relation is it? The Humean answer that we have given is that R, or more accurately the relative product R3, that is, (R|R)|R, is the relation that a bears to d just in case that a belongs at an earlier time to the very same body to which d belongs to at a later time. If we assume that we have a non-problematic account of the identity of bodies through time (which is a safe assumption), it follows that we now have a non-problematic account of the identity of persons through time. And once we have personal identity, the ordinary practice that we noted above of individuating mental states – believings, passions, etc. – and actions by predicating them of persons can be sustained. G. E. M. Anscombe has made a point,26 similar to that of Wittgenstein, noted above,27 that ‘I’ and the proper name referring to the person who uses the ‘I’ have different uses. It is her claim, which goes much beyond Wittgenstein’s, that ‘I,’ unlike the proper name, is used to express the fact of our being in a conscious state but that, unlike the proper

388

Body, Mind and Self in Hume’s Critical Realism

name, is not a referring expression. This is paradoxical, as she herself admits, since she grants (p. 64) that the rule for truth for statements with ‘I’ as subject is If X makes assertions with ‘I’ as subject, then those assertions will be true if and only if the predicates thus used assertively are true of X. She draws her conclusion from a special feature of the use of the word ‘I’ which, in her view, can be explained only if one assumes either a Cartesian substantial self – which has intolerable problems and must be rejected –, or that ‘I’ is not a referring expression. But, as I – F.W. – shall now (consciously) suggest, if one recognizes the special way in which conscious states belong to a body, then there is third alternative, and so, like Wittgenstein, we need not accept her paradoxical conclusion. Anscombe draws her conclusion from an important feature of the use of the word ‘I’. But her starting point is a contrast between the use of ‘I’ and what would be the use of a proper name ‘A’ which everyone used of him or herself (a human being, a person with a body), though each person also had another name which others used of him or her.28 Anscombe suggests that the contrast she intends is hinted at in the term ‘selfconsciousness,’ which “is something real...which ‘I’-users have and which would be lacking to ‘A’-users, if their use of ‘A’ was an adequate tool for their consciousness of themselves” (p. 51). The question for her is whether this “something real,” our self-consciousness, which the use of ‘I’ somehow gets at but which an ordinary proper name does not, is referred to by ‘I.’ The use of ‘I’ requires the presence of self-consciousness, and therefore if ‘I’ is in fact a referring term, then it is “...secure against reference-failure. Just thinking ‘I ...’ guarantees, not only the existence but the presence of its referent” (p. 55). She concentrates on the use of ‘I’ in such thoughts as “I am sitting,” or, as she puts it, “examples of reflective consciousness of states, actions, motions, etc....of this body. These Ithoughts...are unmediated conceptions (knowledge or belief, true or false) of states, motions etc. of this object here” (p. 62). The “conception” through which ‘I,’ if it is a referring term, either a name or a demonstrative, attaches to its object can only be “the thinking of the I-thought, which secures this guarantee against reference-failure” (ibid.). In contrast, the proper name ‘A,’ also a referring term, while it would in use also be guaranteed a referent, would not be guaranteed to have the right referent, since what I take to be part of my own body may not be so. As a consequence, “if ‘I’ is a referring expression, then Descartes was right

389

Five: Hume’s Positive Account

about what the referent was” (p. 58). Its referent can be nothing but a (stretch of) Cartesian Ego. But Descartes’ position “has...the intolerable difficulty of requiring an identification of the same referent in different Ithoughts” (p. 58). But this is impossible: “How do I know that ‘I’ is not ten thinkers thinking in unison” (p. 58). This she takes to be a reductio ad absurdum of the notion that ‘I’ is a referring expression. Getting hold of the wrong object is excluded, and that makes us think that getting hold of the right object is guaranteed. But the reason is that there is no getting hold of an object at all. With names, or denoting expressions (in Russell’s sense) there are two things to grasp: the kind of use, and what to apply them to from time to time. With ‘I’ there is only the use (p. 59).

It follows from this, on her view, that “No problem of the continuity or reidentification of ‘the I’ can arise” (p. 62); it is merely that the same human being has different I-thoughts at different times. A “lapse of selfconsciousness” is just a lack of “unmediated agent-or-patient conceptions of actions, happenings, and states.” But these conceptions, Anscombe maintains, “do not involve the connection of what is understood by a predicate with a distinctly conceived subject”; the notion that there is such a subject is nothing more than a “(deeply rooted) grammatical illusion” (p. 65). It is no doubt true that, from the fact that the use of ‘I’ expresses the fact of self-consciousness, it does not follow that the ‘I’ refers to a simple mental substance: Anscombe is quite correct, there is nothing about the use of ‘I’ that requires this. However, it does not follow that there is no problem of personal identity. From the fact that ‘I’ does not refer to a simple self, it does not follow that it does not refer. Anscombe admits that children sometimes use their own names instead of ‘I,’ and further that this quite conceivably could be the standard practice. But, she claims, what is semantically a name or referring expression when other people use it will not be such when one uses it oneself: “it will not signify like a name” in one’s own utterances (p. 64). But from the fact that it is used to express the awareness that constitutes our conscious state, it does not follow that it is not also a referring expression, referring to the entity, a person or human being endowed with self-consciousness, that is the same entity that others refer to using one’s name or a second- or third-person pronoun. But this entity is, of course, complex. Which is the problem: the task is to unravel the sort of complexity which it has. For Anscombe, however, no reference other than the Cartesian one could guarantee against reference

390

Body, Mind and Self in Hume’s Critical Realism

failure. Once there is complexity, then reference failure is possible, and there is nothing in the use of ‘I’ that guarantees that the supposed referent must be simple. So the Cartesian position does not follow from the uniqueness of the use of ‘I.’ Now, concerning reference failure, there are two kinds, the lack of a referent and the risk of wrong reference. The second form of reference failure involves the risk that what I take to be me might not be me after all: I could have glombed onto the wrong I. This is how ‘I’ differs from ‘A’, the hypothetical second name that every human being uses of himself. Alas, Ms. Anscombe is only human: she does not consider alternatives other than those that favour her argument. Specifically, she does not consider that there could be other ways to guarantee against reference failure, even in the case that the reference is being made to a complex entity. But, in the first place, the rule for the use of ‘I’ is fairly simple: each item that is in the conscious state of a person at a time is to be ascribed in an I-statement to the person at that time. Since conscious states as a matter of fact always belong to one body that defines the person, this rule guarantees that whatever I take in this way to be I, that is, whatever body viewed from the inside is in fact among the contents of the conscious state, is I. And even where I am deluded, when, for example, I assert that I am walking, or talking to Wittgenstein, when I really am not, there will still be a successful reference, to the human being about whom I am making the false assertion. And second, neither is there any risk of wrong reference, not because there is some sort of incorrigible intuition, but simply because as a matter of fact only one body viewed from the inside is among the contents of a conscious state. So the facts about the way in which conscious states belong to bodies, together with the rule for the use of ‘I,’ guarantee that reference will not fail. But they do so in a way that does not commit one to the substantial self of the Cartesian. At the same time, one does not have Anscombe’s all-too-easy way of avoiding the problem of personal identity: since ‘I’ does refer, one can ask what (complex) entity it is that is referred to, what is its nature and what individuates it. If I am shown a leaf from a tree and asked what sort it is, then I will look at it, examine it in detail, and report, say, that it is the leaf of a sugar maple. But if I assert a statement like (+) I feel proud that I then did so-and-so. then it is unlikely indeed that this assertion is the upshot of an examination of the state of affairs that it describes; rather, the assertion simply

391

Five: Hume’s Positive Account

expresses the conscious state that it describes. In this sense, Strawson29 is correct when he insists that “When a man (a subject of experience) ascribes a current or directly remembered state of consciousness to himself, no use whatever of any criteria of personal identity is required to justify his use of the pronoun ‘I’ to refer to the subject of that experience” (pp. 164-5). Yet he is correct, we have been arguing, in also holding that when ‘I’ is thus used in “criterionless self-ascription,” it does not “lose its role of referring to a subject.” This is because it issues from the mouth of a person who can be identified empirically, or, if used in soliloquy, is used by a person who would acknowledge that he or she satisfied the criteria employed in determining that he or she was identical with the person who performed some previous act. ‘I’ can be used without criteria of subject-identity and yet refer to a subject because, even in such a use, the links with those criteria are not in practice severed (p. 165).

Normally, one asserts an I-statement not as a consequence of one’s having gone through a process of identifying the I, but for all that in such a use the I to which reference is made is identified with a self that has preceded and will endure longer. In this sense, “our ordinary concept of personal identity does carry with it empirically applicable criteria for the numerical identity through time of a subject of experiences (a man [sic] or human being)...which involve an essential reference to the human body” (p. 164). What Strawson does not do, and what is essential, is to give an account of the relation by virtue of which a conscious state belongs to the unique body that provides the criterion of identity for the person. This we have tried to do, following John Stuart Mill, as we try to explicate Hume’s account of personal identity. This account of personal identity depends upon the contingent fact that throughout one’s history one is aware of one and only one body from the inside. But given this fact, then it follows that a conscious state cannot belong to two persons at the same time. Don Garrett30 has suggested that Hume’s doubts about self identity arise from the fact that his discussion has the strange consequence that “whenever two perceivers have the same experience – whether impression of sensation, passion, or idea – at the same time, they are in fact literally sharing the same perception” (p. 354). I am not so certain that Hume saw this particular problem; his concerns in the “Appendix” are rather different, if our discussion above is correct. Nonetheless, conscious states are individuated by reference to persons, and

392

Body, Mind and Self in Hume’s Critical Realism

so unless we can give an adequate account of personal identity we will in fact have problems of the sort that Garrett proposes. The point is that once we recognize the special role played by the awareness of our body from the inside, then, as we saw in the preceding chapter, Garrett’s problem does not arise. Hume does recognize that there is no self apart from the body, as when he tells us about “the qualities of our mind and body, that is, self” (T, p. 303). In somewhat similar fashion, Barry Stroud argues that Hume’s puzzlement arises because he cannot, upon his empiricist and associationist principles, explain why perceptions should not come in more than one causally-discrete bundle, each with its own interrelations of causation and resemblance. It is therefore clear that Hume’s explanation of the origin of the idea of the self or mind is not necessarily deficient in failing to give an account of how a certain idea arises from certain ‘data’, but that it leaves completely unintelligible and mysterious the fact that those ‘data’ are as they are...[W]e find it simply taken as a given fact about the universe of perceptions that the range of reflective vision of any one of them does not extend to all the rest....What accounts for the fact that one cannot survey in the same way all the perceptions there are?31

What I am aware of is those conscious states that belong to me; I am not aware of those that belong to others. Stroud is suggesting that Hume’s puzzlement arises because he cannot explain why I am not aware of the conscious states that belong to others. We have just shown in the preceding chapter, however, that a good Humean explanation can after all be given for that fact: it turns upon the lawful connection between my conscious states that is established by the special relation that they all bear to a unique body. Both Garrett and Stroud are attempting to find the problem with which Hume is concerned in places where it does not really exist. The problem that Hume is concerned with in the “Appendix” to the Treatise is not so much the nature of the “quality” of our perceptions that generates the awareness of personal identity but rather the Plotinian problem of how a part of a series could be aware of the whole. Because they do not attempt to situate Hume in the philosophical tradition, Garrett and Stroud both fail to recognize what most deeply concerns him. John Stuart Mill understood the issues more accurately. But the present point is the non-problematic account of personal

393

Five: Hume’s Positive Account

identity that we have just adumbrated. We have stated that account in nonHumean terms, simply in terms of a relation that we called “R”; it can easily be re-stated in Hume’s own language, using his way of speaking about relations, as entities at once philosophical and natural. Things that stand in the relation R have a certain “quality” in common, that of belonging to one and the same unique body. This relation creates an association among the ideas of things that stand in that relation. That is, if X’s and Y’s as X’s and Y’s stand in the relation R, then an idea of an X will introduce the idea of a Y and the impression of an X will introduce the idea of a Y, together in each case with the feeling of unity. But a and d have the quality of belonging to the same body, and therefore stand in relation R to one another. Hence, the impression of that body which is among the contents of conscious state d will introduce the idea of the conscious state a together with the feeling that a and d are a unity. As Hume would think about these things, there are three entities present. There is, first, the present impression of the body which picks out the person. Second, there is the idea of the past states of that person as picked by that body. Third, there is the feeling that the former states, as presented by the idea, are part of a unity that includes the present conscious state which belongs to the same body. Hume makes these several connections when he speaks of “the idea, or rather impression, of ourselves [which] is always intimately present with us...” (T, p. 317). We can restate some at least of these points within the framework of Hume’s associationist psychology. The conscious state d, the final state of our wee series “a, b, c, d,” will contain an impression of the body to which it belongs. By virtue of established associations, the presence of this “quality” in consciousness will introduce ideas of all the other states that constitute the self. But this notion that ideas of all these antecedent states are contained within the present state is clearly implausible. What one should say – as John Stuart Mill’s re-thinking of the notion of psychological analysis makes clear – is that d introduces the idea of the self which is an awareness of the whole as such, but does not include an awareness of the parts of that whole as really present in the conscious state. These parts are present, to be sure, but only dispositionally, not really. Here is at least a major part of the concept of personal identity that a Humean can provide. But there are other aspects of the tale of the “quality” that determines our personal identity, and how we come to acquire it. Of these, more later.

394

Body, Mind and Self in Hume’s Critical Realism

(2) The Bodily Criterion I am sitting at my desk writing on the topic of personal identity in Hume. A few minutes ago I was shuffling my papers, seated at the same desk. Prior to that I talked on the ’phone to my daughter. As I now compose this on my computer, I am half listening to the radio. I am also thinking about the time; I’m a bit hungry but lunch time is not for a while yet. I’ll make myself a coffee instead, in a few minutes. But mostly my attention is on developing an argument. This argument will be part of a larger project of writing a book on Hume on personal identity. I will eventually submit the completed manuscript to some publisher or other. To do that I will have to print it off from my computer, and when that is done I will have to straighten it out and then stuff it in a mail sack to be couriered to the press. Or perhaps I can send it the new way, by an attachment to an e-mail. Now, the point of this little tale is to draw attention to the fact that this is a series of actions which are part of a larger action. Each of these actions involves my body. I remember shuffling my papers, I hear the radio, I feel the hunger, when I anticipate getting coffee I anticipate going through a series of bodily actions eventuating in the physical act of drinking coffee, I will arrange using my hands the sheets of paper that will make up the completed manuscript, I will stuff those papers in an envelope, I will hand that envelope to a courier. It is not surprising that each of these actions of mine involve my body: that, after all, is what gives me my identity as a person, at least if Hume is right. This continuity of the person depends upon the continuity of a material object, namely one’s body.32 As Locke and Hume argued, a material object has a certain nature. This nature is the nominal definition of this material object, that set of properties which are associated together in experience and which continue through time to be associated with each other. Many of these properties are dispositional, and indeed dispositions to acquire dispositions. These dispositions record tendencies and tendencies to acquire tendencies, together describing how the continuing object does and would interact with other bodies. This material object is one that continues as itself to exist through time. It is mine in the sense that we have discussed: Throughout this body is experienced from the inside. In the conscious states that include awareness of this body from the inside are rememberings of things like shuffling my papers, rememberings of consciously moving this body, this same body of which I am now conscious. In the same conscious state are anticipations of other things that

395

Five: Hume’s Positive Account

shall be done through deliberate, conscious, actions done through this body, the same body that I am now conscious of from the inside. This continuity, these continuities, are regularities, patterns. Because these patterns involve the regular and continuing coexistence of a set of properties, we as it were gather (the names of) those properties together to define the nominal definition of that sort of material object. The criterion for introducing such complex concepts is that of significance; we define as we do because the laws or regularities in the world are what they are: we define as we do in order to express the patterns that we discover in experience. The nominal essence defines a natural kind; we record kinds as natural kinds in our language by virtue of the significance that those concepts have. In this sense, the nominal definition of the material object with which we are concerned is that of a man (as Locke puts it [Essay, II, 27, vi]). The concept of man is not that of a rational animal but that of a material object which is organized teleologically into a whole of sympathetically organized working parts and of which consciousness is an important part. ...it is not the idea of a thinking or rational being alone that makes the idea of a man in most people’s sense, but of a body, so and so shaped, joined to it: And if that be the idea of a man, the same successive body not shifted all at once, must, as well as the same immaterial spirit, go to the making of the same man. (Essay, II, 27, vii)

The concept of a person is not that of a natural kind, in this sense. Rather, the concept of a person is introduced not simply to record regularities but to record certain moral judgments that we make about which entities to include as members of civil society. It is, as we have seen Locke put it, a “forensic” notion. But whatever else a person is, he or she is also a man (or woman). Being a person involves being that sort of material object, though it also involves other defining criteria. That at least, is what we have seen Hume argue. Personal identity presupposes the identity of a body. Some have recently argued that this is not so, and that we have to recognize that the continuity depends wholly upon consciousness. In other words, a person is, as Descartes argued, a thinking being: as he put it, sum res cogitans. The argument goes like this.33 Consider two persons, Brown and Brownson, and we suppose that there has been a surgical removal of Brown’s brain from Brown’s body and Brownson’s brain from Brownson’s body, the former brain going into the latter body, and

396

Body, Mind and Self in Hume’s Critical Realism

conversely. The brain of the person who was Brownson is now in the body of the person who was Brown, while the brain of the person who was Brown is now in the body of the person who was Brownson. Who is who? Suppose that one can anticipate that the body which was Brownson will, a while after the operation of brain exchange, be subjected to something which is excruciatingly painful. Brown can prevent this though some action which is to be executed before the transplant. At the same time, one can anticipate that the body which was Brown will, after the transplant, be the subject of an exquisitely pleasant experience. Brown can assure this by performing a different sort of action prior to the transplant. But Brown cannot perform both actions, so that either the Brown body will enjoy something pleasant or the Brownson body will endure something painful. Which action would Brown, before the transplant, perform? How would Brown reason? There is an identity of consciousness between Brown before and Brown after: they share memories, anticipate the same things, and so on. So Brown reasons that the consciousness which he is now is the consciousness that will experience the pain. He therefore acts to prevent the pain. This means that his body will not, in the future, experience the pleasant event. But the consciousness that is being denied that experience has no connection with Brown’s: it is the consciousness of Brownson. This implies that it is identity of consciousness which is the criterion of person identity, not continuity of body. For, though the body to which the consciousness of Brown belongs before the transplant, is not the body to which that consciousness belongs after the transplant, nonetheless the former consciousness acts to prevent pain that will come to him after the transplant. Brown reasons now that it is I who will experience the pain. The identity of the “I” before and after the transplant, the “I” that anticipates the pain and acts to avoid it, cannot be a matter of bodily identity. For Brown’s memories and anticipations which were in Brown’s body before the transplant come to be in the body to which formerly the consciousness of Brownson belonged. If this is taken seriously then we must conclude that bodily identity cannot be essential to personal identity, and that the latter must be a matter of continuity of consciousness alone. So much the worse for Hume’s account of personal identity. Some suggest that we must revise the bodily continuity criterion of personal identity by another one, one which does not make personal identity depend solely upon consciousness, but makes such identity depend

397

Five: Hume’s Positive Account

rather upon a physical continuity. The proposal is based on the idea that, though the Brown/Brownson case seems to establish that personal identity is not constituted by identity of the full body, nonetheless there is a continuity of the brain. The suggestion is that personal identity is constituted by identity of the brain. Then it is pointed out that consciousness does not depend upon the whole brain. The brain can in fact be split in half, down the corpus callosum and separated into two parts, each of which controls some bodily functions.34 It seems possible that one half of the brain could be removed and yet consciousness still proceed in the usual way: the existence of the person is not impaired by the severing. It is then suggested that one half the brain be transplanted to another body. In this case again there will be, it is supposed, continuity of consciousness: the person that exists after the transplant recalls all that happened to him or her as a person prior to the transplant. But the crucial point on this view is not the continuity of consciousness but the continuity of sufficient parts of the brain. Now, the Brown/Brownson case was supposed to call into question the claim that personal identity depends upon bodily identity, and that personal identity is a matter of consciousness alone. The discussion of brains that was just given is taken to suggest that the supposed possibility of the Brown/Brownson case does not force us to accept a criterion of continuity based solely on consciousness. We can still have a physical criterion. This criterion is the continuing existence of some relevant portion of the brain. This being so, one must give up Hume’s bodily criterion of personal identity, but can retain a modified version of it to hold that there is in fact a physical criterion of personal identity, namely, the continuity of the brain or part of the brain that carries the electrical and neuronal patterns that underlie the consciousness of the person.35 The argument against the bodily criterion can be modified so that it attacks any form of the thesis that personal identity depends upon physical identity. The idea of transplantation can be extended even further. Continue our fantasy. Suppose that there is electronic or better circuitry capable of storing brain patterns. We can then suggest that all the electric and neuronic patterns of Brown’s brain are transferred to such a device, where they are stored for a period of time. After that, they are transferred back to the brain which has in the interval been transplanted, at least to the relevant extent, to another body.36 In such circumstances there will be a continuity of consciousness, and one presumes that Brown will recognize that he or she is the person

398

Body, Mind and Self in Hume’s Critical Realism

who was previously in another body and is therefore the same person. So there is personal identity even without any identity of body or relevant bodily parts. But must we take this sort of argument seriously? How seriously do we take the hypothesis that brains have been transplanted or that brain patterns can be stored for periods of time in electronic devices? Take only the easiest case, the one that has the most of whatever plausibility these hypotheses might have, the case of Brown and Brownson. The difficulties that arise for this case, arise even more for the more sophisticated versions. If these difficulties make is impossible to take seriously the Brown/Brownson challenge to Hume’s bodily criterion of personal identity, then they make it even more difficult to take seriously the other challenges. When Brown, prior to the transplant, looks around himself, he is looking through eyes separated by so much, requiring such and such an effort to focus. He has a certain shaped nose, lining his visual field in a certain way, and wears eyeglasses to adjust for nearsightedness. His walk has a certain gait, determined in part by the fact his left leg is a bit shorter than the other, not noticeably so, but there if you look. His is a consciousness habituated through his body (Brown’s before the transplant), its central and peripheral nervous systems, to be that sort of person. And to be a piano player but also someone quite inexpert at pool. This consciousness and this habituation to a physically unique body, with a unique combination of culturally created traits, could only be a combination “not at home” in the Brownson body, with its different characteristics, and different set of habits and traits. For Brownson has an aquiline nose, compared to Brown’s snub nose. Brownson needs no glasses, his eyesight is normal. Brownson’s gait is steadier by a bit than Brown’s since Brownson’s legs do not differ so much in length. Brownson plays poker, not pool, and the best he does in music is the comb and tissue paper. There are many physical and habituated differences, great and small, between Brown and Brownson. Just the sort of differences that one is bound to find between any two individual and unique persons. The argument supposes that we can legitimately assert the contrary to fact conditional: If Brown’s brain were transplanted to Brownson’s body, then what was formerly Browns’s consciousness would be a functioning consciousness in what was Brownson’s body

399

Five: Hume’s Positive Account

In asserting contrary to fact conditionals, one asserts there to be a connection such that the consequent can be inferred from the antecedent. But normally, the lawful structure of the world is such that one can infer the consequent from the antecedent only if one suspends certain other beliefs about the world.37 In order to assume that the Brown consciousness is functioning normally after it is re-located in what was Brownson’s body, it is necessary to suspend all those beliefs about Brown’s consciousness that would prevent the transplanted brain/consciousness from functioning normally. We must assume the consciousness is modified to fit the shape of its new head, to accommodate the difference in gait, and so on. But what would the consciousness then be like? Would it genuinely be Brown’s? Indeed, since the consciousness that was Brown now inhabits, as it were, a different body and is thus oriented quite differently to the world, would that consciousness recognize itself as the same person? Finding itself unable to see the world or to interact with it in the ususal ways, the remembered ways, forced indeed to act and interact in ways that it finds totally impossible, quite beyond its own capacities, would it feel that it is the same person? It seems to me all that one can say is that one cannot say. If I were Hannibal, would I have fought Scipio this way rather than that? Is it really possible for one to make the necessary suspensions of belief that are required for me to become Hannibal? Probably not. Do we give up the fact that Hannibal was and we are not persons who approve of human sacrifices to Baal? Is that belief part of my identity? If we give up the idea that Hannibal so approves because we do not approve and we are Hannibal, then has Hannibal lost a sufficient part of his identity as a Carthaginian and worshipper of Baal that this person ceases to be Hannibal? Or, if we must retain it if Hannibal is to be Hannibal, then do we lose a crucial part of our own identity? Who can say? Who cares? It matters not for the enjoyment of our fantasy of being Hannibal. Our imagination lets us to a certain extent fantasize about being Hannibal. But it is fantasy, and one can therefore ignore all the real difficulties. One doesn’t have to face those difficulties. But this is not so in the case of Brown/Brownson. Here one is supposed to be serious, beyond mere fantasy; one is supposed to be contemplating a real possibility. It is, however, simply fantasy, simply not real. In imagining Brown’s consciousness being transplanted into the Brownson body, one can in fact no more say anything serious about this case than one can about the Hannibal case. One simply cannot say what all one must give up for it to be really conceivable that Brown’s consciousness has been transplanted to the Brownson body and is in the

400

Body, Mind and Self in Hume’s Critical Realism

world roughly as it was previously in the world. But if it really is fantasy – as it is – then one can think anything that one cares to fantasize about the capacity of the Brown consciousness to inhabit the Brownson body. One can fantasize that there really is a continuity between the consciousness as it is transplanted from the one body to the other. So of course in our fantasy life there is such a continuity – just as one can fantasize that one is the famous Carthaginian and anti-Roman general Hannibal. The point is that fantasy is one thing, reality another. We in fact cannot say what would happen. How reality would have to be adapted if we were to have the transplanting of the Brown consciousness is simply beyond careful calculation: it is too complex, the various other consequences of such changes too manifold, and besides that we don’t even know that much to make even plausible guesses. It is not just a matter of suspending certain initial conditions; we would also have to suspend beliefs in various laws describing the consequences of the various changes. And in addition to the new initial conditions, it would be necessary to assume a new lawful structure of the universe. What exactly would be the concepts we need to describe the new world is difficult indeed to say: it is beyond our capacity to say. Our concept of personal identity has the structure it has because it is designed to be used in our world, one in which brain transplants are hardly feasible, and beyond our capacity to genuinely imagine. Our concept of personal identity is not a concept that would be used in the radically different world in which brain/consciousness transplants occur. So, we cannot say whether there would be an identity between the consciousness that was in one body is the same person as the consciousness in the new body in which the brain of the first consciousness was transplanted. But the whole argument that personal identity does not depend upon bodily identity presupposes that we can have a continuity of consciousness through the transplant. But we cannot make such a presupposition. So the argument against bodily continuity being necessary for personal identity dissolves. Another version of the split brain argument suggests that it is possible to divide a person as one functioning consciousness into two functioning consciousnesses. We can think in terms of fission on the model of fission in amoebae. A person is there in one body and this body gradually divides into two persons.38 Let us begin with Brown and let one side of the fission be Brownson and the other Browndottir. By the criterion of bodily identity Brownson is identical with Brown, and also Browndottir

401

Five: Hume’s Positive Account

is identical with Brown. But the relation of “identity” is transitive. So Brownson is identical with Browndottir. But Brownson and Browndottir have different bodies and so cannot be identical. That shows that personal identity cannot be based upon bodily identity. Or so it is claimed. What it shows, I think, is that two persons, Brownson and Browndottir share a common history, and that makes them in part identical. And to the extent that their histories are not shared they are not identical. But even here it is beyond our capacity to conceive how we would react. Suppose that Brown does something that deserves punishment, but that this is not discovered until after the fission. Whom do we punish? Brownson? Browndottir? Or both? One cannot simply answer these questions. There are too many unknown factors. Punishment has several points, one is bringing it about that the wrong acts are not to be performed. Would punishing both bring about this end? Would the threat of punishing both Brownson and Browndottir be enough to deter Brown? Why not punishing just one? But there is also the issue of fairness. If Brownson is to be punished then surely it is unfair that Brown’s other successor, Browndottir, should escape retribution. Moreover, how does Brown make plans? Ordinarily, in the real world, we plan and launch our projects with the knowledge that we will be there, in the future, to further those projects and to watch them come to fruition. What does Brown do? Is he or she going to let the project come to fruition through the efforts of Brownson? Or Browndottir? Who does he or she expect to carry out the later stages of the probject? Brownson? Or Browndottir? Or both? Will there be a competition between Brownson and Browndottir for the fruits of the project which Brown initiated? Who knows? Such a world, one in which fission occurs, would clearly require a re-thinking of what we are as persons and how we interact with the world in which we find ourselves. Again, maybe Brown is a sensible knave. That would, one presumes, make Brownson and Browndottir sensible knaves also. So maybe Brownson will conspire with whomever or however to avoid the duties incurred by himself when he was simply Brown: Brownson will try to get Browndottir to bear the whole burden. And Browndottir will similarly conspire. “It is you, Browndottir,” Brownson will try to argue, “who is really Brown and who therefore must bear the responsibility for the actions undertaken by Brown.” In effect, the sensible knave will argue that he of

402

Body, Mind and Self in Hume’s Critical Realism

she is not the same person as Brown, that it is Browndottir who is really Brown, or what Brown has become. Conversely, of course, Browndottir will argue the same point, that it is not he or she who is Brown; it is Brownson who is really Brown. And suppose that fission as it were sneaks up on persons. In that case Brown will never know precisely when he or she is going to divide. That would throw confusion in all one’s larger projects. In such a world one would not be able to plan projects in the way one plans them in the real world, the world of our ordinary experience in which persons are identified by their bodies. It is clear that to raise these relatively simple points shows that there are many more issues that would arise and require the adaptation of our practices if fission were a reality. Among those practices is the practice of determining what we will count as determining the identity of persons. But once again, that is not the world in which we live. And given the laws of this world, we have the concept of a person that we have, the concept that fits what we require for our being able to get along in this world and with each other in civil society. In this world, the world of ordinary experience, we can safely base, as Hume argued, personal identity on bodily identity. (3) Humean Persons As we have indicated, there is little doubt that Hume would find something along the lines that we have sketched congenial as an account of personal identity: Conscious states are identified as belonging to the same mind by virtue of their belonging to the same body. But in fact, this is not the central theme in Hume’s reply to those, like Butler and Clarke, who held that all stability of personality would disappear if we were to abandon the notion that the self is a simple substance. The self, Hume holds, is also an entity with a certain nature. This nature is the person’s character. The crucial Humean point is that this nature, one’s character, is not given once for all in an innate structure, but is, rather, a social construct. That is, a stable personal identity arises in, and as a result of, a social context. As Capaldi has emphasized, in the discussions of the passions and virtues, “the self in a public and social world is already presupposed in Hume’s analysis.”39 The sort of personal identity one can have is dependent upon the social structure, the norms that Hume speaks of when he discusses the “artificial virtues,” and the stability of that identity depends upon the

403

Five: Hume’s Positive Account

stability of that social structure.40 Now, in Hume’s way of speaking, when the statement (+) I feel proud that I then did so-and-so is used in its normal way, the subject of the feeling of pride is my action of doing so-and-so. The object of pride is always oneself; the subjects are “either parts of ourselves, or something nearly related to us” (T, p. 285). The pride in an action often derives from its virtue – “the good and bad qualities of our actions and manners constitute virtue and vice, and determine our personal character, than which nothing operates more strongly on these passions [of pride and humility]” (T, p. 285); we may assume that this is so in the case of (+), but at any rate its use implies that the action is so related to me that I take pride in it. This relation is not merely that of belonging to my body. I do not take pride in the reflex acts of my arms and legs; reflex acts are not actions, they do not relate to me. Nor, indeed, are certain intentional acts mine in the relevant way. This, surely, is the point about the acts of Locke’s drunk: his actions when drunk, while certainly his in the sense of being consciously intended by him, are nonetheless not his in the sense of being “out of character.” The actions of the drunk are, to use Hume’s terminology, the subjects of shame rather than of humility, though being a drunk may be the subject of humility. It is the element of character that here provides the linkage to self that is recorded in (+). (+) moreover records a knowledge of the connection of action to the self unified not only as belonging to a certain body but also as having a certain enduring character as its nature; and this knowledge would normally be taken as the exercise of the capacity for self-knowledge. There are two points here. The first is that one’s identity as a person is constituted by one’s character. This must be added to bodily identity as part of the relation “R” that links the various parts of a person into a unity. The second is that self-consciousness is to be understood not only in terms of the awareness that constitutes my present conscious state; it is to be understood also in terms of the exercise of the capacity to survey one’s actions and dispositions and to relate them to oneself qua having a certain character. Self-consciousness in the first sense is the immediate awareness that one always has of the contents of one’s conscious states; selfconsciousness in the second sense is the capacity to recognize one’s various actions and dispositions as expressing and as connected to oneself as an enduring unity. The former is a simple awareness. The latter is the capacity to come to know certain patterns in experience. And this coming

404

Body, Mind and Self in Hume’s Critical Realism

to know patterns in experience is to be understood in the Humean terms that we have already noted. We must look at each of these points – character as providing the linkage that defines personal identity, and our own reflective knowledge of our character – in turn. But as part of this discussion we must first look once again at Hume’s moral theory and, more generally, his account of norms and standards of behaviour, including linguistic behaviour [Subsection (i)]. We then turn the notion of character that provides the linkage that defines personal identity [Subsection (ii)]. After this we proceed to the issue of knowledge of self. This involves the notion of attention to one’s self, in the sense of the present contents of one’s consciousness; and also the issue of the knowledge, one’s own and that of others, of one’s character [Subsection (iii)], including the reflective selfawareness that defines the person at his or her best. In the final section [Subsection (iv)] we look at the doubts that a sceptic can raise about our knowledge of our selves. (i) Hume on Norms: His Account of “Ought” We have already looked at aspects of Hume’s views on norms. But it will pay to come at these things from a slightly different perspective, one that will locate some of these thoughts in the context of Hume’s account of the self. Hume’s procedure in ethics is parallel to his procedure in epistemology. In the latter, he argues on the one hand against the rationalists, and, on the other hand, proposes a human and humane standard of practice. Hume rightly supposes that morality connects to human action: morality is pointless if it does not, or cannot, guide behaviour. This means that morality must determine human motives, or, in Hume’s terminology, the passions. Now, traditionally, natures of substances were not only descriptive and explanatory, but also normative. Natures explained because they were active, and for the same reason the norms defined by natures determined the behaviour of substances. But Hume has already exorcised the active powers of the Aristotelians. As a consequence, he has severed the traditional linkage that these gave between description, given by citing the nature of a substance, and motivation, given through the active, that is, moving or motivational, aspect of the nature. Hume therefore asks whether there is any linkage between matters of fact and human passions. This

405

Five: Hume’s Positive Account

question he answers in the negative. Hume considers two views. One is the view of Samuel Clarke and the other Newtonians, following the lead of the Cambridge Platonists, that moral facts were a priori truths about relations of (moral) fitness. These relations they likened to the a priori relations that hold in geometry and algebra. The other view is that of Shaftesbury and his followers, that relations of moral fitness are akin to matters of sensible fact and are known by a sense that is parallel to our ordinary senses. In either case the moral truth is one of fact, either an objective a priori fact or an objective feature of things known by something akin to sensible experience. Hume argues that neither of these positions is tenable: in neither case do the alleged moral facts determine the passions. For that to happen, the fact in question would have to be intrinsically motivating. It would, in other words, have to create the relevant motive. But that would require a necessary connection between the fact and the passion. However, there are no such necessary connections. The accounts of the Newtonians and the moral sense theorists thus do not stand. Hume’s basic argument is the absence of any necessary connections. On the one hand, he makes the point about the separability of our passions and both a priori facts and facts of sense experience –, this is ontological use of PA to exclude objective necessary connections of all sorts; he makes the point about the separability of our passions and other matters of fact, on the other hand, by means of a variety of examples drawn from the works of such defenders of these views as Clarke and Shaftesbury. In this way Hume is able to establish that there is always a gap between “is” and “ought,” that is, ordinary facts and the moral standards that move us to action. Nonetheless, there is in fact a real linkage between “is” and “ought.” Knowing certain facts does often induce a moral feeling or sentiment that moves one to action. The issue that faces Hume is determining what this linkage is and ought to be. Hume makes two points. First, he argues that people are moved by pleasure (and the avoidance of pain): thus he tells us that we are moved by “good and evil, or in other words, pleasure and pain” (T, p. 439). This is not to say that all motives are selfish; there are also unselfish pleasures. There are a variety of objects that attainment of which gives pleasure; some of these are selfish, some unselfish. Moreover, the pleasures themselves come in different species; the pleasures of the flesh are very different from the pleasures of social intercourse with one’s friends. But

406

Body, Mind and Self in Hume’s Critical Realism

whatever is the object of our actions, it is an object solely by virtue of the pleasure that its attainment brings. This is a fact, a lawful fact, of human psychology: pleasure is as a matter of lawful psychological fact the sole end that moves us. Since it would be pointless, given this psychological fact, to urge people to aim at any other end, it is this which will have to provide the standard for human morality. Pleasure ought to be our standard. Secondly, Hume argues that if we make a catalogue of things that people denote as virtues, then we discover that as a matter of fact those things so denoted are patterns of action that tend to bring pleasure either to the actor or to others. Pleasure is in fact our standard. “Pleasure and pain,” we are told, “if not the causes of virtue and vice [are] at least inseparable from them” (T, p. 296). If a virtue is that which tends to give pleasure, then the virtuous action, that which we ought to do, will be that which will maximize pleasure overall. Now, within a family there is no neat division of things into mine and thine. Instead of the neat division of property that one has in civil society, one has instead a community of goods. Each member of the group, the family, feels a concern for the common good of that family. For, one naturally responds generously to the concerns of others in the family. The goods of the family are devoted to the common welfare, rather than one having rights over some of those goods to the exclusion of others. There are no rules of property in such circumstances, nor therefore any of the virtues, e.g., respect of property, or vices, e.g., theft, associated with property. There are nonetheless virtues in such a context, e.g., prudence, gratitude, good humouredness, etc. These are virtues because they tend to give pleasure to oneself or to others. Now, there is in the human breast, according to Hume, and not implausibly, a natural mechanism of sympathy. This mechanism is such that when another feels some sort of pain, then I too feel that sort of pain. Thus, if I see another fearing for her life after falling into deep water, then I too feel fear for that person's life, and, often enough, will act to relieve that fear by removing the person from the present danger: I dive in and rescue her. Again, I give my child the dessert I have not yet finished because I enjoy the pleasure that she takes in it. Through the mechanism of sympathy, what distresses one distresses all; what gives pleasure to one gives pleasure to all. Generosity is a virtue, for, when I acted generously, the other to

407

Five: Hume’s Positive Account

whom I have so act takes pleasure in my action. But then, through the mechanism of sympathy others will also take pleasure in my action. Indeed, through the same mechanism of sympathy I too will take pleasure in my action. The same mechanism will bring about general feelings of disapproval of vicious actions, those that cause pain to myself or others. These sympathetic responses by myself and others to the pleasure (or pain) created by my virtuous (or vicious) acts constitute our moral sentiments, our feelings of moral approval and disapproval. Similarly, in a context in which there is no scarcity, e.g., with respect to the air that we breathe, there again is no distinction of mine and thine. Nonetheless, in such a context good humouredness would remain a virtue, giving pleasure to others. And again, the mechanism of sympathy will create a general feeling of moral approval of this virtue. Hume thus argues that in situations of extended generosity and in situations of plenty, there are virtues but those associated with property are not among them. The virtues that we have in these situations where either generosity is extended or goods are plentiful he refers to as “natural” virtues. But we also have rules of property. For any object that is my property, I have a sense that it is morally right that others not interfere with my use of it, and others have the sense that it is morally obligatory that they not interfere with my use of it. These are passions or sentiments that move us to action. These sentiments in fact coordinate our behaviour, ensuring that each respects the property of every other. These moral rules, Hume argues, will not be present in situations where there is no scarcity, or where one’s natural generosity extends to all members of the community. In conditions of scarcity there will be insufficient goods to satisfy fully the desires of all. If we also have restricted generosity then rather than a willingness to share with others one will instead attempt to satisfy one’s desires, or the desires of one’s family or local group, at the expense of others. Then, supposing that there is no one individual or subgroup that can arbitrarily impose its will on others, that is, supposing what will generally be the case, that there will be a rough equality among the participants, then there will be a struggle of each with every other for possession and use of the scare goods. But there will be no security of possession, nor any capacity to engage in long range planning for the satisfaction of one's desires. Life will be not very pleasant, and devoid of all the rational human practices that presuppose long range considerations. One will survive, but only just, and not for very long. Life

408

Body, Mind and Self in Hume’s Critical Realism

will be, as Hobbes put it, nasty, poor, brutish and short. Each desires a life that is pleasant, humane, and a reasonable length. But the conditions of scarcity and restricted generosity seem to preclude it. Hume argues, however, that art intervenes. If a convention could be established that each would respect the property of each other, sharing in this way the scarce resources, then, while one could not have one’s fill of the scarce goods, one will nonetheless be able to enjoy what one does have in security for one’s natural life. If one could establish such a convention to coordinate the behaviour of each with every other, then one could have a better life than the nasty, brutish and short one that alone is possible in the absence of any such convention. How does such a convention arise? The desire for such a convention is there in each member of the group. Yet so is the desire to appropriate more than others from the scarce resources. If one person desired to conform to the convention, and respected the property of others where the others did not respect his or hers, then that person would be the loser. What is needed for the convention to be instituted is the sense that one can trust others to conform to it while one is oneself conforming. So our question is, how can such a sense of trust arise? Hume argues that this can be acquired through experience. Hume gives the example of two men in a boat who both wish to cross to the opposite shore. Neither knows how to row. But after a time where each pulls on an oar in a random way, they will discover that if they both pull simultaneously in the same direction they will be acting in a way that will enable each to satisfy his or her longer range goal of getting to the other side. That is, they will learn that they can achieve their longer range goal by coordinating their behaviour. Trial and error will soon bring about the expectation in each that the other will pull in the appropriate way and the intention in each to pull in the appropriate way. These expectations and intentions are the passions that move those involved in the cooperative behaviour. They will not be originally present but will be acquired in the learning process. And as part of this learning process each will also learn that the intention of the other to conform to the convention is as strong as his or her own, strong enough that is to ensure that conformity. He or she will learn in other words that as his or her own intention to conform is strong so is he justified in his or her expectation that the other will conform. Each will learn to trust the other in the process and each will learn that his trust in the other is justified. And now: reflection. Reflection upon one’s behaviour and actions

409

Five: Hume’s Positive Account

alone could not create the shared intentions and expectations that make cooperative behaviour possible. The latter is, rather, due to the associative process. Trial and error will produce behaviour at times in both that is conducive to each advancing towards the goal that each has. The pleasure each such modest success will produce will become associated with behaviour of that kind. Eventually the association will produce in each a motive strong enough to generate behaviour of the cooperative sort, to the exclusion of random and uncooperative behaviour. Here is a new motive to conform to these standards. [The rules of civil society] are the conventions of men, which create a new motive, when experience has taught us, that human affairs wou’d be conducted much more for mutual advantage, were there certain symbols or signs instituted, by which we might give each other security of our conduct in any particular incident. (T, Bk, III, Pt. ii, sec. v, p. 522)

Pursuing these ends that the new motives give them will enable the two persons to also pursue effectively the longer range goal they both have. Reflection alone will not create the required passions, but after the experiences that create them through association, one can now reflect upon these acquired passions. Reasoning about means to achieve the longer range goals could not generate these passions, but once they are present reason can see that they are a good means to achieving one’s own end, along, of course, with the end of the other person. Moreover, we see that the means for bringing about these means, that is, these passions, is through processes of learning, that is, processes in which one both inculcates in others these passions and so disciplines oneself as to strengthen these passions. Since we have acquired the expectations and intentions that ensure the cooperative behaviour as a means to achieving other goals, and since reasoning about means and ends can justify our encouraging and developing these passions, we can reasonably characterize them as “artificial passions.” Social conventions exist as a consequence of developing, as a means to achieving shared goals, artificial conventions that motivate the cooperative behaviour. What makes these conventions possible is not so much reason as the fact that persons can acquire habits and motives through processes of association. Nonetheless, reason can justify using and relying upon these processes to produce as means to further, shared ends passions that motivate the cooperative behaviour. The struggle of each against every other that occurs under conditions of equality, scarcity, and restricted generosity, is ended, or at least never

410

Body, Mind and Self in Hume’s Critical Realism

occurs when people establish the convention of property. This convention enables them to share scare resources in a way that benefits everyone. To be sure, no one is fully satisfied with his or her share alone; each would like more of the resources. But in return for giving up the opportunity of satisfying that desire each gains something else that they desire, to wit, the possibility of enjoying with some security that share of the resources that they do obtain. This desire for security which makes possible the convention of property will be strong enough to over-ride short term desires and impulses. The institution of the convention of property is not the result of reasoning alone. This cooperative behaviour, like the cooperative behaviour established by the rowers in Hume’s analogy, is established because, first, people are such that useful behaviour can become motivated as it becomes associated with the pleasure that it produces, and because, second, each respecting the property of every other is useful in producing the better life that each desires. Reason and reflection alone cannot bring us to these conventions. We need also experience of people to do that. We need to know that people are such that they can become motivated to conform to these conventions. And we need to know, through experience, that such conventional behaviour on the part of all is useful to each person as he or she pursues the ultimate end of leading as decent a life as possible. But, given that we know this, then reason can show that the conventions are justified, and that it is reasonable to encourage the relevant motives in others and in ourselves. However, we are so far only at expectations and intentions; we have not yet got morality. What we have is self-interest leading us to cultivate in ourselves and others certain passions that can motivate cooperative behaviour. But we have not yet obtained the moral sentiments that are also characteristic of justice, that is, the convention of property. However, it is clear that once again the mechanism of sympathy can account for these sentiments. When I behave cooperatively relative to another, she is motivated to behave reciprocally, and the passions that move her to this action are fulfilled by my action. She thus takes pleasure in it, as, indeed, I do in hers. Through sympathy others respond to her pleasure and morally approve of my act; in fact, so do both I and she. Similarly, all of us also morally approve of her reciprocal action. Thus, in this way, through the mechanism of sympathy, actions which are just, in the sense of conforming to the conventions of property, and which are originally justified by selfinterest, are moralized, and come to enjoy the moral approval of all members of the community.

411

Five: Hume’s Positive Account

Moral approval is itself a pleasure. As Hume says, “Moral distinctions depend entirely on certain peculiar sentiments of pain and pleasure excited by a mental quality in ourselves or others” (T, p. 574).One has non-moral ends and one acts to secure those pleasures. We are moved to seek such non-moral pleasure, but morally approving of such an act provides an additional motive for one’s performing that act. Self-interest leads to, and reinforces the passions that incline us to just acts. Moral approval in turn reinforces these artificial passions that motivate just acts, acts that conform to the conventions of property. Where self-interest is strong enough to incline one to go against the artificial passions that move one to respect the property of others, we are nonetheless constrained by our own moral sentiments to conform to the conventions of property. (Long term) self-interest that cultivates our cooperative passions, on the one hand, and sympathy, on the other, thus combine to increase the likelihood that one will act justly, even where one’s short term self-interest inclines one to be a knave – increase the likelihood, but there is no guarantee. Moreover, since cooperative behaviour is itself reasonable relative to the long range goal of living as decent, that is, as pleasant, a life as one can, it follows that passions which promote such behaviour are themselves enjoined as reasonable. The moral sentiments do promote such behaviour. Reason, reflecting upon these passions that arise from the mechanism of sympathy, will judge them to be rationally justified relative to that end. Since acting upon our moral sentiments will, in general, produce pleasure, these habits will be virtues. Being a moral person is thus not only virtuous but, reflection tells us, it is also reasonable that it be so.

We now also see how we do in fact make inferences from “is” to “ought.” The inference is one of habit. I recognize that the nuts in that pile have been appropriated by you from the tree in the common. That fact then brings about in me the moral sentiment that I am obliged not to interfere in your use of them and in you the sentiment that it is right that I not interfere in your use of them. These habits are acquired; they conform to the conventions of property. Moreover, these habits are reasonable; they are, therefore, standards to which I, as a reasonable person, ought to conform. In the end, inferences from “is” to “ought” are no more mysterious that causal inferences – though in both cases the Humean picture is hardly likely to satisfy those who continue to demand, with Aristotle, Descartes, Clarke, and Shaftesbury, some sort of standard that transcends the

412

Body, Mind and Self in Hume’s Critical Realism

capacities that we discover through reflection upon our ordinary experience of the world and of ourselves as part of that world. Moreover, some inferences from “is” to “ought” can, like some casual inferences, be shown to be more reasonable than others. It is the reasonable ones that Hume endorses, in the case of causation those that conform to the “rules by which to judge of causes,” and in the case of ethics those of our natural virtues and those such as the rules of property that define civil society. On this Humean account of the inferences of morality, our moral duty is determined by our moral sentiments. But here we can always raise the question that G. E. Moore taught us to ask: Is it really right that we act upon the moral sentiments that we in fact have? Is it really the case that we ought to do what our moral sentiments move us to do? Does Hume have a response? If the question is a demand for an objective justification of the sort given by philosophers like Samuel Clarke or Shaftesbury – or G. E. Moore –, then the answer is simply that it cannot be answered: Hume has established that there is no such objective justification for our moral sentiments – the appeal to PA does that, it shows that there is no objective basis for our moral values in the ontological structure of the universe; and if there is no such objective standard, it is unreasonable to demand one. Since the demand for that sort of justification is unreasonable, the reasonable person must settle for a different kind of justification. With the rejection of objectivism in ethics, one must critically examine the standards of justification in moral theory, just as one must examine critically our cognitive goals and standards after the rejection of the Aristotelian and rationalist positions. One must reflect upon experience and determine what are our human goals, and inquire which among our possible moral sentiments best serve these goals. Reflection upon our human experience shows that goals proposed by, say, monks or pillar saints are unreasonable –.”monks and dervises” are Hume’s examples, directing our attention to their “extravagances of conduct” (T, p. 291). Serious mortification of the flesh – seeking pain rather than pleasure – is in fact impossible – people by their nature seek pleasure; also impossible is the total overcoming of human pride – the psychological mechanisms are such that persons inevitably take pride in something or other that stands in some relation or other to themselves. (“Can we imagine it possible, that while human nature remains the same, men will ever become entirely indifferent to their power, riches, beauty or personal merit, and that their pride and vanity will not be affected by these

413

Five: Hume’s Positive Account

advantages?” [T, p. 281]) Reflection shows that persons cannot in fact pursue these goals, and that the end that persons pursue is in fact pleasure rather than, say, self mortification. Indeed, reflection upon experience reveals that insofar as these monkish goals are pursued, they are pursued as parts of pleasure; the mortification of the flesh has become, through processes of association, part of the monks’ pleasure – surely the pillar saint finds satisfaction, i.e., takes pleasure in, his service to God and his community, sitting on his fragile and open post in order that he could early announce the second coming of the Lord. If he seeks pain it is only because he has come to find that that is what constitutes his pleasure, he is comfortable in his discomfort. As for pride, it is part of our natural constitution that we take pride in certain things; the pillar saint no doubt takes pride in his capacity, lonely as it is, to be the first to see, and announce to others the second coming of the Lord. The pillar saint and the monk, and also the good Calvinist, no doubt strive in their own breast to overcome the pride they feel, but equally without doubt each is as likely to take pride in the strenuousness of his attempts at overcoming pride as the young student is likely to take pride in his or her accomplishments in his or her school exercises. Given human nature, the goals of truly mortifying the flesh and truly overcoming pride are impossible of fulfilment; to attempt to achieve them is to attempt the impossible – just as the attempt in epistemology to meet the Cartesian standard is humanly impossible. As in the case of our cognitive goals, so in the more general case, reflection upon experience determines what is impossible and what is a reasonable goal, and we discipline ourselves to pursue the reasonable – for, if we pursue the impossible we are doomed to a struggle against ourselves and we will subject ourselves to the constant and self-imposed pain of inevitable failure, living with a constant regret for our being deeply sinful and vicious, and for our constant failure to be virtuous. Reason and ends here go together: reason – reflection upon our selves and on the world in which we find ourselves – reason in this sense adapts our ends to the means available, as well as adopting means to our ends. What sort of person, morally and cognitively, we ought to be depends upon what sort of person we can be – we cannot be someone who judges causation infallibly because the world is as it is, with separable parts, and our mind is as it is (we are not omniscient, we know only a sample, not a population) –; and we cannot be a person who perfectly overcomes feelings of pride, our human nature being what it is. Reason in the sense of reflection also determines which patterns of

414

Body, Mind and Self in Hume’s Critical Realism

behaviour are efficient means to the end that defines our good. These means include those habits that are the natural virtues. And they include those habits of conformity to the conventions of property, what Hume calls the artificial virtues. They do not include the habits of the monk, the “dervise,” the superstitious, or the enthusiast. These are unreasonable standards. But the means that reflection selects also include the habits of disciplining ourselves to behave morally when these strengthen our tendencies towards those virtues which are natural and those which are artificial, provided that the latter are, like those of property, reasonable. Reflection thus declares that behaving morally is reasonable behaviour relative to goals that reflection has determined to be reasonable. We might, of course, settle for less. As in the case of cognition, the passions of our “spleen and indolence” may incline us to settle upon what is directly given to us, without reflecting further upon what else might be achieved. But the philosopher is concerned to find the truth in ethics as in metaphysics and epistemology. I cannot forbear having a curiosity to be acquainted with the principles of moral good and evil, the nature and foundation of government, and the cause of those several passions and inclinations, which actuate and govern me. I am uneasy to think I approve of one object, and disapprove of another; call one thing beautiful, and another deform’d; decide concerning truth and falsehood, reason and folly, without knowing upon what principles I proceed. I am concern’d for the condition of the learned world, which lies under such a deplorable ignorance in all these particulars (T, pp. 270-1).

This concern leads the philosopher to inquire into alternatives to our present moral sentiments, to see whether those we now feel, those which now move us, are in fact the best for us as human beings. The philosopher carries on the reflective enterprise to discover how we can best achieve our human goals and discover how to discipline him- or herself to better achieve those goals; and, moreover, the philosopher helps and encourages others, to conform to these standards. That is what the reasonable human being will do. In answer to the question, then, whether we ought to act upon our moral sentiments, the Humean reply is that we ought to do so just in case those habits of so acting are reasonable relative to our human goals. This may not be as much as ethical objectivists like Clarke, Shaftesbury, the monks or the superstitious would like to obtain, but for all that it is as much as is humanly possible. To demand more is as unreasonable as it is to

415

Five: Hume’s Positive Account

settle for less. We have seen previously that the mind arrives at standards of rationality through a process of reflecting upon the world as it is experienced and, equally importantly, upon itself as it is experienced. We set cognitive goals, and adopt means for achieving those goals. Through experience we discover our capacities for attaining these goals, and efficiencies of the means we use. Reflecting upon this experience we adopt goals that are attainable and means that are efficient. Self-reflection leads to standards of practice that define the (cognitive) virtue of rationality; it leads to standards that are attainable and efficient. In other words, selfreflection leads to a reasonable standard of rationality. Or at least, it does so if one is wise. The case of moral standards, we now see, is parallel. Here too reflection leads to standards of practice that are reasonable. Once again we see that it is through the self-reflective capacity of the mind, its ability to reflect upon its own experience, that leads to reasonable standards of practice. And to emphasize once again, this capacity for self-reflection is a central ingredient in Hume’s concept of the person. Hume’s notion of a person, and his concept of personal identity, are thus bound up with his accounts of human rationality, which we knew before, and of human morality, which we now know, that are at once empiricist, humane, and social.

There is another aspect of Hume’s account of the artificial virtues that requires attention. Hume extends the account of the artificial virtues from the conventions of property to those of contract, or promising, and then to those of political allegiance. The conventions of promising extend those of property by allowing us to exchange property. The conventions of political allegiance establish the role of the chief magistrate, which provides a mechanism for the protection of property and the enforcement of contracts. These conventions reinforce those of property and promising by adding the motives of respect for the magistrate and fear of his or her power. These conventions are justified by reason reflecting upon the conditions for living a decent life. And, as in the case of property, they are rendered moral by the mechanism of sympathy. One obeys the magistrate not only because so doing protects our interest in property and contract, and not only because of fear of the magistrate’s power, but also because we are constrained to

416

Body, Mind and Self in Hume’s Critical Realism

conform to the rule of the magistrate by our moral sentiments that judge such behaviour to be morally fit. Conventions of contract or promising and of political allegiance thus become virtues, that is, in Hume’s way of speaking, artificial virtues. Reflection again shows them to be reasonable. These extensions of Hume’s basic points in the case of the conventions of poverty are all obvious enough; we need not pursue them in detail. There is one important point that Hume makes in the context of promising, however, that bears elaboration. This has to do with the role of language. Philosophers like Samuel Clarke argued that we have an innate a priori knowledge of the objective standard that establishes that keeping promises is morally fit. Hume has argued that there is no such a priori knowledge of moral principles, nor, therefore, is there such knowledge in the case of promises. Hobbes had argued similarly that the rule regarding promises had no ground in the ontological structure of the world, and then went on further to argue that the obligation to keep promises (contracts) derived from an agreement. But this latter won’t do, for such an agreement can only be a promise to keep promises. What this shows is that there must be a prior convention to keep promises if we are to agree to keep promises. I.e., promising presupposes rather than establishes the convention to keep promises. What, then, is this convention? Hume argues that the convention consists of the following (acquired) habits: Jones saying to Smith ‘I promise to do x’ brings it about that Jones forms the intention to do x and Smith forms the expectation that Jones will do x Saying ‘I promise’ brings about the intentions and expectations; it causally creates the intentions and expectations that motivate the coordinated behaviour of promisor and promisee relative to the promised action. (There will, of course, be further motives for making a promise, that is, for intending to create the intentions and expectations that coordinate the behaviour of promisor and promisee.) Sympathy will then moralize this conventional behaviour, bringing about the moral sentiments in the promisor that he or she has a moral obligation to do what was promised. Is the convention justified? Are the moral sentiments associated with justified? Hume’s case here is parallel to that which he gives for the convention of property. Self-interest in living a decent life requires not

417

Five: Hume’s Positive Account

only property but the possibility of exchange, including those exchanges which address the future. This is possible only if there is a convention for contracts, that is, promising. Self-interest therefore justifies the institution of this convention. What makes this convention possible is not merely reason, but rather the possibility that we can, via the associative mechanisms, discipline ourselves and others to conform to the convention. Given that reason tells us that it can be done in this way, self-interest justifies our so acting, relative to others and to ourselves, that we come to conform to the convention. Sympathy then moralizes the convention. Reflection upon our human goals establishes that instituting the convention of promising is reasonable as a means to those goals, and that moralizing it makes it a more efficient means. Conforming to the convention and accepting our moral sentiments about such conformity are therefore reasonable, that is, justified in the only way that we, as human beings and as reasonable persons, can possibly justify any moral standard.

The institution of promising, Hume has here argued, is intrinsically linguistic: its institution requires a linguistic convention governing the use of the words ‘I promise.’ Making ourselves and others persons who conform to the convention of promising involves making ourselves and others conform to the linguistic convention for the use of the words ‘I promise.’ Since the institution of promising is inseparable from this linguistic convention, self-interest relative to certain very important shared ends justifies conformity to it by ourselves and others. Sympathy will then work on this convention, moralizing it too. The linguistic convention thus becomes a linguistic rule or norm concerning speech acts to the effect that saying ‘I promise’ ought to bring about the relevant intentions and expectations The point that needs here to be emphasized is that this provides a model for all linguistic rules. Hume himself makes clear that he holds that the rules of language are analogous to the conventions of civil society. ...experience assures us still more, that the sense of interest has become common to all our fellows, and gives us a confidence of the future regularity of their conduct: And ’tis only on the expectation of this, that our moderation and abstinence [from taking the property of others] are founded. In like manner are languages gradually establish’d by human conventions without any promise (T, p. 490; italics added).

418

Body, Mind and Self in Hume’s Critical Realism

The analogy, and in particular the rule for the use of ‘I promise’, makes evident that he holds something like the following about language: There is a deep human need for communication. Trial and error, coupled with the human capacity to learn through practice those strategies that tend to serve our shared ends, generates patterns of linguistic behaviour that serve the need to communicate. These patterns involve a coordination of behaviour, including linguistic behaviour, on the part of speaker and hearer, and also, on occasion, between speaker and him or herself at a later time as respondent. Self-interest justifies these conventions for coordinating behaviour. And then sympathy will moralize them, transforming them into linguistic norms or rules, standards by which to evaluate the behaviour of ourselves and others.

We thus see how our shared interest in communication, as a basic human need, lies behind all our linguistic conventions. What we must now recognize is that this illuminates Hume's account of abstract ideas. In that discussion Hume elaborates upon one portion of this set of conventions, namely, that portion which covers language entry (worldword) transitions. Here we have the habits of use in which words are applied to things, either as individuals or as members of a resemblance class of things. These conventions are governed by rules like ‘Pierre’ means Pierre and ‘red’ means red These rules presuppose, of course, that those whose thought and behaviour they govern have already the idea of Pierre and the idea of red, as the occurrence of the words on the right hand side indicate. These rules, often called designation rules, do not create the habits of usage which they govern but rather in motivating conformity to the patterns they oblige, serve to maintain and strengthen those habits of usage.41 One should add, however, that these rules, in motivating other persons to bring about general conformity to the patterns they oblige, also bring it about that others so educate me that I come to acquire the habits of conforming to those patterns, habits which are subsequently maintained by those rules which I myself have come, through the mechanism of sympathy, to internalize as obligatory. There are further conventions, of course. Besides the world-word transitions that govern meaning, there are world-word transitions that govern the assent, on the basis of experience, to basic observation sentences, e.g.,

419

Five: Hume’s Positive Account

If ‘F’ designates F and ‘a’ designates a, then whenever one assents to ‘a is F’ then, and only then, one has an experiencing in context C of a being F. If this is true then, if the habits given by the designation rules obtain, then there is a habt of assenting to ‘a is F’ when and only when one is experiencing in context C the state of affairs that a is F. The context C needs to be spelled out so as to mention the conditions under which perception is, if not always, then almost always, veridical. It has also to be qualified so as to include information acquired on the basis of legitimate authority. And so on. But again the notion is clear enough. Next there are intra-linguistic (word-word) transitions. These include the rules of syntax narrowly conceived, e.g., the rules of formation for conjunctions and disjunctions, or for universally and existentially quantified sentences. They include the rules of definition by which some expressions abbreviate others. They should also include the rules of socalled “contextual definitions,” e.g., of definite descriptions, though Hume, like all thinkers before Russell, did not recognize these as such. The intralinguistic rules also include the rules of semantics, which give truth conditions for sentences. Then they also include the rules of evidence, e.g., the norms of deductive logic for maintaining consistency in belief, and the norms that Hume calls the “rules by which to judge of causes and effects” for assenting to generalizations which make assertions that go beyond the data one has acquired through observation. In particular, the rules of evidence include one’s ethics of belief. Finally, we should note various conventions governing language exit (word-world) transitions. There are, for example, conventions that one speak the truth: assert ‘p’ in a context where the hearer desires information only if one has evidence that ‘p’ is true. One should make as strong an assertion as one’s evidence allows: if ‘p’ and ‘q’ can both be asserted on the basis of data that one has, and ‘p’ entails ‘q’ but not conversely, then, in a context where the hearer desires information, assert ‘p’. We should of course note in this context the rules with which we began this discussion, and which provide the Humean paradigm for linguistic rules. These are the conventions for the use of such expressions

420

Body, Mind and Self in Hume’s Critical Realism

as ‘I promise’ which do not report but, by virtue of those conventions, create new passions and feelings of obligation. Finally we should note the conventions by which thought, i.e., beliefs, are transformed, in the context of a relevant motive, into actions that are produced by the will as initiating means to achieve the desired end. These are “word-word” or “language-exit” transitions. All these conventions can be altered through deliberate choice in order to satisfy interests that we have. Some changes in syntax and semantics occur slowly, and unconsciously. But others are deliberate. Poets regularly make syntactical innovations to achieve new effects. Semantic rules are changed quite regularly. If one has a cognitive interest in matterof-fact truth, then that will lead one to discipline one’s thought to conform to the rules by which to judge of causes and effects, and to Hume's, rather than the Cartesian, ethics of belief. Or so Hume argues. What the need to communicate generates is the general interest that there be conventions of language. The fact that one is born into a linguistic community gives the accepted rules of that community an advantage at the gate over possible alternatives; but since we learn them so early, they become so deeply ingrained that one can only think of modifying them piecemeal, never of their total replacement. Neither political constitutions, as Livingston has argued,42 nor our ethics of belief, nor even the logical syntax of language can be evaluated from an a priori perspective and the possibility of revolutionary change contemplated. At least, they ought not to be so evaluated: for, if Hume is correct, there is no Plato’s heaven where we can find either an ideal republic, or an objective necessary causal tie, or any other sort of object for our transcendental Reason to grasp.

A central feature in Hume’s case against the Aristotelians and rationalists consists in arguing that thought, and our capacity to think abstractly and generally, is something essentially linguistic. Hume saw this himself when he wrote in his correspondence that “in much of our own thinking, there will be found some species of association. ’Tis certain we always think in language, viz. in that which is most familiar to us; and ’tis but too frequent to substitute words instead of ideas.”43 This point has been grasped fully by Baier, for one, and in this she may be contrasted to many other commentators, e.g., Peter Jones. Jones has remarked on the passage just quoted that

421

Five: Hume’s Positive Account

Unfortunately, the view that we think in language is left unsupported; unless he means by it that the expression of thought requires language, Hume did not mention any such view in his earlier philosophical work, and it may reflect his reading subsequent to it.44

But of course Hume did support the view that thought is language: the argument is constituted by his doctrine of abstract ideas on the one hand and his account of linguistic conventions on the other. Perhaps because these occur respectively in Books I and III of the Treatise, Jones has failed to make the connection. Jones locates Hume’s views on language in the context of French thinkers like Lamy, du Tremblay, and du Marsais, who held that linguistic conventions serve human needs.45 These thinkers were, however, of a piece with the old tradition, also found in earlier thinkers such as Descartes and Locke, in which language derives its signification from its expressing thoughts which are both non- and pre-linguistic. Always eager to locate Hume within a set of precursors, Jones misses Hume’s break with the precursors when he makes the radical innovation that thought does not precede but is language, and instead attributes to Hume the view of the old tradition with which Hume broke, that the role of language is merely that of expressing thought, and that language acquires its signification through expressing thoughts, that is, thoughts which must be prior to the language that merely expresses but does not constitute them. Jones in fact attributes to Hume the position of Descartes and Locke, that we clarify speech by turning to the thoughts or ideas which lie behind it, and give it its signification: No matter how many problems Hume leaves unexplored, his own position is moderately clear: talking is distinct from thinking, and most talk expresses thought; when confused by talk, we have to struggle to identify the thought behind it, and in the rarefied regions of philosophy we often find that such thought is itself confused or incoherent.46

Now of course, thought is different from talking, i.e., overt verbal behaviour or, if you wish, speech acts. Hume agrees: for a thought to be expressed in overt behaviour, an act of volition is, often at least, required. It does not follow, however, that since thought is not talk, i.e., speech acts, therefore it is not linguistic. It is also true that when we are attempting to communicate with others, and we are presented by another with a speech act the meaning of which is unclear to us, then the first step in attempting to understand that event consists in attempting to identify its immediate

422

Body, Mind and Self in Hume’s Critical Realism

causes, that is, the belief and the desire or intention that triggered the volition that produced the speech act. For Hume, however, there is a second step that is required if we are to grasp the signification of language: we must trace out the conventional or habitual links which it has to entities in the world that is given to us in sense experience, that is, to entities that satisfy the empiricist’s Principle of Acquaintance. ’Tis impossible to reason justly, without understanding perfectly the idea concerning which we reason; and ’tis impossible perfectly to understand any idea, without tracing it up to its origin, and examining that primary impression, from which it arises (T, pp. 74-5).

What this means is that, where the Cartesian programme, endorsed on this point by Locke, requires one to turn inward to discover the signification of language, Hume requires one instead to turn outward, to the world that language, through its syntactical and semantical conventions, tries to describe. For Hume we must examine not the fit of our language to our ideas but the fit of our language to the world. Jones refers to the passage just quoted,47 but fails to recognize its significance. As a consequence he ignores the second step that Hume prescribes for grasping the signification of what is said, that is, the tracing out of the habitual or conventional ties of semantics that link terms to what they designate; and instead Jones attributes to Hume the Cartesian and Lockean view that to grasp the sense of what a person says it suffices to turn to the “thought behind it.” But when Jones locates Hume in this tradition of Locke, Descartes, Aristotle and Plato, in which the meaning of language derives from thought which is non- and pre-linguistic, he simply fails to recognize that in Hume the transition from the other-world of Plato and Aristotle, of Descartes and Locke, to naturalism is finally complete. Annette Baier, in contrast, fully recognizes the leap that Hume has made beyond the previous tradition: “without language...there would be no generality, no abstraction...”.48 These conventions are part of the natural causal order, resulting by means of the associative mechanisms from natural causes. At the same time, these conventions are normative, where this normativity also has a naturalistic explanation. Baier emphasizes this naturalism of Hume in her discussion of Hume’s account of causation (p. 89ff). According to this, there are two definitions of cause. The first idea of cause is constant conjunction. The second is that of a determination of the mind to pass from the idea of the cause to that of the effect. Hume offers these definitions after an elaborate causal analysis of the origin of our

423

Five: Hume’s Positive Account

idea(s) of cause. The causal analysis itself exemplifies the patterns mentioned in the two definitions. “Causal inferences are what enable us to trace and recognize constant conjunctions, and their effect on us. Hume’s double definition of the causal relation itself displays a meta-causal relation, and that is what gives it its authority” (p. 91). Hume’s definition of the concept of cause is not confused, unlike those of the rationalists, precisely because he has traced out the causal ties, which are also semantic rules, which link the concept to things of which we actually have experience. In fact the causal analysis not only exemplifies the patterns stated in Hume’s two definitions, but it also exemplifies more specific patterns, to wit, those identified by Hume as the “rules by which to judge of causes and effects,” that is, in effect the rules of eliminative induction. Hume examines the many causes of true belief and error, and concludes that on the whole true beliefs are the causal upshot of inferences that conform to these rules, while other patterns of inference, those of “unphilosophical probability,” have error as their upshot.49 Causal reasoning in conformity to the rules shows us that conformity to those rules is the best means we have for achieving our goal of truth, the passion of curiosity. The rules are human norms tested through reflection upon experience. They are products of an exercise of our capacity to reflect upon our own experience and to devise means to achieve realizable cognitive goals, not a priori standards deriving from God or a world of forms. Once again we see the central role that is played in Hume’s philosophy by our capacity for critical self-reflection. Once again we come to the point where Hume’s concept of a person and his notion of personal identity come into contact with the central concerns of his epistemology and his ethics. (ii) Character and Personal Identity Promising is one way to create new passions, new cases where actions are obligatory. There are other ways in which this can be done. These involve the same performative aspect of language as is involved in the use of the words ‘I promise’ to create these new moral obligations and expectations. Thus, for example, a property owner saying “I bequeath so and so to you” by that speech act creates the rights of ownership of so and so to the other person, the obligations on others not to interfere with the latter person’s use of so and so. That use of “I bequeath” creates rights and obligations not

424

Body, Mind and Self in Hume’s Critical Realism

only in the speaker and the speakee, but also all others in the community. There are other acts, besides speech acts, which create these obligations. Thus, if there are things which are no person’s property, and one, as Locke says, “mixes one’s labour” with those things, then that act creates rights for oneself and obligations in others, the rights that attach to ownership and the obligations to respect that ownership. If one picks nuts from a tree situated on unowned land, then those nuts become one’s property. One may transfer the ownership of goods simply by saying “I give you this.” Or there may be a more complicated contract in which one transfers ownership in certain goods on condition that the contractee will reciprocally transfer at a later date ownership of other goods which he or she then has as his or her own. Exactly how these rules work will vary from culture to culture, society to society. Thus, it could be that the rule of ownership makes it obligatory that one may bequeath one’s landed property only to one’s eldest son – as Hume, a second son, knew full well. But there are other forms of land tenure. In some societies it is obligatory that when bequeathing one’s land one divide it equally among one’s sons. Or, in creating property rights by taking things from what is held in common may create rights only up to a point: convention may limit the amount that one may appropriate from the common for one’s own use. Again, there may be, through convention, limits on those to whom one may transfer ownership of one’s goods. Thus, in some societies one cannot sell rights to certain goods of national heritage to someone who is not a member of that society. In other societies, one cannot transfer property to someone of another clan, even if it lives just over the hill. We thus see how Hume’s basic model of social relationships, the relations of property and promising, provide the outline of a more general account – “model” if you wish – of social relations. The basic model is this. There are two roles, that of promisor and that of promisee. Two persons, ego and alter, enter into the roles of promisor and promisee when ego says to alter that “I promise you that I shall do x.” Humans early acquire the moral sentiment that people ought to keep their promises. That is, there is a general pattern of behaviour (@) Every promisor does for the promisee what was promised. which has attached to it moral sentiments such that people feel that (*) It ought to be that every promisor does for the promisee what was promised. These moral sentiments shared by all people are, of course, learned,

425

Five: Hume’s Positive Account

acquired through association and perhaps other psychological processes of learning, rather than innate as thinkers such as Clarke had claimed; but they are very deeply embedded in our human nature. The point is that the sentiment that people ought to conform to the norm (*) moves people to conform their own behaviour to it. Thus, when the promisor and promisee enter into their roles by ego promising alter to do x, ego comes to feel the moral obligation to do x, what was promised, and also feel that alter has the moral right to expect that x be done; similarly, alter comes to feel that ego is morally obliged to do x and to feel that he or she has a moral right to expect that x be done by ego. Since we feel that is it obligatory that all conform to this norm, we take care to raise children so that they conform to the rule (*); we put them in situations where they come to internalize this standard,50 that is, in situations where pleasure comes to be associated with behaviour conforming to this rule, both their own behaviour and also others’. Moreover, since we feel that it is obligatory that all conform to (*), when a promise is not kept we are liable to take steps to punish the violation, or at least we feel that it is morally appropriate to punish the violator. This basic model can easily be extended to other social relationships, e.g., kinship relations: (**) It ought to be that every person in the bear clan marries a person who is not in the bear clan. What is important about norms like (*) and (**) is that, provided the moral sentiments are in fact efficacious in motivating people, then these sentiments will bring it about that the pattern deemed obligatory will in fact hold in the group. Thus, if the sentiments attaching to (*) are efficacious, then the generalization (@) will as a matter of fact truly describe the people in the group. Similarly, if the sentiments attaching to (**) are efficacious, then it will be a true generalization about the group that (@@) Every person in the bear clan marries a person who is not in the bear clan. Rules such as (*) and (**) provide models that people have of their social relations. If these normative models are efficacious, then they will in fact be true models. Note that (*) and (**) are models in the minds of those in the social group. They are normative for these people. And in so far as they make descriptive claims (@) and (@@), these generalities are explanatory of the observed behaviour of members of the society.51 But these prescriptive

426

Body, Mind and Self in Hume’s Critical Realism

norms, which provide an explanatory model for those in the group, can also constitute an explanatory model for the scientist who is studying the group; the patterns they prescribe are explanatory not only for those in the group but for those studying the group. However, those studying the group must also take these patterns as the object of study; the social scientist thus not only uses the model but also studies it.52 Often he or she studies it in order to understand how change might be effected in the institution being studied. One’s cognitive interest in the institutional structure might be the disinterested concern of the research social scientist or the pragmatic interests of the reformer or administrator, but in any case does not require one to have internalized the norms governing the institution. For those who are participating in the institution the structure is not, or at least not merely, an object of study but a set of norms to which their moral sentiments move them to conform. In recognizing this, we also recognize a further explanatory element. The moral sentiment expressed by the normative statement It is obligatory that p brings it about that p That is, there is a causal relationship to the effect that (c) The sentiment that it is obligatory that p brings it about that p In this sense, (*) and (**) explain, respectively, the patterns (@) and (@@). In other words, we can explain the observed patterns by appeal to the causal efficacy of the moral sentiments that make such behaviour felt to be obligatory. The model in the minds of the members of the group thus not only explains in so far as it is descriptive, it also explains in so far as it is prescriptive. In fact, the latter explanation, since it is based on a more comprehensive causal principle (c), which takes into account more relevant factors, provides a fuller explanation of the observed behaviour. Yet more comprehensive explanations are possible. (c) holds only because of certain learning situations. What we in fact have is something like the following: (L) Whenever a person is in a learning situation of such and such a sort, then that person comes to be such that he or she feels the sentiment that it is obligatory that p. In order to explain why persons feel the sentiment, we turn to their past history, the learning experiences they have had. These experiences, together with the law (L), provide a (covering law) explanation of why

427

Five: Hume’s Positive Account

they feel the sentiment. This, when conjoined with (c), explains (via the covering law model) why they act as they do. The law (L) yields a yet more comprehensive explanation than (c). Of course, (L) as stated is just a sketch of a law or theory, specifically a sketch of a psychological theory of learning. For Hume, of course, the relevant theory of learning is associationism, or, behaviouristicaly re-stated, classical conditioning – with more than a touch of reinforcement theory. One should add, however, that, while associationism is the official theory, there are in addition roles to be played by the mechanism of sympathy, as we see in greater detail later. But the point that needs here to be emphasized is that Hume’s implicit account of social theory and of the methods for justifying its acceptance for purposes of explanation and of practice do not presuppose the commitment to any specific theory of learning. The account of social theory, of its justification, and of its relation to psychological theory does not depend upon the details of Hume’s own theory of learning: for these things all that is required is that there be some learning theory or other, some account of how the norms for social roles become internalized, how our moral sentiments become attached to these patterns of behaviour and not others. If the structural patterns (*) and (**), or what, for purposes of explanation, is much the same, the patterns (@) and (@@), are explanatory, then at the same time it is essential to recognize that the events that these patterns describe are the evidential basis for any claim or thought that those patterns in fact describe correctly the social reality. The observed social relations provide the data for the scientist attempting to ascertain the social structure.53 At the same time those relations are a consequence of that structure. Levi-Strauss seems to put the relationships in pretty well the correct order: observed social relations are the raw material in which social structural relations “inhere”54 and out of which non-statistical models are “constructed.” Levi-Strauss distinguishes “mechanical” from “statistical” models; he characterizes this distinction as one of level, but it is clear that with a “mechanical” model on the level of structure, there is associated a “statistical” model with which it is associated and which describes, with some degree of precision, the actual distributions of behaviours that conform and do not conform with the nonstatistical or mechanical model.55 Such structural models explain structural features exemplified in concrete social relations, and are “translatable” into statistical models.56 The structural models serve to explain the social relations that can be observed, with the approximation to reality of those

428

Body, Mind and Self in Hume’s Critical Realism

models measured by the probabilities in the statistical model. It is to be sure an approximation, but if the approximation is close enough then a degree of social control is possible. But why is there only an approximation? Why do the structural models not fit precisely the realities of social behaviour? It remains true that the norms that define social roles do, in their own way, provide explanations of human action and behaviour. But we must also recognize that these norms are not always efficacious. It is for this reason that the models are not always quite true. Most of the time most people keep their promises; to this extent (*) is efficacious and to this extent (@) is true. But not all promises are kept. A promise might be broken for good reasons; it could not be kept, for example, because one was delayed when one saved a drowning child. Or they may be broken for bad reasons; some people are knaves, and some even just forgetful. As Levi-Strauss once put it, models that are prescriptive in the minds of the members of the group may be preferential in practice.57 This means that (@) is only approximately true. What we have instead is (@') p% of the time promisors do for the promisee what was promised. where “p” is some fairly large percentage representing the probability that a promisor will keep his or her promises. The statistical generalization (@') holds because there are in fact variables that are not mentioned in (@) but which are relevant – those variables in individual psychology that in the real world of everyday life transform prescription into preference! What holds in fact is not the generalization (@), which has, as one says, “exceptions,” but rather (@") There are certain factors of a motivational sort such that, for every promisor, if they are absent, then that promisor does for the promisee what was promised. The law (@) attempts to leave no factors unmentioned; the law (@") in contrast asserts the existence of certain factors but does not mention them. Laws like (@) can be said to be deterministic, while laws like (@") have been called “gappy.” to use the term of J. L. Macke58 that we have used before. Laws that are gappy in just the way that (@") is gappy exist in many areas of science, but particularly in the social sciences. What they mark is the existence of certain variables that we do not know, but which we would like to know.59 For, clearly, laws like (@) are better than gappy laws like (@") for purposes of prediction and contrary-to-fact reasoning.

429

Five: Hume’s Positive Account

The ultimate aim of empirical science is a set of laws which have as few gaps as possible.60 Gappy laws therefore pose a research problem; the task of the research is to discover the factors that the gappy law asserts to be there but which we do not yet know, or do not know in all detail.61 Yet, until that task is complete we must make do with gappy knowledge. Or rather, we can at least approach the matter statistically, as in (@'), to enable us to estimate the extent to which we may reasonably bet upon the relationship holding in a given case.62 But Hume’s thought was of course formed too early for him to take seriously into account the developments in statistics that really began to happen only in the nineteenth century. As the case of kinship relations (**) makes clear, the norms defining social relationships can form a complex set. This set of norms defining the institution may itself exhibit a more abstract structure that it shares with a quite different set of norms defining a different set of social relationships. Levi-Strauss discovered that this is indeed the case with respect to certain kinship structures.63 But we cannot here further explore these points in the philosophy of the social sciences. We must return to Hume. Concerning character, Hume holds, quite correctly, that our psychological dispositions, our human sentiments and passions, are dependent upon social and economic conditions, and that changes in those conditions bring about psychological transformations. He characterizes the conventions that make social life possible as “natural artifices” (T, p. 484). These have developed “gradually and acquire force by a slow progression” (T, p. 490). Such changes in the social and economic structure redirect the human sentiments and passions in various ways. There is therefore, as Hume says, a “progress of the sentiments” (T, p. 500) as the artifices that structure society develop. Traits like “selfishness and confin’d generosity” (T, p. 495) are present in natural persons of the sort that exist prior to the emergence of civil society based on property relations. But as civil society forms through the institution of social conventions concerning property, contract (promising) and government, new traits such as honesty, fidelity, and loyalty, and also chastity, come into existence. And as these make possible the division of labour, so our natural abilities also develop (T, p. 606ff). The abilities that pre-social or natural man possesses are “slender” (T, p. 484); they need to be “augmented” (T, p. 485) if life is to be at all decent. This improvement in our abilities is achieved in the social context; it is “by partition of employments, [that] our ability increases” (ibid.). Among the relevant faculties of the soul are the intellectual, for example

430

Body, Mind and Self in Hume’s Critical Realism

the “faculty of placing our present ideas in such an order, as to form true propositions and opinions” (T, p. 612), as well as our manual abilities, for example, “agility, good mein... dexterity in any manual business or manufacture” (T, p. 279). Hume also mentions “industry, assiduity, enterprize, dexterity,” (T, p. 587) and later “industry, perseverance, patience, activity, vigilance, application, constancy” (T, p. 610). Besides “wit and eloquence” he includes “good humour” (T, p. 611). To “cleanliness” he adds “decorum” (T, pp. 611-2). Hume is quite correct in seeing these sorts of attribute as constituting defining characteristics that persons have of themselves. They define what we feel we ought to be, and, assuming – what is generally true – that we act on such moral feelings, what we therefore actually are. Exactly how these attributes come to be included in the idea one has of oneself is another issue – one to which we shall return. But at this point, we do not have to do anything more than note that these attributes do in fact appear as parts of the concepts that people have of themselves: they define one’s self identity. We shall return to this point below. What we have to do more immediately is to look at one specific capacity that defines a human self, namely, the capacity that we have to know ourselves. What, more exactly, is self-knowledge? (iii) Knowledge of Oneself The various dispositions and traits that constitute one’s character are, as we have seen, all to be understood as patterns of thought and action. The permanence of the self, then, if there be such, is the permanence of a pattern, not the permanence of an entity. The objection of Clarke and Butler to the attack on the substantial self was that it eliminated the permanent; for them the question was, as it was for Shaftesbury, “how [is] that subject continued one and the same?” (Characteristics, II, p. 275). Too wed to the substance tradition, too wed to the Christian doctrine of the self that had appropriated the substance account of personal identity, they could not see that there could be any permanence other than the permanence of an entity. What Hume makes clear, however, is that another account of the permanence of the self is possible. The important point is that the habits, the traits and dispositions, that constitute character are themselves learned; these patterns are acquired patterns. The laws of learning are the same for all persons; these patterns define a common human nature, and through them, together with different

431

Five: Hume’s Positive Account

conditions of learning, we can account for the diversity, and the stability of human nature. We acquire the various traits that define our individual characters through a process of learning when we grow into the roles that are open to us in society, when we choose a role like historian or lawyer or pool hustler, or when we come to play a role to which we are led or assigned or even fated by, say, the place of our parents in society. The skin, pores, muscles, and nerves of a day-labourer are different from those of a man of quality: So are his sentiments, actions and manners. The different stations of life influence the whole fabric, external and internal; and these different stations arise necessarily, because, uniformly, for the necessary and uniform principles of human nature. Men cannot live without society, and cannot be associated without government. Government makes a distinction of property, and establishes the different ranks of men. This produces industry, traffic, manufactures, law-suits, war, leagues, alliances, voyages, travels, cities, fleets, ports, and all those other actions and objects, which cause such a diversity, and at the same time maintain such an uniformity in human nature (T, p. 402).

What accounts for the stability of human character is, therefore, the fact of causal determinism, where by ‘cause’ one means, of course, “constant union” (T, p. 405). Whether we consider mankind according to the difference of sexes, ages, governments, conditions, or methods of education; the same uniformity and regular operation of natural principles are discernible. Like causes still produce like effects; in the same manner as in the mutual action of the elements and powers of nature (T, p. 401).

This means, of course, that the self in the sense of that which is permanent through diversity is created in and by the sensible world rather than being something that transcends that world, as in the substance and Christian traditions. Hume does not neglect to point out that both common sense and the Christian tradition tend to agree with his position rather than that of the substance tradition when they recognize the efficacy of rewards and punishments in shaping human character. “’Tis indeed certain, that as all human laws are founded on rewards and punishments, ’tis supposed as a fundamental principle, that these motives have an influence on the mind, and both produce the good and prevent the evil actions. We may give to this influence what name we please; but as ’tis usually conjoin’d with the action, common sense requires it shou’d be esteem’d a cause, and be

432

Body, Mind and Self in Hume’s Critical Realism

look’d upon as an instance of that necessity...” (T, p. 410). Indeed, if we assume that causal determinism does not hold true, then there would be no connection among the actions of our lives. On the view of the libertarians who deny causal determinism, the self would turn out to be a mere bundle with nothing permanent to it, and however good or evil those actions might in themselves be, there would be no person to hold responsible for them or to praise or blame. Actions are by their very nature temporary and where they proceed not from some causes in the character and disposition of the person, who perform’d them, they infix not themselves upon him, and can neither redound to his honour, if good, nor infamy, if evil. The action itself may be blameable; it may be contrary to all the rules of morality and religion: But the person is not responsible for it; and as it proceeded from nothing in him, that is durable or constant, and leaves nothing of that nature behind it, ’tis impossible he can, upon its account, become the object of punishment or vengeance....’Tis only upon the principles of necessity, that a person acquires any merit or demerit from his actions, however common opinion may incline to the contrary (T, p. 411).

Moreover, as Hume has explained much earlier in his discussion of cause in Book I of the Treatise, given that we are presented with no objective necessary connections in experience, the hypothesizing of a transcendental self as a power that produces actions will not help. ...it having already been prov’d, that the power lies not in the sensible qualities of the cause; and there being nothing but the sensible qualities of the cause; I ask, why in other instances you presume that the same power still exists, merely upon the appearance of these qualities? Your appeal to past experience decides nothing in the present case; and at the utmost can only prove, that that very object, which produc’d any other, was at that very instant endowed with such a power; but can never prove, that the same power must continue in the same object or collection of sensible qualities; much less, that a like power is always conjoin’d with like sensible qualities (T, p. 91).

The metaphysical conjecture of a transcendental substantial self is therefore without any practical importance; the only notion of the permanent self that is useful is that which Hume proposes, the permanence of a pattern rather than the permanence of an entity. We can say therefore that the unity of the self is not merely the unity of the body, though that is a crucial part of it. But the unity of the self includes also the unity constituted by the person’s enduring character. As

433

Five: Hume’s Positive Account

we have seen Hume say, for socialized persons, “our reputation, our character, our name are considerations of vast weight and importance” (T, p. 316). And so, in order to define the relation R that links the various stages of the self into a unity we need to take into account not only the laws that connect the states of the body with their predecessors and successors and states of mind with states of the body but also the permanent patterns of thought and action that define our character and our name, that is, make us who we are, the individual person we are – the laws that, in other words, describe the causes that give us our identity. The point is that these characteristics that constitute the self and its identity are patterns among patterns, features of the causal texture of the world and the persons in it. It is these that we must come to know if we are to know ourselves. For the substance tradition, knowledge of who or what we are is fairly easy to obtain. All we need is a rational intuition of our own form or nature. Thus, for Aristotle persons are rational animals. These are the dispositions and tendencies that yield the patterns of behaviour that are characteristic of us: one grasps the patterns by grasping the form or nature. It involves the insight of a moment, more akin to perception than to the inductive inferences of natural or empirical science. For Descartes self knowledge is much the same. The only difference is that he drops the animal part of our nature: for Descartes we are simply rational, that is, beings whose nature it is to grasp the reasons of things. This Descartes discovers shortly after becoming aware of himself in the cogito; a few paragraphs after his cogito, ergo sum he is able to conclude that sum res cogitans – ‘sum’= I am – ‘sum res’ = I am a substance – ‘res cogitans’ = a thinking thing, a rational being. For Hume, in contrast, knowing oneself amounts to knowing the regularities that describe and explain one’s behaviour. In order to discover the sort of person one is one must put oneself to the experiment, as it were. One must observe how one responds to various situations and in various contexts. One must form hypotheses about the sort of being one is, that is, the sort of character that defines one, and then put these to the test. It does not suffice to grasp one’s inner sense that one is, say, a generous person; one’s intentions are often an unreliable guide to the sort of person one is. One must also see how such intentions and feelings work themselves out in practice. Otherwise, the moral models that we use to describe human behaviour, including our own behaviour, would not be gappy regularities. One has the sense that one is such and such a person. Do others think so too? If not, then maybe I am wrong; via sympathy I tend to take up the

434

Body, Mind and Self in Hume’s Critical Realism

beliefs of others. It is no doubt to speak loosely to the effect that these conflicting beliefs constitute a set of hypotheses about oneself, hypotheses which can be put to the test. One can do this well, or poorly. Those who do it poorly likely know what the patterns are, but deceive themselves as to what sort of person they are, how they act towards and with others. Self knowledge is the acquisition of the capacity to correctly describe that patterns that govern one’s behaviour. Nor is such knowledge the achievement of a moment that is then completed, over and done with. For the laws of learning still apply and when one comes to be situated differently that different situation will causally affect and shape the patterns that correctly describe one’s behaviour. As one learns one must re-learn who one is; as one grows, one’s self knowledge, if we do achieve it, must change. Hume makes the passions central to the understanding of human nature. In having this general perspective on human nature, Hume is located more centrally in his age than if he is seen either as a mere sceptic or a mere utilitarian or contractarian. But it turns out that, if we locate Hume in this way, we see that the standard view of him as a conservative in the Burkean tradition is wholly misplaced: if there are conservative aspects to Hume's moral philosophy, there are also reformist aspects. As usually turns out with a great philosopher, his views are in fact more complex than standard commentaries will allow. What is crucial is in the first place, as we have indicated, Hume’s emphasis upon the passions. Now, in taking up the passions, Hume had been preceded by thinkers such as the Cambridge Platonists. Thus, Henry More, in his Enchiridion Ethicum,64 argued that the passions were good in themselves and “singularly needful to the perfecting of human life...”.65 He speaks of “these Natural and Radical Affections,” contrasting them to things that are acquired, particularly those things that are “the result of freethinking or speculation.”66 They are, he proposes, innate: “’tis manifest they are from Nature and from God.” From this it can be inferred, given the goodness of the Creator, that “whatever they dictate as Good and Just, is really Good and Just”67; in fact, they are part of the Law of Nature that “bears sway in the animal Region,” as “a sort of confused Muttering, or Whisper of ... Divine Law.” The passions, in other words, reflect the dictates of the objective moral order of the universe. To be sure, the same message concerning what we ought to do is “more clear and audible to the intellect.”68 Reason therefore can judge the less clear passions, but with

435

Five: Hume’s Positive Account

that qualification we can reasonably conclude that “we are bound to embrace and prosecute the same [i.e. what the passions dictate as Good and Just].”69 There were other thinkers who not only emphasized the passions, but also took up our sympathetic responses to the passions of others as central to any account of human sociability. Thus, William Wollaston argued that There is something in human nature resulting form our very make and constitution ... which renders us obnoxious to the pains of others, causes us to sympathize with them, and almost comprehends us in their case. It is grievous to see or hear (and almost to hear of) any animal whatever, in torment. This compassion appears eminently in them, who upon other accounts are justly reckoned amongst the best of men: in some degree it appears in almost all....It is therefore according to nature to be affected with the sufferings of other people and the contrary is inhuman and unnatural.70

Hume is squarely in this tradition that makes the passions, and, as part of the account of the passions, sympathy, central features of human nature. For Hume, sympathy is central to the passions, the passions merge into moral evaluation where sympathy is again central in transforming conventions into norms, and such conventions made normative by sympathy play a central role in Hume's positive account of reason and understanding. But sympathy also serves to locate Hume, as few other aspects of his philosophy do, centrally in his own age. In drawing attention to the crucial role of sympathy in Hume’s thought, Páll Árdal has done us a real service in helping us to see how Hume fits into the broader history of thought. But when we do this we also see how it is that the standard characterization of Hume as a sort of Burkean conservative is largely mistaken. The doctrine of sympathy places Hume in the middle of many important efforts at social reform. We can bring out the relevant point by noting Hume’s support for John Hume’s tragedy Douglas. John Hume, secretary to the Earl of Bute, had taken his play to London to offer it to Garrick for production at Drury Lane, but Garrick turned it down. It was subsequently successfully performed in Edinburgh, in spite of objections from the more Calvinist ministers of the Kirk. David Hume puffed it by dedicating an edition of the Four Dissertations (1757) to John Hume.71 As a consequence of the publicity concerning the Edinburgh production and David Hume’s efforts at publicity, the play was subsequently successfully performed at Covent Garden in London. It was reserved to John Hume, and to him alone, wrote

436

Body, Mind and Self in Hume’s Critical Realism

David, “To redeem our stage from the reproach of barbarism”72. “I am perswaded,” he wrote, “that there is not any Tragedy in the English language so well adapted to your theatre [his correspondent is the French Abbé Le Blanc] by reasons of the Elegance, Simplicity, & Decorum, which run thro’ the whole of it.”73 Douglas is “very pathetic, elegant, natural; entirely in the Taste of Greek Tragedy, & very wide of the Fustian and Rant, which have so much prevail’d in our theatre.”74 For our purposes, however, what is of most significance are the remarks Hume made in the “Dedication,” where he remarked that ... the unfeigned tears which flowed from every eye, in the numerous representations which were made of it in this theatre [in Edinburgh]; the unparalleled command, which you appeared to have over every affection of the human breast: These are incontestable proofs, that you possess the true theatric genius of Shakespear and Otway,A refined from the unhappy barbarism of the one, and licentiousness of the other.75

“... the unfeigned tears which flowed from every eye ...”: the characters in Douglas display certain feelings, and the audience, feeling those emotions, those passions, responds appropriately. Nor was David Hume alone in so responding to Douglas. A young Henry Mackenzie attended and he later recalled that I was then [1756] a boy, but of an age to be sometimes admitted as a sort of page to the tea-drinking parties of Edinburgh. I have a perfect recollection of the strong sensation which Douglas exited among its inhabitants. The men talked of the rehearsals; the ladies repeated of what they had heard of the story; some had procured, as a great favour, copies of the most striking passages, which they recited at the earnest request of the company. I was present at the representation; the applause was enthusiastic; but a better criterion of its merits was the tears of the audience, which the tender part of the drama drew forth unsparingly.76

The central place that sympathy has in Hume’s account of human nature is clear from the Treatise, but there are other places where Hume makes the same point. Thus, for example, he asserts of the “true sage and patriot” that in he (or she) “are united whatever can distinguish human nature, or elevate mortal man to a resemblance with the divinity.” The softest benevolence, the most undaunted resolution, the tenderest A

Otway was thought better of in Hume’s day than in ours.

437

Five: Hume’s Positive Account

sentiments, the most sublime love of virtue, all these animate successively his transported bosom77

The emphasis upon sympathy places Hume in the centre of a series of important developments in the history of human nature. This is the sentimental movement. The most important factor basic to this movement has correctly been said to be “the concept of the individual’s emotional state.” It postulated, and so encouraged, an ideal sensitivity to – and spontaneous display of – virtuous feelings, especially those of pity, sympathy, benevolence, of the open heart as opposed to the prudent mind.78

Henry Mackenzie was one of the leaders in this literary movement; his The Man of Feeling (1771) was a major work of the genre. Sir Walter Scott, in his “Life” of Mackenzie, speaks of the “tone of exquisite moral delicacy and of refined sensibility,” and indicates how Mackenzie “imagined a hero constantly obedient to every emotion of his moral sense.”79 The “principle object of Mackenzie,” we are told, ... has been to reach and sustain a tone of moral pathos, by representing the effect of incidents, whether important or trifling upon the human mind, and especially on those which were not only just, honourable, and intelligent, but so framed as to be responsive to those finer feelings, to which ordinary hearts are callous.80

Mackenzie’s novel was a great success. Such success could have been achieved only if his readers in fact had the capacity to respond sympathetically. This development of the capacity for sympathetic response became a central characteristic of the 18th century as it developed. It was important not only for how standards in literature changed as the century moved from the Augustans to the Romantics. It was important for a large number of significant changes in social thought and institutions, changes which are hardly conceivable apart from this change in human sensibility. This changed conception of human nature was important in the process of social reform in the late 18th and the 19th century. There were arguments for reform of various sorts in the society of England and of the United Kingdom. These arguments took various forms, among them the appeal to sympathy. It is this appeal to sympathy that is important for our purposes: reformers would not have used it unless it unless they reckoned

438

Body, Mind and Self in Hume’s Critical Realism

that it would likely have some effect in persuading their listeners. Consider some examples. There is, for one, the doctrine of eternal damnation of the sinful in Hell. It was the traditional doctrine that both God and the blessed enjoyed the pain suffering of those damned for their sins to eternal punishment in hell. This was criticized on the grounds that neither ordinary persons nor a benevolent God, as persons feeling sympathetically the pain, ought to be able to enjoy, let alone tolerate the eternal suffering of the damned. As caring persons we ought to feel misery upon the thought of the suffering of the damned, and ought therefore to reject any dogma that entailed such a monstrosity. Then there is, for two, the use of hanging as a punishment. This was defended as a deterrent. Such a defence was condemned on the grounds that with regard to those who viewed such spectacles the capacity for sympathetic response to the suffering of others was reduced. Hume had used precisely this argument to attack the institution of slavery as it was practiced in the ancient world and as it was practiced in the American colonies. He wrote that The chief difference between the domestic economy of the ancients and that of the moderns consists in the practice of slavery, which prevailed among the former, and which has been abolished for some centuries throughout the greater part of EUROPE.... As much as submission to a petty prince, whose dominions extend not beyond a single city, is more grievous than obedience to a great monarch; so much is domestic slavery more cruel and oppressive than any civil subjection whatsoever. The more the master is removed from us in place and rank, the greater liberty we enjoy; the less are our actions inspected and controlled; and the fainter that cruel comparison becomes between our own subjection, and the freedom, and even dominion of another. The remains which are found of domestic slavery, in the AMERICAN colonies, and among some EUROPEAN nations, would never surely create a desire of rendering it more universal. The little humanity, commonly observed in persons, accustomed, from their infancy, to exercise so great authority over their fellow-creatures, and to trample upon human nature, were sufficient alone to disgust us with that unbounded dominion. Nor can a more probable reason be assigned for the severe, I might say, barbarous manners of ancient times, than the practice of domestic slavery; by which every man of rank was rendered a petty tyrant, and educated amidst the flattery, submission, and low debasement of his slaves.81

We here see Hume using the appeal to sympathy to argue against certain forms of social order. Then, for a third example, we find similar arguments being used to defend humane treatment for animals.

439

Five: Hume’s Positive Account

The point is, to repeat, that these arguments all presuppose the premise not only that people can and do respond sympathetically to others, but further that it is a good thing, socially and psychologically, that they be able so to respond to others. The development of the human capacity for sympathy became a major characteristic of accounts of human being in the 18th and 19th centuries. Hume was central in bringing this human capacity to a place where it was central to our view of human being. This recognition in turn became one of the major grounds for arguing for social reform. Thus, if we do not notice the central importance of this capacity in Hume’s account of human nature, then we fail to locate this philosopher within the mainstream of the reforming movements of the 18th and 19th centuries. Hume is often portrayed as simply a conservative thinker: what else could one be who insisted that the roots of social living lay in convention and habit? But in fact the case for social reformer was built to a large extent upon an appeal to sympathy, an appeal that took for granted that the capacity for sympathetic response was central to our being and our social being. It is important to see Hume as providing a theoretical account of human nature in which sympathy plays a central role. For that places Hume in the very centre of a movement of social reform that is anything but conservative. In reading Hume we should give sympathy a sufficiently central role that we can begin to properly locate Hume in his age. Páll Árdal, in his discussion of punishment, in the essays “Some Reflections upon a Standard Definition of Punishment” and “Does Anyone Ever Deserve to Suffer?”82, brings out in detail the case against moralistic retributivism that was developed in the context of arguments about the death penalty and the suffering in hell. The view of Hume and of the 18th and 19th century activists who argued for reform is brought back to our attention by Árdal when he states that “An individual who always does the right thing, but never gets upset by cruelty and injustice, is not a human being, but a moral monster.” He goes on, David Hume was right in stressing the importance of the emotions in our moral attitudes, although he may not have given a correct account of these emotions. But the fact remains that the morally better a person is the more likely he is to suffer at the thought of ugly cruelties and injustices in this world.83

And, one might add, the next. So much the worse for Aquinas’ view that the blessed enjoy the suffering of the sinful in hell. Árdal goes beyond this to suggest one of the origins of the concept of

440

Body, Mind and Self in Hume’s Critical Realism

Hell. In the essay, "Does Anyone Ever Deserve to Suffer?", he points out that The fact of the matter is that it is contrary to the nature of a decent person to deliberately torture or make to suffer a sentient creature and in particular another human being. As we all know only too well this aversion to imposing suffering on others is indeed a tender plant too easily killed if not given suitable soil in which to grow. Those decent people, and there are many, persuaded of the legitimacy of the concept of ill-desert try to remove themselves as far as possible from actual involvement in the practice of making others suffer. They thus hope that there is a Hell where retributive justice is done.

That hope is the father of the belief. There are further connections that Árdal makes. The notion that suffering in hell must be eternal is tied to the Christian idea that God had to sacrifice his only Son in order to redeem the wicked. Since the Son is in effect God himself, equally and wholly God within the Trinity, it follows that the sacrifice that God must make in order to redeem the wicked is infinite. The sin is therefore infinite. too, if the balance is to be maintained, and the punishment of unredeemed sin must be eternal. In the late 17th century, the progress of empirical reason led such English thinkers as Locke and Newton to move towards unitarianism, and therefore towards an Arian view of Jesus. Since Jesus was less than God, and therefore less than infinite, it follows that the sacrifice of Jesus was not infinite nor therefore need the punishment for unredeemed sin be construed as eternal. The progress of empirical reason about God and the Trinity thus leads directly to a position that enables thinkers to re-evaluate the traditional doctrine of eternal punishments. But Locke and Newton still clearly assumed the Christian doctrine that God did indeed sacrifice Jesus. Árdal, however, has raised the issue whether this doctrine of another paying for my sins makes any moral sense at all. Anyone can pay my debts on my behalf. But does it, in the ultimate analysis, make any sense to suggest that another person can wipe away my moral guilt by suffering on my behalf? Even in the case of legal punishment it is deemed necessary to secure that the guilty should be punished. We do not take communal responsibility for illegal actions comparable to the way in which any member of a hockey team can serve a bench penalty for too many men on the ice. But our moral tradition contains such a strong link between guilt and suffering that someone else is deemed to be able to gain salvation for us through his suffering.84

The notion of Jesus dying for one’s sins makes no sense, nor therefore the

441

Five: Hume’s Positive Account

idea that we must go to hell and suffer if we do not give our souls to God and allow Him to redeem them with His sacrifice of His only Son. Árdal continues: I can only repeat that this seems to me, from a moral point of view, an irrational doctrine and that a change in character is the only way in which a person can make up for evil perpetuated.85

What is important, then, is reform of the guilty, a change in their character that will have the consequence that they will not commit their past sins and crimes again. You compensate for sin by doing what is necessary to make you such that you will no longer commit those sins. Now, the traditional doctrine of Hell is that our fate is determined by the short span, compared to eternity, during which we live on earth. Nothing subsequent that happens can alter our fate: if we die unredeemed, unaccepting of God’s sacrifice, then we suffer for eternity, justly suffer. There is no further chance for redemption, no further chance for reform. But, Árdal is arguing, what is morally important about punishment is precisely the possibility of reform. Since the traditional doctrine of hell allows for no such possibility, so much the worse for that doctrine. Such arguments no doubt now have rather less of the urgency that they had in Hume's day, partly due to the influence of Hume himself. But they remain important. There are many who still accept the traditional doctrine of hell, many more who accept crude retributivist accounts of justice and suffering. It is important to keep in mind, with Árdal, and (one would like to think) with Hume, the Humean moral point of view that critically rejects these views as not merely wrong but deeply immoral. What is important to these considerations is the role of sympathy. Our human capacity to respond sympathetically to others, central to a Humean ethics, is in fact a feature of human nature which, while morally relevant, is often neglected by contemporary moral theorists. Standardly, Hume is taken to be merely a contractarian or merely a utilitarian. Our capacity to respond sympathetically to others is important for a variety of issues, but for medical ethics in particular. Here it is important to notice some of the limitations of Hume’s explicit account of sympathy. Thus, Hume’s detailed account of the mechanisms of sympathy does at times imply that when we respond sympathetically to others we do experience just the pain or just the pleasure that they are experiencing. It important to recognize that “If understanding other people’s suffering through sympathetic imagination necessarily involved having in some form

442

Body, Mind and Self in Hume’s Critical Realism

the experience sympathized with, this would clearly impose an intolerable burden upon most health-care workers.”86 As we earlier noted from Al Gazzali, the physician aims to understand the drunken man, but for that he or she does not need to be drunk. At the same time, one must also recognize that “Death cannot be shared. But through the exercise of sympathetic imagination, the experience of dying can be entered into sympathetically and this helps both to make dying easier and teaches the one who is going to be left behind something about life”, and that “sympathy with sorrow makes it easier to bear.”87 It is also important to note that, contrary to what Hume himself sometimes seems to suggest, sympathy has a cognitive function in making known to us the felt pains and pleasures, the felt passions and sentiments of others. There is a general point here, but reference can be made in particular to care-givers in a health context. The Humean perspective, on the importance of sympathy, is relevant to several more recent concerns in this area. I have in mind in particular the concerns raised by Carol Gilligan in her study In a Different Voice.88 These concerns have recognized as important for the ethics of care and for biomedical education, for example by Alisa L. Carse in her essay “The ‘Voice of Care’: Implications for Biomedical Education.”89 Gilligan argues that one can distinguish two orientations within moral thinking. The first of these is the “voice of justice,” or the justice orientation in ethics. Gilligan indicates four essential features to this perspective. First, it construes the moral point of view as an impartial point of view. Second, it considers particular moral judgments to be justified only if they can be derived from abstract and universal moral principles. Third, it views morality as essentially dispassionate rather than passionate, disinterested rather than requiring emotional involvement. And fourth, it emphasizes individual rights and norms of formal equality and contract or reciprocity as the basic models for understanding moral relations. It is clear that the basic thrust of this point of view, as it would be put by philosophers, is that of contractarianism. To this Gilligan contrasts the “voice of care” or care orientation in ethics. The essential features of this orientation are these. First, moral judgments are best construed as perceptions in specific situations of the needs of others, and as requiring a steady adjustment in conformity with the dynamics of the social interactions. Second, such judgments are not disinterested but require, rather, both concern and sympathy or empathy. Third, moral judgments are justified not by universal rules but by the needs

443

Five: Hume’s Positive Account

that are made evident in the sympathetic responses to concrete situations. And fourth, it emphasizes individual responsibility and the norm of being responsive to the needs and feelings of others as the basic model for understanding social relations. In particular, it ought to be noted that the care orientation emphasizes responsiveness to individuals. It is possible to take a utilitarian orientation in ethics, as well as a purely contractarian one. In fact, one can argue that the contractarian’s rules find their justification in the requirements of general utility or, what is the same, general benevolence. This sees the rules of the contractarians as justified only because they promote the wellbeing of all. But the general principle of utility or benevolence requires that we go beyond these rules of the social contract where it is required by the situation for the end of relieving the distress of others and, more generally, of increasing their welfare. In this respect, the general utilitarian position goes beyond contractarianism. Nonetheless, the care orientation is also critical of utilitarianism as well as contractarianism. What the care orientation emphasizes is not a concern for a person simply as a way of increasing the general good; it is not concerned with others simply as abstract objects as it were, objects whose welfare is being promoted only because it promotes the general welfare. Rather, the care orientation insists that what is important is the sympathetic response, the caring attention, to concrete others in concrete but changing and developing situations. Carse points out that this care orientation in ethics can find its roots in Aristotle, insofar as, upon Aristotle’s view, virtue consists in both dispositions to have certain passions as well as to act in certain ways; it requires that one feel the right emotions “at the right times, with reference to the right objects, towards the right people, with the right aim, and in the right way.”90 But she also points out that this orientation has a “kindred spirit” in David Hume, who she says “views corrected sympathy, not principled reason, as our basic moral capacity.”91 Now, this positioning of the care orientation within the context of Hume’s thought is very useful. As simply stated by Carse, however, it is decidedly misleading. For, Hume emphasizes the role of contractarian issues in ethics as well as those of sympathy. If the latter are important, so are the former. Indeed, insofar as the latter can form the basis of a moral community, a set of moral relationships, it is only because, according to Hume, they are in fact not only useful but also rooted in sympathy. The relevant point can be brought out by considering another remark of Carse:

444

Body, Mind and Self in Hume’s Critical Realism

Whereas we are, on the justice orientation, viewed as individuals first, and in relationship to each other only secondarily, through choice, we are, on the care orientation, understood as essentially in relationship, though no single kind of relationship is endorsed as alone morally paradigmatic.92

The distinction between “essential” and non-essential appears to be that the former is not a matter of choice, where the latter is. But this is misleading so far at least as Hume’s account of the social relations of justice (property) and promising and contract are concerned. It is of course true that one can choose whether to enter into a contract to transfer property, and one can choose which property to transfer. But neither the social relations of property nor the social relations of promising and contract are a matter of choice. In fact, it is central to Hume’s argument against Hobbes and Locke that the conventions of civil society are not promises, not matters of contract. And not being matters of contract, they are not matters of choice. Thus, it is a matter of choice whether I say “I promise that I will do X,” but, upon saying those words, it follows directly, without any matter of choice, that one is thereby obliged to do as one promised. Similar rules hold for property or justice. In this sense, the bonds of civil society as a moral community are as essential as those that derive from sympathy. Moreover, the supposed contrast between the justice ethics, which relates one only to an abstraction, and the care orientation, in which one is related to an individual, is misleading. It is true that on the justice orientation, moral judgements are justified by reference to general rules: “I ought to do so and so because everyone ought to do so and so.” The rules are general because, as Hume insists, the conventions of civil society would not do their job unless they are extended to, and upheld by, all. But when I justify an action towards you by reference to a general rule – “I ought to provide you with health care because I so contracted, and people ought to keep their contracts” – I do not cease to be oriented in my action towards you, rather than some abstract part of you. I am indeed oriented towards you by virtue of some characteristic that you have, one (complex) characteristic from among a much larger and inter-related set. But that does not mean that I am not oriented towards the person who is defined as an individual by that much larger set. It is the same with care based on sympathy. When I respond sympathetically to you and act to relieve your distress, that response is made with regard to only one (complex) characteristic – the characteristic of being in a certain sort of distress, a

445

Five: Hume’s Positive Account

characteristic which in its complexity includes certain passions or feelings. But that does not mean that I am not oriented towards the person who is defined as an individual by that much larger set. The two cases are parallel: the judgments are no more related to mere abstractions in the one than in the other. Gilligan writes that “As a framework for moral decision, care is grounded in the assumption that the other and self are interdependent.”93 But this holds equally for the justice orientation, at least upon Hume's account. For, upon this account we conform to the conventions of civil society, the conventions of justice and contract, only because by so conforming we can all of us achieve a life that is in terms of material conditions tolerably decent. It, too, assumes that “the other and self are interdependent.” Relationships and interdependence are part and parcel of both the care and justice orientations. Contrary to what she suggests, they cannot in these terms be distinguished. Nonetheless, having said that, it must also be said that there is a point to what Gilligan and Carse say, even from Hume’s point of view. For, after all, the conventions of civil society are artificial, the product of art, and are in that sense a mater of choice. In contrast, the social relations that derive from sympathy are not artificial but natural. The conventions of civil society arise because people live in conditions of scarcity. As Hume argues, if these conditions were absent, then we would have no need for these conventions nor, therefore, for the artificial virtues of civil society. But we would still respond sympathetically to others, and would still form various social relations. We would, in other words, to use Hume’s terms, still have the natural virtues. Since these virtues presuppose only the mechanism of sympathy and not the conventions of civil society, naturally virtuous action is prior to choice, prior to art, prior to rational calculation. In that sense it is, perhaps, “essential”, in Carse’s sense. Nonetheless, civil society, Hume argues, naturally becomes a moral community through precisely the same mechanism that makes the natural virtues virtues, namely, sympathy. It must be emphasized, as Hume insisted against Hobbes and Locke, that the rules of civil society, the conventions of the contractarian, are in the end not merely conventions but also moral, and that this making of conventions into moral rules is rooted in the very same fundamental feature of human being that leads us to respond sympathetically to the pains and the pleasures of others. Gilligan suggests that the justice and care orientations “denote different ways of organizing the basic elements of moral judgment: self, others, and the

446

Body, Mind and Self in Hume’s Critical Realism

relationship between them.”94 The point is, however, that they are not just different ways of organizing our moral judgments, but rather ways which are part of a unitary whole. This whole consists of the passions and our sympathetic responses to the passions of others, it consists of needs and interests and sympathetic responses to the needs and interests of others, it consists of conventional behaviour recognized by art as conducive to satisfying the needs of all and sympathetic responses to sentiments and actions that conform to these conventions. Passions, needs, interests, conventions: all essential to human beings, all tied together in the consciousness of human beings by the basic mechanism of sympathy. In this sense, in contrast to what Carse suggests, there is no contrast between the self of the justice orientation in morality, on the one hand, and the self of the care orientation, on the other. Both are misleading abstractions from the single self created by the unifying mechanism of sympathy. By linking passion and value in Hume’s Treatise, Árdal has in fact brought out this point.95 Having said this, there are other things that also need saying. Human beings are complex, and so are the conventions governing their behaviour. Árdal has made the following important point in his essay “Of Sympathetic Imagination”: We are generally taught not to express certain emotions, particularly distress, except in grave emergencies. I heard not so very long ago a university professor say that he thought that the sight of a grown-up man crying was disgusting. He obviously did not think this was a point likely to be disputed, and I suspect that most of the audience probably agreed with him.96

There are, then, social conventions that can lead to suppression of the behaviour that evokes sympathetic responses, and moreover conventions that require us to respond unsympathetically to the expression of certain passions. This is certainly true. Whether it is useful is another thing. Árdal goes on to remark: But why should we not cry if we have something to cry about? Those who are able to have a good cry seem to be the better for it.97

Again, he brings out, in the same essay, another point relevant to care givers, that Those who favor weaving a web of deception around the dying very often

447

Five: Hume’s Positive Account

forget that it is those who are going to be left behind that may be ultimately needing to be comforted. To be able to comfort those who are going to be left behind may give meaning and dignity to the dying person. He or she needs to understand that the memory of how you die is going to live on after your death. Thus the dying as well as those who are left behind need to exercise their sympathetic imagination.98

Our morality of care and convention may form a unified whole, but it is a whole with tensions and conflicts. The conventions of civil society can lead to the distortion of sympathy. But this was already known in the 18th century. Bishop Butler was one of the most shrewd of the English moralists, and he clearly recognized that the repeated occurrence of a compassionate response has a double effect. On the one hand, by creating and strengthening a habit of providing relief and solace it perpetuates a virtuous activity. On the other hand, such repetition will also have the effect of weakening the sympathetic response that is the original impulse to care giving. For Butler, this is the result of the general principle of human psychology that repetition strengthens active habits while weakening passive impressions. “[P]assive impressions,” he points out, “by being repeated, grow weaker.” Repetition may therefore ...harden the mind in a contrary course, and render it gradually more insensible, i.e. form a habit of insensibility, to all moral considerations.99

The result is that compassionate care becomes insensitive routine. Perception of distress in others is a natural excitement, passively to pity, and actively to relieve it: but let a man set himself to attend to, inquire out, and relieve distressed persons, and he cannot but grow less and less sensibly affected with the various miseries of life, with which he must become acquainted; when yet, at the same time, benevolence, considered not as a passion, but as a practical principle of action, will strengthen: and whilst he passively compassionates the distressed less, he will acquire a greater aptitude actively to assist and befriend them.100

The result here is disinterested habit, mere routine and convention. It is in fact perhaps something like this that Gilligan and Carse have in mind when they speak of the “justice orientation” in ethics. For, as they characterize this orientation, it requires value judgements to be disinterested and dispassionate. For Hume, this is misleading. On his view,

448

Body, Mind and Self in Hume’s Critical Realism

the conventions of civil society are, of course, based on certain material interests – which means that they cannot be understood apart from their utility in satisfying certain needs and interests, that is, certain passions. But further, these conventions, Hume argues, are also moral, and in that sense they also involve our moral sentiments, our moral passions. So there is little sense to the claim that the rules of justice and promising and contract are dispassionate and disinterested. Such conventions, such habits, come to have those features only when repetition leads to their becoming mere routine. But insensitivity and lack of passion are precisely those features that attach to moral standards upon the justice orientation, according to Gilligan and Carse. So perhaps they are mistaking these habits of insensitivity for the rules of justice required by the more complex account given by Hume. This suggestion is supported by the remark of Carse that “On the care interpretation, what is sought in ethical debate is not so much theoretically neat, universally justifiable solutions to moral conflicts as shared interpretations of problems and collective success in mediating and balancing the different moral claims and concerns various parties to a case express.”101 The implication is that the rules of the justice interpretation must perforce be insensitive to the felt needs and sentiments of the persons involved in the situation. This is of course not true of the rules of civil society as Hume develops them. Moreover, it is also true that those rules do not apply in all cases. As Hume indicates, besides the artificial virtues of civil society there are also the natural virtues. Resolving conflicts can be a demanding task. One would suppose the justice orientation to require anything other than this only if one mistook the orientation to insist that the norms of justice are insensitive routines, rather than what they in a more fully worked out version, such as Hume’s, which places the moral rules of justice within the context of a full account of human nature. In any case, Butler makes clear one of the great problems of human nature, how to maintain both our compassionate responses and moral sentiments, on the one hand, and, on the other, convention, routine and bureaucracy. Árdal makes a similar point when he remarks in his essay “Of Sympathetic Imagination” that No doubt sympathy with suffering may become easier if one has suffered a great deal and learnt to bear it with fortitude. Quite the contrary. If one has suffered a great deal and learnt to bear it with fortitude, one may become less understanding and tolerant of the suffering of others.102

449

Five: Hume’s Positive Account

He also makes the point of Carse that the care orientation is of particular relevance in medical ethics. As Árdal puts it, “No doubt doing you duty from duty in good Kantian fashion is a vital element in all efficient health care, but I do believe that something like the sympathetic imagination is needed both to make the caring for the sick more thoughtful and more emotionally adequate.”103 The point is, justice and care are both essential. The virtues of Árdal’s approach to sympathy are those of` Hume. Where Gilligan and Carse take a very limited view of human nature and contrast an ethics of care, on the one hand, which is supposed to deal with individuals, with an ethics of justice, on the other, which is supposed to deal only with abstractions, Árdal develops the Humean position that these are simply two aspects of morality and must both be fit into a unified picture of human nature. Árdal has here made an important contribution to current debates in ethical theory – debates which have relevance to issues in medical ethics but also in broader concerns. Rather than separating aspects of human being as do Gilligan and Carse, Hume, and following him, Árdal brings them together so that the tensions and conflicts to which Gilligan and Carse draw our attention can find a reasonable solution. It is in fact hard to imagine how we can make any progress in settling the conflicts unless we follow this line of thought about human being. In no other way does it seem likely that we will be able to find a solution that will be, as Hume insisted that it should be, reasonable, that is, reasonable in the Humean sense. This sense of a reasonable solution requires a sort of balance. This balance can derive from reason understood as a calm passion – “that reason which is able to oppose our passions, and which we have found to be nothing but a general clam determination of the passions, founded on some distant view or reflection” (T, p. 583). The person reflects upon him- or herself, and asks about the sort of character he or she is, about his or her identity. Perhaps he or she finds it wanting in some respect, relative to the longer aim of personal happiness and contentment. It then becomes possible for that person to act upon this long term desire and re-shape his or her identity, an exercise in character formation. Thus, for example, the person might reflect upon the self that he or she is and recognise that his or her conflicts with others arise not from any basic immorality, not from any serious tendency to violate moral rules, but rather through insufficient sympathy with others. There is perhaps a tendency of the sort noted by Butler of one’s practice of morality to become routine, a matter of habit – mere habit. Nothing immoral is being

450

Body, Mind and Self in Hume’s Critical Realism

done, nothing essentially vicious. But morality has in a way become external to the feelings, and is done without passion and even without thought. There is a coldness of heart in one’s moral practice. In fact, one can imagine one’s set of moral principles that one lives by, one’s moral code, becoming in this way automatic, becoming stale and lifeless, without passion and without compassion.104 That this is so, or could become so, is implied by the view of many moral theorists that the point of developing a moral theory is to provide a rule or algorithm that can decide any moral decision, or indeed, any decision whatsoever: everything comes down to morality and there is a rule for deciding any moral issue. The notion that this could be so is suggested by the complaint that moral theory draws too rigid a distinction between our passions or motives on the one hand and our reasons for action on the other.105 This reminds one of the distinction made in the area of medical ethics by Carse and Gilligan between acting out of care and acting for reasons of justice, only made more general. The distinction is on the surface spurious. For, reasons are motives and like (other) motives are rooted in the passionate or feeling side of human being. But the suggestion is that when we act solely on those motives we call reasons, we are acting with a certain coldness and indifference, without passion and without compassion: we are acting as if we are not really motivated. This then becomes an objection to what is called moral theory, and perhaps it is. But it is more an objection to any concentration on a set of principles of justice to the exclusion of our human feelings, including most importantly our human feelings of sympathy. If Hume is correct, then moral rules of justice apply primarily in areas where there are scarce resources, primarily in the realm of material goods. But in fact there has accrued to these rules a welter of other injunctions. Often these injunctions are rooted in superstition. The human mind, fearful of the unknown and fearful of death craves protection from the ravages of nature and from the ravages of our fellow humans. The result is the creation of the illusions of religion. In our fantasy life we invent the idea of protector entities, gods, who can be manipulated but must also be placated. And in our anxiety we treat these illusions as reality. These gods or God will protect us from nature and from those in other tribes provided we conform to His or Her rules defining our relations to Him or Her and to our fellows in society, that is, our own society. Since these rules are rooted more in fantasy than reality, they are fairly arbitrary relative to the distribution of scarce resources. (However, as Hume would point out, priests in those religions that are characteristically superstitious,

451

Five: Hume’s Positive Account

often require more goods than are allotted to the ordinary people of that society.) Moreover, not only are rules of this sort fairly arbitrary, but they extend for no good reason to cover many if not all our social and religious interactions. However, arbitrary though they may be, they come to have deep moral force. And they, too, can become routine and mechanical, constraining those who conform to them, constraining with the sense that they are not really one’s own, but coming form the outside. (Think about Kant on the deep evil of masturbation. Can it really be true that it is condemned by our pure practical reason? Can it be true that this reason finds out that built into the objective structure of the universe is the rule that would forbid one the simple harmless pleasures of masturbation?) It is reason that can free us from these arbitrary bonds. Reason can show us that many of the rules that bind us are not really necessary for living together in civil society. Reason can show us that the gods that transcend the world of ordinary things but nonetheless act into that world to shape our affairs are illusions: there is no world beyond the world that we know by means of our senses. And scientific reason can show us the roots of these irrational practices in the depths of our human being. Reason – Humean or scientific reason – can explore scientifically – using what Locke called the “historical plain method” (Locke, Essay, Bk. I. Ch. 2, sec. 2) – the causes of our fantastic and illusory beliefs in religious entities. The result is a natural history of religion,106 as Hume called it, or a genealogy of morals, as Nietzsche called it.107 Those roots of all moralities are in the human passions. Not only is it the passions that yield our moral sense, and felt moral rightness of the rules and conventions of civil society, but there are also rules that determine one’s behaviour relative to God or the gods, and the latter ideas can also be found to be rooted in our passions, e.g., the fear of death. Hume argued this point, and Nietzsche recalled it: “It has gradually become clear to me what every great philosophy up till now has consisted of – namely, the confession of its originator, and a species of involuntary and unconscious auto-biography; and moreover that the moral (or immoral) purpose in every philosophy has constituted the true vital germ out of which the entire plant has always grown”108 The genealogy or natural history of morals uncovers those sources of morality in the human passions. Many of the rules of ordinary morality are odd and, it would seem, hardly necessary for human well-being – including those features of morality that are supposed by the philosophers to be found in some spiritual realm of absolute values. Tracing the roots of these prejudices scientifically, we can, it is to be hoped, free ourselves from the bondage of

452

Body, Mind and Self in Hume’s Critical Realism

these rules and create for ourselves a more truly human life and society. Nietzsche109 found the roots of his thinking in the thought of Hume.110 Like Hume, he rejected as non-entities anything that violated PA, and claimed a purer reality that transcends ordinary experience. Philosophers introduce Forms of things and things-in-themselves as the realm of the really True, the Truth that judges our knowledge of the ordinary world as second best and, when compared to the Truth, fundamentally misleading about reality, error in other words. Nietzsche puts the point in a way that Hume would never put it, suggesting that since we reject Truth, that is, transcendent realities, we therefore live in error and falsehood. For, it is error only when compared to the transcendent, and if that does not exist it can provide no standard by which one could, as a reasonable person, judge that the world in which we live our ordinary lives is but a shadow, a fake world and a deceit. The truth that we can achieve in our ordinary lives is the only truth there is and therefore will not be rejected by the reasonable person as less than real. But Nietzsche liked the paradoxical. So he tells us how philosophers and defenders of the social order locate our moral code “In the womb of being, ... in the intransitory, in the hidden god, in the ‘thing in itself’...” (BGE, 2, p. 14). But “why not rather untruth? And uncertainty? Even ignorance?” (BGE, 1, p. 13). Only for one who argues for absolute value in eternal forms and natures of things will argue that the world of experience is a world of untruth and falsehood. Abolish that world of forms and the world of ordinary experience is all that there is: there is nothing to contrast it with that could lead one to judge that it is a realm of untruth. So, for one who rejects the world of forms, the world of ordinary experience is not in itself and as a whole a world of error and untruth Nietzsche’s way of putting his point is not just an exaggeration, it is mere rhetoric, and, indeed, misleading rhetoric. Nonetheless, there is a point to that rhetoric: one can recognize in it what are the roots of the moral indignation which greets the empiricist rejection of transcendent realities. Among the transcendent realities that Nietzsche rejects is the simple self of the Cartesian.111 He does not, however, characterize the empirical self in the Humean way as a “bundle of impressions.” “One must,” Nietzsche tells us, “... first of all finish off the other and more fateful atomism which Christianity has taught best and longest, the soul atomism” (BGE, 11, p. 42). It is, rather, a bundle of impulses, which are rooted in our material being, and which are, moreover, socially structured: “In all willing it is absolutely a question of commanding and obeying, on the basis ... of a

453

Five: Hume’s Positive Account

social structure composed of many ‘souls’...” and is intimately tied to our body – in fact, not only is “our soul” a “social structure of drives and emotions” (BGE, 12, p. 44), but further so is “our body ... only a social structure composed of many souls” (BGE, 19, p. 49). With this characterization of the self as having an identity rooted in our body and constituted by a teleological organization determined by our passions in a social context, Hume would not disagree. He might not approve of the prose – his was a more sober age – but he would not disagree. Nor would he find himself in any fundamental disagreement with Nietzsche’s argument that what is needed is a critique of ordinary morality. “Let us articulate this new demand: we need a critique of moral values, the values of these values must first be called in question.”112 In their respective critiques of ordinary values, both Hume and Nietzsche argued that there is no objective standard of moral value: that must be the starting point of a critique of the received morality, that can be changed, corrected, modified, re-valued, only if it is not eternal, not laid down in the ontological structure of the world. And so for both Hume and Nietzsche, and contrary to Plato and the Platonists, there is no Form of virtue, laying down once for all which among all the social conventions is the true morality. In Nietzsche’s terms, both are nihilists. However, even though there are no absolute or objective values, there are relative values. One ought not to mourn the loss of absolute values, but rather accept one’s fate that all there are relative values. But if that is the best that one can do, then, if one is reasonable, with that one ought to be satisfied. But even relative values, some of them, are, when it comes right down to it, hardly satisfying. That is certainly true of those rules that deal with our relations with a deity. But to say that those values are valueless is, of course, a value judgement, and therefore relative like the rest. It is one’s fate to be what one is, and this includes having the values one has: the reasonable person accepts it, taking a stance from where one is and judging from there how things ought to be. As Nietzsche puts it, “The noble type of man feels himself to be the determiner of values, he does not need to be approved of, he judges ‘what harms me is harmful in itself,’ he knows himself to be that which in general first accords honour to things, he creates values” (BGE, ¶ 260, p. 195). In particular, much of our ordinary morality will be judged to be worthless and even just wrong, harmful to decent civil society. These values, these conventions, ought to be replaced: the world needs “new Philosophers,” “spirits strong and original enough to make a start on antithetical evaluations and to revalue and invert ‘eternal values’” (BGE, ¶

454

Body, Mind and Self in Hume’s Critical Realism

203, p. 126). Hume himself suggests such re-valuations. This would make him one of these “new philosophers.” Whether he would be counted as such by Nietzsche himself is hard to say. Hume uses the standard of utility to separate those norms of civil society which one needs, e.g., property and prmosomg, and those which are to be rejected, e.g, the monkish virtues. Both Nietzsche and Hume object to the values of the ascetic. Hume does so on grounds that human well being is not secured by those virtues; other virtues do a better job. Nietzsche apparently rejects this Humean standard, for he does object to a view that makes pleasure the aim of human beings: “Wellbeing as you understand it – that is no goal, that seems to us an end” (BGE ¶ 221, p. 155). This might seem to be in contrast to what Hume proposes, that pleasure and utility is the criterion of morality. On the one hand, Nietzsche argues that English morality is stifling to what ought to be demanded of us, as when he notes the “English-mechanistic stultification of the world” (BGE ¶ 252, p. 183). The English moral philosophers, he suggests, propose themselves as defenders of a non-religious view of the world and of a secular humanist morality. But in fact, when it is examined more closely, the rules that they propose turn out to be the moral system of the Christians. Their moral stance contains within itself an element of self deception: they have not really freed themselves from the constraints of habitual morality. They have not put themselves in a position where they can take morality “beyond good and evil”, that is, beyond the good and evil of conventional morality. On the other hand, in proposing that pleasure and happiness are the true goals of human action, proposing that this is what people are to rest content with, the “English philosophers” in fact demean human being. If human being can be satisfied with mere happiness and pleasure, then what human being becomes is something idly complacent and incapable of rising above a present contentment to something more worthy. He objects to a morality in which happiness as an end is understood as “a sedative ... medicine and mode of thought, pre-eminently as the happiness of repose, of tranquillity, of satiety, of unity at last attained ...”(BGE ¶ 200, p. 101).Hume also decides he ought not to rest content with principles based in nothing more than the “sentiments of my spleen and indolence” (T, p. 27). In this latter case, it is not hard to discern what it is that he is objecting to. Consider James Thomson’s description of the happy and contented self, which he takes to be the highest expression of human

455

Five: Hume’s Positive Account

aspiration. In his good, though not great, poem on “The Seasons” (1726, last edition 1746) he describes, in “Summer,” the situation of the satisfied person. Thrice happy he, who on the sunless side Of a romantic mountain, forest-crowned, Beneath the whole collected shade reclines; Or in the gelid caverns, woodbine-wrought And fresh bedewed with ever-sprouting streams, Sits cooly calm, while all the world without, Unsatisfied and sick, tosses in noon. Emblem instructive of the virtuous man, Who keeps his tempered mind serene and pure, And every passion aptly harmonized, Amid a jarring world with vice inflamed.113

But if much of the 18th century felt, following Thomson, this way, it is hardly the ideal that Hume proposes. To be sure, while Hume argues that there are basic forms for moral codes in civil society, he does not suggest that there is a single pattern for humans to live by. In his own case he makes clear the motivation. Like Nietzsche he rejects a rule of life based on the sentiments of one’s “spleen and indolence” (T, p. 270); not for him the contented imperturbability of Thompson’s sage. For himself, Hime proposes to explore human nature by means of the new empirical science of Bacon and Newton. His inclinations are both personal and social. He wishes to serve both himself and his community: he is moved by the passion of curiosity and by the feeling that he would like to learn more – he is “naturally inclin’d” in that direction (T, p. 270) – he “cannot forbear having a curiosity” about the principles of philosophy and morals (T, p. 270) – he is “uneasy” about those principles, and is “concerned about the condition of the learned world” and is in a state of “lamentable ignorance” in respect of them (T, p. 271) – he feels inclined to do good for and win praise from the community in which he lives – “I feel an ambition arise in me of contributing to the instruction of mankind, and acquiring fame by my inventions and discoveries” (T, p. 271). This work prompted by these “sentiments” – the composition of the Treatise – was no doubt hard work, hardly something one would undertake out of a desire for indolence. It is safe to say that it involved pain and effort – suffering, if you wish to speak with Nietzsche about the difficulties that accompany any truly creative effort. But that pain and effort itself can often be rewarding, more rewarding than the contentment of which Thomson speaks: hard work is

456

Body, Mind and Self in Hume’s Critical Realism

hardly something done for its own sake but it can be rewarding, especially if it is work towards a desired end. “You want ... to abolish suffering ...?” Nietzsche asks (BGE ¶ 225, p. 155), knowing the answer that one truly reflective on human morality will give. There are, however, kinds of suffering. Compare St. Simon Stylites: he sat on his pillar in a life that could only have been one full of discomfort, but the suffering was worthwhile, for him and for his community, as he aimed to be the first to see from his perch the second coming of the Saviour, and to inform his community of it. The effort at composing the Treatise could hardly be compared to what Simon went through – but as John Stuart Mill said, the pillar saint is an example of what a man can do, not what he ought to do. There is suffering for the sake of a noble end, and there is suffering for the sake of an illusion. “The discipline of suffering, of great suffering – do you not know that it is this discipline alone which has created every elevation of mankind hitherto?” (BGE, ¶ 225, p. 155). Not only is there the effort and pain of work but as one aims, as Hume was aiming, to reform the values of his society, eliminating or attenuating the religion and the moral rules rooted in superstition rather than reason, but there will also be the conflict with one’s community and its moral rules as this community sees one attacking good and supporting evil: it was clear to his community that Hume’s philosophy was atheistic and was morally bad, subverting the values upon which that society saw itself to be based. Such a conflict between the moral code and one’s aim as an individual can be a challenge to one’s integrity as a person: there will be, on the one hand, the project to which one is committed, and, on the other hand, the demands of the moral code of one’s society, and then the conflict between them. The following, rather dramatic, example has been given.114 Jim, on a botanical expedition, arrives in an Indian village in Guatemala. There he finds ten Indians lined up against a wall, and in front of them several other men armed and clearly preparing to shoot the others, with the aim of suggesting to other local people that there are advantages to not making trouble. The commander of the firing squad proposes to Jim that he will honour the latter as a special guest. Jim is given the opportunity of shooting one of the persons lined up against the wall. If he does then the rest will be spared. But if he does not, then all will be shot. What is Jim to do? On grounds of utility, he ought to shoot the one Indian. But strongly within his own felt morality, there is the sense that he is morally not the sort of person who could coldly kill another person, and in particular kill

457

Five: Hume’s Positive Account

simply in order to deter others from what he (Jim but not the commander) easily sees can be reasonable protest. This sort of example is used to criticize a strict utilitarian position in moral theory, the view that an action, any action, is right if and only if it maximizes happiness. With such a rule, Jim’s choice is apparently easy. But in fact we should expect it to be hard. It should be hard to the extent that it places Jim in an internal conflict between his care for those who would survive, on the one hand, and, on the other hand, his own integrity as a person who recognizes the dignity of every human and in particular abhors the idea of killing anyone for any reason like that of the commander. Too, Jim must be angry, angry at the world if you wish, at its having put him into such an intolerable position. He may well have to choose, but he does not wish to have such a choice. All these deep feelings ought to come into play; that is what we would expect of any decent human being, one secure in his or her own self about the moral principles that he or she is prepared to defend. Indeed, one can argue on utilitarian grounds that it is a good thing to have people of this sort, who are prepared to arrange the world, if they can, so as not to place others in the sort of position that Jim is in, a position in which one must make an intolerable and anguishing choice. And it is a good thing to have persons who are strongly committed to their own moral integrity and to the dignity and human worth of others. All these character traits make for a better society and a better world. The point is not here to mount a defence of utilitarianism, but simply to recognize that such character traits can easily be fit into a Humean vision of human being: it is evident that all these character traits are Humean virtues, including both the virtue of moral integrity and the virtue of maximizing happiness or well-being. Normally, of course, these do not create problems: it is only in extreme and unexpected situations, such as that in which Jim finds himself, that they come into conflict. One can imagine his internal conflict. On the one hand, he is there is the virtue of acting for the well-being of others. On the other hand, there is the virtue of not involving oneself in the killing of another. And there is Jim reflecting on – not reflecting on, that is a suggestion of calm on these things, – but agonizing over his quandary, agonizing over himself, but also agonizing about a world that would put him is position where such a choice was forced upon him. But in another sense, this is the human lot: while one cannot expect a perfect world that is free from such situations of conflict, at the same time, one cannot arrange human feelings so as to enable

458

Body, Mind and Self in Hume’s Critical Realism

people, well fitted for normal situations, to cope easily and without conflict in such extreme situations. Moreover, even if it is possible in principle to raise persons with virtues so structured that the choice is easy in all contexts, the costs that would be involved in such a process of social engineering would undoubtedly be too great. One might be able to eliminate the need for choices such as Jim must make, but one could not eliminate all conflict and every case where reflection on becomes agonizing over. We recognize, then, that one’s projects as a person may well come into conflict with moral principles, understanding the latter in a narrow sense. Indeed, while there certainly are cases like that of Jim, where the conflict rages over the need to make a choice where both alternatives are evil, it may in other cases be a good thing that there is such conflict. For, it may lead to the reform of those codes, better adapting them to circumstances and to changing and more complicated circumstances. As elsewhere, so in morality: creativity can be good thing, and increase human well-being and human freedom to choose for oneself the life that one finds good. Power is the capacity to achieve what one wants. Hobbes and Hume both recognize that in the absence of the social conventions of civil society, all lack power: life in the “state of nature” is poor, nasty, brutish and short. Power comes only when the institutions of justice, contract and allegiance come into being. We all have, as Nietzsche said, a will to power: we all aim to maximize our capacity to choose what we want to have and to be. The will to power is not the will to be able to tyrannize over others – that is hardly the good life – and, though it is at times possible to achieve it, such a life is inevitably one of insecurity, as Plato already saw in the Republic. The will to power is precisely that which leads us to learn to limit our desires and to share scarce resources. The security of tenure and contract which comes from civil society ensures that in the end we secure more of what we want and ensures that we can more likely become what we want to be. The will to power wills civil society. But the civil society that results may well involve many virtues that are everywhere replete with forms of behaviour that society deems virtuous but which utility would not require. The monkish virtues are virtues in the society that includes monks. To the extent that the moral code that constitutes civil society involves more than mere utility can justify they come into conflict with the will to power. If nothing else, they can become boring and mechanical, constraining rather providing for the expression of what is best in human

459

Five: Hume’s Positive Account

being. The will to power will in such circumstances aims, or, in the case of some persons, aims at release from morality – that is, not from all morality but from the arbitrariness and/or mechanical aspects of received morality. How exactly one goes about this will be a matter of reflective selfawareness: who am I and where am I and where shall I go? Often enough, of course, there will be no clear direction. How does one go from here, where we are now, to a better moral code and better institutional arrangements? One may recognize that something better is possible, but not know exactly what nor how to get there. One will recognize that the social conventions to which one habitually and out of a sense of duty conforms, are not as useful to oneself and others as and that the social world could be better than it is. And the aim of going beyond current good and evil to some other better set of arrangements, may itself turn out to be a decision full of stress. One will have one’s concept of oneself as a moral person, accepting of the need to behave virtuously. And one will have another concept of oneself, as a person who is to treat those virtues as in fact vices, and aiming to change them in the name of as not yet clearly located new virtues. These actions will also be virtuous, but the tension will be there, one part of one’s identity in conflict with another. The situation will be one that calls into question one’s moral integrity. The decision will be agonizing, and full of self-doubts, the effort one of suffering rather one that could in any way be called a pleasure or could result in a sense of well-being. Nietzsche emphasizes suffering. Not that he wishes pain as an end in itself: he does not propose the disinterested malevolence of a de Sade. Recall Nietzsche’s remark, “The discipline of suffering, of great suffering – do you not know that it is this discipline alone which has created every elevation of mankind hitherto?” (BGE, ¶ 225, p. 155). Suffering is always a means to a further end, and is never for its own sake.115 Nor is such suffering worthwhile in attempts to achieve easy comfort. Suffering is only as noble as the end which it serves. In particular, there is the suffering that comes from breaking out of the established and religiously based moral rules, and going beyond good and evil – the good and evil of the established moral code – that is why Nietzsche speaks of “We immoralists” (BGE, ¶ 226, p. 156). The aim is to creatively develop a new and more noble system of values: the task of the true philosopher, of the philosophical labourer, “demands that he create values” (BGE, ¶ 211, p. 142). It is often enough the mechanism of sympathy that generates conflict

460

Body, Mind and Self in Hume’s Critical Realism

within ourselves. I am moved in a certain direction by passions contrary to the moral principles that shape society. These principles are artificial virtues, and through the moral approbation that I give to these principles I am moved in the opposite direction. But others also morally approbate those same principles, and so they contradict me when I act in violation of those principles. Through the mechanism of sympathy I am also moved by their disapprobation that they feel towards my actions. This principle of sympathy is of so powerful and insinuating a nature, that it enters into most of our sentiments and passions, and often takes place under the appearance of its contrary. For it is remarkable, that when a person opposes me in any thing which I am strongly bent upon, and rouzes up my passion by contradiction, I have always a degree of sympathy with him, nor does my commotion proceed from any other origin. We may here observe an evident conflict or rencounter of opposite principles and passions. On the one side, there is that passion or sentiment which is natural to me; and it is observable that the stronger this passion is, the greater is the commotion. There must also be some passion or sentiment on the other side; and this passion can proceed from nothing but sympathy. (T, p. 593)

Sympathy makes a conflict with others into a conflict with myself. It leads to the question, am I really after all what I ought to be? All this presupposes that we can reflect upon ourselves, upon the world in which we live, and upon our place in that world. It requires the existence of a common moral language, the same common language that is involved in the establishing of and the adjusting of one’s reputation or in having a “name” (T, p. 582). Having a reputation, having a “name”, having a certain identity, recognizable by both oneself and others all presuppose a social context shaped by a common language. The development of society in the form of social conventions such as those of property, contract (promising) and government also brings with it, in the form of language, the capacity to describe virtues and vices and to communicate one’s judgments of character. Moreover, since an abstract idea is, for Hume, simply the association of a word with a similarity class of individuals, it follows that we can have no abstract ideas, nor, therefore, any thought, apart from language, apart from linguistic conventions. The capacity to think about ourselves, the capacity for reflective self-consciousness, is thus acquired only in the social context. The establishment of one’s “identity,” one’s “reputation,” “character” and “name” is a social process. This includes the establishment of oneself as a knower, the internalization of the cognitive standards that

461

Five: Hume’s Positive Account

define one’s orientation to the world. For some, these cognitive standards are those of the Platonist or the Cartesian. For others the cognitive standards that define one are those of the Humean and the natural scientist. Social utility determines that the latter sort of cognitive virtues are to be preferred. Such at least is the argument Hume offers when he offers his defence of science. As part of one’s capacity to know, there is the capacity to know oneself. This is the capacity for reflective self-awareness, the capacity to come to know one’s own reputation, character and name. One’s identity is acquired in a social context; so is the capacity to know one’s own character. This capacity to know oneself, one’s identity, one’s reputation, character and name, is thoroughly social. This capacity to know oneself, Hume argues, is mediated by the capacity to know these in others; the capacity for selfknowledge presupposes the ability to know the character of others. Though one’s identity is a matter of the regularities that characterize one’s thought and action, our knowledge of this character and identity is not merely inductive, though it is that, and our reflective claims to self-knowledge must always be tested against the reality of myself and others by a sort of experimental method. But the knowledge also comes by way of the mechanism of sympathy. The mechanism that is central to the development of social conventions into artificial virtues is also central to our capacity for self-knowledge. I find my reputation and my name in what others think and feel about me. Through sympathy and reasoning, my own and others, those others form a mirror in which I discover myself. “’Tis certain, then,” Hume tells us, “that if a person consider’d himself in the same light in which he appears to his admirer he wou’d ...receive...a pride or self satisfaction....Now nothing is more natural than for us to embrace the opinions of others in this particular; both from sympathy, which renders all sentiments intimately present to us, and from reasoning, which makes us regard their judgment as a kind of argument for what they affirm” (T, p. 320). Hume does admit that it is more natural for us to embrace the good rather than the bad opinions others have of our character, but “in general we may remark that the minds of men are mirrors to one another” (T, p. 365). The shared practice of establishing reputations and names and of morally assessing characters is the stable public fact which enables individual persons to come to recognize in others, and in themselves, the enduring and permanent virtues and vices, abilities and capacities, that provide the structure that unifies the changing flux of action and perception into the bundles that we call persons. These unified bundles will be

462

Body, Mind and Self in Hume’s Critical Realism

continuing individuals each with a unique organized combination of personal abilities, of virtues and vices, and of purposes and ends that make of the bundle a teleological unity.

Conclusion; or, perhaps not ... Locke recognized that the notion of a person is a forensic notion. Hume accepted this point, but went beyond it to locate the notion of a person in the broader framework of the linguistic and moral practices that define society and the social context in which we live and work. Since the forensic notion of the self defines the latter with respect to responsibility, Locke discovered, in his correspondence with Molyneaux, that the linkage which creates a person out of a series of events need not coincide with the casual chain that links those events. Both bodily identity and memory are crucial to personal identity, but there is more to it than that: there are also the permanent traits of character that unify one’s thought and action as one’s own. But there is yet more. As Locke also recognized, continuing the tradition coming down from Plotinus through the Christian tradition, the responsible self is one that is self-reflective. Locke was not able to free himself sufficiently from the substance tradition to recognize that this capacity had to be understood not as a simple power of a substance but as a learned capacity of a socialized being. This last step was the one that Hume took. Identity of self is not something that one can “find” in “the thought alone” (T, p. 635); reason reflecting upon its own processes cannot discover the principle that constitutes personal identity. Nor is that which constitutes the unity of the self to be found in the passions alone, the sentiments, traits and abilities that define one’s character. What is required is the union of these two in a character that includes the capacity for selfconscious concern, the reflective concern of a mind for the status and being of its own self, the reflective concern of a mind “all collected within itself” (T, p. 270). This mind “all collected within itself” is the product of the social context in which each mind mirroring its fellows discovers itself: “a mind will never be able to bear its own survey which has been wanting in its part to mankind and society” (T, p. 620). The person of thought alone, the reason whose thought is restricted to cognitive activities alone, is a “strange, uncouth monster” (T, p. 264); once we have discovered exactly what it is that constitutes personal identity, we recognize that it is not at all

463

Five: Hume’s Positive Account

surprising that such a one – that is, in effect the reader at this point in Book I of the Treatise – is incapable of discovering his or her own personal identity, that is, to put it another way, he or she cannot recognize that the full solution of the problem of personal identity cannot be given in Book I of the Treatise, and recognize that one must wait for the completion of Books II and III. The man or person of character is one who can “mingle and unite in society” (ibid.) through the establishment of conventions and social artifices. Only when this is added to the person considered as a knowing and reasoning thing do we achieve a person, someone capable of surveying his or her own being and bringing it into harmony with the deepest standards he or she has for him- or herself alone, someone responsible for his or her own identity. This concept of person is not adopted for purely descriptive purposes. Unlike our abstract idea of, say, a natural chemical kind, e.g., gold, we do not adopt it simply because it is useful in stating the regularities that we discover to obtain in the world. That, to be sure, is part of it, since regularities, both those that define our body and those that define our character, are important in determining the utility of our concept of a person. But Locke’s insight remains, that the concept of a person is also a forensic concept, in part defined as it is because it is useful in the social practices of men. Persons are, in the first instance, those entities whose enduring dispositions and traits we morally evaluate and who, on the basis of those evaluations, we reward and punish. But persons are more than entities whose behaviour and whose traits can be shaped by the actions of others through reward and punishment. Persons are not merely responsible for what they do, they are capable of taking responsibility for their actions. They are capable of self-evaluation, and capable of acting to shape themselves, and to determine their own way of being. This capacity is itself useful both to the possessor and to others – it is to one’s own good and that of others that one knows oneself sufficiently to anticipate the responses of others, and, where one finds it desirable, to modify one’s actions and dispositions in the light of this knowledge. And since this capacity has utility, it is, upon Hume’s view, a virtue. The reflective self-consciousness that is a central characteristic of persons, players of social roles, is achieved and maintained in a mind only in the context of social conventions, social roles, and only with the help of the mirroring minds of one’s fellows. This means that those whom we recognize as persons are in the first place those with whom we interact sympathetically. And it means in the second place that there are no persons

464

Body, Mind and Self in Hume’s Critical Realism

apart from the context of the moral community that constitutes civil society. But civil society is constituted in the first place through conventions that define justice. These conventions reckon as persons those with whom we interact sympathetically. But then, the conventions are, as Hume argues, established not by feelings of sympathy and benevolence but by self-interest. In one way, then, the concept of a person and of personal identity is determined by the utility it has in defining what it is to be a role player in our system of social conventions and artificial virtues. But these become virtues only through the work of sympathy: it is sympathy which makes morality out of prudence; and these conventions are maintained as features of a moral community only through the work of sympathy. Since the conventions are maintained by our moral sense, the fact that we respond sympathetically to the feelings and passions of others itself has utility. It is therefore of utility to reckon as persons those with whom we interact sympathetically. Reflective self-consciousness, the recognition by oneself that one is a person with duties and obligations, is of utility in maintaining the system of civil society. We therefore bring up those who are to become members of our civil society to be persons who can reflect upon their own being, and who can, in the light of such reflection, modify their being as appropriate.

But ... There are also sensible knaves. ... according to the imperfect way in which human affairs are conducted, a sensible knave, in particular incidents, may think, that an act of iniquity or infidelity will make a considerable addition to his fortune, without causing any considerable breach in the social union and confederacy. That honesty is the best policy, may be a good general rule; but is liable to many exceptions: And he, it may, perhaps, be thought, conducts himself with most wisdom, who observes the general rule, and takes advantage of all the exceptions.116

The sensible knave is also self-reflective. It is his or her wisdom that there is no moral problem with violating one of the norms of civil society, However, It is hard to tell, whether you hurt a man’s character most by calling him a knave or a coward, and whether a beastly glutton or drunkard be not as odious and contemptible, as a selfish, ungenerous miser.117

465

Five: Hume’s Positive Account

The knave can do his or her knavish acts only if others have no awareness of them. He or she is might violate the rule of honesty, but that can be done with impunity only if it is disguised. The knave must dissimilate, if he or she is to be successful. That also is what wisdom, his or hers, such as it is, dictates. Thus, a capacity for self reflection and self knowledge may in fact be in general a good thing, and its absence a vice, but it cannot be what distinguishes the virtuous from the vicious, the good person from the bad person. The good person must have virtues that are prior to the virtue of self reflection and self knowledge. But if the good person has good ends, then self reflection and self knowledge can help him or her become a better person, one who better conforms to the basic standards of virtue. But wisdom can also help a knave be a better knave. How is it that a firm conformity to the moral rules of civil society is there in many or most persons? Why are most of us not knaves, not even sensible knaves? How come most of us have a conscience firm enough to bring us into conformity with the norms of civil society? It is of course through a process of learning that we acquire these habits; they are hardly innate. Causation alone ensures the embedding of a strong conscience within the characters of most persons, making it, and thereby the habits of civil society, firmly part of our identities. Hume’s basic theory of learning is his associationist account for habit formation. But there is more to virtue that habits; there is also the moral force that these norms have over us, keeping us through our own sense of virtue in conformity with those norms. Few of us are knaves, even sensible knaves, and few of us want to be knaves. We value our name. It is with that value, and the accompanying sense that virtue is a good thing and ought to be pursued, shaping what we are that self reflection itself becomes a virtue, one that is of utility both to myself and to others, helping us to shape ourselves into better persons. These are not merely habits. Just as with reason, so also with virtue. Reason, in the first instance, is a matter of association, of the formation of habits of thought. But those habits of inference are sometimes thwarted; we sometimes find our expectations turn out to be false. But we are guided by the passion of truth, the desire to obtain matter of fact knowledge. Guided by this passion we consciously (re-)form our habits of causal inference. We come to realize that inferring in conformity with the rules by which to judge of causes and effects is a

466

Body, Mind and Self in Hume’s Critical Realism

better way to satisfy our passion of curiosity. So causal inference at its best proceeds in what Hume calls an indirect and oblique manner. In similar fashion we find that simple habits of cooperation do not always work; and so, guided by our sense of what is useful, to ourselves and to others, we adjust our being to better achieve the ends for which civil society exists. Even so, it is not, according to Hume, merely a matter of habit formation, even the (re-)formation of habits through self reflection on how we can best achieve a decent life along with the others who are members of civil society. But part of the learning consists in seeing how others see us: we adjust our being, our identity, through the reflection of ourselves that we find in the view that others have of us. Self reflection includes reflecting on our reflection in the eyes of others. It is this way of discovering what we are and ought to be through the gaze of others that implies there is more to forming that identity than association and habit formation, even self conscious habit formation based on a reflective knowledge of ourselves. There is, as Hume indicates in the “Appendix” to the Treatise, more to the “quality” that unites the bundle of impressions which we are into our identity than can be provided by the mechanism of association, even when that is supplemented by our capacity to adjust ourselves in the light of our reflective self knowledge. We should note, too, that there are situations and social contexts in which treachery would be useful both to oneself and to others, and would therefore be counted as a virtue. So would the taking of goods that would be reckoned theft in other societies, perhaps even by those who were relieved of those goods. Robert MacGregor, history’s “Rob Roy,” made his reputation through his exploits in capturing cattle from his neighbours, and switched allegiances between the Earl of Montrose and the Duke of Argyll, as best suited his fortunes and those of his men. His acts were no doubt considered virtuous by the clansmen whom he led. What was virtuous in the Highlands was hardly that in the Lowlands. When reflecting on one’s moral qualities, and how one might do better, it is natural enough to make comparisons with how others do things, not merely in one’s own community, but also in others. This sort of reflection upon what one does and might do calls into question the validity of one’s moral judgments, and, above all, oneself, what one is as a person. Even if one decides that one’s values are indeed better than those of the next fellow or of the next community, reflection can raise the issue about whether one could do better. If not a different standard, then perhaps a higher standard is that for which one ought to aspire. That we are virtuous

467

Five: Hume’s Positive Account

self reflection calls into question. Hume suggests in one of his essays that there are two “means commonly employed to destroy the conclusion” that we are beings that meet a decent level of virtue. The first is “by making an unfair representation of the case, and insisting only upon the weaknesses of human nature.” The second is “by forming a new and secret comparison between man and beings of the most perfect wisdom.”118 And so we reflect upon ourselves we recognize that we are selfish and egoistic, knavish and vicious. And even if we notice that this is only sometimes the case, when we put before us a standard of a person who is infinitely just and virtuous, we fall infinitely short of what we ought to be. In fact, we also often reflect upon our cognitive capacities, at how we are fallible, and often fall into error. Even if we allow that we often correct our errors, and that we have often uncovered the causes of things, when we put before us a standard of a person who is infinitely knowing, omniscient and infallible, then we fall infinitely short of what we ought to be. Recall Pascal119 (who was undoubtedly known by Hume), who compared our capacity to know with the infinity that is beyond us: I want to show him [the inquirer] a new abyss. I want to depict to him not only the visible universe, but all the conceivable immensity of nature enclosed in this miniature atom…. [He is] equally incapable of seeing the nothingness from which he emerges and the infinity in which he is engulfed…. What else can he do, then, but perceive some semblance of the middle of things, eternally hopeless of knowing either their principles or their end? All things have come out of nothingness and are carried towards infinity. Who can follow these astonishing processes? The author of these wonders understands them: no one else can. (Pensées, # 199, pp. 60-61)

Neither reason nor the senses are sources of real, that is, incorrigible knowledge, knowledge of the sort that Descartes, for example, sought: Man is nothing but a subject full of natural error that cannot be eradicated except through grace. Nothing shows him truth, everything deceives him. The two principles of truth, reason and the senses, are not only both not genuine, but are engaged in mutual deception. (Pensées, # 45, pp. 12-13)

+ The sceptics win the argument: Let us then concede to the sceptics what they have so often proclaimed, that truth lies beyond our scope and is an unattainable quarry, that it is no earthly only in so far as it pleases him to reveal it. (Pensées, #131, pp. 34-25)

468

Body, Mind and Self in Hume’s Critical Realism

Truth and genuine knowledge cannot be attained through our natural faculties; turning to God and faith in God is the only solution. If we know it is only through innate principles and “we cannot be sure that these principles are true” through our “natural intuition”; “faith and revelation” alone suffice for knowledge,” they alone enable us to escape the threat that the Cartesian demon really is behind all appearances and all innate ideas. There is no certainty, apart from faith, as to whether man was created by a good God, or an evil demon, or just by chance, and so it is a matter of doubt, depending on our origin, whether these innate principles are true false or uncertain (Pensées, # 131, p. 33)

Self reflection upon our capacities, both moral and cognitive, thus leads, often enough, to the thought of how imperfect we are, how sinful and wretched: we can achieve neither virtue nor truth. And never knowing the truth about anything, we can never know the truth about ourselves. What sort of freak then is man! How novel, how monstrous, how chaotic, how paradoxical, how prodigious! Judge of all things, feeble earthworm, repository of truth, sink of doubt and error, glory and refuse of the universe…. Let us then conceive that man’s condition is dual. Let us conceive that man infinitely transcends man, and that without the aid of faith he would remain inconceivable to himself, for who cannot see that unless we realize the duality of human nature we remain ignorant of ourselves. (Pensées, #130, pp. 34-35)

Thus, self reflection might well be a virtue, but it can also lead us to a situation where we despair of ourselves, as moral beings and as rational beings, and indeed as beings. Hume understands this process towards which self reflection leads. He considers the abstract thought of philosophy, the questioning that it involves of all our basic principles. For with what confidence can I venture upon such bold enterprizes, when, beside those numberless infirmities peculiar to myself, I find so many which are common to human nature? Can I be sure that, in leaving all established opinions, I am following truth? and by what criterion shall I distinguish her, even if fortune should at last guide me on her footsteps? (T, p. 265)

When we do thus reflect upon ourselves and upon our capacities, we recognize how imperfect are our faculties. It is only experience that teaches

469

Five: Hume’s Positive Account

us about ourselves and about the world about us. After the most accurate and exact of my reasonings, I can give no reason why I should assent to it, and feel nothing but a strong propensity to consider objects strongly in that view under which they appear to me. (T, p. 265)

How can we allow it reasonable that our knowledge of fact and value rests upon such a feeble foundation when we think of the aspirations of such thinkers as Descartes, who insisted that can and must know infallibly the existence of an omniscient and infinitely just God in order to know anything at all, that I am sitting on a chair and even the sort of being that I am. I am in fact a wretch that knows nothing, not even the being that I, as a person, am. I do not know what I am, the sort of being that I am: there is merely a point of consciousness, a point of consciousness which may or may not be me. Certainly, I do not know that I am a knower, or how to become one. I have no sense of how to recover myself, whoever I am, nor to discover my place in the world, if there be a world, or, if there is, what it really is like the one I think that I remember. The result, he knows, is despair. He compares the person who has carefully examined the possibility of knowledge, who has discovered that they are subject to error, that infallible knowledge is beyond us, with one who “having struck on many shoals, and having narrowly escap’d ship-wreck in passing a small frith, has yet the temerity to put out to sea in the same leaky weatherbeaten vessel…” (T, p. 263). …The impossibility of amending or correcting these faculties, reduces me almost to despair, and makes me resolve to perish on the barren rock, on which I am at present, rather then venture upon that boundless ocean, which runs out into immensity.

Despair and melancholy is the upshot. This sudden view of my danger strikes me with melancholy; and ’tis usual for that passion, above all others, to indulge itself; I cannot forbear feeding my despair, with all those desponding reflections, which the present subject furnishes me with such abundance. (T, p. 264)

Pascal used the same metaphor of the person stranded on a shoal or island: the sceptical arguments leave one with no way clear way out of the situation, with regard to both cognitive and moral principles. And the result for Pascal, as for Hume, is despair.

470

Body, Mind and Self in Hume’s Critical Realism

When I see the blind and wretched state of man, when I survey the whole universe in its dumbness and man left to himself without knowing who put him there, what he has come to do, what will become of him when he dies, incapable of knowing anything, I am moved to error like a man transported in his sleep to some terrifying desert island, who wakes up quite lost and with no means to escape. (Pensées, # 198, p. 59)

As he later realizes, the despair derives, as Pascal saw, not just some sense of scepticism, but the realization that if this be true, then we cannot even know ourselves. The intense view of these manifold contradictions and imperfections in human reason has so wrought upon me, and heated my brain, that I am ready to reject all belief and reasoning, and can look upon no opinion even as more probable or likely than another. Where am I, or what? From what causes do I derive my existence, and to what condition shall I return? Whose favour shall I court, and whose anger must I dread? What beings surround me? and on whom have I any influence, or who have any influence on me? I am confounded with all these questions, and begin to fancy myself in the most deplorable condition imaginable, environed with the deepest darkness, and utterly deprived of the use of every member and faculty. (T, p. 268)

How does one retrieve from such despair a sense of self and of oneself being in the world? We must begin again: more is required before we obtain a complete sense of the ontology of the Humean self. (4) Becoming Our Selves In fact one finds in Voltaire’s commentary on Pascal120 what is the basis of an answer to the latter, an answer which obviates the claim that one can find oneself, who one is, only through the faith that is granted to one through the saving grace of God. Pascal suggested, as we know, that without faith self-knowledge is impossible: …without the aid of faith he would remain inconceivable to himself, for who cannot see that unless we realize the duality of human nature we remain ignorant of ourselves. (Pensées, #130, pp. 34-35)

To which Voltaire replied

471

Five: Hume’s Positive Account

Man is not an ænigma, as you [Pascal] figure him to yourself to be, merely to have the pleasure of unriddling it. Man seems to have his due place in the scale of beings….Man is like everything around us, a composition in which good and evil, pleasures and pains are found. He is inform’d with passions to rouze him to act; and indued with reason, to be as the director of his actions. If man was perfect, he would be God; and those contrarieties, which you call contradictions, which you call contradictions, are so many necessary ingredients to the composition of man, who is just what he ought to be. (p. 124)

But this is only the sketch of an answer, without the detail that might establish clearly that, despite the fallibility of our reason and the relativity of our moral principles, despair and melancholy is an unreasonable response. What must be established is that we do know enough about ourselves and about the world in which we find ourselves thrown to carry on living in a reasonable human way. Hume provides such an argument in his account of our selves and our knowledge of them. In its main structure Hume’s account of our self identity turns on his theory of learning, that is, his associationist psychology. But there is, of course, a further mechanism to which Hume appeals when he treats of our feelings and passions. This mechanism is, as we have seen, that of sympathy, and it is not irrelevant to Hume’s account of our sense of self identity. The self endures, and it has an identity: the identity of the self is the complex relational structure that binds the parts of the bundle into a unified whole, something with a past, present and future. It is causality in particular that constitutes that structure. The continuity of the body is of course part of how the self is defined. The passions also are relevant.121 Many of our passions, hope and fear, for example, refer to the future (T, pp. 442-45), and help determine (in the sense of cause) the future being of one’s self. It is our conception of what the future will be and could be that affects the will (T, p. 430), and therefore again is something that contributes to the shaping of our identity as a self. But the passions and the will, that is, the patterns one finds in these, flow from one’s character Moreover, one’s reason, which guides these passions to their fulfilment, is also part of one’s character. It is this, one’s character, as a set of settled patterns that is important. It is often part of what one is, what binds one into a whole with a past, a present and a future, that one is a member of a clan, the MacGregors, say, or a nation, a Scotsman, perhaps, a profession, a lawyer, say, or a philosopher, say, as part of the Republic of Letters, or a member

472

Body, Mind and Self in Hume’s Critical Realism

of civil society, a citizen. These parts of one’s character are patterns of behaviour, of thought and of action; and they, therefore, are dispositions in Hume’s sense.122 As dispositions they often explain our actions, but they do so, not because they are somehow unanalyzable powers, as an Aristotelian might have held, but because they are, as all dispositions must be on Hume’s account, patterns or regularities. Thus, at one point he tell us that “...a spectator can commonly infer our actions from our motives and our character,” (T, p. 408) and he says elsewhere that “the knowledge of these characters is founded on the observation of an uniformity in the actions, that follow from them.” (T, p. 403) These dispositions, these settled patterns or regularities, which are regularities in action, behaviour of one’s body, are elements of one’s character, and provide (some of) the glue that binds the parts of the bundle that is my self into that self. They do this by virtue of their being regularities. For, a characteristic of this sort as a regularity describes a general pattern of behaviour, one that is repeatable, marking the bundle as the same or identical with that bundle as it was earlier and as it will become. Hume makes the point in this way, in his consideration of moral evaluations, where he argues that such elements of character are “durable principles of the mind,” and further notes how they help define the identity of one’s self – they “enter into [one’s] personal character,” he says. If any action be either virtuous or vicious, ’tis only as a sign of some quality or character. It must depend upon durable principles of the mind, which extend over the whole one’s conduct, and enter into the personal character. Actions not proceeding from any constant principle, have no influence on love or hatred, pride or humility, and consequently are never considered in morality. (T, p. 575)

These patterns that “enter into [one’s] personal character” are the real components of the Humean self, and provide a basis for the ascription of moral qualities that avoids the substantial self beloved of the critics of Hobbes. But of course, there are also aspects of character which we judge negatively: besides generosity there is miserliness, besides friendship there is mean-spiritedness, besides reasonableness there is dogmatism and superstition. Vice as well as virtue is part of human nature: for the Christian critics of Hobbes vice was somehow always unnatural – unless one allowed for original sin as a stain inevitably infecting the human soul and the very nature of humankind. We may have our vices, but Hume’s view is not only more reasonable but also the more humane: for Hume, if

473

Five: Hume’s Positive Account

we do have a vice, it is at least not something we have inescapably imposed on us as a matter of metaphysical necessity. Be that as it may, we should note about these aspects of character, that the regularities which define them as dispositions are not strictly universal, they are not without exceptions: while these patterns are regularities, they are gappy regularities. “We know,” Hume said, “in general, that the characters of men are, to a certain degree, inconstant and irregular.”123 But gappy though they are, they nonetheless, like any gappy law, provide explanations. One can do something “out of character”, casually, or in a fit of temper – even the miser can act generously at times, but that does not prevent what he does being for the most part explainable in terms of his miserliness. More importantly, these characteristics are for the most part not native but acquired. The story of the learning process by which such a character is acquired is generally to be found in the associationist theory that Hume accepts, but also in the mechanism of sympathy. We know how sympathy works. When my friend is in the water and in danger of drowning, she feels fear for her life. Seeing her struggle and expressing that fear in her struggles, I come to feel the same fear as she feels, namely, a fear for her life. This fear that I experience is a passion that moves me action; I do what is necessary to get her out of danger. Or, to use Hume’s examples, “A cheerful countenance infuses a sensible complacency and serenity into my mind; as an angry or sorrowful one throws a sudden damp on me” (T, p. 317). We observe first the outward signs of the emotion in the other person, the expression of that emotion in behaviour. Observing these signs we come to have the idea of that emotion. “This idea is presently converted into an impression, and acquires such a degree of force and vivacity, as to become the very passion itself, and produce an equal emotion, as any original affection” (T, p. 317). Whether Hume has correctly stated the mechanism through which sympathy works is no doubt an important issue. Whether, for example, Hume’s account is overly “mechanical,” as Árdal has suggested,124 is an interesting question. Given that Hume allows that we can, through determined self discipline, develop our capacity for sympathetic response, this is likely a criticism due to an over-simplified view of Hume’s position. But we need not worry such issues too much. What is important is that something like what Hume describes is a genuine psychological fact: we do, often, respond sympathetically to others. That is the important point. We have seen how Hume uses the appeal to sympathy to improve on

474

Body, Mind and Self in Hume’s Critical Realism

the Hobbesian picture of humankind, and make of ethics a theory of morality and not just a theory of prudence. But it plays other roles in Hume’s account of each person’s human nature. In particular, he appeals to the phenomenon of sympathy to give an account of the passions of pride and humility, on the one hand, and love and hate, on the other. The emphasis is upon pride and humility, with the account of love and hate running parallel to this. In his discussion of these passions, Hume begins by distinguishing the object of the passions of pride and humility and the cause of these passions. Both passions always have the same object, namely, “the self, or that succession of related ideas and impressions, of which we have an intimate memory or consciousness” (T, p. 277). That pride and humility have this object is taken to be primitive feature of the human mind, not subject to further explanation (T, p. 286). Similarly, that pride is a pleasant passion and humility painful is something else that is taken to be basic, and not subject to further explanation (T, p. 286). That these passions have the self as their object is important for Hume’s full account of our sense of self identity. Of that, more later. But first, more must be said about how these passions work, and in particular about what Hume calls their cause. Specifically, the cause of these passions, according to Hume, not implausibly, is whatever evokes them: we distinguish “betwixt that idea, which excites them, and that to which they direct their view, when excited” (T, p. 278). As for the cause itself, we must further distinguish the quality that evokes the passion from the subject in which that quality inheres (T, p. 279). Hume gives the example of the person who takes pride in his or her beautiful house, Ninewells, perhaps, or a new house on St. David’s Street in Edinburgh, of perhaps Chateau Lafitte. The object of the pride is the person, the cause is the beautiful house; the quality is beauty, and the subject is the house, considered, as Hume emphasizes, as the person’s property. The subject must have both the quality and the relation to the person who takes pride in it. As Hume puts it, using this example, Beauty, consider’d merely as such, unless plac’d upon something related to us, never produces any pride or vanity; and the strongest relation alone, without beauty, or something else in its place, has as little influence upon that passion (T, p. 279).

Pride and humility have their causes, but there is nothing innate or original about what causes these passions, either with regard to the quality or the subject. But they do all have one thing in common in the case of pride, and

475

Five: Hume’s Positive Account

that is that the cause insofar as it is related to the self produces feelings of pleasure in that self; while in the case of humility they have in common that they produce feelings of pain in the self. This the crucial feature to which Hume appeals in his explanation of how we come to take pride in things. Here is his explanation: That cause, which excites the passion, is related to the object, which nature has attributed to the passion; the sensation, which the cause separately produces, is related to the sensation of the passion: From this double relation of ideas and impressions, the passion is deriv’d. The one idea is easily converted into its correlative; and the one impression into that, which resembles and corresponds to it... (T, pp. 286-7).

The account is associationist, but in a complicated way. As we know, on the associationist theory of learning, if A’s and B’s are repeatedly presented as standing in some relation R, then a habit is formed in the mind such that if one has an impression of an A then that will evoke an idea of a B, or if one has an idea of an A then that will introduce an idea of a B. Thus, in the case of causation, if the cause A is repeatedly experienced in the relation of contiguity with the effect B then the habit is formed such that an impression of an A or an idea of an A will introduce the idea of a B. This, as we know, is Hume’s account of causal inferences. Similarly, in the case of resemblance, if an impression A resembles an impression B and these are repeated viewed together, then an impression of an A or an idea of an A will introduce the resembling idea of a B. As we have seen, this provides Hume with an account of how ideas become general in their representation, that is, his account of abstract ideas. In the case of pride, there are two relations. (Humility is analogous.) The first is the relation, call it “R,” of the cause to the self. Some of the things in which we take pride are simply qualities that we have as human beings, “qualities of our mind and body” (T, p. 303), such as virtue (T, p. 294ff) as a quality of mind, and beauty (T, p. 298ff) as a quality of the body. But we also take pride in things which are “extrinsic” to us and “distant” (T, p. 303). Hume explains that “this happens when external objects acquire any particular relation to ourselves and are associated or connected with us” (ib.) We found a vanity upon houses, gardens, equipages, as well as upon personal merit and accomplishments... (ib.)

476

Body, Mind and Self in Hume’s Critical Realism

In order to see how the mechanism of pride (and humility) works, we should consider a particular example, say pride in one’s house. If one takes pride in one’s house, then R is the relation of ownership. The cause introduces the idea of the self as R-ed by the cause. The cause, in the case of pride, produces a feeling (in this case) of pleasure in the self. (In the case of humility, the cause produces feelings of pain in the self.) These feelings of pleasure are evoked by the cause always in conjunction with the feelings of pleasure. The cause therefore introduces both the idea of the self and the feelings of pleasure. The idea of the self as R-ed by the cause is constantly conjoined with the feelings of pleasure. This constant conjunction fuses into the feeling of pride: pride is the pleasant feeling we have when we contemplate the object that causes it insofar as the cause R’s the self. (Objects can stand in a relation R to the self where the object does not cause a feeling of pleasure: one can be indifferent to some, or even maybe all, of one’s possessions, one can be indifferent to the fact one belongs to Clan MacGregor, and so on. In such cases there is no feeling of pride [nor one of humility].) Repetition of the sort that is necessary to generate the habit comes through sympathy. Hume notes how praise and pride go together (T, p. 320), and discusses the importance of praise in generating pride. Praise is the expression of love. Love (and hate) run parallel to pride (and humility). Love has another as its object, where pride has oneself as its object (T, p. 329). As before one must distinguish the cause from the object (T, p. 330), and the cause is further distinguished into the subject and the quality that evokes the love (ib.). Thus, when David praises someone, his Welsh friend Jones, say, for a quality, say the excellence of his (Jones’) house, then David loves that person in virtue of that quality. Jones, recognizing in the praise the outward signs of this love, forms the idea of that love, which in turn is, by means of the mechanism of sympathy, converted into the passion. Since David’s love is directed at Jones, the passion that Jones comes to feel is likewise directed at Jones, i.e., at himself. Since David’s love is pleasant, so is the passion that Jones experiences: he experiences a pleasant passion directed at himself. But David loves Jones by virtue of a quality that Jones has, or more specifically, by virtue of another object which has a certain quality and which stands in a certain relation R, in this case, that of ownership to another object which in turn has the quality of excellence. This quality of excellence itself produces pleasant feelings in Jones, the owner, which pleasant feelings. These feelings resemble the

477

Five: Hume’s Positive Account

pleasant feeling of pride, so the latter too is introduced. The pleasant feeling that Jones feels towards his house is thus strengthened That is, the feeling of pride is strengthened. The mechanism of sympathy can thus produce repeated cases of pride, and thereby strengthen that feeling. It is important for our purposes to notice what else can become associated in this context. The cause of pride is repeatedly presented as standing in relation R to the self, and that cause also introduces feelings of pleasure. Thus, the idea of the cause, qualified in the appropriate way, and the idea of the self and feelings of pleasure. But further, the idea of the cause and the idea of the self come themselves to be associated in the mind, and in a pleasantly approving sort of way: the concept of the self is expanded and comes to have a further defining characteristic. In this way the concept of the self that a person has is generated by the passion of pride (and humility).125 It might be argued that the notion of pride presupposes the notion of a self rather than accounts for it. The feeling of pride is of course owned by a self. This is what Hume has in mind when he says that “pride and humility have the qualities of our mind and body, that is self, for their natural and more immediate causes...”. (T, p. 303) But then there is also, as we have seen, the self in the sense of “self or that succession of related ideas and impressions of which we have an intimate memory and consciousness” (T, p. 277; italics added), or “that connected succession of perceptions which we call self” (T, p. 277; italics added). What are these connections? Clearly, there are the connections that are established by bodily continuity, and those established by memory. But Hume has in view other qualities that establish connections between the perceptions and give an enduring quality to the self. These are those regularities that form aspects of character which constitute, as we have seen him say, “durable principles of the mind,” and which help define the identity of one’s self – they “enter into [one’s] personal character” (T, p. 575) Making a contrast to these regularities that define one’s self, he points out that “accidental blessings and calamities are in a manner separated from us and are never consider’d as connected with our being and existence” (T, p. 302).126 Clearly, Hume has in mind that there are things which contrast to these accidental blessing and calamities, which are “solely or certainly fix’d” in oneself (ib.), and these are “connected with our being and existence”: these stable characteristics define one as the individual one and others conceive oneself to be and, to the extent that it is a characteristic “peculiar to ourself” (ib.), it determines one’s identity, one’s individuality, how one and

478

Body, Mind and Self in Hume’s Critical Realism

others distinguishes one as a person from other persons. As we know full well, Hume distinguishes “betwixt personal identity, as it regards our thought or imagination, and as it regards our passions or the concern we take in ourselves” (T, p. 253). We now see how identity as regards our passions and the concern that we take in ourselves comes to be: the qualities that enduringly through time define our being and determine our individuality become part of our self-concept through the mechanisms of pride and sympathy. In becoming part of our self-concept, part of the idea that we and others have of ourselves, they thereby become the relations that structure our perceptions into the self we and others conceive ourselves to be, the relations that as it were transform a mere bundle of impressions into a person. Perhaps what is most important about this account of the self is that it involves social relations. It has been pointed out that Hume’s account of pride ensures that the passion is intimately involved with what Hume refers to as its cause. In the jargon of more recent philosophy, the cause is not only that but also the intentional object of the passion.127 But Hume is not just working back from the effect – the passion – to its cause. It is correct to emphasize Hume’s general point that “’tis not the present sensation or momentary pain or pleasure, which determines the character of any passion, but the general bent or tendency of it from the beginning to the end” (T, pp. 384-5).128 But there are causes which are as it were antecedent to the thing that Hume refers to as the cause of the passion. It is important to also note, as we have, the role of sympathy and the love of others towards us in generating the pride that we feel. Otherwise the cause would evoke pleasure in us, the passion that Hume refers to as “joy,” but it would not generate pride. For, without the sympathetic response to others loving us for our qualities the idea of self, the object of pride, would come to be included in the context in a way that would allow the associative mechanisms to work. Pride is a socially based passion. We can also see why concepts of the self are often fairly similar within a society and relative stable. They are sustained by the approval of others, working by way of sympathy. Hume discusses this point in his essay “Of National Characters.” National characters are formed, Hume argues not from physical causes such a climate, but from “moral causes,” by which he meant “all circumstances, which are fitted to work on the mind as motives or reasons, and which render a peculiar set of manners habitual to us. Of this kind are, the nature of the government, the revolutions of public affairs, the plenty or penury in which the people live,

479

Five: Hume’s Positive Account

the situation of the nation with regard to its neighbours, and such like circumstances.”129 He suggests that his conclusion is indeed obvious since “a nation is nothing but a collection of individuals, and the manners of individuals are frequently determined by these causes.”130 The mechanisms by which uniformity of national character are of the sort Hume describes in the Treatise, though less technically put – so that, for example, he speaks of “imitation” rather than “sympathy”, but the point is the same. The human mind is of a very imitative nature; nor is it possible for any set of men to converse often together, without acquiring a similitude of manners, and communicating to each other their vices as well as virtues. The propensity to company and society is strong in all rational creatures; and the same disposition, which gives us this propensity, makes us enter deeply into each other’s sentiments, and causes like passions and inclinations to run, as it were, by contagion, through the whole club or knot of companions. Where a number of men are united into one political body, the occasions of their intercourse must be so frequent, for defence, commerce, and government, that, together with the same speech or language, they must acquire a resemblance in their manners, and have a common or national character, as well as a personal one, peculiar to each individual. Now though nature produces all kinds of temper and understanding in great abundance, it does not follow, that she always produces them in like proportions, and that in every society the ingredients of industry and indolence, valour and cowardice, humanity and brutality, wisdom and folly, will be mixed after the same manner.131

Hume argues that we can similarly account for the fact that in the different professions characters are different: The same principle of moral causes fixes the character of different professions, and alters even that disposition, which the particular members receive from the hand of nature. A soldier and a priest are different characters, in all nations, and all ages; and this difference is founded on circumstances, whose operation is eternal and unalterable.132

He puts the point in the Treatise this way: To this principle [sympathy] we ought to ascribe the great uniformity we may observe in the humours and turn of thinking of those of the same nation; and ‘tis much more probable, that resemblance arises form sympathy, than from an influence of the soil and climate, which, tho’ they continue invariably the same, are not able to preserve the character of a nation the same for a century together (T, pp. 316-17).

480

Body, Mind and Self in Hume’s Critical Realism

When we recognize how pride is produced by and in turn produces the idea of the self that determines our being and existence, several things that Hume says about pride (and humility) fall into place. In particular, we can see more clearly what Hume has in mind when he sets out to distinguish pride from joy. Hume makes five points in this context. The first is this. According to Hume, pride has the self as its object, where joy has no such object. But joy like pride is a pleasant feeling, one which is evoked by some cause. The first point that Hume makes is that the cause of pride is more closely related to oneself than is the cause of joy. Hume suggests that “everything related to us, which produces pleasure or pain, produces likewise pride or humility.” However, “there is not only a relation requir’d, but a close one, and a closer than is required for joy” (T, p. 291). Hume is relatively vague on what the relevant relations are. This is because these relations are in fact socially determined, varying from nation to nation, culture to culture, profession to profession, and even person to person. The second point that Hume makes is that the quality of the subject that evokes pride is “peculiar to ourselves or at least common to us with a few persons” (T, p. 291). There is no such limitation on the quality that evokes joy. The reason for this, Hume suggests (T, p. 292), is that the cause must not only cause pleasant sensations, as with joy, but also evoke the idea of the self which is the object of the passion of pride. It will do this easily only if the quality which causes the pleasant sensations is unique or closely unique to the self the idea of which must be evoked in the case of pride. Hume notes elsewhere that what is not distinctive in one context may well be distinctive in another, and that what may relate closely to us in one context need not involve a close relation to self: We love company in general; but ’tis as we love any other amusement. An Englishman in Italy is a friend: A European in China: and perhaps a man wou’d be belov’d as such were we to meet him in the moon. But this proceeds only from the relation to ourselves; which in these cases gathers force by being confined to a few persons (T, p. 482).

These same constraints apply to sympathy. Thus, Hume argues that “we find, that where, beside the general resemblance of our natures, there is any peculiar similarity in our manners, or character, or country, or language, it facilitates the sympathy” (T, p. 318). Conversely, dissimilarity will hinder the sympathy, lessening the strength of the response, perhaps preventing it

481

Five: Hume’s Positive Account

entirely. The fact that the qualities that evoke pride are unique to the self or nearly so is tied to the fact that pride produces our concept of the self. These unique qualities serve to individuate the self, make it distinct from others. Turn now to the third point that Hume makes when he contrasts pride and joy, “that the pleasant or painful object be very discernible and obvious, and that not only to ourselves, but to others also” (T, p. 292). Pride unlike joy requires the support of others, communicated by sympathy. The “causes of pride,” Hume tells us, “have little influence, when not seconded by the sentiments and opinions of others” (T, p. 316). Indeed, as we have argued, pride often arises as a sympathetic response to the love others bear towards us in respect of the quality in question. So both ourselves and others must be able to discern the quality that evokes pride. This condition is part of the social nature of pride. The fourth point that Hume makes is that the quality that evokes pride, in contrast to a quality that evokes joy, must be relatively enduring. “What is casual and inconstant gives but little joy, and less pride” (T, p. 293). Since pride marks an “excellency in ourselves” (ib.) and the self is something enduring, something “more durable” (ib.), some other thing which is only momentarily connected to us will hardly constitute an excellency of the self. We can see that this features of pride also fits with the Humean thesis that pride generates our idea of our self. That idea is the idea of a series of perceptions structured by relations. Among the relations are the continuing presence of certain qualities. If a quality is not continuing, not “durable,” then it will not be able to perform the function of uniting impressions into a self. The fifth point that Hume makes is that “general rules have a great influence upon pride and humility” (T, p. 293). Thus, for example, ... we form a notion of different ranks of men, suitable to the power or riches they are possest of; and this notion we change not on account of any peculiarities of the health or temper of the persons, which may deprive them of all enjoyment in their possessions (ib.)

Since these latter persons do not enjoy their power or riches, it would seem that they could take no pride in them: a quality evokes pride only if it also causes pleasure independently of that pride. They might nonetheless take pride in the rank established by their power or riches. This is because we

482

Body, Mind and Self in Hume’s Critical Realism

restrain our passions in conformity with the general rule that persons take pride in the rank they obtain through power or riches. The passions, we know “are often vary’d by very inconsiderable principles; and these do not always play with a perfect regularity, especially on the first trial” (T, p. 294). But we have “general maxims” which “custom and practice” have established, and these “contribute to the easy production of the passions, and guide us” in the judgments we make about what ought to be preferred. Deriving from “custom and practice” these rules are a matter of convention, not different significantly from the conventions of property, and like the latter are made normative through sympathy. This point is also connected with the generation of the idea of the self, much as the requirement that the qualities in which we take pride be relatively enduring. But following general rules in the formation of such passions as pride and humility we achieve the sort of constancy that is essential to the formation of the stable idea or identity of a self, the idea for one’s own self and also the idea of one’s self that others have. Hume takes seriously this point about the self having stable features – a character – that endure through time because, as he argues, this is essential to the person as a moral being. Without such character traits, there could be no moral responsibility. Hume defends the doctrine of necessity, the thesis that actions are all caused, against the doctrine of liberty, the thesis that actions are uncaused (T, Bk. II, Pt. iii, §§ 10-11). He argues that it is persons as enduring beings that are held to be morally responsible. The constant and universal object of hatred or anger is a person or creature endow’d with thought and consciousness; and when any criminal or injurious actions excite that passion, ’tis only by their relation to the person or connexion with him (T, p. 411).

The doctrine of liberty makes nonsense of moral responsibility. As Hume puts it, ... according to the doctrine of liberty or chance, this connexion is reduc’d to nothing, nor are men more accountable for those actions, which are design’d and premeditated, than for such as are casual and accidental (ib.).

The reason is this: Actions are by their very nature temporary and perishing; and when they proceed not from some cause in the characters and disposition of the person, who perform’d them, they infix not themselves upon him, and can neither

483

Five: Hume’s Positive Account

redound to his honour, if good, nor infamy, if evil (ib.).

It is only if the person who acts is one with stable or enduring features from which actions flow that moral responsibility makes sense. The idea of his or her own self which a person has must contain these enduring features, features which provide a structure to the self, linking its momentary stages to one another over time. The same point can be put another way. Suppose that there was no concept of the self, but that persons were nonetheless moved by the direct passions. These momentary impulses would move the person one way and another according to their immediate force. But there would be no putting one aside for the present in order to achieve a greater gain in the future. There would be no planning and no projects. In order to plan, the person needs a concept of him- or herself that enables that person to envision something in the future happening to him or her, the same person. Self love organizes our impulses over time; it arranges them in a structure that denies the fulfilment of certain immediate impulses in order to secure greater fulfilment at later times. But for this to work, the later benefits must be benefits to oneself. So the concept of the self as enduring is central to one’s being responsible for projects extending over longer periods of time. The concept of the self also enables our sympathy to be enlarged. Hume contrasts how we respond to a person who inflicts injury on another with the response we have to the injured person. We feel pity for the injured person. This pity evokes in us a hatred of the person who did the injury and a love of the person injured. The misfortune produces the hatred in the usual way. But it also produces love because we view the injury suffered in the context of the whole life of the injured person: “... in considering the sufferer we carry our view on every side, and wish for his prosperity, as well as are sensible of his affliction” (T, p. 389). The love is evoked not just by the present pain but by the more extended view we have of his or her well being over time. We can carry our view beyond the present affliction to his future prosperity precisely because we have an idea of him (or her) as something that endures, remains essentially the same, over time. The concept of the self also provides standards for action. Many of the enduring features of the self will be there through pride and humility. They will be there, in other words, because we reckon them to be qualities of excellence or degradation. These qualities will therefore provide standards in terms of which we evaluate for ourselves our future goals and

484

Body, Mind and Self in Hume’s Critical Realism

projects, our possible future actions and their consequences. This ordinary concept of the self as enduring through time provides an identity as “regards our passions or the concern we take in ourselves” Hume of course distinguishes this identity from identity “as it regards our thought or imagination” (T, p. 253). The question of personal identity is “more abstruse” than any other in philosophy (T, p. 189). Indeed, Hume proposes, So far from being able by our senses merely to determine this question, we must have recourse to the most profound metaphysics to give a satisfactory answer to it... (ib.).

In fact, of course, as Hume argues, metaphysics can give no answer to the question. The point is that, if we stick to the second of the two concepts of personal identity, then we can provide no answer to the question of the “nature of the uniting principle which constitutes a person” (ib.). But if we have no such concept then moral responsibility disappears as does the possibility of action and projects that look forward to the future beyond our momentary direct passions. Faux de mieux, then, we must rely upon any concept of the self that we can obtain by means of our “senses.” However, Hume suggests that there are problems with the first of his two concepts of personal identity. ... in common life ’tis evident these ideas of self and person are never very fix’d nor determinate (T, pp. 189-90).

But we must see this point in contrast. Of course when the idea of the self generated by pride is compared to the sort of principle of identity to which the metaphysician aspires, then it is not fixed and determinate. From the viewpoint of the metaphysician, then, the ordinary concept is worthless. But is it worthless from the point of view of the “concern we take in ourselves”? By metaphysical standards, it might fall far short, but by standards of being a useful concept for purposes of our moral life it could turn out to be useful indeed. Hume, we have seen, argues that this is so. The concept of the self that is generated by pride and sympathy is, to be sure, changeable. Experience in the context of social interaction with our fellow human beings will at times undermine parts of our concept, extend it, cause it to contract, but also confirm it. This experience of common life will in general not only change our idea of our self but will also support it. The ordinary concept is indeed changeable, not fixed and determinate, but

485

Five: Hume’s Positive Account

nonetheless is held sufficiently stable for getting on with life among our fellows. It is held stable by custom and habit, by the habits of others which re-inforce our habits of conceiving ourselves and eliminate not our individuality but our idiosyncrasies, and by general rules that we learn over time and apply to both others and ourselves. So much the worse, then, for the metaphysical concept: we have no need of it. There is no problem of personal identity, only the practical task of developing as decent persons in a social context.133 Or, it should be added, there is no problem of personal identity unless one succumbs to the extreme scepticism that calls into question all our ordinary judgments, including judgments about who one is, and even if one exists. Recall the anguished questions of the person in the Treatise who succumbs to the depression and despair consequent upon falling to radical scepticism: “Where am I, or what? From what causes do I derive my existence, and to what condition shall I return?” (T, p. 269) But of that, more later. Return for now to our ordinary sense of our identity, of who we are. Among the features in which we take pride is that of virtue. “To [morally] approve of a character,” we are told, is to feel an original delight upon its appearance. To disapprove of it is to be sensible of an uneasiness. The pain and pleasure, therefore, being the primary causes of vice and virtue, must also be the causes of all their effects, and consequently of pride and humility, which are the unavoidable attendants of that distinction (T, p. 296).

The motive of virtue or the sense of justice has been found by persons experienced in the world to be socially useful, productive of pleasure to others and to the person so motivated. Others approve of one so motivated, loving that person for it. The person so loved will take pride in being so motivated. That motive will in consequence come to be incorporated as a settled trait in our character, in the idea that we have of our self. In fact, it will be something that we ought reasonably to cultivate in ourselves. ... who can think of the advantages of fortune a sufficient compensation for the least breach of social virtues, where he considers, that not only his character with regard to others, but also his peace and inward satisfaction entirely depend upon his strict observance of them; and that a mind will never be able to bear its own survey, that has been wanting in its part to mankind and society? (T, p. 620)

486

Body, Mind and Self in Hume’s Critical Realism

Virtue or the sense of justice is a trait that we ought, reasonably, to take pride in, and consciously develop and encourage in ourselves, or, what is the same, endeavour to make a central ingredient in the idea of our self. But virtue is not the only feature that forms part of the idea of our self. “Our reputation, our character, our name are considerations of vast weight and importance” (T, p. 316), and among the causes of pride. Similarly, we can be “vain of our country,” and others “boast of their ... ancestors...” (T, p. 307). This last is what is at present of importance to us. Given the Hobbesian assumptions about human nature, human beings with character traits such as these are simply not possible. But Hume enlarges the picture of human nature. Human beings do not consist simply of direct passions organized by self love, as in Hobbes. Rather, human nature includes above all sympathy, but also love of others. These yield motives which are disinterested, or, if you wish, altruistic. According to Hume’s account of love, this passion, like pride, requires both a cause and an object. In some cases, however, the cause may itself be the object. The object of love may be a parent, say. This person is also the cause of the love: being a parent is the quality that excites the pleasant feelings that in turn help evoke the passion of love. As Hume puts it, Whoever is united to us by any connexion is always sure of a share of our love, proportion’d to the connexion, without enquiring into his other qualities (T, p. 352).

This means, in particular, that we especially love not only our parents but also our countrymen, our professional colleagues, our fellow chess or backgammon players, our fellow citizens, and others who stand in such relations to us. We love our country-men, our neighbours, those of the same trade, profession, and even name with ourselves (ib.).

This love that is evoked by the quality of being like oneself a Scotsman, or a MacGregor, or a Hume of Ninewells, or a lawyer or librarian, or a chess player, or gardener, of a resident of new Edinburgh opposite the old, and so on, is of course communicated by sympathy, becoming in the object of the love a pride in the quality of being a Scotsman or a MacGregor or a Hume of Ninewells or a lawyer or a Scotsman, or whatever. And this pride in being a Scotsman or a MacGregor or a Hume of Ninewells comes, in the

487

Five: Hume’s Positive Account

person who feels such pride, to be incorporated into the idea they have of themselves, the idea that determines our “being and existence,” as we saw Hume put it. Thus, being a Scotsman or a librarian or whatever as a disposition or set of dispositions comes to be a defining feature of one’s self and a tie, which, through sympathy, love and pride, becomes part of one’s character and binds one, that is, one’s self, to other members of the relevant group. What holds for the mentioned groups holds, of course, for other groups – class or family or country or club or profession, as, for example, a lawyer or an historian or a librarian, or owner of Chateau Lafitte, or descendent of Macbeth, or whatever – you name it –.it becomes part of what defines oneself and binds together in a unifed bundle its manifold parts, past, present, and future, and binds that self into the relevant groups with the varying sets of norms for the members of the group, both oneself and others. Where Hobbes cannot provide an explanation of how such a society and such selves are possible, Hume, with his richer view of human nature, can do that. For our purposes what is important is that the relations that define one’s personal identity include the aspects of one’s character which constitute the patterns that place one in various social groups. (5) Conclusion – The Final One Hume, as we know, follows Hobbes, in making justice and the other norms for civil society a matter in the first instance of prudence, that is, a matter of organizing our selves and our characters better to serve our material interests. But he goes beyond Hobbes, to argue that justice is a virtue and injustice a vice. Hume proposes the question in this way, “Why [do] we annex the idea of virtue to justice, and of vice to injustice?” (T, p. 498). In a society limited to one’s family, sympathy extends as it were only to the limits of one’s family. Sympathy among family member ensures the sharing of scarce resources and prevents conflict over their distribution. But the limitation of sympathy to the members of the family guarantees conflict among family groups. This is, as we have seen, because there is little similarity between oneself and the members of other clans that could facilitate sympathetic responses. The institution of property and the formation of civil society is the artifice designed to overcome the limitations of sympathy. But the norms of property and of civil society are not yet moral rules. In effect this is achieved by sympathy; with the

488

Body, Mind and Self in Hume’s Critical Realism

formation of civil society there comes to be bonds of similarity that enable sympathy to extend beyond the narrow bounds of the family to include others, not quite indiscriminately, but at least to the bounds of the group conforming to the norms of property and of civil society. Once the institution of the ownership of property by individuals is instituted, then we do share something with members of other families: we are both covered by the conventions of justice. We both share the characteristic of being owners of property, and loss of that property would pain us. Since we are similar in being owners of property that similarity will facilitate the working of sympathy. And so, if the member of another family loses his or her property through theft, the similarity which that person bears to me will facilitate my responding sympathetically to the other’s pain. It is true that simply out of self interest we do have an interest in maintaining throughout society conformity to the rules of property. But our capacity to trace out consequences of violating these rules is often difficult. Interest therefore will often not provide strong enough sentiments to lead us to enforce these rules among persons to whom we have little connection. But this failure on the part of interest finds a remedy in sympathy. [W]hen society has become numerous, and has encreas’d to a tribe or nation, this interest [that is, the “public” or “common interest” (T, p. 500, p. 490)] is more remote; nor do men so readily perceive, that disorder and confusion follow upon every breach of these rules [of justice], as in a more narrow and contracted society. But tho’ in our actions we may frequently lose sight of that interest, which we have in maintaining order, and may follow a lesser and more present interest, we never fail to observe the prejudice we receive, either mediately or immediately, from the injustice of others.... Nay when the injustice is so distant from us, as no way to affect our interest, it still displeases us; because we consider it as prejudicial to human society, and pernicious to every one that approaches the person guilty of it. We partake of their uneasiness by sympathy ... (T, p. 449).

It is this that transforms the conventions of justice, and the other norms of civil society, from rules based on interest to rules conformity to which is a virtue and non-conformity to which is a vice. Since, Hume argues, “...every thing, which gives uneasiness in human actions, upon the general survey, is call’d Vice, and whatever produces satisfaction, in the same manner, is call’d Virtue; this is the reason why the sense of moral good and evil follows upon justice and injustice” (ib.). Once again, general rules come to play a role. We generalize and regulate our own conduct in conformity to that rule. Others do the same, applying the same rule to both their own self

489

Five: Hume’s Positive Account

and yourself. Those others love us when we conform to the rule, and we sympathize with that love in the form of pride, being pleased with ourselves for being virtuous, or being humble and pained at ourselves for being vicious. The general rule reaches beyond those instances, from which it arose; while at the same time we naturally sympathize with others in the sentiments they entertain of us (ib.).

The moral sentiments thus function to strengthen the tendency in each to conform their actions to the convention of justice. In our deliberations, we may be tempted to reason as does the person Hume calls the “sensible knave”: Treating vice with the greatest candour, and making it all possible concessions, we must acknowledge, that there is not, in any instance, the smallest pretext for giving it the preference above virtue, with a view to self-interest; except, perhaps, in the case of justice, where a man, taking things in a certain light, may often seem to be a loser by his integrity. And though it is allowed, that, without a regard to property, no society could subsist; yet, according to the imperfect way in which human affairs are conducted, a sensible knave, in particular incidents, may think, that an act of iniquity or infidelity will make a considerable addition to his fortune, without causing any considerable breach in the social union and confederacy. That honesty is the best policy, may be a good general rule; but is liable to many exceptions: And he, it may, perhaps, be thought, conducts himself with most wisdom, who observes the general rule, and takes advantage of all the exceptions.134

The problem with the sensible knave is his or her lack of sympathy and the sense of moral virtue consequent upon that. If, as with most persons, he or she were to have an active moral sense, his or her feelings that theft is wrong would, as he or she rationally deliberated, prove to provide a motive strong enough to outweigh the direct passion that tempts him or her towards the theft. Hume summarizes his account of justice in this way: ... self-interest is the original motive to the establishment of justice: but a sympathy with public interest is the source of moral approbation, which attends that virtue (T, p. 490-500).

What holds for property, also holds for the institution of government, and the rule of conforming to the constitution of one’s nation and of

490

Body, Mind and Self in Hume’s Critical Realism

conforming to the decisions of a chief magistrate. The task of the chief magistrate is to enforce the rules of civil society, through threat of punishment for violations, in conformity with the constitution, and in particular to ensure ownership of property and the honouring of contracts. The first motive for the formation of government is interest, our interest in a more strict observance of the fundamental laws of property, transference and contract. So we have our civic duties in order to secure the better conformity to the three fundamental conventions. “So far, therefore, our civil duties are connected with our natural, that the former are invented chiefly for the sake of the latter; and that the principal object of government is to constrain men to observe the laws of nature” (T, p. 543). Government thus far is prudential: “[o]ur interest is always engag’d on the side of obedience to magistracy...” (ib.). But the same mechanism works here as in the case of the norms of property and promising or contract, and allegiance thereby becomes not only prudential but moral. Disobeying the civil magistrate is “highly prejudicial to public interest,” and, as Hume says, “to my own in particular.” He continues, This naturally gives us an uneasiness, in considering such seditious and disloyal actions, and makes us attach to them the idea of vice and moral deformity (T, p. 545).

That magistrate is politically legitimate to whom we are bound by the moral ties of allegiance. Finally, we can say that these norms come to be embedded in the idea one forms of oneself: we view ourselves as virtuous persons and loyal citizens.135 Conformity to these norms, that is, civic virtue, comes to be an aspect of one’s character, patterns of thought and action that define one’s identity as a self. This pattern is general, we have been arguing. Other norms, defining other groups, come similarly to be incorporated into our character and therefore to be among the “durable” patterns of behaviour that bind together the parts of that bundle that is our self. As we have seen Hume point out, We love our country-men, our neighbours, those of the same trade, profession, and even name with ourselves (T, p. 352).

This love generates pride through sympathy. And so we come to be not only loyal British subjects, but also Scotsmen. We also incorporate into our

491

Five: Hume’s Positive Account

self identity the norms of our profession or trade. Our identity as a self is thus made stable and secure. With these things in mind we can now return to the melancholy and despair into which Hume (or Pascal) supposes one can fall when one pursue the sceptical arguments of the Pyhrronist (or the Cartesian) to their ends where we conclude that we really do know nothing. We are now in a position to identify the cause of this melancholy. The cause, as Hume sees it, is the attempt to establish a priori a secure starting point for reason. This is, contrary to Descartes, something that is impossible to do: contrary to Descartes, reason a priori cannot provide us with a starting point that is absolutely secure and infallible: if we take up that task we shall never complete it, for it is something that is impossible to do. And to attempt an impossible task leads to melancholy and despair over one’s incapacities. Implicitly Hume is arguing that the way to avoid the melancholy is to avoid attempting this impossible task. As Hume sees it, what initially provides a cure for this melancholy, this “splenetic humour” (T, p. 269), is being out in the world. Hume puts it this way: Most fortunately it happens, that since reason is incapable of dispelling these clouds, nature herself suffices to that purpose, and cures me of this philosophical melancholy and delirium, either by relaxing this bent of mind, or by some avocation, and lively impression of my senses, which obliterate all these chimeras. I dine, I play a game of back-gammon, I converse, and am merry with my friends; and when after three or four hours’ amusement, I wou’d return to these speculations, they appear so cold, and strain’d, and ridiculous, that I cannot find in my heart to enter into them any further (T, p. 269)

The remedy is getting out into the world, that is, business on the one hand and diversion, backgammon for example, on the other. Hume does not recommend a total abandonment of philosophy, what is to be given up is the search for an infallible a priori starting point, not the search for empirical knowledge, satisfying to a restrained curiosity, and which could be useful to us, if only for recreation. Hume allows that there are times when he is tired of mere amusement, and finds himself moved by the passion of curiosity. I cannot forbear having a curiosity to be acquainted with the principles of moral good and evil, the nature and foundation of government, and the cause of those several passions and inclinations which actuate and govern me. I am uneasy to

492

Body, Mind and Self in Hume’s Critical Realism

think I approve of one object, and disapprove of another; call one thing beautiful, and another deformed; decide concerning truth and falsehood, reason and folly, without knowing upon what principles I proceed. I cannot forbear having a curiosity to be acquainted with the principles of moral good and evil, the nature and foundation of government, and the cause of those several passions and inclinations which actuate and govern me. I am uneasy to think I approve of one object, and disapprove of another; call one thing beautiful, and another deformed; decide concerning truth and falsehood, reason and folly, without knowing upon what principles I proceed. I cannot forbear having a curiosity to be acquainted with the principles of moral good and evil, the nature and foundation of government, and the cause of those several passions and inclinations which actuate and govern me. I am uneasy to think I approve of one object, and disapprove of another; call one thing beautiful, and another deformed; decide concerning truth and falsehood, reason and folly, without knowing upon what principles I proceed (T, pp. 270-1).

Notice that, where until now Hume had being limiting himself to the case of principles of reason, he now makes clear that it was not just reason with which he was concerned. He here shows that moral principles were equally threatened. The despair and melancholy Hume is describing would concern not merely an incapacity to know matters of fact but also the incapacity to have a secure and infallible starting point for one’s moral principles, an incapacity that would imply a lack of any sound basis for one’s actions, an incapacity to recognize, and therefore do, what is morally right. Any aspiration I might have to be moral is thwarted through my incapacity to grasp what is moral, my lack of a moral sense. What this means is that the despair that Hume has been painting is not merely his own despair, nor even the despair that was the disease of the learned, but the despair of the person (like Pascal) who cries with the Psalmist, “My God! My God! Why hast thou forsaken me?” This scepticism about God produces despair, embitteredness, spiritual listlessness, and uncertainty. Knowing nothing, there is nothing that can be done. At the same time, it is also true that everything goes, everything is permitted. Nothing can be done, but the grossest immorality is possible. For, with no moral principles to hand, one comes to be in the position of Hume’s knave. Lack of a moral conscience can in effect let one give oneself permission to indulge affections such as avarice, greed, immorality and selfishness, leading to the vicious behaviour of the knave. There is, Hume is arguing, a cure for this sort of despair consequent upon an extreme scepticism. What we need, of course, is knowledge of moral principles. Hume, in describing this despair in Book I of the

493

Five: Hume’s Positive Account

Treatise, is setting himself to provide a reasoned justification for a set of moral principles. This reasoned justification will appear in Book III of the Treatise, after the necessary background has been given in Book II. What he will do, however, is aim at something that can be achieved. He will not aim to do, as Pascal tried to do, the impossible task of providing an incorrigible foundation for morality, or for knowledge, one that requires a use of faculties (which we do not have) that must somehow take one beyond the world of ordinary experience, “the sphere of common life” (T, p. 271). He will, to the contrary, limit himself to this ordinary world. It is certain that superstition is much more bold in its systems and hypotheses than philosophy; and while the latter contents itself with assigning new causes and principles to the phenomena which appear in the visible world, the former opens a world of its own, and presents us with scenes, and beings, and objects, which are altogether new. Since, therefore, it is almost impossible for the mind of man to rest, like those of beasts, in that narrow circle of objects, which are the subject of daily conversation and action, we ought only to deliberate concerning the choice of our guide, and ought to prefer that which is safest and most agreeable. And in this respect I make bold to recommend philosophy, and shall not scruple to give it the preference to superstition of every kind or denomination (T, p. 271).

In the sphere of knowledge, Hume will defend principles conformity to which will, so far as one can tell, contribute to one’s end of “curiosity, or love of truth,” a passion which he discusses later in Book II of the Treatise (Bk. II, Pt. iii, sec. x). He is now no longer simply indulging in an “indolent belief in the general maxims of the world” (T, p. 269); this was the attitude that he had when he first discovered that he could not find a priori any secure foundation on which to build the edifice of knowledge (and morality). Instead Hume – and with him, the reader – is now actively searching out for the means, the cognitive means, that are best suited to fulfilling this passion of curiosity, to the extent that we can fulfill that passion. These means consist in conforming to certain principles of thought. These principles are, of course, the “rules by which to judge of causes and effects” (T, Bk. I, Pt. iii, sec. xv) that Hume has earlier defended in the Treatise (T, Bk. I, Pt. iii, secs. xi, xiii).136 We no longer search for some sort of absolute certainty. Rather, ... we might hope to establish a system or set of opinions, which if not true (for that, perhaps, is too much to be hop’d for) might at least be satisfactory to the human mind, and might stand the test of the most critical examination (T, p.

494

Body, Mind and Self in Hume’s Critical Realism

272).

This active search is for knowledge that we can achieve, recognizing that although we can achieve certain things, we are for all that capable of error and cannot secure ourselves a priori from the possibility that we are wrong. As Hume now describes the cognitive stance of a person who adopts this attitude, The conduct of a man, who studies philosophy in this careless manner, is more truly sceptical than that of one, who feeling in himself an inclination to it, is yet so over-whelm’d with doubts and scruples, as totally to reject it (T, p. 273).

Similarly, just as there is no a priori starting point for knowledge, so there is no a priori starting point for ethics. In the sphere of morals, the principles are those such as justice, promise keeping, and allegiance that contribute to the well being of humankind. This solves the sceptical problem that generated the melancholy and despair. It solves the problem by rejecting the position created by the Cartesian of demanding an a priori starting point in morals and in knowledge, that is, by demanding a starting point which is beyond our human capacities to achieve. Above all, there is a solution to the problem (to which Pascal pointed, as had Hume himself) of losing one’s identity and sense of self, the problem of the unbinding of the self into something that is nothing more than a bundle – the bundle dissolves into its parts and really is no longer a genuine self. The self as an empirical and social construct, and as a moral subject, can be subjected to the same general sceptical critique of the Cartesian. Descartes of course has a way out: he has (he claims) no problem with his identity – sum res cogitans. But, as Pascal saw, the sceptical argument cannot be avoided, there is no substantial self that can bind the bundle of perceptions into a self and give it an enduring identity: “Where am I, or what? From what causes do I derive my existence, and to what condition shall I return?” asks the person of the Treatise who has fallen into profound scepticism. The “I” remains, the centre of consciousness. But the empirical self, the “I” who has an identity in the world, is called into question, just as all causal and empirical reasoning is called into question, just as all matter-of-fact judgment is called into question: where am I, or what? The solution to this question of the self, for Hume, is to adopt the commonsense view, that the self is indeed a bundle of impressions, but an ordered bundle, and that order is not to be rejected simply because various uncertainties attach to it, just because we might not

495

Five: Hume’s Positive Account

be fully pellucid to ourselves. This is solving the problem of personal identity by providing the self with an identity in a civil society: it is this solution that Hume recorded in the Treatise. There is of course another solution. Pascal too argued that the Cartesian programme lead to a sceptical conclusion in which knowledge of the world, of morality, and of the self is impossible. Cartesian infallibility and certainty can be achieved only by finding a world that is beyond our ordinary world, one that is timeless and unchanging, solid and enduring. But the task is impossible. If one searches after a world that is beyond the world of common life, then it cannot be successful: the reasoner finds that he or she cannot achieve the only sort of solution that can be satisfying. But the Pascalian solution of faith and a God and a soul or self that are beyond time is only a pretended solution, an illusion. Precisely because it is an illusion it is unstable. If it is capable of producing moral behaviour, it can also produce much that is, if not silly, then simply horror. This point was made by Robert Burton in what was perhaps the most well known of the anatomies of melancholy. This is of course Robert Burton’s Anatomy of Melancholy,137 originally published in 1632, but repeatedly enlarged in subsequent editions. Burton spends much time on faith as the cure for melancholy. For Burton, the prognostics of religious melancholy are none “otherwise than folly, dotage, madnesse, grosse ignorance, despaire, obstinacy, a reprobate sense, a bad end.” (III.4.1.4, vol. 3, p. 189) As a consequence, “Wee heape upon ourselves unnecessary troubles, observations; wee punish our bodies ....” As for “all honest recreations” of which it can be said that “God hath ... indulged them to refresh, ease, solace and comfort us,” ...we are some of us too sterne, too rigid, too precise, too grossely superstitious, and whilst we make a conscience of every toy, with touch not, tast not, &c....; we tyrannize over our brothers soul, loose the right use of many good gifts, honest sports, games and pleasant recreations, punish ourselves without a cause, lose our liberties, and sometimes our lives.(Ibid.)

When “evill is expected, we feare: but when it becomes certaine, we despaire.” (III.4.2.2, vol. 3, p. 408) The certainty of evil can result in what are in effect heroic efforts to overcome the perceived evil. “Because they cannot obtaine what they would, they become desperate, and many times either yeeld to the passion by death it selfe, or else attempt impossibilities, not to be performed by men.”(Ibid.. p. 409) To cope the afflicted often resort to superstition, blasphemy, or heresy. This is indeed a remedy of

496

Body, Mind and Self in Hume’s Critical Realism

sorts, but one that itself has terrible effects, since it can lead at worst to persecution and war, or at least to the private torment of souls. “What else,” Burton asks rhetorically, “can superstition, heresie produce, but wars, tumults, uproars, torture of soules, and despaire, a desolate land ... when they commit Idolatry, and walk after their own waies: how should it be otherwise with them? What can they expect by blasting, famine, dearth, and all the plagues of Egypt ... ?”( III.4.1.4, vol. 3, p. 389). Hume agrees with Burton, that war and persecution are among the consequences of the religious solution, or, alternatively, foolish practices such as increased fasting or denying oneself and others simple amusements. Hume makes Burton’s point this way: ... as superstition arises naturally and easily from the popular opinions of mankind, it seizes more strongly on the mind, and is often able to disturb us in the conduct of our lives and actions. Philosophy, on the contrary, if just, can present us only with mild and moderate sentiments; and if false and extravagant, its opinions are merely the objects of a cold and general speculation, and seldom go so far as to interrupt the course of our natural propensities. The Cynics are an extraordinary instance of philosophers, who, from reasonings purely philosophical, ran into as great extravagances of conduct as any monk or dervise that ever was in the world. Generally speaking, the errors in religion are dangerous; those in philosophy only ridiculous (T, pp. 271-2).

This, then, is Hume’s conclusion about the claim that the despair and melancholy consequent upon the discovery that reason ends up in an incurable scepticism, finds its cure in a faith that takes one from this world to a God that supposedly inspires in one decent and moral behaviour. But faith doesn’t do that and in fact it’s dangerous. Backgammon and making ourselves busy in the world of ordinary business soon shows that one does have an identity, that one knows who one is and where one is at, that there is an achievable cognitive goal for reason, namely, the goal of improving our ordinary matter of fact truth, that there is a method for doing this systematically, namely, the method of empirical science, and that there is a foundation, one that is reasonably secure, in human nature for the moral rules that permit us to live together with our fellow persons. This conclusion is reached by a careful use of the process of self examination that in a way was central to the Cartesian programme. This process of self examination has been turned by Hume into something wholly of this world of ordinary empirical fact, and which is also something we can do. This is the use of the scientific method to delineate

497

Five: Hume’s Positive Account

the structure of Human Nature. The reasonable person will have as his or her project that of introspective psychology, the project of uncovering the associationist causes of expectations, of our passions, and of our moral sense. And it has become the problem of locating empirically the processes that account for the origin of our sense of personal identity; and thereby discovering the ties that, on the one hand, bind the events in the bundle which is our history into a structured self, and that, on the other hand, bind this self to others in the civil society in which we find ourselves – and the discovery of how we can go about creating for oneself this identity that defines one’s self.

Endnotes to Chapter Five

1Compare his remark in the posthumously published essay “On the Immortality of the Soul”: Nothing in the world is perpetual. Every being, however seemingly firm, is in continual flux and change: The world itself gives symptoms of frailty and dissolution: How contrary to analogy, therefore, to imagine that one single form, seemingly the frailest of any, and from the slightest causes, subject to the greatest disorders, is immortal and indissoluble? What a daring theory is that! How lightly, not to say, how rashly entertained! (p. 597) 2Hume, Dialogues concerning Natural Religion, pp. 156-57. 3Ibid., p. 159. 4Ibid. 5Ibid., p. 160. Jane McIntyre has convincingly argued that the bundle view was consistently Hume’s throughout his career, and throughout his philosophical corpus from beginning (the Treatise) to the end (the Dialogues); see her essay on “Hume’s Underground Self.” 6G. Vesey, Personal Identity: A Philosophical Analysis, pp. 108-9. 7Compare the account of conscious states in G. Bergmann, “Acts.” 8N. Capaldi, “The Historical and Philosophical Significance of Hume’s Theory of the Self,” p. 279.

498

Body, Mind and Self in Hume’s Critical Realism

9Certainly, the Plotinian and quasi-Hegelian conclusions that Capaldi (ibid., pp. 280-1, pp. 282-4) draws in no way follow, nor does the further conclusion Capaldi draws (p. 280ff) that we must abandon an empiricist and positivist reading of Hume. 10G. Ryle, The Concept of Mind, pp. 162-3. 11B. Russell, “On the Nature of Acquaintance,” p. 166ff. 12J. S. Mill, “Auguste Comte and Positivism,” p. 296. 13Ibid. The reference to Hamilton is to his Lectures on Metaphysics and Logic. 14Hamilton, Lectures, p. 248-54. As Mill, “Auguste Comte and Positivism,” notes, Hamilton’s number of six is a bit silly; on points like this, “numerical precision is out of the question” (p. 296n). 15J.-J. Séverin de Cardaillac, Études élémentaires de philosophie, vol. I, pp. 4-5; his italics. 16In fact, it turns out that the ontology of substances cannot account for this fact, where the non-substantialist ontology of Hume and the Mills can; cf. F. Wilson, “Mill and Comte on the Method of Introspection.” 17For this notion of ‘linguistic role’ and ‘meaning’, see W. Sellars, “Reflections on Language Games,” and “Notes on Intentionality.” Also F. Wilson, “Marras on Sellars on Thought and Language,” and “Effability, Ontology and Method,” both of which can be found in F. Wilson, Acquaintance, Ontology and Method. 18L. Wittgenstein in The Blue and Brown Books, p. 67. 19R. Chisholm, “On the Observability of the Self.” 20A. Shalom, The Body/Mind Conceptual Framework and the Problem of Personal Identity. 21It is precisely at this point we need an adequate ontology of relations, if Hume’s bundle view of the self is to have a reasonable defence. For this account, one needs the notion of the relative product of a relation with itself, an idea which is incoherent on the nominalist account of relations, but not, of course, on the account given to us by Russell and James. That is why chapter Two, above, which devotes considerable space to the ontology of relations, is important to our present discussion of how bundles are bundled. 22J. S. Mill, Examination of Sir William Hamilton’s Philosophy, p. 205n.

499

Five: Hume’s Positive Account

23 A. J. Ayer, The Concept of Person and Other Essays, p. 116. 24But as we have seen, and contrary to what is claimed in an interesting paper by David and Alan Hausman, “Hume’s Use of Illicit Substances,” the fact that all that we experience is in a sense “mine” is not the difficulty that Hume is discussing when he worries in the “Appendix” to the Treatise that he has not solved the problem of personal identity. Nor is it a difficulty for Hume’s critical realism. 25Ian Gallie, “Mental Facts,” and J. R. Jones, “The Self in Sensory Cognition.” 26As if, perhaps, she had not fully grasped what Wittgenstein was trying to say. 27Wittgenstein in The Blue and Brown Books, p. 67. 28G. E. M. Anscombe, “The First Person.” 29P. F. Strawson, The Bounds of Sense.. 30Don Garrett, “Hume’s Self-Doubts about Self Identity,” pp. 337-59. 31Barry Stroud, Hume, p. 138. 32Compare Hume’s remark in the posthumous essay “On the Immortality of the Soul”: Everything is in common between soul and body. The organs of the one are all of them the organs of the other. The existence therefore of the one must be dependent on that of the other. (p. 596) 33Cf. S. Shoemaker, Self-Knowledge and Self-Identity. Also B. Williams, “The Self and the Future.” 34The operation has been performed to produce in particular a treatment for certain epileptic symptoms. Cf. R. W. Sperry, “Hemisphere Disconnection and Unity in Conscious Awareness.” 35Cf. D. Wiggins, Identity and Spatiotemporal Continuity. 36Cf. B. Williams, “Are Persons Bodies?” 37Cf. F. Wilson, Laws and Other Worlds. 38Cf. D. Parfit, “Personal Identity.” 39N. Capaldi, David Hume, p. 138.

500

Body, Mind and Self in Hume’s Critical Realism

40For what follows, see the important essay of Annette Baier, “Hume on Heaps and Bundles.” 41This point has been vigorously defended by Wilfrid Sellars; see, for example, his “Reflections on Language Games.” 42D.Livingston, Hume’s Philosophy of Common Life. 43 Hume, Letters, ed. J. Y. T. Grieg, vol. I, p. 201. 44Jones, Hume’s Sentiments, p. 146. 45 Ibid., Ch. 4. 46 Ibid., p. 146. 47 Ibid., p. 144. 48Baier, A Progress of Sentiments, p. 100. 49This long argument has been examined in detail in F. Wilson, Hume’s Defence of Causal Inference. 50I am not suggesting that “internalization” is an explanation of this event; it merely describes it. The explanation is to be found in the learning theory that provides a law or generalization that relates learning experiences to the fact of internalization. 51Levi-Strauss, Structural Anthropology, pp.273-4. 52Levi-Strauss, Structural Anthropology, p. 121; also Elementary Structures, Preface. 53 “...social reactions consist of the raw material out of which the models are built, while social structure can, by no means, be reduced to the ensemble of social relations to be described in a given society” (Social Anthropology, p. 271). 54Cf. Levi-Strauss, Elementary Structures, p. 483. 55Levi-Strauss, Structural Anthropology, pp. 275-6. 56Levi-Strauss, Structural Anthropology, p. 292. 57 “Rather let us own that the notions of prescriptive and preferential are relative: a preferential system is prescriptive when envisaged at the model level; a prescriptive system must be preferential when envisaged at the level of reality” (Levi-Strauss, Elementary Structures, p. xxxiii).

501

Five: Hume’s Positive Account

58The term is due to J. L. Mackie, “Causes and Conditions.” 59The role of gappy laws in the covering law model of explanation is discussed in G. Bergmann, Philosophy of Science, Ch. II; M. Brodbeck, “Explanation, Prediction, and ‘Imperfect’ Knowledge”; and in considerable detail in F. Wilson, Explanation, Causation and Deduction and F. Wilson, The Logic and Methodology of Science and Pseudoscience. 60Cf. Gustav Bergmann, Philosophy of Science, Ch. II. 61Cf. F. Wilson, Laws and Other Worlds, Ch. I. 62On the role that statistics thus plays in science, see F. Wilson, Laws and Other Worlds, Ch. I, sec. i, and Empiricism and Darwin’s Science, Ch. VII and VIII. 63Cf. Structural Anthropology, pp. 271-3. See also algebra of kinship constructed by André Weil which Levi-Strauss included as Ch. XIV of his Elementary Structures. 64.Latin, 1666; English, 1690. 65.Enchiridion Ethicum, the English edition of 1690 (New York: 1930), p. 41. 66.P. 78. 67.P. 54. 68.Pp. 78-9. 69.P. 54. 70.William Wollaston, The Religion of Nature Delineated (London: 1726), p. 165. This was a very popular work. The first edition was in 1722, with eight further editions before 1759. 71.See Hume’s letter of 1757 to Adam Smith in J. Y. T. Grieg, ed., The Letters of David Hume, vol. II, p. 244-, pp. 255-6. 72Hume, Letters, ed. Greig, vol. II, p. 215. 73Ibid., vol. II, p. 261. 74Ibid., vol. II, pp. 40-1. 75 Quoted, Hume, Letters, ed. Greig, vol. II, p. 41n. 76.Henry Mackenzie, Account of the Life and Writings of John Hume (Edinburgh,

502

Body, Mind and Self in Hume’s Critical Realism

1822), p. 38. 77 “The Stoic,” in David Hume, The Philosophical Works, ed. T. H. Green and T. H. Grosse, vol. 4, p. 209. 78.Brian Vickers, “Introduction” to Henry Mackenzie, The Man of Feeling, p. ix. 79.Sir Walter Scott, Lives of the Novelists, p. 164. 80.Ibid., p. 171. 81David Hume, “Of the Populousness of Ancient Nations,” p. 387. Hume is in fact arguing against the views of Andrew Fletcher of Saltoun who argued against the Union of Scotland and England and proposed instead a form of serfdom for most Scots in order to secure a minimum of economic security for all. Hume argues that “Some passionate admirers of the ancients [that is, Fletcher of Saltoun], and zealous partizans of civil liberty, (for these sentiments, as they are, both of them, in the main, extremely just, are found to be almost inseparable) cannot forbear regretting the loss of this institution [slavery]; and whilst they brand all submission to the government of a single person with the harsh denomination of slavery, they would gladly reduce the greater part of mankind to real slavery and subjection. But to one who considers coolly on the subject it will appear, that human nature, in general, really enjoys more liberty at present, in the most arbitrary government of EUROPE, than it ever did during the most flourishing period of ancient times” (p. 387). 82P. Árdal, “Some Reflections upon a Standard Definition of Punishment” and “Does Anyone Ever Deserve to Suffer?” 83Árdal, “Does Anyone Ever Deserve to Suffer?” p. 60. 84“Does Anyone Ever Deserve to Suffer?”, p 61. 85Ibid. 86P. Árdal, “Of Sympathetic Imagination,” p. 68. 87Ibid ., p. 67. 88.Carol Gilligan, In a Different Voice: Psychological Theory and Women’s Development. 89.Alisa L. Carse, “The ‘Voice of Care’: Implications for Biomedical Education.” 90.Aristotle, Nichomachean Ethics, 1106b21-23. 91.Carse, p. 18.

503

Five: Hume’s Positive Account

92.Carse, p. 6. 93.Gilligan, In a Different Voice, p. 24. 94.Carol Gilligan, “Moral Orientation and Moral Development,” p. 23. 95Árdal, Passion and Value in Hume’s Treatise. 96Árdal, “Of Sympathetic Imagination,” p. 69. 97Ibid. 98Ibid. 99.Joseph Butler, The Analogy of Religion, I, v, 8, in The Works of Joseph Butler, ed. W. E. Gladstone, vol. I, p. 92. 100.Ibid., I, v, 9, vol. I, p. 93. 101.Carse, p. 21. 102Ibid., p. 66. 103Ibid., p. 67. 104Cf. B. Williams, Ethics and the Limits of Philosophy; A. Baier, “Theory and Reflective Practices,” and “Doing without Moral Theory”; Philippa Foot, “Are Moral Considerations Overriding?” and Owen Flanagan, “Admirable Imperfection and Admirable Immorality.” But see also Robert B. Louden, “Can We Be Too Moral?” 105In the well-known essay, Michael Stocker, “The Schizophrenia of Modern Ethical Theories” (1976). 106Hume, The Natural History of Religion. 107Friedrich Nietzsche, On the Genealogy of Morals. 108Friedrich Nietzsche, Beyond Good and Evil (hereafter “BGE”), Ch. 1, on the “Prejudices of Philosophers”, 6, pp. 10-11. 109We cannot underestimate the impact of Nietzsche; but the impact is that of a Humean: Nietzsche simply re-states Hume for the nineteenth century. For a just evaluation of Nietzsche’s impact, see Thomas Mann, “Nietzsche’s Philosophy in the Light of Recent History.” 110It came by way of F. Lange’s History of Materialism – really a history of

504

Body, Mind and Self in Hume’s Critical Realism

empiricism – which Nietzsche read near the start of his career. I have used F. A. Lange, The History of Materialism, third edition, trans E. C. Thomas, with an Introduction by Bertrand Russell (London: Routledge and Kegan Paul, 1925). The first German edition was published in 1865; volume I of the second edition came out in 1873, volume II in 1875. The English translation is from this second German edition. Nietzsche’s connections with Lange are explored in detail in G. Stack, Lange and Nietzsche. 111Cf. N. Davey, “Nietzsche and Hume on Self and Identity,” and “Nietzsche, the Self, and Hermeneutic Theory”; and C. Cox, “The ‘Subject’ of Nietzsche’s Perspectivism.” 112Nietzsche, On the Genealogy of Morals, Preface, 6, p. 20. 113James Thomson, The Seasons and the Castle of Indolence, ed., J. Sambrook, “Summer,” ll. 458-478, pp. 49-50. 114B. Williams, “A Critique of Utilitarianism,” p. 98ff. 115Cf. Brian Leiter, “Nietzsche and the Morality Critics.” 116Hume, Enquiry concerning the Principles of Morals, p. 282. 117Ibid., p. 315. 118David Hume, “Of the Dignity or Meanness of Human Nature,” pp. 882-83. 119

Blaise Pascal, Pensées, first published 1670. References are to numbered Pensée and page of the Penguin edition.

120

Voltaire, Letters concerning the English Nation, Letter XXV, “On Paschal’s Thoughts concerning Religion, &c.” This letter was omitted from the first edition (in English) of 1733, and appears first in the popular first French edition of 1734.

121Compare J. McIntyre, “Personal Identity and the Passions.” 122For discussion of Hume on one’s own character, see J. McIntyre, “Character: A Humean Account.” 123Hume, Enquiries, p. 88. 124Páll Árdal, Passion and Value in Hume’s Treatise, p. 45f. 125This point has been made by A. Rorty, “‘Pride Produces the Idea of Self’: Hume on Moral Agency.” Rorty does not, however, spell out the associative mechanisms that

505

Five: Hume’s Positive Account

generate the idea of the self. Moreover, it is important, as we shall see, to take into account the role of humility in forming one’s concept of oneself. This point Rorty simply ignores. 126Cf. D. Ainslie, “Scepticism about Persons in Book II of Hume’s Treatise.” 127Cf. Rorty, op. cit., p. 260. Also D. Davidson, “Hume’s Cognitive Theory of Pride,” and A. Baier, “Hume’s Analysis of Pride.” 128As does Rorty, op. cit., p. 258. 129Hume, Essays: Moral, Political and Literary, ed. Green and Grose, vol. I, p. 244. 130Ibid. 131Ibid., p. 248. 132Ibid., p. 245. 133To make this point is not to say that there are no problems with Hume’s psychological account of how we develop our concept of our self. 134Hune, Enquiry concerning the Principles of Morals, pp. 282-3. 135Cf. D. Ainslie, “The Problem of the National Self in Hume’s Theory of Justice.” 136Cf. F. Wilson, Hume’s Defence of Causal Inference. 137 Robert Burton, The Anatomy of Melancholy, ed. T. Faulkner, N. Kiesling, and R. Blair, with an Introduction by J. B. Bamborough, 3 vols. The Anatomy of Melancholy first appeared in1621, and has been reprinted many times. Burton divides his massive work into numbered Partition, Section, Member, and Subsection. References will be to the numbered division, as well as volume and page of the cited edition. For a helpful discussion of Burton, see Patricia Vicari, The View from Minerva’s Tower.

506

Bibliography Addis, L. The Logic of Society (Minneapolis: University of Minnesota Press, 1976). Addis, L., “Pains and Other Secondary Mental Entities,” Philosophy and Phenomenological Research, 47 (1986), pp. 59-74. Addis, L., “Ryle’s Ontology of Mind,” in L. Addis and D. Lewis, Moore and Ryle: Two Ontologists (The Hague: Nijhoff, 1965), pp. 1-101. Ainslie, D., “The Problem of the National Self in Hume’s Theory of Justice,” Hume Studies, 21 (1995), pp. 289-313. Ainslie, D., “Scepticism about Persons in Book II of Hume’s Treatise,” Journal of the History of Philosophy, 37 (1999), pp. pp. 469-92. Allaire, E. B., “Berkeley’s Realism,” in E. B. Allaire et al., Essays in Ontology, pp. 92-105. Allaire, E. B. et al., Essays in Ontology (The Hague: Nijhoff, 1963). Allestree, Richard, The Practice of Christian Grace; or Whole Duty of Man, Laid Down in a Plain and Familiar Way for All but Especially the Meanest Reader, with Private Devotions for Several Occasions. With a Preface by the Rev. William Bentinck Hawkins (London: William Pickering, 1842; first published 1658). Allison, Henry E., “Locke on Personal Identity: A Re-Examination,” Journal of the History of Ideas, 27 (1966), pp. 41-58 Anderson, F. H., “The Influence of Contemporary Science on Locke’s Method and Results,” University of Toronto Studies in Philosophy, vol. 2, no. 1 (Toronto: University of Toronto Library, 1923). Anscombe, G. E. M., “Hume Reconsidered,” (Review of A. Flew, Hume’s Philosophy of Belief), Blackfriars, 43 (1962), pp. 187-90. Anscombe, G. E. M., Intention, 2nd ed. (Oxford: Blackwell, 1963). Anscombe, G. E. M., “The First Person,” in S. Guttenplan, ed., Mind and Language (Oxford: Oxford University Press, 1975), pp. 45-65. Arbuthnot, J., J. Swift, A. Pope, and J. Gay, Memoirs of the Extraordinary Life, Works, and Discoveries of Martinus Scriblerus, ed. Charles Kerby-Miller (New

Bibliography

Haven: Yale University Press, 1950). Árdal, P., “Does Anyone Ever Deserve to Suffer?” in his Passions, Promises and Punishments, pp. 50-63. Árdal, P.,“Of Sympathic Imaginiation,” in his Passions, Promises and Punishments, pp. 64-72. Árdal, P., Passion and Value in Hume’s Treatise, (Edinburgh: Edinburgh University Press, 1966). Árdal, P., Passions, Promises and Punishments, (Reykjavik: University of Iceland Press, 1998). Árdal, P., “Some Implications of the Virtue of Reasonableness in Hume’s Treatise,” in D. Livingston and J. King, eds., Hume, A Re-Evaluation (New York: Fordham University Press, 1976), pp. 91-106. Árdal, P., “Some Reflections upon a Standard Definition of Punishment,” in his Passions, Promises and Punishments, pp. 41-49. Aristotle, The Basic Works, ed. R. McKeon (New York: Random House, 1941). Aristotle (Pseudo-Aristotle), Problems, trans. E. S. Forster, in Aristotle, The Complete Works, ed. J. Barnes (Princeton: Princeton University Press, 1984), vol. 2. Arnauld, A., “Objections” to Descartes’ Meditations, in Haldane and Ross, eds., Philosophical Works of Descartes, vol. II, pp. 79-95. Arnauld, A. and P. Nicole, The Art of Thinking, trans. J. Dickoff and P. James (BobbsMerrill: Indianapolis, Indiana, 1964). Augustine, City of God, trans. H. Bettenson, Intro. J. O’Maera (London: Penguin, 1984). Austin, J., “Other Minds,” Proceedings of the Aristotelian Society, supplementary volume 20 (1946), pp. 148-187; reprinted in A. Flew, ed., Logic and Language, 2nd series (Oxford: Blackwell, 1959), pp. 123-58. Page references are to the latter. Ayer, A. J., The Concept of Person and Other Essays (London: Macmillan, 1963). Ayer, A. J., The Problem of Knowledge (Harmondsworth, Middlesex: Penguin, 1956) Babb, Lawrence, “The Elizabethan Conception of Love in the Elizabethan and Early Stuart Drama,” PMLA, 56 (1941), pp. 1020-35.

508

Body, Mind and Self in Hume’s Critical Realism

Bacon, Francis, Essays, Advancement of Learning, New Atlantis, and Other Pieces, selected and Edited by Richard Foster Jones (Garden City, New York: Doubleday, Doran & Co., 1937). Baier, Annette, “Doing without Moral Theory,” in her Postures of Mind, pp. 228-245. Baier, Annette, “Hume on Heaps and Bundles,” American Philosophical Quarterly, 16 (1979), pp. 285-95. Baier, Annette, “Hume’s Analysis of Pride,” Journal of Philosophy, 75 (1978), pp. 2740. Baier, Annette, Postures of Mind (Minneapolis, Minn.: University of Minnesota Press, 1985). Baier, Annette, A Progress of Sentiments (Cambridge, Mass.: Harvard University Press, 1991). Baier, Annette, “Theory and Reflective Practices,” in her Postures of Mind, pp. 207227. Barber, K., Meinong’s Hume Studies: Translation and Commentary (Ann Arbor, Michigan: University Microfilms, 1966) Barry, R., “Personhood: The Conditions of Identification and Description,” Linacre Quarterly, 45 (1978), pp. 64-81. Bayle, Pierre, The Dictionary Historical and Critical, Second Edition (London: J. J. and P. Knapton et al., 1736). Baxter, Richard, The Practical Works of the Rev. Richard Baxter (London: James Duncan, 1830), 23 vols. The Practical Works was first published in London, 1707. Page references are to the 1830 edition. Baxter, Richard, The Saint’s Everlasting Rest, in vols. 22 and 23 of The Practical Works of the Rev. Richard Baxter. Beauchamp, T. and S. Perlin, eds., Ethical Issues in Death and Dying (Englewood Cliffs, NJ: Prentice Hall, 1978). Bennett, J., Locke, Berkeley, Hume (Oxford: Oxford University Press, 1971) Bentley, Richard, Sermons Preached at Boyle’s Lecture, in vol. III of The Works of Richard Bentley, ed. Alexander Dyce (London: Francis Macpherson, 1838).

509

Bibliography

Bergmann, G., “Acts,” in his Logic and Reality, pp. 3-44. Bergmann, G., “The Contribution of John G. Watson,” Psychological Review, 63 (1956), pp. 265-276. Bergmann, G., “The Problem of Relations in Classical Psychology,” in his Metaphysics of Logical Positivism, Second Edition (Madison, Wisc., 1967), pp. 27799. Bergmann, G., “Inclusion, Exemplification, and Inherence in G. E. Moore,” in his Logic and Reality, pp. 158-70. Bergmann, G., “Intentionality,” in his Meaning and Existence, pp. 3-38. Bergmann, G., “Ineffability, Ontology and Method,” in his Logic and Reality, pp. 4563. Bergmann, G., Logic and Reality (Madison, Wisc.: University of Wisconsin Press, 1964). Bergmann, G., Meaning and Existence (Madison, Wisc.: University of Wisconsin Press, 1960). Bergmann, G., “The Ontology of Edmund Husserl,” in his Logic and Reality, pp. 193224. Bergmann, G., Philosophy of Science (Madison, Wisc.: University of Wisconsin Press, 1956) Bergmann, G., “Stenius on the Tractatus,” in his Logic and Reality, pp. 342-71. Bergmann, G., “Synthetic A Priori,” in his Logic and Reality, pp. 272-301. Berkeley, G., Principles of Human Knowledge, in vol. II of the Works, ed. A. A. Luce and T. E. Jessop (London: T. Nelson, 1948-57) Berry, Robert, “The Roman Catholic Position on Abortion,” in Rem B. Edwards, ed., New Essays on Abortions and Bioethics (Advances in Bioethics, vol. 2) (Greenwich, Comm.: Jai Press, 1997). Boas, George, Essays on Primitivism and Related Ideas in the Middle Ages (Baltimore: The Johns Hopkins Press, 1948) Brentano, Franz, Psychologie vom Empirischen Standpunkt (Leipzig, 1874; [ed. O.

510

Body, Mind and Self in Hume’s Critical Realism

Kraus] Hamburg: Felix Meiner, 1955); republished as Psychology from the Empirical Standpoint, trans. A. C. Rancurello, D. B. Terrell, and Linda McAllister, ed. Linda McAlister (London: Routledge and Kegan Paul, 1973). Bricke, John, Hume’s Philosophy of Mind (Edinburgh: Edinburgh University Press, 1980). Boethius, Commentary on Porphyry’s Isagoge, translated in R. McKeon, ed., Selections from Mediaeval Philosophers (New York: Scribner’s, 1957-8), vol. I Boethius, The Theological Tractates, trans. H. F. Steward and E. K. Rand, Loeb Classical Library (Cambridge, Mass.: Harvard University Press, 1962) Boring, E. G., A History of Experimental Psychology, second edition (New York: Appleton Century Crotfs, 1960). Boring, E. G., Sensation and Perception in the History of Experimental Psychology (New York: Appleton-Century-Crofts, 1942). Bowle, J., Hobbes and His Critics (London: Jonathan Cape, 1951). Box, M. A., The Suasive Art of David Hume (Princeton: Princeton University Press, 1990). Bramhall, John, The Catching of the Leviathan, 1658; in his Works (Oxford: J. Parker,1843-5), vol. IV. Bricke, John, Hume’s Philosophy of Mind (Edinburgh: Edinburgh University Press, 1980) Broad, C. D., Mind and Its Place in Nature (London: Routledge and Kegan Paul, 1951). Brodbeck, M., “Explanation, Prediction, and ‘Imperfect’ Knowledge,” in H. Feigl and G. Maxwell, Minnesota Studies in the Philosophy of Science (Minneapolis: University of Minnesota Press, 1962), vol. 3. pp. 231-272 Brodbeck, M., “Meaning and Action,” Philosophy of Science, 30 (1963), pp. 309-324 Brodbeck, M., “Mental and Physical: Identity vs. Sameness,” in P. K. Feyerabend and G. Maxwell, eds., Mind, Matter and Method (Minneapolis, Minn.: University of Minnesota Press, 1966), pp. 40-58.

511

Bibliography

Burton, Robert, The Anatomy of Melancholy, ed. T. Faulkner, N. Kiesling, and R. Blair, with an Introduction by J. B. Bamborough, 3 vols. (Oxford: Oxford University Press, 1989). Burton divides his massive work into numbered Partition, Section, Member, and Subsection. References will be to the numbered division, as well as volume and page of the cited edition. Burtt, E. A., The Metaphysical Foundations of Modern Science, second revised edition (London: Routledge and Kegan Paul, 1959). Butchvarov, P., Resemblance and Identity (Bloomington, Indiana: University of Indiana Press, 1966) Butler, J., The Analogy of Religion, I, v, 8, in The Works of Joseph Butler, ed. W. E. Gladstone, 2 vols. (Oxford: Oxford University Press, 1897). Butler, J., “Dissertation of Personal Identity,” in his Analogy of Religion, vol. II of his Works, ed. J. H. Bernard (London: Macmillan, 1900), pp. 279-285. Campbell, Keith, Body and Mind (Anchor: Garden City, New York, 1970). Capaldi, N., David Hume (Boston: Twayne, 1975). Capaldi, N., “The Historical and Philosophical Significance of Hume’s Theory of the Self,” in A. J. Holland, ed., Philosophy, Its History and Historiography (Dordrecht, Holland: Reidel, 1985), pp. 271-287. Carpenter, William B., Principles of Human Physiology, 5th ed., (London: Churchill 1855). Carpenter, William B., Principles of Mental Physiology (New York: Appleton, 1890). Carse, Alisa L., “The ‘Voice of Care’: Implications for Biomedical Education,” Journal of Medicine and Philosophy, 16 (1991), pp. 5-28. Chisholm, R., “On the Observability of the Self,” Philosophy and Phenomenological Research, 30 (1969), pp. 7-21. Christie, J. R. R., “Ether and the Science of Chemistry,” in G. N. Cantor, and M. J. S. Hodge, eds., Conceptions of Ether (Cambridge: Cambridge University Press, 1981). Cicero, The Nature of the Gods, trans. P. G. Walsh (Oxford: Oxford University Press, 1997). Clarke, Samuel, A Discourse concerning the Being and Attributes of God, the

512

Body, Mind and Self in Hume’s Critical Realism

Obligations of Natural Religion, and the Truth and Certainty of the Christian Revelation, Sixth Edition (London: James Knapton, 1725). Clarke, Samuel, Works (London: J. and P. Knapton, 1738). Collins, A., see Clarke, Samuel, Works. Cullen, William, The Works of William Cullen, M.D., in Two Volumes, ed. John Thomson (Edinburgh: William Blackwood, and London: T. & G. Underhill, 1827). Cox, C., “The ‘Subject’ of Nietzsche’s Perspectivism,” Journal of the History of Philosophy, 35 (1997), pp. 269-291. Cranfield, E. B., A Critical and Exigetical Commentary on The Epistle of Paul to the Romans, 2nd edition (Edinburgh: T. & T. Clark, 1983). Cudworth, Ralph, The True Intellectual System of the Universe, first edition 1678; second edition (London: J. Walthoe et al., 1743). References are to the second edition. Cummins, P., “Perceptual Relativity and Ideas in the Mind”, Philosophy and Phenomenological Research, 24 (1963), pp. 202-14 Davey, N., “Nietzsche and Hume on Self and Identity,” Journal of the British Society for Phenomenology, 18 (1987), pp. 14-29. Davey, N., “Nietzsche, the Self, and Hermeneutic Theory,” Journal of the British Society for Phenomenology, 18 (1987), pp. 272-284. Davidson, D., “Causal Relations,” in his Essays on Action and Events, pp. 149-62. Davidson, D., Essays on Action and Events (Oxford: Oxford University Press, 1982). Davidson, D., “The Individuation of Events,” his Essays on Actions and Events, pp. 163-180. Davidson, D., “Hume’s Cognitive Theory of Pride,” Journal of Philosophy, 73 (1976), pp. 744-757. Davidson, D., “Mental Events,” in L. Foster and J. Swanson, eds., Experience and Theory (Amherst: University of Massachusetts Press, 1970), pp. 79-101. Davidson, D., “The Myth of the Subjective,” in M. Krausz, ed., Relativism (Notre Dame, Indiana: University of Notre Dame Press, 1989), pp. 159-172. Davidson, John, Some Helpes for Young Schollers in Christianity, as they are in use

513

Bibliography

and taught (originally published Edinburgh, 1602); The Ordinary Chatecisme, As it is Taught in the Newe Erected Kirk of Salt-Preston Every Saboth-Day; Ane Breif Commendatioun of Uprichtnes (originally published St. Andrews, 1573). All reprinted in Charles Rogers, ed., Three Scottish Reformers (London: English Reprint Society, 1874). Dennett, D. C., Brainstorms (Boston: Bradford Books, MIT Press, 1978) Descartes, R., Meditations, in Haldane and Descartes, vol. 1..

Ross, eds., Philosophical Works of

Descartes, R., Oeuvres, ed. C. Adam and P. Tannery (Paris: L. Cerf, 1897-1910), 12 vols. Descartes, R., Oeuvres et Lettres (Paris: Gallimard, 1953). Descartes, R., The Passions of the Soul, in vol. 1 of his, Philosophical Works, ed. J. Cottingham, et al. Descartes, R., Philosophical Works, ed. J. Cottingham, et al., 2 vols. (Cambridge: Cambridge University Press, 1984). Descartes, R., Philosophical Works of Descartes, ed. E. S. Haldane and G. R. S. Ross, 2 vols. (Cambridge: Cambridge University Press, 1931). Donne, John, The Complete English Poems, ed. A. J. Smith (London: Penguin, 1986). Ducasse, C. J., Nature, Mind and Death (La Salle, Ill.: Open Court, 1951). Edwards, Rem B. ed., New Essays on Abortions and Bioethics (Advances in Bioethics, vol. 2, 1997), (Greenwich,. Conn.: JAI Press, 1997) Edwards, W., “Foreword” to S. P. Lamprecht, The Metaphysics of Naturalism, pp. v xi. Emerson, Roger, “The ‘Affair’ at Edinburgh and the ‘Project’ at Glasgow: the Politics of Hume’s Attempts to Become a Professor,” in M. A. Stewart and John Wright, eds., Hume and Hume’s Connexions (Edinburgh: Edinburgh University Press, 1994), pp. 122. Encyclopaedia Britannica, Third Edition (Edinburgh: Bell and MacFarquhar, 1797) Ficin, Marsile, Théologie Platonicienne de l’Immortalité des Ames, texte critique établi et traduit par Raymond Marcel (Paris: Société d’Éditions “Les Belles Lettres”, 1964), 3 vols.

514

Body, Mind and Self in Hume’s Critical Realism

Flage, D., ”Hume’s Relative Ideas”, Hume Studies, 7 (1981), pp. 55-73. Flage, D., “Locke’s Relative Ideas”, Theoria, 47 (1981), pp. 142-59. Flanagan, Owen, “Admirable Imperfection and Admirable Immorality,” Journal of Philosopohy, 83 (1986), pp. 41-60. Foot, Philippa, “Are Moral Considerations Overriding?” in her Virtues and Vices (Berkeley: University of California Press, 1978), pp. 181-88. Frankfurt, H., “Freedom of the Will and the Concept of a Person,” Journal of Philosophy, 68 (1971), pp. 239-53. Gallie, Ian, “Mental Facts,” Proceedings of the Aristotelian Society, 37 (1936-7), pp. 191-212 Garrett, D., “Hume’s Self-Doubts about Self Identity,” Philosophical Review, 90 (1981), pp. 337-59. Gauthier, D., “David Hume: Contractarian,” Philosophical Review, 88 (1979), pp. 338. Gay, John, A Dissertation concerning the Fundamental Principle and immediate Criterion of Virtue. Prefixed to William King, Essay on the Origin of Evil (London and Cambridge: W. Thurlbourn, 1731). Geckle, George L., John Marston’s Drama: Themes, Images, Sources (Cranbury, NJ: Associated University Presses, 1980). Giamatti, A. Bartlett, The Earthly Paradise and the Renaissance Epic (Princeton: Princeton University Press, 1966). Gibbon, Edward, The Decline and Fall of the Roman Empire, ed. J. Bury, 3 vols. ((New York: Modern Library, 1995). Gilligan, Carol, In a Different Voice: Psychological Theory and Women’s Development (Cambridge, Mass.: Harvard University Press, 1982). Gilligan, Carol, “Moral Orientation and Moral Development,” in E. Feder Kittay and D. T. Meyers (eds.), Women and Moral Theory (Totawa, N. J.: Rowan and Littlefield, 1987). Gilson, É., “Christian Personalism,” in his The Spirit of Mediaeval Philosophy, trans. Downes (New York: Scribners, 1936)

515

Bibliography

Gilson, É., and T. Langan, Modern Philosophy: Descartes to Kant (New York: Random House, 1963). Grant, George, English-Speaking Justice (Toronto: Anansi, 1985). Grant, George, The George Grant Reader, ed. W. Christian and S. Grant (Toronto: University of Toronto Press, 1998). Grant, George, Philosophy in the Mass Age (Toronto: Copp Clark, 1959). Grant, George, “The Philosophy of Francis Bacon,” in W. Christian and S. Grant, eds., The George Grant Reader, pp. 321-322. Grant, George, “Pursuit of an Illusion,” in W. Christian and S. Grant, eds., The George Grant Reader, pp. 323-334. Green, T. H., Prolegomena to Ethics, ed. A. C. Bradley, 4th edition (Oxford: Oxford University Press, 1899). Grossart, Alexander B., Representative Nonconformists (London: Hodder and Stoughton, 1879). Grossmann, R., “The Dogma of Localization,” in E. B. Allaire et al., Essays in Ontology, pp. 50-63. Grossmann, R., The Structure of Mind (Madison, WI: University of Wisconsin Press, 1959). Gustavson, D., ed., Essays in Philosophical Psychology, (Garden City, N.Y.: Doubleday Anchor Books, 1964). Hallam, Henry, Introduction to the Literature of Europe in the 15th, 16th, and 17th Centuries, Three Volumes, Reprint of the Edition of 1873 (New York: Frederick Ungar, 1970). Hamilton, Sir W., Lectures on Metaphysics and Logic, ed. H. Mansel and J. Veitch (Edinburgh: Blackwood, 1861). Harré, R., and E. H. Madden, Causal Powers (Totowa, N. J.: Rowan & Littlefield, 1975). Hartley, David, Hartley’s Theory of the Human Mind, ed. Joseph Priestley, 2nd edition (London: Jl Johnson, 1790). Hartley, David, Observations on Man (London,1749, reprinted Scholars Facsimiles &

516

Body, Mind and Self in Hume’s Critical Realism

Reprints, Delmar, New York, 1976). Hausman, A., “Adhering to Inherence,” Canadian Journal of Philosophy, 14 (1984), pp. 421-43. Hausman, A., “Solipsism and Berkeley’s Alleged Realism,” Revue Internationale de Philosophie, 23 (1968), pp. 403-412. Hausman, Alan and David Hausman, “Hume’s Use of Illicit Substances,” Hume Studies, 15 (1989), pp. 1-38. Hay, W. H., “Berkeley’s Argument from Nominalism,” Revue Internationale de Philosophie, 7 (1953), pp. 19-28. Heidegger, M., Sein und Zeit, in his Gesamteausgabe, Band 2 (Frankfurt am Main: Vittorio Klosterman, 1976) Hempel, C. G., Aspects of Scientific Explanation and Other Essays (New York: The Free Press, 1965) Hempel, C. G., “Explanation in Science and in History,” in his Aspects of Scientific Explanation. Hempel, C. G., and Oppenheim, P., “Studies in the Logic of Explanation,” in C. G. Hempel, Aspects of Scientific Explanation. Hobbes, T., Elements, Part II, Ch. 11, sec. 1, in English Works, ed. W. Molesworth (London: J. Bohn, 1839-45), 22 (1968), pp. 403-412. Hochberg, H., “Mapping, Meaning and Metaphysics,” in Contemporary Perspectives in the Philosophy of Language, ed. P. French et al., (Minneapolis, Minn.: University of Minnesota Press, 1979), pp. 326-346, Hochberg, H., “Nominalism, General Terms, and Predication,” The Monist, 71 (1978), pp. 460-475. Homer, The Odyssey, trans E. V. Rieu (Harmondsworth, Middlesex: Penguin, 1946). Huet, Pierre-Daniel, Censura philosophiae Cartesiana (Paris, 1692) Huet, Pierre-Daniel, De la situation du paradis terrestre (Paris: Jean Anisson, 1691); translated as A Treatise of the Situation of Paradise (London, 1694). References are to the Paris edition. Huet, Pierre-Daniel, Traité philosophique de la foiblesse de l'esprit humain

517

Bibliography

(Amsterdam: Henri du Sauvet, 1723). Hume, David, Abstract of a Treatise of Human Nature, Intro. J. M. Keynes, ed. J. M. Keynes and P. Sraffa (Hamden, CT: Archon Books, 1965). Hume, David, A Letter from a Gentleman to his Friend in Edinburgh, ed. E. C. Mossner and John V. Price, (Edinburgh: Edinburgh University Press, 1967). Hume, David, Essays: Moral, Political and Literary, ed. T. H. Green and T. H. Grose (London: Longmans, Green & Co., 1889). Hume, David, Letters, ed. J. Y. T. Greig, 2. vols. (Oxford: Oxford University Press, 1932). Hume, David, History of England, 8 vols. (London: T. Cadell, 1794) Hume, David., History of England, 6 vols. (Boston: Little, Brown, 1854). Hume, David, History of England, ed. W. B. Todd (Indianapolis: Liberty Fund, 1987) Hume, David, “On the Immortality of the Soul,” in David Hume, Essays: Moral, Political, and Literary, ed. Eugene F. Miller (Indianapolis IN: Liberty Classics, 1987), pp. 590-598. Hume, David, “My Own Life,” in Hume, Essays: Moral, Political and Literary, ed. Green and Grose, vol. I. Hume, David, The Natural History of Religion, in Essays Moral, Political and Literary, ed. Green and Grose, pp. 309-363. Hume, David, New Letters, ed. R. Klibansky and E. C. Mossner (Oxford: Oxford University Press, 1954). Hume, David, “Of Parties in General,” in his Essays: Moral, Political and Literary, ed. Green and Grose, vol. I, pp. Hume, David, The Philosophical Works, 4 vols., ed. T. H. Green and T. H. Grose (Oxford: Oxford University Press, 1886). Hume, David, “On Suicide,” in his Essays: Moral, Political and Literary, ed. Green and Grose, vol. 2, pp. Reprinted in T. Beauchamp and S. Perlin, Ethical Issues in Death and Dying, pp. 105-110. Page references are to the latter publication. Hume, David, “Of the Dignity or Meanness of Human Nature,” in his Essays: Moral, Political and Literary, ed., Green and Grose, vol. 1, pp. 15-56.

518

Body, Mind and Self in Hume’s Critical Realism

Hume, David, “Of the Dignity or Meanness of Human Nature,” in David Hume, Essays Moral Political and Literary, ed. Eugene F. Millar, revised edition (Indianapolis: Liberty Classics, 1987), pp. 80-86 Hume, David, “Of the Populousness of ancient Nations,” in his Essays: Moral, Political and Literary, ed., Green and Grose, vol.1, pp. Hume, David, “The Stoic,” in Essays: Moral, Political and Literary, ed. Green and Grose, vol. 1, pp. Hume, David, “Of the Standard of Taste”, in D. Hume, Of the Standard of Taste and Other Essays, ed. J. Lenz (Indianapolis: Bobbs-Merrill, 1965), pp. 3-34. Hume, David, Treatise of Human Nature, ed. L. A. Selby-Bigge, revised P. H. Nidditch (Oxford: Oxford University Press, 1968) Husserl, E., “Phenomenology,” Encyclopaedia Britannica, 14th ed.; reprinted in Realism and the Background of Phenomenology, ed. R. Chisholm (Glencoe, Ill.: Free Press, 1960), pp. 118-128. Huxley, T. H., “On the Hypothesis that Animals Are Autonmata and Its History,” Fortnightly Review, n.s. 22, (1874); reprinted in his Method and Results (London: Macmillan, 1912), pp. 199-250. Page references are to the latter. Jackson, Hughlings, Selected Writings, ed. James Taylor, 2 vols. (New York: Basic Books, 19580. Jackson, Stanley W., Melancholia and Depression (New Haven: Yale University Press, 1986). James, William, The Varieties of Religious Experience, Modern Library Edition (New York: Random House, 2002). Johnson, Samuel, Lives of the Poets, (The Six Chief Lives) ed. Matthew Arnold (London: Macmillan, 1892) Jones, J. R., “The Self in Sensory Cognition,” Mind, n.s. 63 (1949), pp. 40-61. Jones, Peter, Hume’s Sentiments (Edinburgh: Edinburgh University Press, 1982). Kenney, Anthony, The Five Ways (New York: Shocken Books, 1969) Köhler, K., The Place of Value in a World of Facts (New York: Liveright, 1938).

519

Bibliography

Lamont, W. M., Richard Baxter and the Millennium: Protestant Imperialism and the English Revolution (London: Croom Helm, 1979). Lamprecht, S. P., The Metaphysics of Naturalism (New York: Appleton-CenturyCrofts, 1967) Lange, F. A., The History of Materialism, third edition, trans E. C. Thomas, with an Introduction by Bertrand Russell (London: Routledge and Kegan Paul, 1925). Lawson, George, An Examination of the Political part of Mr Hobbs his Leviathan (London: R. White for F. Taylor, 1657) Leibniz, G. W. F., Fifth Letter to Clarke, in Leibniz, The Leibniz-Clarke Correspondence, pp. 55-97. Leibniz, G. W. F., Lettres de Leibniz à Arnauld, ed. G. Lewis (Paris: Presse Universitaire de France, 1952) Leibniz, G. W. F., The Leibniz-Clarke Correspondence, ed. and trans. H. G. Alexander (Manchester: Manchester University Press, 1965). Leibniz, G. W. F., New Essays concerning Human Understanding, trans. P. Remnant and J. Bennett (Cambridge: Cambridge University Press, 1981). Leiter, Brian, “Nietzsche and the Morality Critics,” Ethics, 107 (1997), pp. 250-36. Levi-Strauss, C. Elementary Structures of Kinship, second ed., J. H. Bell, J. R. von Sturmer and R. Needham (Boston: Beacon, 1969). Levi-Strauss, C., Structural Anthropology, trans. C. Jacobson and B. G. Schoepf (Garden City, NY: Doubleday Anchor, 1967). Lewis, C. I., and C. H. Langford, Symbolic Logic, Second Edition (New York: Dover, 1959). Lewis, C. S., Mere Christianity (London: Geoffrey Bles, 1952). Livingston, D., Hume’s Philosophy of Common Life (Chicago: University of Chicago Press, 1976). Locke, J., Correspondence with Stillingfleet, in his Works, 11th edition, vol. 4 (London: W; Otridge, 1812). Locke, John, Essay concerning Human Understanding, ed., P. H. Nidditch (Oxford: Oxford University Press, 1976).

520

Body, Mind and Self in Hume’s Critical Realism

Locke, John, Some Thoughts concerning Education, ed. John W. Yolton and Jean S. Yolton (Oxford: Oxford University Press, 1989). Locke, John, Works, 9 vols., 9th ed. (London: T. Longman etc., 1794). Louden, Robert B., “Can We Be Too Moral?” Ethics, 98 (1988), pp. 361-78. Lovejoy, A. O., “The Communism of Saint Ambrose,” Journal of the History of Ideas, 3 (1942), pp. 458-68. Lowde, James, A Discourse concerning the Nature of Man (London: T. Esrren for W. Kettiby, 1694; reprinted New York: Garland, 1979) Loyola, Ignatius, The Spiritual Exercises, trans. Elisabeth Meier Tetlow (Lanham, MD: University Press of America, 1987). Lucretius, On the Nature of the Universe, prose translation by R. E. Latham (Harmondsworth, Middlesex: Penguin, 1951). Lucy, William, An Answer to Mr. Hobbes his Leviathan with Observations, Censures, and Refutations of Divers Errours (London: S. G. and B. G. for R. Mann, 1673). Mach, E., The Analysis of Sensations (New York: Dover, 1959). Mackenzie, Henry, Account of the Life and Writings of John Hume (Edinburgh: J. Constable, 1822). Mackenzie, Henry, The Man of Feeling (London: Oxford University Press, 1967). Mackie, J. L., “Causes and Conditions,” in E. Sosa, ed., Causation and Conditionals (London: Oxford University Press, 1975). Mackie, J. L., Hume’s Moral Theory (London: Routledge and Kegan Paul, 1980). Malcolm, N., “Knowledge of Other Minds,” Journal of Philosophy, 55 (1958), pp. 960-978; reprinted in Gustavson, ed., Essays in Philosophical Psychology Page references are to the latter. Malcolm, N., “Privacy of Experience,” in Epistemology, ed. A Stroll (New York: Harper and Row, 1967). Mandelbaum, M., The Anatomy of Historical Knowledge (Baltimore: Johns Hopkins University Press, 1977)

521

Bibliography

Mandelbaum, M., “The Distinguishable and the Separable: A Note on Hume and Causation,” Journal of the History of Philosophy, 12 (1974), p. 242-47. Mandelbaum, M., Philosophy, Science and Sense Perception (Baltimore: Johns Hopkins University Press, 1964). Mann, Thomas, “Nietzsche’s Philosophy in the Light of Recent History,” in his Last Essays, trans. Richard and Clara Winston and Tania and James Stern,(London: Secker and Warburg, 1959), pp. 141-177. Marston, John, What You Will, in John Marston, The Works, ed. Arthur Henry Bullen (London: John C. Nimmo, 1887), vol. II. Mayor, Joseph B., et al., Virgil’s Messianic Eclogue: Its Meaning, Occasion, and Sources, Three Studies by Joseph B. Mayor, W. Warden Fowler, and R. S. Conway, with the Text of the Eclogue, and a Verse Translation by R. S. Conway (London: John Murray, 1907). McGinn, Colin, “Can We Solve the Mind-Body Problem?” Mind, n.s. 98 (1989), pp. 349-365. McIntyre, Jane, “Character: A Humean Account,” History of Philosophy Quarterly, 7 (1990), pp. 193-206. McIntyre, Jane, “Hume’s Underground Self,” in S. Tweyman, ed., Studies in Early Modern Philosophy, vol. III (Delmar, NY: Caravan Books, 1993), pp. 110-126. McIntyre, Jane, “Personal Identity and the Passions,” Journal of the History of Philosophy, 27 (1989), pp. 545-557. Mill, James, Analysis of the Phenomena of the Human Mind, Second Edition (London, 1869; reprinted New York, Augustus M. Kelley, 1967) Mill, J. S., “Auguste Comte and Positivism,” in Essays on Philosophy and the Classics, vol. X of his Collected Works, ed. J. Robson (Toronto: University of Toronto Press, 1969) Milll, J. S., “Bain’s Psychology,”in vol. XI of his Collected Works, ed. J. Robson (Toronto: University of Toronto Press, 1979). Mill, J. S., Examination of Sir William Hamilton’s Philosophy, in vol. IX of his Collected Works, ed., J. Robson, (Toronto: University of Toronto Press, 1959). Mill, J. S., “Notes” to James Mill, Analysis of the Phenomena of the Human Mind, Second Edition

522

Body, Mind and Self in Hume’s Critical Realism

Moody, E. A., The Logic of William of Ockham (New York: Russell and Russell, 1965). Moore, G. E., “Are the Characteristics of Particular Things Universal or Particular?” in his Philosophical Papers (London: Allen and Unwin, 1959), pp. 17-31. Moore, G. E., “The Nature and Reality of Objects of Perception,” in G. E. Moore, Philosophical Studies (London: Routledge and Kegan Paul, 1922), pp. 31-96. More, Henry, An Antidote against Atheism (Cambridge: William Morden, 1657). More, Henry, The Immortality of the Soul (Cambridge: William Morden, 1662). Mossner, E. C., The Life of David Hume, second edition (Oxford: Oxford University Press, 1980). Mossner, E. C., and John V. Price, “Introduction” to David Hume, A Letter from a Gentleman to his Friend in Edinburgh (Edinburgh: Edinburgh University Press, 1967). Nagel, T., “What Is It Like to Be a Bat?” In his Mortal Questions (Cambridge: Cambridge University Press, 1979), pp. 165-180. Nietzsche, F., Beyond Good and Evil, trans. J. Hollingdale, Intro. M. Tanner (Harmondsworth, Middlesex: Penguin, 1970). Nietzsche, F., The Genealogy of Morals, in W. Kaufmann, ed., F. Nietzsche, The Genealogy of Morals and Ecce Homo (New York: Vintage Books, 1967). Orme, William, “Life and Times [of Richard Baxter]” and “Life and Works,” in vol. I of The Practical Works of the Rev. Richard Baxter. Overton, Rcihard, Foundations of Freedom [also attributed to John Lilburne] (London: published for the satisfaction of all honest interests, 1648). Overton, Richard, Mans Mortallitie (Amsterdam [ = London]: John Canne, 1643); second edition Man Wholly Mortal (London: s.n., 1655). Overton, Richard, Vox Plebis: or, The Voice of the Oppressed Commons of England against Their Oppressors (London: s.n., 1653) Ovid, Metamorphoses, trans. M. Innes (Harmondsworth, Middlesex: Penguin 1955). Pappas, G., “Ideas, Mind and Berkeley,” American Philosophical Quarterly, 17 (1980), pp. 181-94.

523

Bibliography

Parfit, D., “Personal Identity,” Philosophical Review, 80 (1971), pp. 3-27. Pascal, Blaise, Pensées, trans. A. J. Krailsheimer (Harmondsworth, Middlesex: Penguin, 1966) Passmore, J., Hume’s Intentions, Third Edition (London, 1980) Passmore, J., Philosophical Reasoning (New York: Scribner’s, 1961).. Peden, Alexander, The Lord’s Trumpet Sounding Alarm Against Scotland. First published 1682. References are to the Glasgow (?) edition of 1720. Penelhum, t., “Hume on Personal Identity,” Philosophical Review, 64 (1955), pp. 57189. Penelhum, T., “Self-Identity and Self-Regard,” in Amélie Rorty, ed., The Identities of Persons, pp. 253-280. Plantinga, A., God and Other Minds (Ithaca: Cornell University Press, 1967). Plantinga, A., “Induction and Other Minds,” Review of Metaphysics, 19 (1966), pp. 441-461. Plotinus, The Enneads, trans. Stephen MacKenna, Fourth Edition, revised B. S. Page (London: Faber and Faber, 1969). Price, H. H., Hume’s Theory of the External World (Oxford: Oxford University Press, 1940). Pratt, J. B., “Critical Realism and the Possibility of Knowledge”, in Durant Drake, A. O. Lovejoy, J. B. Pratt, A. K. Rogers, George Santayana, R. W. Sellars and C. A. Strong, Essays in Critical Realism (New York: Macmillan, 1920), pp. 85-116. Raleigh, Walter, History of the World, 11th edition (London: J. J. and P. Knapton etc., 1733). Rees, Goronwy, A Bundle of Sensations, Sketches of Autobiography (New York: Macmillan, 1961). Reid, T. Essays on the Intellectual Powers of Man, in his Philosophical Works. Reid, T, Inquiry into the Human Mind, in his Philosophical Works. Reid, T., Philosophical Works, ed. Sir W. Hamilton, 8th ed., (Edinburgh: Maclachlan

524

Body, Mind and Self in Hume’s Critical Realism

and Stewart, 1880). Rollock, Robert, Certaine Sermons upon severall texts of Scripture (Edinburgh: Andro Hart, 1616). Rollock, Robert, Five and Twentie Lectures (Edinburgh: Andro Hart, 1619). Rollock, Robert, Lectures upon the Epistle of Paul to the Colossians (London: Felix Kyngston, 1603). Rollock, Robert, Lectures upon the First and Second Epistles of Paul to the Thessalonians (Edinburgh: Robert Chateris, 1606). Rorty, A., ed., The Identities of Persons (Berkeley: University of California Press, 1976) Rorty, A. “A Literary Postscript: Characters, Persons, Selves, Individuals,” in A. Rorty, ed., The Identities of Persons, pp. 301-224, Rorty, A., “‘Pride Produces the Idea of Self’: Hume on Moral Agency,” Australasian Journal of Philosophy, 68 (1990), pp. 255-69. Russell, B., Inquiry into Meaning and Truth (London: Allen and Unwin, 1940) Russell, B., Logic and Knowledge, ed., R. C. Marsh (London: Allen and Unwin, 1956) Russell, B., “On the Nature of Acquaintance,”in his Logic and Knowledge, pp. 125174. Russell, B., “On the Notion of Cause,” in his Mysticism and Logic (New York: W. W. Norton, 1929), pp. 180-208. Russell, B., “On the Relations of Universals and Particulars,” in his Logic and Knowledge., pp. 103-124. Russell, B., Principles of Mathematics, Second Edition (London: Allen and Unwin, 1937) Rutherford, Samuel, Joshua Redivivus (Rotterdam, 1664). Ryle, G., The Concept of Mind (London: Hutchinson, 1949). Sartre, Jean-Paul, Being and Nothingness, trans. H. Barnes (New York: Philosophical Library, 1956).

525

Bibliography

Scott, Sir Walter, Lives of the Novelists (Henry Frowde, Oxford University Press: London, 1906). Scott, Sir Walter, Old Mortality, ed. J. Robinson and P. Davidson (Oxford: Oxford University Press, 1993). Sellars, R. W., Critical Realism: A Study of the Nature and Conditions of Knowledge (Chicago: Rand, 1916). Sellars, W., “Notes on Intentionality,” in his Philosophical Perspectives, pp. 308-320. Sellars, W., Philosophical Perspectives, (Springfield, Ill.: Charles C. Thomas, 1967). Sellars, W., “Some Reflections on Language Games,” in his Science, Perception and Reality, pp. 321-358. Sellars, W., Science, Perception and Reality (London: Routledge and Kegan Paul, 1963). Sergeant, John, Solid Philosophy Asserted against the Fancies of the Ideists (London: R. Clavel, 1697). Séverin de Cardaillac, J. J., Études élémentaires de philosophie (Paris: Firmin Didot Frères, 1830) Shalom, A., The Body-Mind Conceptual Framework and the Problem of Personal Identity (Atlantic Highlands, N J: Humanities Press International, 1985) Shoemaker, S., Self-Knowledge and Self-Identity (Ithaca, NY: Cornell University Press, 1963). Siebert, Donald, “‘Ardor of Youth’: The Manner of Hume’s Treatise,” in Robert Ginsberg, ed., The Philosopher as Writer: The Eighteenth Century (Selinsgrove: Susquehanna University Press, 1987), pp. 177-198. Sievert, D., “Historical Cross-Section Laws and Austin’s Scepticism in ‘Other Minds’,” Philosophy of Science, 30 (1970), pp. 146-52. Shaftesbury, Third Earl of, (Anthony Ashley Cooper), Characteristics of Men, Manners, Opinions and Times, Intro. S. Grean, 2 vols. (New York: Bobbs-Merrill, 1964). Smith, John, Select Discourses (London: J. Flesher, 1660); fourth edition, ed. H. G. Williams (Cambridge: Cambridge University Press, 1859). References are to the latter.

526

Body, Mind and Self in Hume’s Critical Realism

Sorabji, R., Necessity, Cause and Blame: Perspectives on Aristotle’s Theory (London: Duckworth, 1980). Sperry, R. W., “Hemisphere Disconnection and Unity in Conscious Awareness,” Journal of Medicine and Philosophy, 2 (1986), pp. 107-25. Spencer, Herbert, Principles of Psychology, 2 vols. (London: Longman & Co., 1855). Stack, G., Lange and Nietzsche (Berlin and New York: Walter de Gruyter, 1983). Strawson, P. F., The Bounds of Sense (London: Methuen, 1966). Strawson, P. F., “Persons,” in Minnesota Studies in the Philosophy of Science, vol. 2. pp. 330-353; reprinted in D. Gustavson, ed., in Essays in Philosophical Psychology. Page references are to the latter. Stocker, Michael, “The Schizophrenia of Modern Ethical Theories,” Journal of Philosophy, 73 (1976), pp. 453-464.. Stroud, Barry, Hume (London: Routledge and Kegan Paul, 1977). Suarez, F., On the Various Kinds of Distinctions, trans. C. Vollert (Milwaukee, Wisc.: Marquette University Press, 1947) Sumner, L. W., “Moderate View of Abortion,” in Rem B. Edwards, ed., Advances in Bioethics: New Essays on Abortion and Bioethics, pp. 203-226. Sydenham, Thomas, His Life and Original Writings, ed. K. Dewhurst (Berkeley: University of California Press, 1966). Thomas Aquinas, Summa Theologica, trans. by the Fathers of the Dominican Province ((London: Benziger Bros., 1947). Thomson, James, The Seasons and the Castle of Indolence, ed., J. Sambrook (Oxford: Oxford University Press, 1972). Tooke, John Horne, The Diversions of Purley (London, 1798; reprinted Menston, England: Scolar Press, 1968). Turnbull, R. G., “Aristotle’s Debt to the ‘Natural Philosophy’ of the Phaedo,” Philosophical Quarterly, 8 (1963), pp. 131-43. Tuveson, E. L., The Imagination as a Means of Grace (Berkeley: University of California Press, 1960).

527

Bibliography

Tweyman, S., “Hume on Separating the Inseparable,” in W. B. Todd, ed., Hume and the Enlightenment (Edinburgh: Edinburgh University press, 1974), p. 30-42. Valla, Lorenzo, On Pleasure: De voluptate, trans. A. Kent Hieatt and Maristella Lorch, Intro. by Maristella Lorch (New York: Abaris Books, 1977). Van Iten, R. J., “Berkeley’s Alleged Solipsism,” Revue Internationale de Philosophie, 16 (1962), pp. 447-452. Vendler, Zeno, “A Note to the Paralogisms,” in G. Ryle, ed., Contemporary Aspects of Philosophy (Stocksfield, 1977) Vesey, G., Personal Identity: A Philosophical Analysis (Ithaca, NY: Cornell University Press, 1974). Vicari, Patricia, The View from Minerva’s Tower (Toronto: University of Toronto Press, 1989). Vickers, Brian, “Introduction” to Henry Mackenzie, The Man of Feeling (London: Oxford University Press, 1967). Virgil, Messianic Eclogue: see Mayor, Joseph B. Vlastos, G., “Reasons and Causes in the Phaedo,” Philosophical Review, 78 (1969), pp. 291-325. Voltaire (François-Marie Arouet), Letters concerning the English Nation, ed. Nicolas Cronk (Oxford: Oxford University Press, 1994). Walker, Patrick, Six Saints of the Covenant: Peden, Semple, Welwood, Cameron, Cargill, Smith, ed. with an Introduction by D. Hay Fleming, and a Forward by S. R. Crocket, in two volumes (London: Hodder and Stoughton, 1901). Webber, Joan, The Eloquent “I” (Madison: University of Wisconsin Press, 1968). Weinberg, J., Abstraction, Relation, and Induction (Madison, Wisc.: University of Wisconsin Press, 1965). Weinberg, J., Ockham, Descartes, Hume (Madison, Wisc.: University of Wisconsin Press, 1977) Weinberg, J., “The Nominalism of Berkeley and Hume,” in his Abstraction, Relation, and Induction, pp. 3-60. Wiggins, David, Identity and Spatiotemporal Continuity (Oxford: Blackwell, 1967).

528

Body, Mind and Self in Hume’s Critical Realism

Wiggins, David, Sameness and Substance (Cambridge, Mass.: Harvard University Press, 1980). William of Ockham, Summa Logicae (Notre Dame, Indiana: University of Notre Dame Press, 1974). Williams, B., “A Critique of Utilitarianism,” in J. C. C. Smart and B. Williams, Utilitarianism: For and Against (London: Cambridge University Press, 1973), pp. 77150. Williams, B., “Are Persons Bodies?” in his Problems of the Self (Cambridge: Cambridge University Press, 1973), pp. 69-81. Williams, B., Ethics and the Limits of Philosophy (Cambridge, Mass.: Harvard University Press, 1985). Williams, B., “Persons, Character and Morality,” in A. Rorty, ed., The Identities of Persons, pp. 197-216. Williams, B., “The Self and the Future,” Philosophical Review, 79 (1970), pp. 161180. Williams, G. H., Wilderness and Paradise in Christian Thought (New York: Harper, 1962) Wilson, F., “Acquaintance, Ontology and Knowledge,” The New Scholasticism, 54 (1970), pp. 1-48; revised version in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 1-108. Wilson, F., Acquaintance, Ontology and Knowledge (Frankfurt: Ontos Verlag, 2007). Wilson, F., “Addis on Defining Disposition Concepts,” Inquiry, 28 (1985), pp. 247-72. Wilson, F., “Burgersdijck, Coleridge, Bradley, Russell, Bergmann, Hochberg: Six Philosophers on the Ontology of Relations,” in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 275-328. Wilson, F., “Definition and Discovery,” British Journal for the Philosophy of Science, 18 (1967), pp. 287-303 and 19 (1967), pp. 43-56. Wilson, F., “Dispositions Defined: Harré and Madden on Analyzing Disposition Concepts,” Philosophy of Science, 52 (1985), pp. 591-607. Wilson, F., “Dispositions: Defined or Reduced?” Australasian Journal of Philosophy,

529

Bibliography

47 (1969), pp. 184-204. Wilson, F., “Effability, Ontology, and Method,” Philosophy Research Archives, 9 (1983), pp. 419-70; also in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 537-592. Wilson, F., Empiricism and Darwin’s Science (Dordrecht: Kluwer, 1991). Wilson, F., Explanation, Causation and Deduction (Dordrecht, Holland: Reidel, 1985). Wilson, F., “Herbert Spencer,” Dictionary of Literary Biography, vol 262, British Philosophers, 1800-2000, ed. P. Dematteis, P. Fosl, and L. McHenry (“A Brucolli Clark Layman Book,” Farmington, MI: Thomson Gale, 2002), pp. 252-265. Wilson, F., “Hume and Derrida on Language and Meaning,” Hume Studies, 12 (1986), pp. 99-121; also in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 109-128. Wilson, F., “Hume and Ducasse on Causal Inference from a Single Experiment,” Philosophical Studies, 35 (1979), pp. 663-68. Wilson, F., “Hume’s Cognitive Stoicism,” Hume Studies, 1985 Supplement, pp. 52168. Wilson, F., “Hume’s Defence of Causal Inference,” Dialogue, 22 (1983), pp. 661-94. Wilson, F., Hume’s Defence of Causal Inference (Toronto: University of Toronto Press, 1998) Wilson, F., “Hume’s Defence of Science,” Dialogue, 25 (1986), pp. 611-628. Wilson, F., “Hume’s Fictional Continuant,” History of Philosophy Quarterly, 6 (1989), pp. 171-88 Wilson, F., “Hume’s Theory of Mental Activity,” in D. F. Norton, N. Capaldi, and W. Robison, eds., McGill Hume Studies (San Diego: Austin Hill Press, 1979), pp. 101-120 Wilson, F., “Is Hume a Sceptic with regard to the Senses?” Journal of the History of Philosophy, 27 (1989), pp. 49-73. Wilson, F., “Is Operationism Unjust to Temperature?” Synthese, 18 (1968), pp. 394422. Wilson, F., “Is There a Prussian Hume?” Hume Studies, 8 (1982), pp. 1-18.

530

Body, Mind and Self in Hume’s Critical Realism

Wilson, F., “Kuhn and Goodman: Revolutionary vs. Conservative Science,” Philosophical Studies, 44 (1983), pp. 135-50. Wilson, F., Laws and Other Worlds (Dordrecht, Holland: Reidel, 1986). Wilson, F., “The Lockean Revolution in the Theory of Science,” in S. Tweyman and G. Moyal, eds., Early Modern Philosophy: Epistemology, Metaphysics and Politics (New York: Caravan Press, 1986), pp. 65-97 Wilson, F., The Logic and Methodology of Science and Pseudoscience (Toronto: Canadian Scholars Press, 2000) Wilson, F., The Logic and Methodology of Science in Early Modern Thought (Toronto: University of Toronto Press, 1999) Wilson, F., “Marras on Sellars on Thought and Language,” Philosophical Studies, 28 (l975), pp. 91-102; also in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 519-536. Wilson, F., “Mill and Comte on the Method of Introspection,” Journal for the History of Behavioral Science, 27 (1991), pp. 107-129. Wilson, F., “Mill on the Operation of Discovering and Proving General Propositions,” Mill News Letter, 17 (1982), pp. 1-14 Wilson, F., “Mill’s Proof that Happiness Is the Criterion of Morality,” Journal of Business Ethics, 1 (1982), pp. 59-72. Wilson, F., “Moore’s Refutation of Idealism,” in P. Coates and D. Hutto, eds., Current Issues in Idealism (Bristol: Thoemmes Press, 1996), pp. 23-58; also in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 225-274. Wilson, F., “A Note on Hempel on the Logic of Reduction,” International Logic Review, 13 (1982), pp. 17-29. Wilson, F., “A Note on Operationism,” Critica, 2 (1968), pp. 79-87. Wilson, F., “The Notion of Logical Necessity in the Later Philosophy of R. Carnap,” in A. Hausman and F. Wilson, Carnap and Goodman: Two Formalists, Iowa Studies in Philosophy, Vol. III (Martinus Nijhoff: The Hague, 1967), pp. 97-225. Wilson, F., “On the Hausmans’ ‘New Approach to Berkeley’s Ideal Reality’,” in R. Muehlmann, ed., Berkeley’s Metaphysics: Structural, Interpretative, and Critical Essays (University Park, PA.: Pennsylvania State University Press, 1995), pp. 67-87; also in F. Wilson, Acquaintance, Ontology and Knowledge, pp. 169-194.

531

Bibliography

Wilson, F., “Placing Bergmann,” in L. Addis, Greg Jesson, and Erwin Tegtmeier, eds., Ontology and Analysis: Essays and Recollections about Gustav Bergmann (Frankfurt: Onto Verlag, 2007). Wilson, F., Psychological Analysis and the Philosophy of John Stuart Mill (Toronto: University of Toronto Press, 1990). Wilson, F., “Resemblance, Universals and Sorites: Comments on March on Sorting Out Sorites,” Canadian Journal of Philosophy, 17 (1987), p. 175-84. Wilson, F., Review of E. Prior, Dispositions, Canadian Philosophical Reviews, 7 (1987), pp. 14-16. Wilson, F., Socrates, Lucretius, Camus – Two Philosophical Traditions on Death (Lewiston, NY: Edwin Mellen Press, 2002) Wilson, F., “Some Controversies about Method in Nineteenth- Century Psychology,” Studies in History and Philosophy of Biological and Biomedical Science, 30C (1999), pp. 91-142. Wilson, F., “Was Hume a Sceptic with regard to the Senses?” Philosophy Research Archives, 14 (1989), pp. 247-82. Wilson, F., “Weinberg’s Refutation of Nominalism,” Dialogue, 8 (1969), pp. 460-74. Wilson, F., “Wordsworth and the Culture of Science,” The Centennial Review, 33 (1989), pp. 322-392. Wittgenstein, L., The Blue and Brown Books (Oxford: Basil Blackwell, 1958) Wittgenstein, L., Philosophical Investigations, 2nd ed. (Oxford: Oxford University Press, 1958). Wollheim, R., “Privacy,” Proceedings of the Aristotelian Society, n.s. 61 (1950-1951), pp. 83-104. Wright, John, “Metaphysics and Physiology: Mind, Body, and the Animal Economy in Eighteenth-Century Scotland,” in M. A. Stewart, ed., Studies in the Philosophy of the Scottish Enlightenment (Oxford: Clarendon Press, 1990), pp. 251-301. Wright, John, The Sceptical Realism of David Hume, (Minneapolis, Minn.: University of Minnesota Press, 1983) Wright, Thomas, The Passions of the Minde in Generall (London, 1630 [first

532

Body, Mind and Self in Hume’s Critical Realism

published in 1601]). (1604 edition, Reprinted, Urbana, IL: University of Illinois Press, 1971), Yolton, J., “Hume’s Ideas,” Hume Studies, 6 (1980), pp. 1-25.

533

Index of Names Adam, 35, 55, 59 Addis, L., 102, 183, 328ff, 334, 342, 350, 366, 367, 369, 370, 372f, 374 Addison, J., 88, 102 Ainslie, D., 506 Alcinous, 52 Al Gazzali, 300, 443 Allaire, E. B., 363, 364 Allison, H. E., 361 Ambrose, St., 54, 60, 62 Anaxagoras, 36, 96 Anderson, F. H., 246 Anscombe, G. E. M., 186, 361, 388ff, 500 Arbuthnot, J., 90, 102 Árdal, Páll, 43, 99, 249, 440ff, 447f, 503ff, 505 Aristotle, 9f, 19ff. 26ff, 32ff, 34ff, 73, 103f, 113ff, 128, 179, 184, 191, 194ff, 205, 213, 252, 256, 375, 405f, 412f, 444, 503 Argyll, Duke of, 467 Arnauld, A., 122, 162, 163, 188 Augustine, 53f, 79, 101 Austin, J., 369, 372 Ayer, A. J., 374, 385f, 499 Bacon, F., 64, 100 Baier, A., 43, 99, 421, 422f, 500, 501, 504, 506 Bain, A., 95, 320, 368 Basil, 59, 60 Bayle, P., 42f, 99 Baxter, R., 65f, 100 Beauchamp, T., 101 Bennett, J., 269ff, 363 Bentham, J., 187 Bentley, R., 191, 194, 290f, 305, 365 Bergmann, G., 97, 111, 126, 140, 173, 182, 184, 185, 188, 189, 249, 328, 334, 362, 364, 369, 370, 372, 501f Berkeley, G., 122, 183, 223, 277, 312

Berry, R., 101 Boas, G., 99 Boethius, 7, 16, 36, 171, 188, 253 Boring, E. G., 305, 362, 363, 366, 367 Bourgeois, W., 16 Boyle, R., 241ff Bradley, F. H., 95, 187 Bramhall, J., 35, 98 Brentano, F., 363, 370 Bricke, J., 367 Broad, C. D., 366 Brodbeck, M., 366, 369, 502 Burgersdijck, R., 187 Burke, E., 436 Burton, Robert, 496ff, 506 Burtt, E. A., 246 Butchvarov, P., 187 Bute, Earl of, 436 Butler, J., 87f, 104, 182, 255, 361, 363, 378, 403, 432f, 444f, 449f, 504 Butler, R. J., 267 Calvin, J., 2, 436 Cambridge Platonists, 36, 40f, 253, 406, 435 Campbell, K., 323, 369 Capaldi, N., 379f, 403, 498f, 500 Cardaillac, J. Séverin de, 380f, 499 Carnap, R., 183 Carpenter, W. B., 320f Carse, A. L., 443ff, 503 Cebes, 78 Chambers, E., 249 Charles, I, 66 Charles II, 66 Chisholm, R., 280, 364, 382f, 499 Christie, J. R. R., 189 Cicero, 365 Clarke, S., 44, 51, 56f, 63, 88, 92f, 99, 102, 191, 194, 246, 289, 305, 365, 378, 403, 406, 412ff, 417, 425f, 431f Clerselier, C., 188

Coleridge, S., 187 Collins, A., 88, 92f, 102 Comte, A., 16, 380, 499 Constantine the Great, 53 Cox, C., 506 Critical Realists, 216ff, 220ff, 239ff, 272 Crousaz, R., 235, 258 Cudworth, R., 40f, 98 Cullen, W., 179f, 323 Cummins, P., 361 Darwin, C., 101, 368, 502 Davey, N., 505 Davidson, D., 186, 287ff, 295, 303f, 364, 365, 506 Dennett, D., 41, 98 Descartes, R., 7f, 10ff, 14, 21, 25, 39ff, 96, 118, 121ff, 162f, 176, 183, 188, 206, 243, 249, 253, 272, 277, 297f, 308, 320, 324f, 361, 375, 381, 389f, 396, 412, 414, 422f, 434, 461, 468, 469f, 492, 495ff Ducasse, C. J., 16, 185 Durandus, 51 Edwards, T., 108, 182, 185 Elijah, 65 Epicurus, 36, 40, 79, 86ff Eve, 35, 59 Ficino, M., 39f, 98 Feigl, H., 306f, 366 Flage, D., 249, 365 Flanagan, Owen, 504 Fletcher, A., of Saltoun, 503 Flew, A., 372 Foot, Phillipa, 504 Foucher, S., 176f Frankfurt, H., 5 Galllie, I., 386f, 500 Garrett, D., 393f, 500 Garrick, R., 436 Gauthier, D., 99 Gay, J., 90, 102, 316, 318, 367 Giamatti, A. B., 99 Gibieuf, P., 124, 183 Gilligan, C., 443ff, 503f

536

Gilson, É., 88, 102, 361f God, 52f, 59, 61, 63, 65, 78, 123, 128, 176ff, 191ff, 194, 200, 305, 424, 439, 441, 451, 467ff, 471, 493ff his triune nature, 88 alleged paternity of Jesus, 441 Goodman, N., 248 Grant, G., 13, 69ff, 85, 94, 101 Gregory of Nazianus, 50 Green, T. H., 94, 259, 358ff, 361, 374 Grossart, A. B., 100 Grossmann, R., 185, 370 Guericke, Otto, 241 Gustavson, D., 372 Hallam, H., 189 Hamilton, Sir W., 362, 367, 372, 373, 380, 499 Harré, R., 107f, 144ff, 182, 186 Hartley, D., 305ff, 316ff, 366, 367 Hausman, A., 363, 364 Hausmans, A. & D., 259, 274ff, 279ff, 362, 361, 363, 500 Hegel, G. W. F., 498 Heidegger, M., 5, 16 Hempel, C. G., 26, 97, 368 Hippocrates, 195 Hobbes, T., 11, 34ff, 39ff, 42ff, 51f, 57ff, 68ff, 72f, 84f, 184, 459, 573, 487, 488 Hochberg, H., 187, 364 Homer, 52f Hooker, R., 32 Huet, P.-D., 48ff, 51ff Hume, David, passim Hume, John, 436ff Husserl, E., 189, 370 Huxley, T. G., 321ff, 366, 368 Isaiah, 65 Jackson, Hughlings, 368 James, W., 365 Jesus, 441 alleged filiation to God, 441 alleged capacity to Save, 457 John the Baptist, 65 Johnson, Samuel, 54, 99

Jones, J. R., 386f, 500 Jones, Peter, 421ff, 501 Joseph, H. W. B., 95f Jupiter, 55 Kames, Henry Home, Lord, 270, 363 Kant, I., 13, 32, 100, 189, 361 Kenney, A., 186 Kerby-Miller, C., 102 Kim, Jaegwon, 309f, 367 Köhler, W., 339f, 371 Kuhn, T., 248 Lactantius, 54, 63f, 66 Lamont, W. M., 100 Lamprecht, S. P., 139ff, 182, 185f Lamy, B., 422 Langan, T., 259, 361f Lange, F., 504 Langford, C. H., 185 Lawson, G., 35, 98 Leibniz, G. F. W., 165, 171, 185, 245 Leiter, B., 505 Levi-Strauss, C., 428ff, 501f Lewis, C. I., 185 Livingston, D., 246f, 248, 501 Locke, J., 24, 26, 75f, 85ff, 90ff, 92ff, 94, 103ff, 108ff, 121ff, 126ff, 133ff, 195ff, 205, 223, 235f, 241, 251ff, 277, 291ff, 300f, 305f, 307f, 316ff, 354, 357,361, 365, 366, 371, 378, 393f, 422f, 441, 452, 463 Louden, R., 500 Lovejoy, A. O., 99 Lowde, J., 35, 98 Lucretius, 36ff, 40f, 86ff, 94, 98, 101 Lucy, W., 35, 98 MacGregor, Robert, 467, 472, 487 Mach, E., 339, 370 Mackenzie, H., 437ff, 503 Mackie, J. L., 333, 364, 369, 501 Madden, E. H., 107f, 144ff, 182, 186 Malcolm, N., 372f, 374 Malebranche, N., 11, 178 Mandelbaum, M., 140ff, 153ff, 161ff, 167f, 174, 185f, 249 Mann, Thomas, 504

Marras, A., 499 Marsais, C.C., Sieur du, 422 Megaric school, 20, 203f Metrie, H. de la, 320 McGinn, C., 283ff, 289ff, 299ff, 303f, 312, 364, 365 McIntyre, Jane, 498, 505 Mill, J. S., 1, 13, 16, 24, 187, 209, 248, 259ff, 266, 308f, 314ff, 316, 342f, 357ff, 362, 353, 366, 367, 371, 374, 380, 387, 392f, 457, 499 Molière, 27 Molyneaux, H., 255, 355f, 378, 463 Montaigne, M. de., 39 Montrose, Earl of, 467 Moore, G. E., 184, 365, 371, 372, 373, 374, 413 More, H., 191ff, 194, 199, 202, 246, 435f Moses, 49ff, 54ff Nagel, T., 299f, 365 New Realists, 217, 220, Newton, I., 90, 191, 194, 242, 246, 307f, 318, 367, 406, 441 Nicole, P., 122, 162, 188 Nidditch, P, H., 16 Neitzsche, F., 452ff, 504f Novimagus, 51f Ockham, William, of, 171f, 184, 188f Oppenheim, P., 97 Origen, 51 Otway, T., 437 Overton, R., 40, 98 Ovid, 52f, 59, 99 Parfit, D., 500 Pascal, B., 468ff, 471f, 492, 493ff, 505 Paul, St., 44, 51 Passmore, J., 227f, 248, 269ff, 363, 372 Peden, Alexander, 66f Penelhum, T., 374 Perlin, S., 101 Plantinga, A., 315ff, 334, 342, 367, 373 Plato, 11, 17ff, 30, 33, 34ff, 36, 73, 94, 96, 103f, 181, 254, 256, 375, 459, 461 Plotinus, 11, 21f, 36f, 94, 97, 98, 191,

537

251, 253f, 256, 353, 357, 381, 383, 387, 463, 498 Pope, A., 90 Popkin, R., 231 Postellus, 51 Pratt, C. W., 248, 249 Price, H. H., 218f, 231, 246, 247ff Priestly, J., 320 Putnam, H., 367 Pyrrho, 48, 231, 242, 247, 492 Raleigh, W., 48ff, 51ff, 63, 68 Rees, G., 94ff, 102, 175 Reid, T., 258, 259, 270, 361, 363 Rob Roy: see MacGregor, Robert Robison, J., 180f Rorty, A., 7, 16, 505f Russell, B., 72f, 126, 140, 165, 173, 185, 187, 189, 300, 359, 499, 505 Rutherford, Samuel, 66 Ryle, G., 351, 370, 371, 372f, 374, 499 Sade, Marquis de, 460 Sartre, J.-P., 339, 370 Saturn, 55 Scriblerus, Martinus, 90, 102 Scott, Sr W., 101, 438, 503 Seivert, D., 369, 372 Selby-Bigge, L. A., 16, 185 Sellars, R. W., 221, 247ff Sellars, W., 368f, 499, 501 Sergent, J., 361 Shaftesbury, Third Earl of, 87f, 102, 406, 412ff, 431f Shalom, A., 23ff, 94, 97, 381, 383f, 499 Shoemaker, S., 500 Simon Stylites, St., 457 Smith, Adam, 502 Smith, J., 40f, 98 Socrates, 17ff, 76ff, 252 Sophists, 30 Sorabji, R., 97 Spencer, H., 368 Sperry, R. W., 500 Spinoza, B., 183 Stack, G., 505 Stevenson, D., 100

538

Stella, 227 Stenius, E., 189 Stillingfleet, Bishop, 198 Stocker, M., 500 Stout, G. F., 184 Strabus, 51 Strawson, P. F., 335, 336, 345, 372, 392 Stroud, B., 258, 361, 393f, 500 Suarrez, F., 128, 162f, 187f Sumner, L.W., 101 Swift, J., 90, 102, 227 Sydenham, T., 195f, 241, 246 Tertulian, 51, 54, 59 Theophilous, 60 Thomas Aquinas, St., 76f, 101, 288f, 361f, 365 Thomson, J., 455, 505 Tooke, J. Horne, 266, 363 Tremblay, F. de, 422 Turnbull, R. G., 17, 97 Tuveson, J., 93, 102 Tweyman, S., 187 van Iten, R., 364 Valla, L., 38f, 98 Vendler, Zeno, 5, 16 Vesey, G., 376ff, 498 Vicari, P., 506 Vickers, B., 503 Virgil, 52f, 99 Vlastos, G., 17, 96 Voltaire, 471, 505 Walker, P., 101 Wallis, W., 241 Watson, J. B., 372f Watts, I., 236 Weil, André, 502 Weinberg, J., 183, 184, 185, 186, 188 Wiggins, D., 182, 500 Wilkins, J., 241 Williams, B., 16, 500, 504 Williams, G. H., 100 Wilson, F., passim Wittgenstein, L., 335ff, 372, 374, 382, 388f, 391, 499, 500

Wollaston, W., 436, 502 Wollheim, R., 367 Woodsworth, W., 363 Wren, C., 341 Wright, J., 189, 249 Yolton, J., 258, 361 Zeno of Citium, 288 Zeno the Eleatic, 42

539

PhilosophischeAnalyse PhilosophicalAnalysis 1 Herbert Hochberg Russell, Moore and Wittgenstein The Revival of Realism

8 Rafael Hüntelmann Existenz und Modalität Eine Studie zur Analytischen Modalontologie

2

9 Andreas Bächli / Klaus Petrus Monism

ISBN 3-937202-00-5 334 pp., Hardcover € 94,00

Heinrich Ganthaler Das Recht auf Leben in der Medizin Eine moralphilosophische Untersuchung

ISBN 3-937202-01-3 167 pp., Hardcover € 58,00

3 Ludger Jansen Tun und Können Ein systematischer Kommentar zu Aristoteles’ Theorie der Vermögen im neunten Buch der „Metaphysik“ ISBN 3-937202-02-1 302 pp., Hardcover € 70,00

4 Manuel Bremer Der Sinn des Lebens Ein Beitrag zur Analytischen Religionsphilosophie ISBN 3-937202-03-X 134 pp., Hardcover € 58,00

5 Georg Peter Analytische Ästhetik Eine Untersuchung zu Nelson Goodman und zur literarischen Parodie ISBN 3-937202-04-8, 332 pp. Hardcover € 94,00

6 Wolfram Hinzen / Hans Rott Belief and Meaning Essays at the Interface ISBN 3-937202-05-6 250 pp., Hardcover € 58,00

7 Hans Günther Ruß Empirisches Wissen und Moralkonstruktion Eine Untersuchung zur Möglichkeit von Brückenprinzipien in der Natur- und Bioethik ISBN 3-937202-06-4 208 pp., Hardcover € 58,00

ISBN 3-937202-07-2 189 pp., Hardcover € 58,00

ISBN 3-937202-19-6 340 pp., Hardcover € 70,00

10 Maria Elisabeth Reicher Referenz, Quantifikation und ontologische Festlegung ISBN 3-937202-39-0 ca. 300 pp., Hardcover € 89,00

11 Herbert Hochberg / Kevin Mulligan Relations and Predicates ISBN 3-937202-51-X 250 pp., Hardcover € 74,00

12 L. Nathan Oaklander C. D. Broad's Ontology of Mind ISBN 3-937202-97-8 105 pp., Hardcover € 39,00

13 Uwe Meixner The Theory of Ontic Modalities ISBN 3-938793-11-2 374 pages, Hardcover,€ 79,00

14 Donald W. Mertz Realist Instance Ontology and its Logic ISBN 3-938793-33-3 252 pp., Hardcover, EUR 79,00

15 N. Psarros / K. Schulte-Ostermann (Eds.) Facets of Sociality ISBN 3-938793-39-2 370 pp., Hardcover, EUR 98,00

16 Markus Schrenk The Metaphysics of Ceteris Paribus Laws ISBN 13: 978-3-938793-42-8 192pp, Hardcover, EUR 79,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier

PhilosophischeAnalyse PhilosophicalAnalysis 17 Nicholas Rescher Interpreting Philosophy The Elements of Philosophical Hermeneutics ISBN 978-3-938793-44-2 190pp., Hardcover € 89,00

18 Jean-Maurice Monnoyer(Ed.) Metaphysics and Truthmakers ISBN 978-3-938793-32-9 337 pp., Hardcover € 98,00

19 Fred Wilson Acquaintance, Ontology, and Knowledge Collected Essays in Ontology ISBN 978-3-938793-58-9 XX, 726., Hardcover, EUR 159,00

20 Laird Addis, Greg Jesson, and Erwin Tegtmeier (Eds.) Ontology and Analysis Essays and Recollections about Gustav Bergmann ISBN 978-3-938793-69-5 312 pp., Hardcover, EUR 98,00

21 Christian Kanzian (Ed.) Persistence ISBN 978-3-938793-74-9 198pp., Hardcover, EUR 79,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier