Reexamining Berkeley's Philosophy 9781442684751

Citation preview

REEXAMI NI NG BE RKE L EY’ S PH IL OS OP H Y

This page intentionally left blank

Reexamining Berkeley’s Philosophy

Edited by Stephen H. Daniel

U N I V E R S I T Y O F TO R ON TO P RE S S Toronto Buffalo London

www.utppublishing.com © University of Toronto Press Incorporated 2007 Toronto Buffalo London Printed in Canada ISBN 978-0-8020-9348-6 (cloth) Toronto Studies in Philosophy Editors: Donald Ainslie and Amy Mullin

Printed on acid-free paper

Library and Archives Canada Cataloguing in Publication Reexamining Berkeley’s philosophy / edited by Stephen H. Daniel. (Toronto studies in philosophy) Includes index. ISBN 978-0-8020-9348-6 1. Berkeley, George, 1685–1753. I. Daniel, Stephen H. (Stephen Hartley), 1950– II. Series. B1348.R43 2007

192

C2007-900475-X

University of Toronto Press acknowledges the financial assistance to its publishing program of the Canada Council for the Arts and the Ontario Arts Council. University of Toronto Press acknowledges the financial support for its publishing activities of the Government of Canada through the Book Publishing Industry Development Program (BPIDP).

Contents

Preface

vii

Contributors Abbreviations

ix xiii

Introduction 3 stephen h. daniel Berkeley, Ideas, and Idealism michael r. ayers

11

Berkeley’s Assessment of Locke’s Epistemology 29 george s. pappas The Problem of the Unity of a Physical Object in Berkeley richard glauser

50

Why My Chair Is Not Merely a Congeries: Berkeley and the Single-Idea Thesis 82 marc a. hight Berkeley on Visible Figure and Extension 108 ralph schumacher Perceiving and Berkeley’s Theory of Substance phillip d. cummins

121

vi

Contents

Berkeley’s Actively Passive Mind genevieve migely

153

Berkeley’s Four Concepts of the Soul (1707–1709) bertil belfrage

172

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism roomet jakapi

188

Berkeley’s Criticism of Shaftesbury’s Moral Theory in Alciphron III laurent jaffro Berkeley Poetized 214 wolfgang breidert Index

231

199

Preface

For three hundred years George Berkeley’s provocative claim that the existence of things depends on their being perceived has generated agreement, puzzlement, and ridicule. More often than not, he is interpreted in the context of contemporaries with whom he either disagreed or whose ideas he considered as simply beside the point. We are still debating what he meant, but instead of interpreting his remarks solely on how they address specific issues in epistemology or metaphysics, we now at least also consider how his insights are part of a larger, cohesive set of doctrines that includes scientific, religious, and ethical dimensions. This collection of essays acknowledges the ways in which Berkeley’s philosophy hangs together only if its various components are seen as mutually supportive. That sense of how aspects of Berkeley’s thought complement one another was evident in April 2003, when more than twenty-five scholars from the United States, Canada, and Europe met for three days at Texas A&M University to commemorate the 250th anniversary of Berkeley’s death. Out of those discussions in College Station arose most of the essays included in this volume. Neither the conference nor this book would have been possible without the support of the Texas A&M University Office of Graduate Studies, the Melbern G. Glasscock Center for Humanities Research, the College of Liberal Arts, the Franklin J. Matchette Foundation, the International Berkeley Society, Bertil Belfrage, Richard Glauser, Genevieve Migely, Hampden-Sydney College, and several departments and programs at Texas A&M University. I would like to thank all who participated in the conference for providing inspiration for the ideas that have come to fruition in this collection.

viii Preface

This book is dedicated to my wife and best friend Breaux, whose professional efficiency and sense of decorum are balanced by a whimsical fascination with life. To me, she is joy. stephen h. daniel

Contributors

Michael R. Ayers, emeritus professor of philosophy at Oxford University, is the author of Locke: Ideas and Things (1997) and Locke: Epistemology and Ontology (1991); editor of Rationalism, Platonism and God (2007) and Berkeley’s Philosophical Works (1975, 1996); and co-editor of The Cambridge History of Seventeenth Century Philosophy (1998) and Philosophy and Its Past (1978). Bertil Belfrage, research fellow in the history of science and ideas at Lund University, Sweden, is the author of numerous essays on Berkeley and the editor of George Berkeley’s Manuscript Introduction (1987). He is currently preparing a new edition of Berkeley’s works. Wolfgang Breidert, professor of philosophy at the University of Karlsruhe, has translated most of Berkeley’s writings into German (1980–96). He is the author of George Berkeley: 1685–1753 (1989) and essays on Berkeley’s mathematics, and editor of Die Erschütterung der vollkommenen Welt: die Wirkung des Erdbebens von Lissabon im Spiegel europäischer Zeitgenossen (1994). Phillip D. Cummins is emeritus professor of philosophy at the University of Iowa. Known especially for his work on early modern empiricism, he is the co-editor of Minds, Ideas, and Objects: Essays on the Theory of Representation in Modern Philosophy (1992) and author of several essays on Berkeley, Hume, Bayle, and Reid. Stephen H. Daniel, Fasken Chair in Distinguished Teaching and professor of philosophy at Texas A&M University, is the president of the

x Contributors

International Berkeley Society. He is the author of essays on Berkeley’s treatment of the mind–idea relation and The Philosophy of Jonathan Edwards (1994); Myth and Modern Philosophy (1990); and John Toland: His Methods, Manners, and Mind (1984). He is also the editor of Current Continental Theory and Modern Philosophy (2005), New Interpretations of Berkeley’s Thought (2007), and senior editor of Berkeley Studies. Richard A. Glauser is professor of philosophy at the University of Neuchâtel, Switzerland. Among his publications are Berkeley et les philosophes du XVIIe siècle: perception et scepticisme (1999) and essays on how issues in metaphysics and moral philosophy affect one another in the works of Descartes, Spinoza, Leibniz, Locke, Berkeley, Shaftesbury, and other early modern philosophers. Marc A. Hight is Elliott associate professor of philosophy at Hampden-Sydney College in Virginia. He is the author of several essays on Berkeley and Locke, including discussions of Berkeley’s treatments of divine archetypes and bodily resurrection. He is the coordinating editor of Berkeley Studies. Laurent Jaffro is a professor of philosophy and department chair at Blaise Pascal University in Clermont-Ferrand, France. He is the author of Éthique de la communication et art d’écrire: Shaftesbury et les lumières anglaises (1998) and editor of Le sens moral: une histoire de la philosophie morale de Locke à Kant (2000). He has also edited works by John Toland (2003) and Shaftesbury (1993). Roomet Jakapi teaches at the University of Tartu, Estonia. He has translated Berkeley’s writings into Estonian and is the author of articles on the relation of Berkeley’s emotive theory of meaning to propositions about religious truths. Genevieve Migely teaches philosophy at Cornell College in Mt Vernon, Iowa. In 2006 she completed her PhD at Claremont Graduate University, writing her dissertation on Berkeley’s philosophy of mind. George S. Pappas is a professor of philosophy at Ohio State University. He is the author of Berkeley’s Thought (2000) and numerous essays on Locke’s and Berkeley’s epistemology and metaphysics; editor of Justification and Knowledge (1979); and co-editor of Essays on Knowledge and Justification (1978).

Contributors xi

Ralph Schumacher teaches philosophy at the Humboldt University in Berlin. He is the author of essays on Locke, Berkeley, Leibniz, and Kant; editor of Perception and Reality: From Descartes to the Present (2004), Perspectives on Colour Perception (2006), and Idealismus als Theorie der Repräsentation? (2001); and co-editor of Kant und die Berliner Aufklärung (2001).

This page intentionally left blank

Abbreviations

Alc

CSM

DHP Essay

IN

NB NTV PHK

PP PW

George Berkeley, Alciphron, or the Minute Philosopher (1732), as found in W, vol. 3. Cited by dialogue number, section, and page. René Descartes, The Philosophical Writings of Descartes. 3 vols. Ed. and trans. John Cottingham, Robert Stoothoff, Dugald Murdoch, and (vol. 3) Anthony Kenny. Cambridge: Cambridge University Press, 1984–5, 1991. Cited by volume and page. George Berkeley, Three Dialogues between Hylas and Philonous (1713), as found in W, vol. 2. Cited by page. John Locke, An Essay concerning Human Understanding (1690). Ed. Peter H. Nidditch. Oxford: Clarendon Press, 1975. Cited by book, chapter, section, and page. George Berkeley, Introduction to A Treatise concerning the Principles of Human Knowledge (1710). Cited by section number, as found in PW. George Berkeley, Notebooks, also known as the Philosophical Commentaries (1707–8). Cited by entry number, as found in PW. George Berkeley, An Essay towards a New Theory of Vision (1709). Cited by section number, as found in PW. George Berkeley, A Treatise concerning the Principles of Human Knowledge, Part I (1710). Cited by section number, as found in PW. René Descartes, Principles of Philosophy (1644), as found in CSM, vol. 1. Cited by part, section, and page. George Berkeley, Philosophical Works, including the Works on Vision. Ed. Michael R. Ayers. London: J.M. Dent, 1992.

xiv Abbreviations

TDS

THN

TVV W

Peter Browne, Things Divine and Supernatural, Conceived by Analogy with Things Natural and Human (1733). Reprint. Bristol: Thoemmes, 1990. Cited by chapter or page. David Hume, A Treatise of Human Nature (1739–40). Ed. L.A. Selby-Bigge and P.H. Nidditch, 2nd ed. New York: Oxford University Press, 1978. Cited by book, chapter, section, and page. George Berkeley, The Theory of Vision ... Vindicated and Explained (1733). Cited by section number, as found in PW. George Berkeley, The Works of George Berkeley, Bishop of Cloyne. 9 vols. Ed. A.A. Luce and T.E. Jessop. London: Thomas Nelson and Sons, 1948–57. Cited by volume and page.

R E E X A M I N I N G B E R KE L E Y ’ S P H I L OS OP H Y

This page intentionally left blank

Introduction s t e p h e n h . dan i e l

It is not surprising that topics relating to George Berkeley’s intriguing claims about the immaterial nature of reality, knowledge, vision, the self, morality, and religion still provoke debates 250 years after his death. The traditional strategy of treating his philosophy simply as a stepping stone from Locke to Hume is now being supplanted by more refined research into his doctrines and relations to his predecessors and contemporaries. Indeed, recent scholarship asks us to reconsider Berkeley on his own terms, and in some cases that has meant reconfiguring how we frame our understanding of the central themes that inform his thought. As examples of such a development, the essays in this collection invite us to reassess what Berkeley means by saying that our knowledge of the world is nothing other than our experience of ideas. Typically, this way of speaking – designated at times as either immaterialism or idealism – has been understood as the concern of epistemology or metaphysics. But these essays show how Berkeley’s depiction of things in the world as ideas also highlights presuppositions about science, morality, and religion that are much broader than questions about the nature of physical objects or minds might indicate. For in explaining how our ideas are related to physical bodies and God’s ideas, Berkeley provides a nuanced account of reality that, once understood in the context of his doctrine of ‘the reality and perfection of human knowledge,’ no longer can be mistaken for a theologically modified version of solipsism. At the heart of this volume is the question that haunts Berkeley scholarship: namely, how can we really know anything about the world if all we know are our ideas? That is, how is Berkeley’s realism compatible with his idealism? If we say (with John Locke, for example) that ideas

4 Stephen H. Daniel

represent objects or are the things that are before the mind when it thinks, we risk importing a way of speaking about minds, ideas, and objects that prejudices our reading of Berkeley and precludes our appreciating his insights on their own terms. If we say that objects can exist apart from individual minds – for example, in the mind of God – then we risk opening up the possibility that such objects exist ‘in themselves’ apart from our ideas. These essays (when read together) point to a way out of such a quandary by noting how, for Berkeley, the existence of real things depends on their being caused, but they are caused to be exactly those things in precise relations to others only by being both known and willed in terms of mind. Unless they are known as ordered (consistently with God’s will), they will not be recognized as real or identified even as objects, for it is the recognition of the order of ideas that allows them to be seen as unities. Accordingly, a finite mind’s activity of perceiving and willing – like the total complex of all finite acts of perception and willing – is an expression of God’s will that there be such differentiation and order. Things are thus perceived in certain ways not because that is the way they are; rather, things are the way they are because that is the way God wills that they be and be perceived. This point indicates how we cannot simply assume that something other than what we know exists, for that begs the question of how such a reality could mean anything apart from the framework of experience. As Berkeley recognizes, such an assumption undermines the integrity of experience and opens the door to scepticism. In Berkeley’s more holistic account, the moral/natural structure of divinely ordained experience supplants fragmented, atomistic views in which ‘simple’ ideas lack any intrinsic connections. It is not surprising, then, that this volume ranges from epistemology and metaphysics to morality, religion, and even the aesthetic (specifically, poetic) apprehension of Berkeley’s thought. Indeed, the shifts in interpretive focus represented by these essays embody the ways in which Berkeley himself points to the need for different (though complementary) methodologies in handling different topics. Specifically, the first five essays in this collection reveal how ideas are real in virtue of being objects of mind; the last five essays reveal how minds are real in virtue of expressing the harmony of ideas; and the middle (sixth) essay reveals how this juxtaposition of epistemological concerns, on the one hand, and metaphysical, moral, and religious concerns on the other, occurs in Berkeley’s development of a view of reality

Introduction 5

that is based on activity rather than on the static existence of substances. In that view, we do not say that we perceive things because of the way they are; rather, we say that the way things are is a function of our perceiving them in certain ways. By insisting on the volitional character of such constitutive perception, Berkeley associates the order of physical objects with the order of created minds in one ongoing creative act. In the first essay, Michael Ayers argues that to examine Berkeley’s idiosyncratic idealism is to raise questions about what idealism is and how it is expressed in other thinkers (e.g., Kant). Ayers uses Descartes’s (scholastic) distinction between the formal and objective existence of ideas to argue that unlike Descartes or Kant, Berkeley does not allow for things to exist ‘in themselves,’ that is, apart from their being objects identified in our ideas. This way of thinking leads Berkeley to conclude that our ideas of things do not reveal how objects are shaped by forms of sensibility and thought. In fact, for Berkeley, the distinction between things ‘in the mind’ and those ‘in reality’ is suspect, because for him ideas do not represent things in themselves and are related to one another only in a contingent (albeit divinely ordered) way. According to Ayers, then, Berkeley should not be identified with the transcendental idealism of Kant, the problematic idealism of Descartes, or even the conservative idealism of Richard Burthogge, because Berkeley does not think that our fixed conceptual schemes determine the things we conceive. Furthermore, the affective consistency of the order of our ideas is not enough of a commitment to realism to make Berkeley’s pragmatism a proper form of idealism. The issue of how sensible objects can be known without being caused by any thing outside our experience is a concern of George Pappas’s essay. He considers two Berkeleian arguments that purport to reveal the failure of Locke’s epistemology to provide a justified account of how we can know objects perceptually. The first assumes that we can have knowledge only if we are sure of a conformity between our ideas and objects, and the second assumes that we can have knowledge only if we are justified in drawing inferences from our ideas to beliefs or statements about objects. Pappas argues that Locke’s theory of perceptual knowledge is not vulnerable to Berkeley’s critique because it does not make either of these assumptions. The fact that we do not know with certainty whether our ideas conform to objects does not preclude our knowing those objects, for that would require that in order to know anything, we would have to know that we know (which Pappas points

6 Stephen H. Daniel

out is both epistemically unnecessary and unfair as an interpretation of Locke). For Locke, our perceptual knowledge of objects is via ideas, but that does not mean that we infer anything about objects from our ideas, only that there are objects that cause our perceptions. The ‘delicate’ point at issue, then, turns on how the perception of an object can be identified with the object that causes it, and that is where Pappas suggests that a non-inferential, reliabilist version of representative realism can avoid Berkeley’s criticisms. Reframing Locke’s position in this way also opens the door for reinterpreting Berkeley’s strategies along the same lines, in which case it becomes possible to think of Berkeley’s idealism in terms of the divinely ordered complex of real objects identified by ideas. Richard Glauser tackles the thorny topic of how, for Berkeley, ideas are unified in and as particular objects. He argues that such unity cannot be due to the activity of finite minds because such a view is inconsistent with Berkeley’s commonsense realism and God’s role in creating the system of signs embodied in nature. Instead, for Berkeley, a physical object is a divinely instituted combination of sensible ideas that is discovered by finite minds through experience. Of course, no finite mind will be able to perceive the infinite combination of qualities that identify an object. But we can represent a collection of ideas (through memory and imagination) as having a unity in virtue of their arrangement according to certain sorts, kinds, or causal relations, even if such a judgment does not confer unity on objects. Glauser concludes that even though physical objects are neither unified by an ontological principle nor constituted by finite minds, sortal representations of sensible ideas as objects are justified because, in Berkeley’s view, they are objectively (i.e., really) ‘suitable to our ends and purposes.’ Although he is also concerned with Berkeley’s treatment of how ideas of sense comprise physical objects, Marc Hight takes a different tack. He argues that when Berkeley says that ideas of sense are ‘observed to accompany each other’ and marked by one name, he does not mean that a commonsense object is a collection of ideas signified by a name. Rather, an object is a single idea represented by a name that signifies a set of sensory ideas (each of which ‘suggests’ the single idea). In this way, Hight is able to defend the view (challenged in Pappas’s Berkeley’s Thought) that a physical object is still an idea. He agrees with Glauser about the importance of the imagination in identifying a physical object, but he adds that such an identity must itself be an idea. Glauser resists this move because it shifts the activity whereby an object is identified

Introduction 7

away from God to finite minds. But Hight suggests that the process of imagining a collection of ideas as a single idea does not usurp the role of God as much it reinforces the importance of the imagination. Where Glauser, then, explains why sensory ideas form certain collections, Hight focuses on how ideas of imagination are implicit in our perception of objects even though the different sensible ideas that comprise a physical object are not perceived by any one sense. Like Pappas and Hight, Ralph Schumacher investigates what Berkeley means by immediate objects of perception – in particular, objects of sight, visual ideas of light and colour. But instead of examining the principle of unity for such ideas, Schumacher discusses Berkeley’s explanation for how visible figures and extended shapes can be objects of immediate perception when light and colours are supposedly the only real objects of sight. Admittedly, patterns of light and colours can be regarded as visible shapes and marks of tangible shapes. But the issue is how such patterns are recognized in the first place, since even the recognition of connections between patterns of visible and tangible ideas assumes that those patterns have been identified and can be mapped on one another. In order to identify a pattern of visible ideas (i.e., a visible figure) as representing a particular pattern of tangible ideas, it appears that we would have to think of visible shapes as intrinsically related to tangible figures. But that seemingly would contradict Berkeley’s claim that visible figures and extension are immediate objects of sight and not inferential judgments or associations based on experience. Schumacher observes that we can escape this impasse by saying that we immediately (in a non-inferential way) perceive the real qualities of an object (e.g., shape) by perceiving visible figures in relation to other patterns of light and colours, not as marks of tangible shapes. This way of thinking of immediate perceptions as representations of patterns allows us to distinguish between apparent and real properties, but it still suggests that visible shapes intrinsically represent tangible shapes. Phillip Cummins points out that Berkeley’s distinction between perceptions and volitions helps us understand how sensible objects cannot be substances that are ontologically independent from mind. This, Cummins notes, reveals how Berkeley reconfigures the ‘substance tradition’ (e.g., as propounded by Descartes) by proposing that perception is the fundamental relation in terms of which minds are considered the substances on which sensible objects (i.e., ideas) depend. Even though minds are necessarily linked to their objects, they are ontologically prior to – and can be said to ‘support’ – their objects. However, by invoking

8 Stephen H. Daniel

the distinction between perceiving and willing, Berkeley is able to claim that sensible objects are not causally dependent on created minds. In Cummins’s view, the distinction between ontological and (efficient) causal dependence thus qualifies the scope of Berkeley’s idealism by highlighting how what the mind does differs from what it wills. Such a distinction is more radical than that found in Berkeley’s Notebooks (to which Cummins does not refer), but it does explain why scholars find it difficult to link Berkeley’s epistemology to his morality. Genevieve Migely resolves the seeming conflict between the essential activity of the mind and its passivity in perceiving by showing how Berkeley’s view of mind is consistent with his commonsense realism. She notes that, according to Berkeley’s doctrine of mediate perception – where, for example, the sweetness of a cherry or the cherry as a whole is perceived through the immediate perception of its shape – the mind is actively engaged in linking ideas to one another. The mind does not always have a choice about how it actively relates ideas, but it always is engaged in some activity (e.g., desire, assent) regarding its ideas. In that sense, it is active even in immediate perception and is passive only in the sense that it is not the origin or cause of the content of such perceptions. For Berkeley, this way of speaking about mind as essential activity indicates how minds cannot be abstracted from the activities of understanding and will whereby ideas are identified and related to one another. Minds are thus both ontologically distinct and existentially inseparable from their ideas. Bertil Belfrage suggests that such an analysis of mind downplays indications that Berkeley seems to have changed his doctrine of the soul at least four times between 1707 and 1709. According to Belfrage, after initially deciding not to speculate on the religiously sensitive topic of the soul, Berkeley toyed with a ‘bundle theory’ of the self (as a congeries of ideas), but then he recognized that that view – as well as the view adopted by some of his contemporaries, that minds are substances with faculties of intellect and will – does not properly describe the active, willing character of the self. He therefore concluded that ‘persons’ are successions of perceptions and volitions. Indeed, as Berkeley worked on the New Theory of Vision (1709), he came to recognize that the attempt to provide an account of the soul as a passive recipient of ideas and passions needed to be replaced by one in which the soul is described in terms of law-like activities of combining qualities into objects. In Belfrage’s view, Berkeley’s subsequent discussions signal his effort to move beyond this purely empirical description of mind to a metaphysical conception of the soul as a separate active being.

Introduction 9

As commentators have always recognized, Berkeley’s statements about spiritual realities such as souls pose problems because of his insistence that we limit our claims about meaningful utterances to experience. However, Berkeley explicitly notes that metaphysical, scientific, religious, and ethical discourse is often filled with language that is not intended to communicate information or to be descriptive. Actually, non-cognitive (instrumentalist, emotive) uses of words such as gravity or grace might at times apply to Christian mysteries as well. But Roomet Jakapi argues that even though Berkeley considered central Christian beliefs (e.g., about the Trinity) as mysteries, he nonetheless thought of them cognitively as true propositions about supernatural realities. According to Jakapi, Berkeley is able to do this by recognizing that despite the fact that different domains of discourse (such as science or religion) rely on different syntaxes and appeal to different semantic principles, they can still be held to the same philosophic standards of intelligibility. Accordingly, Berkeley avoids having to invoke the kind of ungrounded analogical reasoning (such as his contemporary Peter Browne proposes) by adopting a strategy in which religious propositions are understood in terms of their appropriate spiritual contexts. Laurent Jaffro highlights the distinctly moral context of Berkeley’s thought by considering how Berkeley appeals to the relativist objections of Bernard Mandeville to critique the ethical views of Anthony Ashley Cooper, third earl of Shaftesbury. According to Jaffro, instead of showing how Shaftesbury’s effort to divorce moral judgment and motivation from any religious context ends in superficialities and relativism, Berkeley proposes Christianity as a substitute for morality that removes the need for any natural or autonomous ethical theory. In Berkeley’s view, a ‘minute philosopher’ like Shaftesbury has to rely on a moral sense based on a subjective ability to appraise the value of an action; whereas Christians can rely on the universal, rule-governed, God-given dictates of conscience. In contrast to the ‘passion’ of the moral sense, conscience is capable of enhancement through education. Berkeley acknowledges that Shaftesbury ultimately comes to replace this arbitrary moral sensibility with a cultivated moral taste that is perfected through Stoic self-improvement. But this later ideal of heroic self-achievement strikes Berkeley as unattainable to the average human being, who is much more likely to develop moral dispositions as a result of common habits enforced by divine laws and validated by the attainment of general happiness. Wolfgang Breidert concludes this volume by recounting how poets (e.g., Lord Byron, W.B. Yeats) have invoked Berkeley’s doctrines in ways

10

Stephen H. Daniel

that are witty and insightful, even if often misleading. Such poetic invocations challenge our usual ways of thinking of Berkeley and prompt us to reconsider the traditional formulations of his ideas. For example, in the hands of Sully Prudhomme, Berkeley’s view of an external world ‘in’ the mind becomes an invitation to consider the inherent paradoxes of idealism and immaterialism; and Paul Muldoon reveals how poetry makes thematic the mystery of Berkeley’s talk of unperceived objects and spirits who are unconscious or no longer exist. Other twentiethcentury poets (e.g., Irving Layton, Charles Sisson, Donald Davie) allay these worries by recalling Berkeley’s appeal to God as the cause of our ideas; but the poetic framing of those justifications all the more highlights the ways in which Berkeley’s statements about objects of experience, minds, and God point beyond what a simple reading calls for. As the essays in the second half of this volume attest, for Berkeley, the impulse to limit claims of knowledge only to our ideas (in an almost solipsistic way) is always balanced by references to the spiritual and moral principles that underlie experience as the causes and rationale of existence. These principles embody Berkeley’s belief that even if we do not recognize the unity and integrity of all reality (because of our fallen natures), we are nonetheless capable of appreciating the rationality (i.e., the mentality) of the world. We can do so, however, only if we assume that all things are intelligible in relation to one another. And that is where those interested in Berkeley’s epistemology have to appeal to other aspects of his philosophy. In sum, for Berkeley, the metaphysical, religious, moral, and pragmatic principles whereby things come into being and are initially cognized are not subject to the same strategies of intelligibility as those that characterize our ordinary ways of thinking. Indeed, each of the essays in this collection suggests that the key to understanding specific features of Berkeley’s thought lies in appreciating how they assume ways of thinking that cannot be reduced to one simple strategy. That is why these essays are best read together, for only by seeing how they highlight their own limits by pointing to themes beyond what they address can we appreciate the unity of Berkeley’s philosophy by means of respecting his emphasis on both the variety and harmony of experience.

Berkeley, Ideas, and Idealism m i ch a e l r . ay e r s

If someone asks whether a philosopher is a ‘rationalist’ or an ‘empiricist’ or an ‘idealist,’ the issue is not, or should not be, how near the philosopher comes to articulating one of a number of distinct eternal, ideal philosophies. The question makes more sense if we take it to presuppose only what might be called ‘traditions,’ a more flexible, fuzzy web of doctrinal affinities, debts, and loyalties, not to speak of recognizably familial conflict.1 In this sense of ‘tradition,’ there is characteristically interplay and overlap between traditions. Conflict can be the mother of invention, not only by stimulating new arguments and counter-arguments but by inspiring some to look for a new direction that very often involves reconciliation or synthesis. This, I believe, was repeatedly true of Plato’s ‘battle of the gods and giants’ (Sophist 246a), the open dispute between the traditions that came to be known in their early modern manifestations as ‘rationalism’ and ‘empiricism.’ Aristotle’s system itself looks like a conscious attempt at such a synthesis between what he saw as right in Platonism on the one hand and in empiricist materialism on the other, a reconciliation of love of the forms and teleology with respect for bodily existence and for the foundational role of sense experience in the first steps towards science. If in an utterly different way, Spinoza too appears to be consciously aiming for synthesis, in his case in the form of that somewhat paradoxical achievement, a thoroughly naturalistic, mechanistic Platonism which gives at least one kind of priority to sense and imagination.2 Thomas Lennon plausibly presents another philosopher, Robert Desgabets, as, in yet another way, a sort of Cartesian empiricist.3 The key to Berkeley’s thought, I believe, is that, while deeply and unswervingly committed to some of the main metaphysical conclu-

12

Michael R. Ayers

sions of the gods (though not, of course, to the existence of transcendent universals, even in the mind of God), he sought to show that they can be securely based, and can only be securely based, on the very empiricist epistemology, suitably purged, that was traditionally linked with materialism. What about Berkeley’s relationship to an ‘idealist’ tradition? Such a tradition becomes easy enough to identify after Kant, but who, if anyone, should be included before Kant? The Eleatics and Plato, after all, can be seen as proto-idealists as well as proto-rationalists. Arthur Lovejoy claimed in 1908 that seventeenth-century English Platonists had already enunciated the main theses of Kantian idealism.4 Yet Berkeley, who has generally been considered a paradigm idealist since before Kant wrote, not to say the father of modern idealism, held none of the views discussed by Lovejoy. And indeed, when someone now accuses John McDowell or Donald Davidson of idealism, or offers an idealist interpretation of Spinoza’s Ethics or Wittgenstein’s Tractatus, the theories in question are certainly not being closely assimilated to Berkeley’s. What, then, would someone be claiming of Berkeley, Spinoza, the early Wittgenstein, and McDowell in ranking them together as idealists? I have put the question in this way partly because commentators who talk of Berkeley’s idealism are liable to gloss the reference either simply by referring to his ‘idealism or immaterialism,’ as if the terms meant the same thing, or else by directly stating what is taken to be the fundamental Berkeleian principle or principles constitutive of his idealism. Lisa Downing, for example, writes of Berkeley’s ‘idealism, the positive doctrine that reality is constituted by spirits and their ideas.’5 That gloss (given appropriately explanatory footnotes) is irreproachable as a summary of the immaterialist thesis that is called Berkeley’s idealism. It also has the considerable virtue of directing attention to the primary motivation of Berkeley’s philosophy, which is surely not, for example, to solve some epistemological ‘problem of perception,’ or even to refute scepticism. Yet such a gloss fails to explain what is being said in employing the term ‘idealism’ insofar as it fails to explain the affinities Berkeley’s thesis is supposed to have with other ‘idealist’ philosophies. Let us start with the paradigms that Kant brought under the term ‘idealism’: first, Cartesian scepticism or ‘problematic idealism’; second, the immaterialism or ‘dogmatic idealism’ of Berkeley; and, third, his own ‘transcendental idealism.’ All of these have to do with our ideas of the material world and with what, if anything, they represent. Yet nei-

Berkeley, Ideas, and Idealism 13

ther the first nor the last of these three kinds of idealism could be accurately summarized as the view that reality is constituted by spirits and their ideas. It is true that, as Descartes formulates it, the sceptical argument he uses aims to show that, for all we know, it could be that immaterial spirits are the only substantial things. Yet his sceptical argument is essentially of the same form as the standard argument for brain-in-avat scepticism, which implies no such thing.6 As for transcendental idealism, it may be that Kant believed that our rationality and moral experience point to the fundamentally spiritual nature of things in themselves. Yet, even if so, his view is some way from Berkeley’s metaphysical thesis that spirit is the only substance. For Kant, it seems, even if the order of things in themselves is spiritual, we simply cannot know it. Indeed, as far as objective experience and knowledge can go, material substance has priority in the Kantian system. It may be no accident that idealists have tended to hold that reality is fundamentally spiritual, but that view is not definitive of idealism.7 And that is just as well for the thesis that the position of a present-day philosopher such as McDowell or Davidson is idealist, or for the plausible interpretation of the theory of Tractatus Logico-Philosophicus as a kind of transcendental idealism. It is worth pausing on the platitude that the term ‘idealism’ derives from the word ‘idea’ in its post-Cartesian use. Descartes carefully explained the meaning he gave the word with the aid of some scholastic terminology.8 As historians of seventeenth-century philosophy are now sharply aware, talk of ‘ideas’ was for him entirely commensurable, indeed interchangeable, with scholastic talk of objective being and objective concepts, as these were opposed to formal being and formal concepts. Consideration of the explanation he gives can help clarify the relationship between the three kinds of idealism, and do so in a way that will allow us not only to understand but also to criticize Kant’s conception of idealism and to assess both the characterization of Berkeley as an idealist and the differences between his philosophy and idealism as later developed. Descartes tells us that the representative elements in thought, ideas, can be considered in either of two ways: formally, as mental acts or modes of thought, or objectively, with respect to their objects or content. The expression ‘my idea of the sun’ is therefore ambiguous, referring either to my thinking of the sun (which includes having perceptual experiences of it) or else to the sun itself – not, however, the sun as it is

14

Michael R. Ayers

in reality or in itself (‘in the sky,’ as Descartes once misleadingly puts it)9 but as it is ‘in my understanding,’ that is, as I conceive of it or experience it. All this can be set out as follows: idea

taken formally

act or mode of thought

thing

taken objectively

as it is (exists) objectively

the thing as it is perceived or conceived of, as it is ‘‘in the mind’’

as it is (exists) formally

the thing as it is in itself, in reality

This Cartesian-scholastic schema can be interpreted more generally as a formal account of perfectly ordinary features of natural language, namely of ways in which we ordinarily deal linguistically or conceptually with intentionality, with what has intentional content. For example, to say that Dürer’s picture of a rhinoceros is not much like a rhinoceros is to take the picture objectively, not formally: the comparison is not between some paper with black lines on it and a rhinoceros, but between a rhinoceros as Dürer depicted it and a rhinoceros as it really is. In effect, all that Descartes is telling us about the term ‘idea’ at this point is that it enjoys a systematic ambiguity of the same general kind as such ordinary terms as ‘picture,’ ‘statement,’ and ‘thought,’ not to speak of that other technical term, at least in its scholastic employment, ‘concept.’ All this, so far, is as philosophically neutral and innocuous as a straightforward explanation of what constitutes type-token ambiguity. The systematic ambiguity of such representational or intentional terms exists, as does type-token ambiguity, and it would be as unreasonable, indeed as absurd, to deny the distinctions embodied in this part of our schema as it would be to deny the existence of type-token ambiguity. The other half of the schema, the distinction between things’ formal and objective ways of being, may seem essentially more metaphysical and problematic. Yet the commonplace notion (or way of speaking) of how something was in a dream, or of something’s being so in a story or picture, or of something’s existing only in someone’s mind or imagination, is not, after all, a technical or peculiarly philosophical one. It is just one of the ways in which we commonly deal with the difference between how things actually are and how they are represented as being or appear to be. It is, no doubt, philosophically unsafe to interpret the

Berkeley, Ideas, and Idealism 15

distinction as between different modes of existence or senses of ‘exist,’ and very much better to think in terms of operators or qualifiers: ‘in my dream,’ ‘in the story,’ ‘in appearance,’ and so forth. Such operators serve to mark intentional contexts, with ‘in reality’ employed, roughly speaking, to spell it out that no such intentional operator is in play.10 Yet, however that may be, the distinction itself is embedded in ordinary language and thought. In the terms of the above schema, all three ‘idealisms’ identified by Kant share a common focus on the pig in the middle, ‘ideas’ understood as things as they are ‘objectively’ or ‘in the mind.’ First, in the temporary role of problematic idealist, Descartes takes it that our knowledge of these ideas (or, in the scholastic sense, objective concepts) does not go beyond knowledge of what he calls the forms of our thoughts, that is, their specifying contents as they are (he supposes) necessarily available to the thinker. Insofar as he adopts the stance of sceptic of the senses, Descartes presents us with a gulf between ideas taken objectively and things as they are in themselves, between the apparent world and the real world, a gulf so radical that, allegedly, we cannot be sure of the existence of a real world outside ourselves corresponding to our thoughts of it. This gulf, of course, he later tries and, as most would agree, fails to bridge by an innatist epistemology and a purported proof that a non-deceiving God exists. A main aim of the whole argument, although not its only aim, is to demonstrate that things are not in themselves just as they seem to the senses, but are constituted by matter or body as innately conceived and as described by Cartesian physics, body stripped of purely sensory qualities. Second, as paradigm ‘dogmatic’ idealist, Berkeley in effect agrees that, as far as the sensible or physical world is concerned, we cannot have knowledge going beyond ideas, that is, beyond things as they are perceived or conceived of, as they exist in the mind. As his particular kind of dogmatic idealist, however, Berkeley insists that this is no limitation on our knowledge, since the very notion that there might be a sensible or physical world consisting of material things that are not mind-dependent ideas, but that correspond to and cause ideas or sensations, is incoherent and self-contradictory. The world that is the object of sense perception and the subject matter of natural science lies on this side of the Cartesian gulf, in us, ‘in the mind,’ and is none the worse for that as far as our ordinary notion of reality goes. Rational inference can assure us, moreover, that what lies beyond the gulf is not matter but is an intelligent agent who produces arbitrarily, but wisely and benevo-

16

Michael R. Ayers

lently, ordered ideas of sense. The last of these theses (especially as it is glossed in Siris for the sake of finding anticipations in Platonism) no doubt helps to explain Kant’s conception of Berkeley as a ‘dogmatic’ intellectualist idealist, in the same tradition as Parmenides, if with a rather different conception of the reality behind sense experience. Third (to simplify drastically for present purposes), as transcendental idealist, Kant in effect concurs with the view of both problematic and dogmatic idealists that all that we can have knowledge of through the senses are ideas, and he concurs with Berkeley that that raises no sceptical problems with respect to the objects of ordinary experience or natural science, the objects we ordinarily and, in a sense, rightly think of as real, independent objects for investigation. It is this latter point of agreement, of course, that is the nub of the charge that Kant is more like Berkeley than he admits: at the extreme, that he is a Berkeleian. Yet there are enormous differences. For one thing, Kant retains the category on the right hand of our schema, of things as they are in themselves. For Kant, what is demonstrably paradoxical and absurd is not the notion of things as they are in themselves but the notion that things are in themselves as we conceive of them on the basis of experience (and there is no other basis). A nutshell summary in the terms of our schema might go as follows. For the Cartesian sceptic, that the physical things that exist objectively also exist formally is doubtful. For Berkeley, that the physical things that exist objectively should also exist formally is impossible. For Kant, that the physical things that exist objectively should also exist formally is possible, and, what is more, it can be known when they do; but it is impossible that they should exist formally or fully independently just as, or in the form in which, they exist objectively, that is, as material, physical things. Someone might object to this characterization of Berkeley on the ground that if it were right, Berkeley could not claim, as he does, that he maintains the distinction between things with formal existence and things with only objective existence, that is, the distinction he draws between real things and chimeras. Indeed, at NB 474 he characterizes his own doctrine as I have just characterized it, in closely equivalent terms: ‘N.B. according to my doctrine all things are entia rationis i.e. solum habent esse in Intellectu’ (‘all things are notional beings, i.e. they have being only in the mind’). This entry he later marked with a cross,11 adding the correction, ‘according to my doctrine all are not entia rationis the distinction between ens rationis and ens reale is kept up by it as well

Berkeley, Ideas, and Idealism 17

as by any other doctrine’ (474a). Yet, if he now rejected his former suggestion that no object of perception is ‘real,’ an ens reale, he nevertheless wanted to retain what he had originally offered as a gloss on that suggestion, that every such object exists only in a mind. Accordingly Berkeley is liable to the charge of simply trying to have his cake and eat it by separating the notion of existence in the mind, or objective existence, from its opposition to real or formal existence. These are surely correlative notions, and each has significance only in relation to the other. This may be easier to see if we accept that the distinction is not properly between different kinds of existence, still less between things which have different kinds of existence, but between description or assertion within an intentional context (which may be description of intentional content or, as in creative storytelling, construction of content) and, as we might put it, straight description. ‘Real’ and ‘really’ (‘King Arthur was a real person,’ ‘It looked like a man, but was really a bush’) function like ‘straight’ as employed in the previous sentence. Berkeley does oppose the objective or mind-dependent existence of sensible things to a superior type of existence, but it is not to the formal or independent existence of bodies. It is to the allegedly substantial existence of minds. Since substantial existence is one term of yet another traditional opposition, between the independent being of substances and the dependent being of accidents, the topic of intentionality has become strangely entangled in his theory with the topic of categories and categorical difference. The existence of objects in the mind is explicitly conflated with the existence of accidents in substance, as if the latter notion is to be explained by the former. That is, Berkeley, unlike Descartes, does not take there to be a special sort of accidents of thinking things, namely intentional acts or states, but holds that the only intelligible and possible kind of relation between substances and ‘dependent beings’ or qualities is the perceiver–idea relation (cf. PHK 7). Intentionality is made a fundamental and universal ontological relation, as it were usurping the place of the substance–accident relation. The charge that, in placing all sensible things ‘in the mind,’ Berkeley has helped himself to half a concept, so to speak, clearly has force. For he relies heavily on current philosophical doctrine in explaining and justifying his claim, for example on the popular thesis that colour and other secondary qualities exist only ‘in the mind’ (see, e.g., PHK 49). He could hardly do that and at the same time pretend that he is not using the expression ‘in the mind’ in the usual way; but the usual way presupposes a use for the contrary operator, ‘in reality.’ Moreover, the esse

18

Michael R. Ayers

is percipi principle itself is simply a version of the account given by some scholastics of intentional objects of thought, forms existing intentionally in the mind, that their esse is intelligi. Here lies the previously mentioned difference between Berkeley and Kant, as Kant himself points out. Although Kant’s ‘empirical reality’ can appear to correspond closely to Berkeley’s ‘real things,’ Kant does not incoherently discard the category of things as they are in themselves. If objects of experience, or objects as experienced, are ideas or representations, that means, Kant insists, that there is something unknown that is represented. But Berkeley’s ideas are blank data rather than representations that of their nature point to things existing formally, to things represented that are not themselves representations. This aspect of Berkeleian ideas is not unprecedented and comes to the fore in his arguments that trade on Locke’s conception of sensible qualities as powers to produce simple ideas in us, a conception central both to Locke’s notion of ‘real’ ideas and knowledge and to his account of ‘sensitive’ knowledge. For that Lockean conception, too, treats each simple idea as a blank sensory effect which signifies (i.e., represents) insofar as it signifies (i.e., is an indication of) whatever it is in objects that normally causes such an idea – the power of objects to produce that idea in us. (Descartes, incidentally, held a similarly bad view of sensations as bare data, but Descartes, unlike Locke and Berkeley, distinguishes between sensations and ideas.) Locke attempts to bridge the gulf between ideas and reality in part by postulating a kind of isomorphism between simple ideas received in actual sensation and the powers of their evidently external causes so to affect us. But he also assumes that ideas of primary qualities represent in a stronger sense the constitutional, in particular spatial, properties of things that cause them and that, according to him, they resemble. Locke assumes, too, rightly but without explanation, that we can perceive different qualities as belonging to the same body. Berkeley takes over the ‘blank effects’ model while vigorously rejecting the concomitant, in a way ameliorative but ultimately discordant, Lockean assumptions that the causes of my ideas are external bodies ‘before my eyes’ and that an idea can resemble the quality in bodies that it represents. He has the insight to recognize and exploit the tension in Locke’s epistemology, but he fails to recognize that the real problem lies with the blank-effects model for ideas. If we understand the expression ‘in the mind’ as marking an intentional context – that is, in effect, if we understand the term ‘idea’ more or less as Descartes does – then the assertion that an idea in the mind resem-

Berkeley, Ideas, and Idealism 19

bles its object, something not an idea, is as semantically unproblematic as the assertion that a portrait resembles the sitter or as the claim that Smith’s description of Brown, or Brown as described by Smith, does not resemble the real Brown. Kant himself lays some emphasis on a related difference between his idealism and Berkeley’s: namely, that he, Kant, offers a principled, a priori, unitary explanation or justification of the criteria we employ to distinguish reality from illusion. On Berkeley’s account of ‘real things,’ in contrast, it seems that ‘real’ simply sums up a set of contingently related properties of certain ideas, some of them intrinsic properties and some relational, that together define a class of ideas that experience teaches us are pragmatically important for their connections with pain and pleasure, and as grounds for certain inferences. Given their associations with one another, from some ‘real’ ideas we can infer others and so can anticipate pleasure and pain; and, given their overall purposive character, from this class of ‘real’ ideas we can infer the being of a wise and benevolent God. But the character, significance, and importance of ‘real’ ideas are on this account contingent, a posteriori, due to the grace of God. This difference from Kant is closely related to other differences and to Berkeley’s relentlessly empiricist epistemology. Kant stresses the weakness of Berkeley’s account of space and geometry and is clearly (and surely rightly) opposed to anything like Berkeley’s reductive, subjectivist treatment of time. Moreover Berkeley has no place for an innate scheme of ordering concepts or categories such as are supposed by both Descartes and Kant to play an essential role in the move from mere sensation to the perception or experience of objects. His objects of perception simply are what are given to sense, ‘our own ideas or sensations,’ as he calls ‘sensible things.’ Berkeley says this very emphatically (e.g., at PHK 4) and evidently assumes it in a vast number of other passages. Nevertheless some commentators argue that his account of substantives such as ‘apple’ or ‘stone,’ as standing for dependably associated ‘collections of ideas,’ in fact implies that we have, or construct, notions of physical things such as cannot simply be given as objects in single sense-occurrences, and reasonable belief in the existence of which must be inferential, the inference being to other actual or possible ideas.12 Yet, even if it is true that some of what Berkeley says has such an implication, there is little evidence that the thought occurred to Berkeley fully explicitly or at all insistently, or that his notion of a ‘collection’ in fact moved very far towards the early twentieth-century empiricist theory of logical constructions. In the Notebooks he considers treating qualities

20

Michael R. Ayers

as powers, that is, powers of or in God, and in a number of places he responds to the issue of unperceived existence by means of hypotheticals. There is one well-known, cryptic entry which might be taken to imply that what is a collection for us may be unitary for God: Bodies etc do exist whether we think of them or no, they being taken in a twofold sense, Collections of thoughts and collections of powers to cause those thoughts, these later exist, tho perhaps a parte rei it may be one simple perfect power. (NB 282)

Yet Berkeley here seems to be alluding to the possibility of a single, complete act of creation, a topic evidently just then on his mind. In any case, this and the other passages identifying sensible qualities with powers of God, rather than with immediately perceived ideas, are marked with the cross of possible rejection. Berkeley’s considered position with respect to ‘things’ seems to be that we tend to lump together different but contingently associated ideas or sensible qualities13 as making up the same thing, with the same name, simply as a matter of convenient shorthand, just as we use the same term for the allegedly different qualities or ideas of visual and tactual squareness (cf. NTV 139–45). Clearheaded philosophical reflection will simply distinguish what in ordinary life we conflate, rather than reveal a structured scheme of different conceptual levels as between a thing and its qualities: As to what philosophers say of subject and mode, that seems very groundless and unintelligible ... To me a die seems to be nothing distinct from those things which are termed its modes or accidents. And to say a die is hard, extended and square, is ... only an explication of the meaning of the word die. (PHK 49)

‘Strictly speaking,’ Berkeley notoriously tells us, ‘we do not see the same object that we feel; neither is the same object perceived by the microscope, which was by the naked eye’ (DHP 245). All this fully justifies Kant’s complaint that Berkeley’s ontology of ideas is essentially atomistic, making the connections or relations between ideas, or what appears in sensation, essentially a posteriori. The language that is explained as marking those connections and analogies, whether in talk of ‘things’ perceivable by different senses or in talk by physicists of attractive forces or the like, Berkeley represents as, in the last analysis, useful forms of speech, a veil of words that can be stripped away to reveal naked ‘sensations or ideas,’ the subject matter of ‘strict’ speech.

Berkeley, Ideas, and Idealism 21

I would suggest that, accordingly, despite his role in the history of the term, Berkeley is not a mainstream idealist. That is to say, he is not a philosopher who holds that the objects of our knowledge and thought, things as we conceive of them, are shaped by the ways in which we apprehend and comprehend the world, by the forms of our sensibility and/or thought. I say ‘sensibility and/or thought’ just because our conceptual scheme or theory of the world may, for any particular idealist, be either more or less dependent on, or involved with, the forms of our sense experience. For Kant, that things are spatially and temporarily presented is something prior to categories or concepts, even if the capacity to assign objectively determinate spatial and temporal location presupposes the categories. But for some present-day idealists, such as Davidson, sensibility has a purely causal role in shaping judgment. For post-Kantian idealists, moreover, our conceptual scheme may be either immutable (as Kant believed) or mutable, and either justifiable (as Kant held) or a brute fact of human nature; or it may be thought simply a socially determined product of a particular culture. Yet for all idealists, the shape of our conceptual scheme is imposed on the world we apprehend rather than derived in any respect from the independent world through our apprehension of it. According to idealist criticism, the very idea that the independent world could give us our objects presupposes, absurdly, the possibility that we could climb out of our perceptual and conceptual skins, as it were, or peer round our concepts and modes of perception in order to view the world directly and compare it with the way we experience it and think of it. For example, here are a couple of quotations from a prominent idealist of our own time, Hilary Putnam: Kant ... taught us that the whole idea of comparing our conceptual system with a world of things-in-themselves ... is incoherent. ... [This idea of Gadamer’s] – that objects and reference arise out of discourse rather than being prior to discourse – is rather widespread in twentieth-century philosophy, in both its analytical and ‘continental’ varieties ... ‘Objects’ do not exist independently of conceptual schemes.14

One might think, with Putnam, that such conceptual idealism arrived on the scene with Kant and that Berkeley relates to it ancestrally, a kind of proto-idealist with a coarse resemblance to later conceptual idealists. The famous argument as to the impossibility of conceiving of a tree existing unperceived (sometimes called his ‘master argument’) might seem a crucial antecedent of ‘what Kant taught us.’ Yet the argument is

22

Michael R. Ayers

simply directed to the conclusion (already proved to Berkeley’s satisfaction) that sensible things have an ontologically dependent rather than ‘absolute’ existence. The ontological status of sensible things is for Berkeley a fundamental metaphysical fact, a fact about the structure of reality, so to speak, and the real point of the ‘master argument’ is simply to explain why people so generally fail to recognize this fact.15 It is not concluded that the sensible world is mere appearance, that reality eludes us, or that our objects are distinguished and shaped by our faculties, by the forms of our sensibility and thought. Indeed, the very idea of the form of our sensibility is alien to Berkeley. It is, perhaps, easier to recognize Berkeley’s distance from the heart of the idealist tradition if it is also recognized that a more full-blooded, comprehensive conceptual idealism was up and running well before Berkeley. In the paper previously mentioned, Lovejoy in effect proposes that the view of seventeenth-century Platonists that our way of conceiving of the world is shaped by the innate structure of our cognitive faculties already constitutes the ‘Copernican revolution’ in epistemology.16 Of course there has to be more to idealism than that, since such innatism was commonly combined, as by Descartes, with the realist view that there is a preestablished match between innate conceptual structure and the world. Another Platonic, or rather Neoplatonic, line of thought, however, was to the effect that we are unable to achieve more than partial and inadequate views of reality, if only because our knowledge is always of diversity, whereas reality is in itself one, an unarticulated unity, the One. As the Neoplatonist Robert Greville, Lord Brooke put it in The Nature of Truth (1641), ‘All Being is but one, taking various shapes, sometimes discovering it selfe under one, sometimes under another, whereas it is but one Being.’17 The diversity lies, not in reality, but in our notions. Time and place, Brooke claimed, are unreal appearances: ‘all things did exist in their Being with God ab omni aeterno ... and this succession is but to our apprehension’ (NT 99). If this is not comprehensive idealism, it is because, as in Plato’s Republic, some promise is held out of a transcendent, a priori view of reality in which, from the contemplation of ‘Universall Being and Unity,’ we would be able to grasp how ‘all things are but one emanation from power divine’ (NT 115). A more sober, empirically minded, and yet even more clearly idealist theory was advanced by Richard Burthogge in 1678:18 Every Faculty hath a hand, though not the sole hand, in making its immediate Object; as the Eye makes the Colours it is said to see, the Ear the

Berkeley, Ideas, and Idealism 23 Sounds, the Fancy the Idols, and so the Understanding the Conceptions or Notions under which it apprehends and sees things. So that all the immediate Objects of Human Cogitation (to use the word in its largest sense) are Entia Cogitationis, All Appearances; which are not properly and (may I use a School-term) formally in the things themselves conceived under them. (OVN 13)

Things in themselves are beyond human comprehension, and attempts to penetrate to them will lead only to confusion and paradox: As in Quantity, the common Notion of it, how evident is it! ’Tis evident to all men, and none but knows what is meant by it; and he that looks on Quantity but so, observes a due distance; but whoever looks nearer, looks too near, and is confounded with the composition of the Continuum (and well he may that takes a Phaenomenon, a Spectrum, an Appearance for a Reality). (OVN 40)

Best, then, Burthogge thinks, to stay this side of the Cartesian gap: ‘He that looks for Notions in Things, looks behind the Glass for the Image he sees in it’ (OVN 16). In other words, a certain kind of metaphysics is misguided, since its supposedly external object is really the structure of our own thought reflected in, or projected onto, the reality that appears to us. Kant’s ‘Copernican revolution’ might well seem to have been under way in 1678. Just because we cannot take, in McDowell’s expression, a ‘sideways on’ view of our cognitive relation to the world, comparing scheme with reality (OVN 53) or with the divine view of it (OVN 48), cognitive progress, for Burthogge, can only be assessed by reference to properties internal to our system of beliefs, above all its coherence and comprehensiveness. Where rationalistic Platonism offers the possibility of transcending the partiality of ordinary experiential knowledge and achieving a principled, unitary science, Burthogge’s idealism holds that we must start from sense experience to build as coherent and comprehensive an account of the world as we can. Yet even our very best account necessarily falls short of being an account of things as they are in themselves just because it is framed in ways determined by our faculties, by us rather than by the world. Burthogge does, however, share the Platonist belief in a unified and coherent independent reality, a belief that underlies the role he ascribes to coherence as criterion of truth (OVN 33, 41–3). Berkeley differs from such predecessors as he differs from Kant.

24

Michael R. Ayers

Ideas are not appearances of anything, whether inadequate or adequate appearances. They are what appears, caused in us by the only transcendent reality, God. True, the divine ideas have to be different in kind from our ideas of sense, since God is purely active and outside time. But there is no proposal in the early and main works, as far as I can see (unless, just possibly, one is hinted at in that possibly discarded note about how things may be a parte rei), that our ideas are inadequate temporal representations of an eternal intelligible reality constituted by divine ideas (or one divine Idea) different in kind from our ideas. On the contrary, Berkeley discusses the difference between human and divine ideas purely as if dealing with certain purported difficulties for his assimilating them as he does: the problem of whether God has sensations and the problem of the creation of the sensible world. Each difficulty meets with a theoretically economical defence, with no sign that here at last we are getting to the heart of Berkeley’s philosophy in a contrast between the unity and eternity of all Being and the multiplicity and temporality of our apprehensions of it. Berkeley does not present the order of nature as the ineluctable emanation from, or unfolding of, the Divine Nature. On the contrary, it is for him the arbitrary creation of a supremely wise spirit, contingently law-governed precisely for the good of created spirits and constituting, in effect, both the world in which those spirits can act and live wisely and morally and a language in which the divine wisdom is made manifest. The contingency of laws expressly allows the easy possibility of miracles. If miracles are a problem for Berkeley, it is not because they would have to be contrary to a necessary order of nature, since there is in his view no such necessary order, but because Berkeley’s account of ‘real things’ appears to rule out that contra-nomic events would count as ‘real.’ Berkeley’s ‘idealism,’ moreover, is fundamentally unlike the idealisms of Neoplatonists, of Burthogge and of Kant in another way, since the latter are universal in scope, including our notion of ourselves. Berkeley, on the other hand, in this like Leibniz and indeed Descartes, is clear that we have immediate, reflexive, intuitive knowledge of ourselves as simple, undivided, active (if also passive) beings. He seems ready to follow Locke in understanding reflexive intuition as a form of ‘experience,’ although he also follows tradition (and Descartes) in ascribing it to pure intellect. This is of ontological and theological as well as epistemological significance, since it was an important issue whether finite spirits are true monads or simply aspects of the One, as was held

Berkeley, Ideas, and Idealism 25

by some Neoplatonists, including Spinoza and Lord Brooke. I know of no evidence that the latter view had any attractions for Berkeley.19 Pre-Kantian idealism sprang from a number of sources, not only from Platonism. Gabriel Nuchelmans has argued plausibly that Burthogge’s idealism was a development of the old notion that the distinctions and categories that logical theory deals with are a product of the mind as it shapes its subject matter for the purposes of thought and reasoning.20 This idea had already served as a stick with which to beat the Aristotelians, who were accused of confusing merely logical distinctions and entities for real ones. The step into idealism taken by Burthogge is constituted by the thought that all our distinctions are merely logical or intentional, all our objects are ‘beings of reason’ with no more than objective existence. We can best understand his taking this step in the light of his advocacy of religious toleration and his consonant desire to formulate an anti-dogmatic epistemology without falling into scepticism. He was not alone in being so motivated, and anti-dogmatism remains a motive of some forms of idealism still. Berkeley’s motives, and his philosophy, were very different. He also appears much further from the centre of the idealist tradition as it was already developing. Paradigm ‘idealist’ as he was for the eighteenth century, his philosophy was by no means the prototype for later conceptualist idealism. That had roots in earlier thought and, for good or ill, blooms prolifically in present-day philosophy ‘in both its analytical and “continental” varieties.’ Notes 1 That is not to rule out historically unconnected traditions within different cultures (e.g., within Buddhism and modern European philosophy), where both are reasonably described as versions of ‘idealism’ on the basis of similarities of doctrine and argument. The question of which is closer to ‘pure’ idealism is the one I would want to reject. 2 Spinoza combines a Platonic emphasis on the unity and perfection of Being from which everything flows with an almost Hobbesian account of human knowledge tucked into a rationalist shell. Cf. my ‘Spinoza and Platonism,’ in Rationalism, Platonism and God, ed. Michael R. Ayers (Oxford: Oxford University Press, forthcoming). 3 Thomas M. Lennon, ‘The Cartesian Dialectic of Creation,’ in The Cambridge History of Seventeenth-Century Philosophy, ed. Daniel Garber and Michael Ayers (Cambridge: Cambridge University Press, 1998), 351–3.

26

Michael R. Ayers

4 Arthur O. Lovejoy, ‘Kant and the English Platonists,’ in Essays, Philosophical and Psychological: In Honor of William James, ed. George S. Fullerton et al. (New York: Longmans, Green, 1908), 263–302. 5 Lisa J. Downing, ‘Berkeley,’ in A Companion to the Philosophers, ed. Robert L. Arrington (Malden, MA: Blackwell, 1999), 169. 6 Admittedly, not all of us would now call the Cartesian sceptic an ‘idealist,’ if only because Descartes allows the possibility of the commonsense view that independent objects exist more or less as we experience them, apart from the occasional illusion. 7 In 1955 Colin Turbayne published a paper that stimulated a thirty-year debate, collected by Ralph Walker in The Real in the Ideal (New York: Garland, 1989), on the old question of the relation between Kant and Berkeley: in particular, the question whether their theories are not more alike either than Kant was aware or else, to his discredit, than he was ready to admit. Alterations to the first Critique in its second edition, apparently stimulated by a reviewer’s suggestion that the author was in effect a Berkeleian, have been taken together with other references to Berkeley as evidence that Kant had little understanding of Berkeley’s philosophy. The consensus of most contributors to the debate, however, a view with which I still concur, is that these references and criticisms not only suggest that Kant understood Berkeley rather well, in some respects (as I myself argued) better than some modern commentators, but also largely justify his insistence on the distance between their views. I will not now rehearse the arguments. I want rather to approach the issue, and the question of the relation between Kant’s three kinds of idealism, from a bit further away. 8 See my ‘What Is Realism?’ Aristotelian Society Supplementary Volume 75 (2001): 92–5. 9 ‘Author’s Replies to the First Set of Objections,’ in CSM 2: 75. The phrase is in effect a category mistake in Gilbert Ryle’s sense. The sun is in the sky, whether ‘in my mind’ or ‘in reality.’ 10 The situation is interestingly more complicated than ‘roughly speaking’ might suggest. In philosophical talk of ‘things as they are in reality,’ ‘things in themselves,’ etc., the operator or qualifier might not seem simply to be signalling the absence of any intentional qualifier. But consider the philosophical claim, ‘Nothing is coloured in itself.’ This can indeed be taken to imply that ‘Rubies are red’ is false unless an intentional qualifier is presupposed (‘In appearance, rubies are red’). Here, of course, there is room for the familiar complaint that this qualifier has other work to do, distinguishing, for example, the moon’s being red only in appearance from the grass’s really being green. But this is like arguing that it is a misuse of language to

Berkeley, Ideas, and Idealism 27

11

12 13 14

15 16 17

18

19

say that a ring round the moon, or a rainbow, is a matter of appearance, since we need to distinguish there really being a ring round the moon or a rainbow from there merely appearing to be one because of the spectacles we are wearing (or whatever). The issue of whether the + mark indicates a position which Berkeley later rejected is a point of debate among Berkeley scholars. In my view it is better regarded as a general mark of exclusion from intended publication rather than specifically one of intellectual rejection, although it seems clear that rejection and change of mind are often reasons for exclusion. I think it is therefore unwise to base an interpretation of Berkeley’s overall position on any entries marked in this way. Cf. the essays by Richard Glauser and Marc Hight in this collection. Berkeley makes no distinction between the two, of course. Hilary Putnam, Realism and Reason (Cambridge: Cambridge University Press, 1983), 177, xvi; and idem, Reason, Truth and History (Cambridge: Cambridge University Press, 1981), 52. Cf. DHP 200, where the point of the argument is to convict Hylas of a ‘pleasant mistake.’ See Lovejoy, ‘Kant and the English Platonists,’ 269. The Nature of Truth: Its Union and Unity with the Soule, Which Is One in Its Essence, Faculties, Acts; One with TRUTH (London: Samuel Cartright, 1641), 43. Henceforth, NT. In Organum vetus et novum (London: Samuel Crouch, 1678), 13. Henceforth, OVN. Burthogge is discussed by Lovejoy (‘Kant,’ 278), but as if he were a Platonist. Lovejoy’s knowledge of Burthogge was admittedly second-hand, since no copy of his work could then be found in New York! All this raises the question of the significance of Siris, a work in which Berkeley may seem to endorse just the kind of Neoplatonic theory that has a historic connection with later conceptualist idealism. There are roughly three possible views of the relation of Siris to the earlier works: first, that Berkeley has radically changed his philosophical position, with or without a softening of the brain; second, that Berkeley is rambling through the Platonic tradition and (non-accidentally) finding there glimmerings of his own published philosophy and of Christian doctrine, as well as of a sort of metaphysical explanation of the alleged efficacy of tar-water; third, a view that Stephen Daniel has recently proposed [in ‘Berkeley’s Christian Neoplatonism, Archetypes, and Divine Ideas,’ Journal of the History of Philosophy 39 (2001): 239–58], that the endorsement of Neoplatonism in Siris is the revelation of what has been underlying Berkeley’s philosophy since before the first entries in his notebooks, a Neoplatonist theology centring on an inter-

28

Michael R. Ayers

pretation of the Trinity that finally arrives on stage only in the concluding sections of Siris. I find the second of these by far the most plausible, in general agreement with T.E. Jessop. Daniel’s argument is interesting but in my view far-fetched. It runs up against almost everything Berkeley says in his main works about the nature of spirits and their relation to God, not to speak of the arbitrary contingency of the natural system. At the same time, it relies heavily on tendentious readings of just those entries in Berkeley’s early notebooks that he marked with a cross (cf. note 11 above). 20 See Gabriel Nuchelmans, Judgment and Proposition from Descartes to Kant (Amsterdam: North-Holland Publishing, 1983), ch. 6. Nuchelmans takes the main influence on Burthogge to have been Arnold Geulincx.

Berkeley’s Assessment of Locke’s Epistemology ge o r g e s . pap pa s

Among the many different criticisms that Berkeley has of Locke’s philosophy are a number of epistemic arguments, each designed to show that Locke cannot account for sensitive knowledge. More specifically, they are supposed to show that if Locke’s theory of perception is correct, then we do not have any perceptual knowledge of objects. These arguments have not received as much attention as, for example, Berkeley’s attack on abstract ideas, though it seems to me that they are equally important. Berkeley’s claim that his philosophy can explain how we come to have knowledge of objects gets some of its force from the supposed failure of Locke’s theory. In this essay I will examine anew two of Berkeley’s main epistemic arguments against Locke. The first, what I will call the conformity argument, holds that if Locke’s theory of perception is correct, then we have knowledge of objects only if we are sure that a conformity obtains between currently experienced ideas and some features of the object. Berkeley thinks we are never in a position to be sure that the needed conformity holds, so that Locke’s theory leads to the result that we have no knowledge of objects. The second argument, what I will call the inference argument, contends that on Locke’s theory of perception, one has knowledge of objects only if one can make a warranted inference from one’s currently experienced ideas to some belief or statement about the object. Berkeley thinks that this inference would not be warranted, so that, once again, we reach the conclusion that on Locke’s theory of perception, we never attain perceptual knowledge of objects. Berkeley’s arguments have generally been regarded as decisive. Philosophers from Hume to the present day have tended to accept them,

30

George S. Pappas

particularly the inference argument.1 I believe, however, that these arguments are much less forceful than they might initially appear. Indeed, the conformity argument as it usually has been construed does not strictly appear in Berkeley’s texts. What he actually presents is a somewhat different argument that completely fails to engage what Locke had defended. The inference argument also fails to engage Locke’s actual position, I maintain, because it wrongly asserts that Locke holds that perceptual (sensitive) knowledge of objects requires an inference from currently experienced ideas to a belief about the object. It also fails as a philosophical argument, taken separately from what Locke actually held. What Berkeley does not see is that the representative realist theory of perception that he takes Locke to hold is compatible with perceptual knowledge of objects being non-inferential. 1. Context Berkeley’s target with the epistemic arguments to be discussed, I have said, is Locke. I do not want to deny that Berkeley may have had other philosophers in mind as well when he formulated these arguments – perhaps he was also thinking of Descartes or Malebranche, for example. Here I take no position on these possible additional targets of Berkeley’s arguments and thus have nothing to say about the effect of his arguments on their respective positions. Further, even identifying Locke as a principal target of Berkeley’s criticisms on these epistemic points is a matter that calls for close textual analysis and interpretation in which I will not here engage. I will instead assume that this case can be successfully made. Connected to this is the fact that Berkeley’s Locke is reckoned to hold a representative realist theory of perception. On this point one might be tempted to assess Berkeley’s criticisms by noting that right at the outset Berkeley has taken a wrong step because Locke does not really defend representative realism. Rather, according to this line of thought, Locke is really a direct realist about perception, and within that position ideas of sensation are not themselves objects, and hence they are not perceived objects. Accordingly, physical objects and not ideas are the things that are typically immediately perceived. Berkeley’s epistemic criticisms are thus ineffective, not because they are weak arguments against representative realism but because they attribute to Locke a theory that Locke simply did not accept. I confess to having great sympathy with this direct realist reading of

Berkeley’s Assessment of Locke’s Epistemology

31

Locke. In this essay, however, I will not follow the lead of the different commentators who have proposed this interpretation, for I am mainly interested in Berkeley’s Locke. That is, I am interested in Locke as Berkeley understood him and in whether that Locke is undermined by the arguments Berkeley presents.2 2. The Conformity Argument The conformity argument is set out in PHK 86, which I here quote nearly in full: And first as to ideas or unthinking things, our knowledge of these has been very much obscured and confounded, and we have been led into very dangerous errors, by supposing a twofold existence of the objects of sense, the one intelligible, or in the mind, the other real and without the mind: whereby unthinking things are thought to have a natural subsistence of their own, distinct from being perceived by spirits. This which, if I mistake not, has been shown to be a most groundless and absurd notion, is the very root of skepticism; for so long as men thought that real things subsisted without the mind, and that their knowledge was only so far forth real as it was conformable to real things, it follows, they could not be certain that they had any real knowledge at all. For how can it be known, that the things which are perceived, are conformable to those which are not perceived, or exist without the mind. (Emphases in original)

There is a companion argument in the Dialogues, presented by Philonous: It is your opinion, the ideas we perceive by our senses are not real things, but images, or copies of them. Our knowledge therefore is no farther real, than as our ideas are the true representations of those originals. But as these supposed originals are in themselves unknown, it is impossible to know how far our ideas resemble them; or whether they resemble them at all. We cannot therefore be sure we have any real knowledge. (DHP 246)

On the face of it, this argument from the Dialogues begs the question against Locke, as it assumes that objects (the ‘supposed originals’) are not known. To avoid this problem, Berkeley’s term ‘unknown’ in the phrase ‘these supposed originals are in themselves unknown’ must mean something else, most likely ‘unperceived.’ Such a reading would

32

George S. Pappas

align this argument with a point made in a related argument presented at the end of the first Dialogue: philonous. But neither is this all. Which are material objects in themselves, perceptible or imperceptible? hylas. Properly and immediately nothing can be perceived but ideas. All material things therefore are in themselves insensible, and to be perceived only by their ideas. philonous. Ideas then are sensible, and their archetypes or originals insensible? hylas. Right. (DHP 206.)

This passage is followed by a statement of a version of the likeness principle: philonous. But how can that which is sensible be like that which is insensible? Can a real thing in itself invisible be like a colour; or a real thing which is not audible, be like a sound? In a word, can anything be like a sensation or idea, but another sensation or idea? (DHP 206; emphases in original)

In all of these passages Berkeley takes the Lockean position to affirm a likeness between ideas and insensible or imperceptible objects, and it may rightly be protested that this is to mistake Locke’s theory. Locke holds only that objects are not perceived in the way ideas are, that is to say immediately, and not that objects are insensible. However, it is doubtful whether Berkeley makes this simple error. For he has Hylas say ‘material things therefore are in themselves insensible, and to be perceived only by their ideas.’ This suggests that he is perfectly aware that Locke’s position is that objects are perceived indirectly and by means of ideas, not that they are altogether unperceived. Interestingly, this point does not matter to Berkeley’s overall argument for, as the last-quoted passage makes clear, the likeness principle is quite general and applies to all putative likenesses between ideas and non-ideas, whether or not these latter items are sensible in some way. These latter passages from the first Dialogue may be taken as some evidence that Berkeley means by ‘unknown originals’ something like ‘unperceived originals,’ where the latter most likely means ‘not perceived immediately.’3 Even with this understanding, however, the conformity argument as presented in the Dialogues differs from that of the Principles because the former attributes to Locke the requirement that

Berkeley’s Assessment of Locke’s Epistemology

33

one know about a resemblance between ideas and objects, where the latter speaks more generally of knowing that there is a conformity between ideas and objects. The latter is preferable, for it more squarely lines up with what Locke held. Only some simple ideas are ever claimed by Locke to resemble some features of objects, while all simple ideas are said to be conformable to objects. So I will concentrate on the Principles version of the argument. In this argument Berkeley mentions two different conformities: one is between our knowledge and real objects, while the second, given in the last sentence of the passage, is between experienced ideas and objects. Further, as he is speaking of real knowledge, it is clear that he has in mind Locke’s discussion of the reality of knowledge in Essay IV.iv. There Locke does speak of a conformity between our ideas and objects but not at all of a conformity between our knowledge and objects. Hence, it appears that Berkeley’s conformity argument is a nonstarter, as it fails to engage anything Locke was actually defending. Moreover, there are additional problems with the argument, which we can see by stating the argument as Berkeley presents it: (1) We have real knowledge of objects only if there is a conformity between our knowledge and objects. (2) We cannot know that there is a conformity between our ideas and objects. (3) Hence, we cannot be certain that we have real knowledge of objects. The first problem, already noted, is that premise (1) does not engage Locke’s actual views. Beyond that, premise (2) does not deny the consequent of (1). One reason is that the two premises are alluding to two different conformities, so the argument strictly commits an equivocation. But even if the conformities alluded to in these premises were the same, (2) would still not deny the consequent of (1). Premise (2) denies that we can know about a certain conformity, not that there is no such conformity of the sort affirmed in (1). Lack of knowledge of a conformity does nothing to deny the existence of a conformity. There is yet another problem with the argument. As the full passage from PHK 86 makes plain, Berkeley’s contention is that Locke’s theory leads to scepticism; that is, to the result that we do not have knowledge of objects. However, this is not how Berkeley states the conclusion of the conformity argument. Rather, Berkeley speaks in (3) of our not being certain that we have real knowledge,4 and that is perfectly com-

34

George S. Pappas

patible with our actually having such knowledge. To see this, imagine that being certain that p is tantamount to knowing that p, not an unreasonable assumption for the early eighteenth century. Then Berkeley’s conclusion in (3) merely asserts that we do not know that we have knowledge of objects. It is of the form ~ K(Kp), and of course many sentences of this general form are perfectly consistent with sentences having the form Kp. Four-year-old Brianna may not know that she knows that her brother’s name is Patrick, because she lacks the concept of propositional knowledge. But certainly she knows that her brother’s name is Patrick. Thus, the conclusion Berkeley states in (3) is completely compatible with the denial of scepticism about objects. So, even if the conformity argument as given in PHK 86 proves something, it hardly proves that Locke’s theory leads to scepticism. 3. Another Try By being charitable to Berkeley we can come up with a different conformity argument. Two instances of charity concern premise (1), which speaks of a conformity between our knowledge and objects. As premise (2) speaks of a conformity between ideas and objects, which is more in line with what Locke accepted, we can change (1) to make it line up both with Locke and with Berkeley’s own second premise. In effect, this is taking Berkeley’s phrase ‘their knowledge was only so far forth real as it was conformable to real things’ so that the term ‘it’ is really not referring back to ‘knowledge’ as we had earlier construed it but is actually standing in for ‘ideas.’ A second case of interpretive charity concerns what we may think of as strong conformity, the sort that demands that one know that one’s ideas conform to objects. This is what is denied in premise (2). Premise (1), however, refers merely to weak conformity, that is, to the mere presence of a conformity between ideas and objects. Therein lies the equivocation noted earlier. One change would be to make use of strong conformity in both premises. Doing that requires another change in (1) and assumes that strong conformity was uppermost in Berkeley’s mind – not an unreasonable assumption since Berkeley certainly alludes to it in PHK 86. On the other hand, Berkeley may have had weak conformity in mind all along, and if he did, we would want to change the argument so that weak conformity was used throughout. Even with these possible changes noted, however, there is the problem that the original conformity argument Berkeley presents is invalid

Berkeley’s Assessment of Locke’s Epistemology

35

because he overstates the conclusion. As we have seen, he claims that we could not be certain that we have real knowledge of objects. However, we can provide reasons to think that this statement is not what Berkeley aims to establish. Berkeley touts his own immaterialist position as avoiding the sceptical results that he thinks attend Locke’s theory. What he claims is that according to his theory, one does have knowledge of objects and that it is acquired in perception. Here is one passage where this point is made: I am of a vulgar cast, simple enough to believe my senses, and leave things as I find them. To be plain, it is my opinion, that the real things are those very things I see and feel, and perceive by my senses. These I know, and finding they answer all the necessities and purposes of life, have no reason to be solicitous about any other unknown beings. (DHP 209)

This passage comes on the heels of a discussion of scepticism, where Hylas is said to hold views that lead to scepticism, while Berkeley holds a view supporting the opposite, non-sceptical position. What is thus suggested is that Berkeley takes scepticism per se to be the thesis that there is no perceptual knowledge of objects. Hence, it would be in keeping with Berkeley’s thinking if we recast the conclusion of the conformity argument so that it states a sceptical position directly rather than the overstated conclusion he seems to endorse. Making all of these changes, both charitable and interpretive, yields two different ways to understand the conformity argument: (1a) We have real knowledge of objects only if we know that our ideas conform to objects. (2a) We do not know that our ideas conform to objects. (3a) Hence, we do not have real knowledge of objects.

(1b) We have real knowledge of objects only if our ideas conform to objects. (2b) Our ideas do not conform to objects. (3b) Hence, we do not have real knowledge of objects.

I am inclined to think that the second of these arguments, making use of weak conformity, is not what Berkeley is presenting in PHK 86. The reason is that when Berkeley states the companion argument in the Dialogues, his claim is that one would not be able to know that a resemblance obtains between ideas and objects. This is a form of strong

36

George S. Pappas

conformity, analogous to that given in PHK 86 and presented here in (2a). Another reason is that Berkeley really gives no reason to think that (2b) is true. Of course, the likeness principle is used by Berkeley to try to show that the alleged resemblance between some ideas and features of objects is mistaken. But resemblance is a special, narrow form of conformity, and showing that no ideas resemble features of objects does not by itself show that ideas fail to conform to objects. More to the point, Berkeley simply does not defend (2b); indeed, he pays it no attention, so far as I am aware. So we can focus attention on the conformity argument just stated in terms of strong conformity.5 Berkeley seems on safe ground with this argument because Locke certainly seems to accept the strong conformity claim in its first premise. Locke says: Where-ever we perceive the Agreement or Disagreement of any of our Ideas there is certain Knowledge; and where-ever we are sure those Ideas agree with the reality of Things, there is certain real Knowledge. (Essay IV.iv.18, 573)

There are, however, two good reasons to be doubtful of Locke’s commitment to strong conformity. First, even this last passage is not a clear endorsement of strong conformity. It would be if we interpreted the term ‘where-ever’ as expressing a bi-conditional, since after all the first premise (1a) of the conformity argument under investigation claims that knowledge of conformity is a necessary condition on real knowledge. But it is also reasonable to interpret the passage above as expressing merely a sufficient condition, to the effect that if we are sure that (i.e., know that) our ideas conform to objects, then we have real knowledge of objects. If this is the correct reading of this passage, then of course it does nothing to express endorsement of (1a). There is, moreover, textual evidence in favour of attributing to Locke at most an endorsement of weak conformity. He says at one point: ’Tis evident, the Mind knows not Things immediately, but only by the intervention of the Ideas it has of them. Our Knowledge therefore is real, only so far as there is a conformity between our Ideas and the reality of Things. (Essay IV.iv.3, 563)

Further, Locke draws a parallel between the reality of knowledge and real truth. He says:

Berkeley’s Assessment of Locke’s Epistemology

37

Though what has been said in the fore-going Chapter, to distinguish real from imaginary Knowledge, might suffice here, in answer to this Doubt, to distinguish real Truth from chimerical, or (if you please) barely nominal, they both depending on the same foundation. (Essay IV.v.8, 577)

Slightly further on in this passage Locke notes what this foundation is: And therefore Truth, as well as Knowledge, may well come under the distinction of Verbal and Real; that being only verbal Truth, wherein terms are joined according to the agreement or disagreement of the Ideas they stand for, without regarding whether our Ideas are such, as really have, or are capable of having an Existence in Nature. But then it is they contain real Truth, when these signs are joined, as our Ideas agree; and when our Ideas are such, as we know are capable of having an Existence in Nature. (Essay IV.v.8, 577–8)

In this passage the necessary condition for real truth is agreement of the ideas with things; and, as this is claimed to be parallel with the case of real knowledge, Locke’s point is that one has real knowledge only if one’s ideas conform to or agree with things in nature. This is just weak conformity rather than the strong variety attributed to him by Berkeley in the conformity argument. This argument, then, fails to engage Locke on his own terms. It is possible, of course, that Berkeley thinks that Locke is committed to strong conformity regardless of whether he actually and openly endorsed it. If so, his thought would doubtless be that some doctrines Locke surely accepts further commit Locke to strong conformity.6 The most likely candidate for this role in Locke would naturally be the representative realist theory of perception that Berkeley takes Locke to hold. On this way of thinking, Berkeley’s thought would be: (A) If the representative realist theory of perception is correct, then one has real knowledge of objects only if one knows that one’s ideas conform to the objects. If this statement is true, then Berkeley’s conformity argument would gain in force against Locke; for we have assumed, with Berkeley, that Locke does accept the representative realist theory. I can think of just one argument for statement (A). Suppose one thinks that the representative realist theory requires that any perceptual

38

George S. Pappas

knowledge there might be would in all cases be inferential. As we find below in the discussion of the inference argument, this is something Berkeley apparently accepts. If perceptual knowledge is inferential, then naturally one premise in such an inference would be a statement indicative of knowledge of currently perceived ideas. An inference from currently experienced ideas alone, however, will not yield either deductively or inductively any statement about an object.7 For any such inference to work, the inference basis would need to be supplemented with a statement concerning knowledge of the conformity between one’s ideas and objects. Hence, if representative realism requires that all perceptual knowledge of objects be inferential, then statement (A) is very likely correct. In that case, Locke would be committed to the first premise of the conformity argument, even though he did not outright endorse it. The flaw in this argument, it seems to me, comes in its first step: representative realism does not require that all perceptual knowledge of objects be inferential. Nor, indeed, does Locke think that perceptual knowledge of objects is inferential. Both of these points will be discussed below in connection with the inference argument. Their relevance here is that if they are correct, then the above argument for statement (A) breaks down. If that is so, then premise (1a) of the conformity argument is left unsupported. It is not accepted by Locke; and we have no good reason to hold that Locke is committed to the truth of that premise. Berkeley’s conformity argument can be dismissed as an ineffective criticism of Locke. 4. The Inference Argument The inference argument as Berkeley presents it can be understood either as a claim about what Locke actually held and the implications of that position or as a claim about what representative realism implies regardless of what Locke may have said. Here is the argument from the Principles: But though it were possible that solid, figured, moveable substances may exist without the mind, corresponding to the ideas we have of bodies, yet how is it possible for us to know this? Either we must know it by sense, or by reason. As for our senses, by them we have knowledge only of our sensations, ideas, or those things immediately perceived by sense, call them what you will: but they do not inform us that things exist without the

Berkeley’s Assessment of Locke’s Epistemology

39

mind, or unperceived, like to those which are perceived. This the materialists themselves acknowledge. It remains therefore that if we have any knowledge at all of external things, it must be by reason, inferring their existence from what is immediately perceived by sense. But what reason can induce us to believe the existence of bodies without the mind, from what we perceive, since the very patrons of matter themselves do not pretend, there is any necessary connection between them and our ideas? (PHK 18)

This compact passage presents an argument that has been exceedingly influential, having been accepted as sound in something like this form by legions of philosophers from Hume down to modern times.8 There is no reason to think that in asking whether we know about bodies by sense Berkeley is making the point that, by themselves, the senses are ‘cognitive silent.’ He acknowledges that we have knowledge of ideas by sense, so he certainly thinks that knowing by sense is a legitimate category. The most likely thing Berkeley would have in mind is that knowing by sense is knowledge gained without inference from some other piece of knowledge or evidence. The key first question then becomes: is Berkeley right in saying that the materialists acknowledge that we do not have non-inferential knowledge of bodies but only have such knowledge of ideas? No doubt Berkeley is right if he is thinking of Descartes and Malebranche, and maybe if he has Hobbes in mind.9 What shall we say of Locke? Is sensitive knowledge of bodies, by Locke’s lights, always inferential and based upon one’s knowledge of currently perceived ideas? Here we are taking Berkeley to be making a straightforward exegetical or interpretive point vis-à-vis Locke and wondering if he is correct. Certainly Berkeley has had a lot of fellow-travellers who have also supposed that sensitive knowledge in Locke is inferential.10 A text in support of this view is this: ’Tis evident, the Mind knows not Things immediately, but only by the intervention of the Ideas it has of them. Our Knowledge therefore is real, only so far as there is a conformity between our Ideas and the reality of Things. (Essay IV.iv.3, 563)11

This passage certainly looks as though the sort of intervention Locke has in mind is an epistemic one, and that would make sensitive knowledge in his estimation inferential. However, the passage actually

40

George S. Pappas

admits of another reading, according to which the intervention is really perceptual rather than epistemic. On that reading, Locke would be saying that objects are not themselves immediately perceived but are only perceived by means of perceptual intermediaries. If this is the point here made, then the passage is actually silent regarding whether sensitive knowledge is inferential or not. At one point in the Notebooks (NB 547) Berkeley claims that contrary to Locke, he can legitimately claim intuitive knowledge of physical objects. We might think, then, that by the term ‘knowledge by sense’ he means ‘intuitively know by sense.’ In that case Berkeley would be completely right to say that Locke, too, concedes that we do not know about objects by sense. Locke takes sensitive knowledge of objects to be quite distinct from intuitive knowledge of ideas and of the self. However, in PHK 18 Berkeley notes that if we do not have perceptual knowledge of objects by sense, then such knowledge would have to be gotten by inference. With the current understanding of ‘know by sense,’ though, this is a mistake. One can know by sense without essential reliance on inference even though one’s knowledge is not intuitive. This option is not closed off just by reading ‘know by sense’ as ‘have intuitive knowledge.’ Another passage where Locke seems to endorse the requirement of inference for sensitive knowledge is this: There can be nothing more Certain, than that the idea we receive from an external object is in our Minds; this is intuitive knowledge. But whether there be any thing more than barely that Idea in our Minds, whether we can thence certainly infer the existence of any thing without us, which corresponds to that Idea, is that, whereof some men think there may be a question made, because men have such Ideas in their Minds, when no such things exists, no such object affects their senses. (Essay IV.ii.14, 537)

Here Locke seems to be saying this: I immediately experience an idea. This event leads me to believe that there is something external corresponding to that idea and no doubt causing me to experience that idea; and so I infer that there is some specific object – a table perhaps – that fills this corresponding, causal role. Thus do I know that there is a table present. We can be sure that the experience of the idea plays the causal role of inducing belief in an object, at least much of the time. The question, though, is whether it also plays an epistemic role, for Locke, of being the

Berkeley’s Assessment of Locke’s Epistemology

41

inference basis for a warranted belief about the existence of an object. On this point, how Locke follows this passage is instructive. He says: But yet here, I think, we are provided with an evidence that puts us past doubting. (Essay IV.ii.14, 537)

He goes on to explain not what the warranted inference would be, but rather in effect why that inference is unnecessary. He speaks of being conscious that a perception in clear light is different from one in the night and that a waking perception is different from that in a dream. To the perhaps Cartesian objection that we cannot know with certainty that a fire is without us, Locke says that we certainly finding, that Pleasure or Pain follows upon the application of certain objects to us, whose Existence we perceive or dream that we perceive, by our Senses, this certainty is as great as our Happiness or Misery, beyond which we have no concernment to know, or to be. (Essay IV.ii.14, 537)

Locke then draws a conclusion from these observations, and it is instructive: I think, we may add to the two former sorts of Knowledge (intuitive and demonstrative) this also, of the existence of particular external Objects, by that perception and Consciousness we have of the actual entrance of Ideas from them. (Essay IV.ii.14, 537–8)

Here Locke’s point is that perception of the objects is sufficient to yield knowledge of the existence of the objects. A similar point is made in the next chapter: Secondly, that we can have no knowledge, farther than we can have Perception of that Agreement or Disagreement: which Perception being, 1. Either by Intuition, or the immediate comparing any two Ideas; or, 2. By Reason, examining the agreement of two Ideas, by the intervention of some others; or, 3. By Sensation, perceiving the Existence of particular things. (Essay IV.iii.2, 539)

So, on balance, the chief passages where Locke seems to insist that sensitive knowledge requires inference from ideas do not support such a

42

George S. Pappas

reading; and, in or near those very same passages, Locke stresses a contrary point, viz., that we gain knowledge of the existence of bodies by perception of those very bodies. Inference from ideas, of course, may in rare cases operate to support a belief in bodies, but not in general, and typically such inference is simply not needed.12 The upshot is that Berkeley is just (again) misunderstanding Locke’s position when he says in the inference argument of PHK 18 that we do not know bodies by sense and that the materialists all agree to this. Locke does not so agree, and he is certainly one of the core materialists Berkeley is discussing and trying to criticize. Now, just as in the case of the conformity argument, Berkeley may not strictly be making an exegetical or interpretive point about Locke when he gives the inference argument. He may, rather, be making a philosophical point to the effect that a representative realist theory of perception is stuck with no better than inferential knowledge of external bodies and that the needed inferences are at best precarious. In what follows I will consider the first of these points and try to show that the representative realist theory properly understood is perfectly consistent with non-inferential knowledge of bodies.13 5. Representative Realism and Knowledge of Objects The representative realist theory holds that objects are perceived mediately or indirectly by means of the immediate perception of one or more ideas that stand in a causal relation to the perceived object. On one standard interpretation of representative realism, indirect perception of an object is actually a judgment that some object is then present. More exactly, perception of the object is a judging that is occasioned by the immediate perception of some ideas. Veridical perception of an object thus has two elements: the sensory element confined to the experience of ideas and the judgmental element pertaining to the judgment that an object is present. Strictly speaking, it is the judgment that is the perception, provided that the judgment is caused by the event of immediately perceiving the relevant ideas. Merely immediately perceiving the relevant ideas, by itself, is not perception of an object, not even if those ideas stand in the right causal relation to the object. If this is the way in which we understand representative realism, then it is quite plausible to hold that the theory itself demands that sensitive knowledge of objects requires inference. The judgment that is the perception of an object, on this reading, in effect is an inference from the

Berkeley’s Assessment of Locke’s Epistemology

43

immediate perception of the ideas to a belief about an object. Thus Berkeley would be right – the representative realist theory requires that all sensitive knowledge of objects be inferential – and then the full force of the inference argument could be brought to bear.14 The standard interpretation is not the only way to understand representative realism. An alternative account is suggested by Locke: To ask, at what time a man first has any Ideas, is to ask, when he begins to perceive; having Ideas, and Perception being the same thing. (Essay II.i.9; emphasis in original)

Locke’s statement here seems to rule out the standard representative realist account, and this for two reasons: first, it makes no mention of judgment as an element of perception; and second, it identifies perception with the sensory element alone. That is, the sensory event, which is the event of immediately perceiving one or more ideas, just is the event of perceiving an object.15 We can think of this way of understanding perception as a constitution theory, because it identifies the event of perceiving an object with the event of experiencing ideas. More exactly, it is a representative realist theory nonetheless, because perception of objects is still dependent on and mediated by the immediate perception of ideas. The event of perceiving an object consists in nothing more than the event of immediately perceiving some ideas. Of course, the converse does not hold. One can experience some ideas without thereby perceiving an object, for example in a case of hallucination. Thus, the experience of some ideas will count as a perception of an object only when those ideas are causally connected in the right way with that object.16 To help situate this version of representative realism we can consider a version of direct realism once defended by Roderick Chisholm. Chisholm advocated the ‘adverbial’ account of the sensory element in perception as opposed to anything like ideas or sensa. On an adverbial theory, adverbial events of sensing are not themselves perceived entities, even though at least one such adverbial event is an ingredient in each perceptual experience. It is the fact that adverbial events are not perceived that allows Chisholm’s theory to be a species of direct realism. As in the constitution version of representative realism, Chisholm identifies the event of perceiving an object with an adverbial event of sensing in a certain way. This is done in two steps. First, Chisholm defines the non-comparative sense of ‘appears’ in terms of sensing:

44

George S. Pappas X appears _____ to S means that S senses _____ with respect to X.17

Here the term ‘X’ stands in for some object, and the blank following the term ‘appears’ is filled in with some term for a sensory quality such as blue colour. The blank following ‘senses’ is filled in with the relevant adverb such as ‘blue-ly,’ and the phrase ‘with respect to’ is a causal one, indicating that the adverbial event is caused by the object X. Thus, if X is a blue door, the above general expressions would be instantiated to ‘The door appears blue to S means that S senses blue-ly with respect to the door.’ Further, the door would be said to be the proper causal stimulus, so that the causal chain going from the door to the event of S sensing blue-ly does not proceed through any deviant route. The second step deals with what Chisholm calls the ‘non-propositional sense of “perceives,”’ reflected in expressions like ‘S perceives O,’ where the perception verb takes a grammatical direct object as complement. He says: We may now define the simplest of the non-propositional senses of ‘perceive’: ‘S perceives X’ means: X appears in some way to S.18

Since the relevant notion of appearing is itself defined in terms of sensing, as noted above, then the identity we find in Chisholm for nonpropositional perception boils down to this: S perceives X means: S senses _____ with respect to X where again the blank following the term ‘senses’ is filled in with a special adverbial modifier. In Chisholm, then, perception of objects is wholly constituted by the sensory component, understood as caused in a certain manner by the perceived object. The sensory component, in turn, is thought of as an adverbial event of sensing, one that itself includes no element of inference or judgment. In structure, Chisholm’s direct realist theory is parallel to the constitution version of the representative realist theory. When one perceives a blue door, this event of perceiving the door is identified in Chisholm’s theory with the event of sensing blue-ly with respect to the door, and it is identified in the representative realist theory with the event of experiencing some ideas that stand in the ‘right’ causal relation to the door. The only key difference between these two theories is that in Chisholm’s theory adverbial events of sensing are not perceived, thereby allowing his account to qualify as a version of direct realism.

Berkeley’s Assessment of Locke’s Epistemology

45

To see that Chisholm’s account of the perception of objects is also compatible with our having non-inferential knowledge of perceived objects, we need only note how the story would be told with a reliable process theory of knowledge. (Here I speak of a reliability theory of knowledge rather than of warrant or justification, as is usually the case.) The perceptual event of sensing blue-ly with respect to the door typically would cause in S the belief that there is a door before her. This counts as knowledge for S provided that the belief is true and provided that the process type which takes the causal relation between the sensing event and the belief as a token is reliable. Such a belief-inducing process type will qualify as reliable when more of its belief outputs turn out to be true than otherwise. If this is what it is for a person to acquire perceptual knowledge, then it is plain that Chisholm’s version of direct realism is perfectly consistent with non-inferential perceptual knowledge of objects. Exactly the same holds for the constitution version of the representative realist theory. That theory, too, is consistent with non-inferential knowledge of objects because it is consistent with a reliable process account of perceptual knowledge. We can now return to the question of Berkeley’s inference argument. We found that, understood as aimed at doctrines Locke actually held, the argument is a failure since Locke did not maintain that sensitive knowledge of objects actually is inferential. But Berkeley may have meant that some feature of representative realism itself demands that all sensitive knowledge be inferential. On one way of understanding that theory, Berkeley would be right. If the representative realist theory is one which includes an element of judgment in all perception of objects, then it is quite plausible to suppose that sensitive knowledge of objects is inferential. However, that is just one version of representative realism. Another version, the constitution theory as elaborated above, does not require that sensitive knowledge of objects be inferential. On the contrary, such a theory is consistent with non-inferential sensitive knowledge, so long as we interpret the acquisition of sensitive knowledge along the lines of the reliable process theory. So Berkeley is ultimately incorrect on this point: the general category, representative realist theory of perception, does not entail that sensitive knowledge of objects is inferential because not all instances of this general category have this entailment. Hence, even when we interpret Berkeley’s inference argument to be making not an exegetical or interpretive point but rather a philosophical one, aimed generally at representative realism, the inference argument must be counted a failure.19

46

George S. Pappas

Notes I am indebted to Martha Bolton for valuable discussion of reliability accounts of knowledge in Locke and Berkeley. I am further much indebted to Margaret Atherton, Georges Dicker, and Richard Glauser for very insightful comments given when this paper was presented at the University of Rennes conference on Berkeley, October 2003. 1 Thomas Reid, for example, defended the thesis that the ‘ideal theory’ – that is, the theory of perception we would think of as indirect realism – leads to scepticism about the external world. Reid’s general indictment of the ideal theory makes up what has come to be called the ‘veil of perception’ problem, which has often been thought to refute indirect realism. 2 The direct realist reading of Locke may be found in E.J. Lowe, Locke on Human Understanding (London: Routledge, 1995); and John Yolton, Locke and the Compass of Human Understanding (Cambridge: Cambridge University Press, 1970). That Locke’s ideas of sensation are not objects of immediate perception is defended in Vere Chappell, ‘Locke’s Theory of Ideas,’ in The Cambridge Companion to Locke, ed. Vere Chappell (Cambridge: Cambridge University Press, 1994), 26–55. 3 This argument involving the likeness principle may be thought of as a third epistemic argument against Locke. For Berkeley uses it, not to conclude that the representative realist theory is false (because its representation claim is not true), but rather that the theory implies that there are no material objects. Hence, he says, Hylas’s principles lead to scepticism about objects, no doubt on the grounds that one cannot have knowledge of objects if there are no such things. This is how Berkeley concludes the argument at the end of the first Dialogue: ‘You are therefore by your principles forced to deny the reality of sensible things, since you make it to consist in an absolute existence exterior to the mind. That is to say, you are a downright sceptic’ (DHP 206). I do not further pursue this argument here. For a brief discussion of it, see my Berkeley’s Thought (Ithaca, NY: Cornell University Press, 2000), ch. 9. 4 At one point Locke speaks of being certain that one has real knowledge, which is of the form S is certain that S knows that p (see Essay IV.iv.6, 565). Perhaps this led Berkeley into stating his conclusion of the conformity argument in a similar fashion. 5 One may think that by the term ‘conform’ in premise (2b) Berkeley really means ‘resemble,’ so that the argument using weak conformity needs greater scrutiny. After all, so construed, we are sure that Berkeley argues

Berkeley’s Assessment of Locke’s Epistemology

47

over and over that (2b) is true. Ideas do not resemble objects, he thinks, because an idea can only resemble another idea. However, such an understanding of this argument also requires changing (1b) so that it, too, uses the term ‘resemble,’ yielding ‘We have real knowledge of objects only if all of our ideas resemble objects.’ Doing this, however, creates a premise that Locke did not hold, for he held that only some simple ideas resemble features of objects. Another defence of Berkeley, due to Richard Glauser, would be to say that he certainly argues for something that implies (2b), or at least does so in conjunction with a truism. The truism is that an idea conforms with an object only if the idea conforms with one or more of the object’s qualities. And surely Berkeley argues, for instance at PHK 14, that there are no mindindependent qualities with which an idea might conform. These two points jointly imply (2b), even in a wide sense of conformity. Further, since Locke accepts (2a), he could not avoid the sceptical result of (3b). The force of this argument depends on the cogency of Berkeley’s argument in PHK 14 for the thesis that there are no mind-independent qualities. There he appeals to perceptual relativity arguments and notes that if they work to show that secondary qualities are not mind-independent, then the same sorts of arguments will show that neither are the primary qualities. However, in PHK 15 Berkeley concedes that the relativity arguments support only a weaker claim, namely that we do not know by sense what an object’s true qualities are. So, this attempt to shore up Berkeley’s use of a weak conformity argument runs aground. 6 It is just barely possible, though hardly likely, that Berkeley thinks Locke is committed to strong conformity either because (1) strong conformity is an epistemic truth to which all theorists are beholden or because (2) strong conformity is a truth that must be accepted by all who endorse the theory of ideas. I do not discuss these possibilities both because there is no textual basis for thinking Berkeley held such views and also because both of these possibilities seem so plainly false and therefore would do nothing to advance Berkeley’s conformity argument. 7 This claim is denied by some Locke commentators, for example, J.L. Mackie, Problems from Locke (Oxford: Clarendon Press, 1980), ch. 2, esp. 62– 9. Mackie tries to show that a broadly explanatory argument will do the trick, and of course this would be an inductive argument. I will not here examine Mackie’s arguments, for, as will emerge, I have another reason to be suspicious of this argument for statement (A). Berkeley anticipates an argument like Mackie’s in PHK 19. 8 See my note 1 above, where the connection to the veil of perception is raised.

48

George S. Pappas

9 Here the importance of being clear about Berkeley’s intended targets, alluded to earlier, becomes pressing. 10 One recent fellow-traveller on this point is Robert Meyers, ‘Was Locke an Empiricist?’ Locke Studies 1 (2001): 63–85. Mackie, in Problems from Locke, also reads Locke this way and then goes on to try to show that the needed inferences can be successfully made. See my note 7. 11 Note that here Locke affirms only weak conformity as a necessary condition on real knowledge. 12 In Essay IV.vi.11, 586–7, Locke takes up concurrent reasons in support of one’s belief about a presently perceived object, and concurrent reasons certainly are inferential. I take Locke’s point there to be that concurrent reasons are needed to establish the general reliability of the senses and not to be an endorsement of the thesis that sensitive knowledge of objects is inferential and based upon concurrent reasons. On this point I am indebted to some correspondence with Georges Dicker. 13 I briefly consider whether these inferences really are precarious in ‘Degrees of Certainty in Locke’s Account of Knowledge’ (unpublished). See further note 7 above. 14 If Locke held representative realism in this form, and also held that sensitive knowledge of objects is typically non-inferential, as I have argued above, then he would be holding inconsistent views. 15 I take the term ‘perceive’ in the phrase ‘when he begins to perceive’ to be referring to perception of objects, a reading that is essential for the alternative account to be here elaborated. At Essay II.ix.3, 143, Locke makes a similar claim, there identifying actual perception of pain with having ideas of pain. This is something like an identity of events: the event of perceiving the pain is the very same event as having ideas of pain. 16 It is a delicate question just how the causal relation is to be understood. Presumably the causal chain starts at the object and ends with the event of experiencing some ideas. Another presumption doubtless would be that the causal chain not be deviant. Locke famously thought that some ideas resemble real qualities of objects, but it is not clear that resemblance would play any role in the perception of an object on the identity construal of representative realism. 17 Roderick Chisholm, Perceiving (Ithaca, NY: Cornell University Press, 1957), 125. 18 Ibid., 149. 19 I have not argued in this paper that the constitution version of representative realism is actually Locke’s theory, as that was beside the philosophical point of whether representative realism requires inferential knowledge of

Berkeley’s Assessment of Locke’s Epistemology

49

objects. But I do think that a plausible case can be made that something like the constitution theory is Locke’s own theory of perception. I discuss the constitution version of representative realism in Locke more fully in ‘On Some Philosophical Accounts of Perception,’ in Philosophy in America at the Turn of the Century, APA Centennial Supplement to the Journal of Philosophical Research (Charlottesville, VA: Philosophy Documentation Center, 2003), 71–82; and in ‘Locke’s Account of Sensitive Knowledge,’ in Contemporary Topics in Philosophy, ed. M. O’Rourke (Cambridge, MA: MIT Press, forthcoming). In the former article the constitution theory is referred to as an identity theory. Interestingly, though Berkeley seems to have completely missed the possibility of connecting representative realism with non-inferential knowledge of objects, he himself seems to have endorsed something like a reliability account of knowledge of objects. This line of thought is suggested in PHK 31, where Berkeley says that the general stability of ‘laws of nature’ holding between experiences of ideas of certain types underwrites our knowledge of objects. I think Locke also adopted a reliability account of sensitive knowledge; this point is defended in my ‘Locke’s Account of Sensitive Knowledge.’ The general question of whether Locke’s theory of perception leads to scepticism requires that we be clear about just what Locke’s theory of perception is. I discuss this as well in ‘Locke’s Account of Sensitive Knowledge’ and in ‘Locke’s Account of Perception,’ unpublished.

The Problem of the Unity of a Physical Object in Berkeley ri ch ar d glauser

Although Berkeley says that ‘strictly speaking ... we do not see the same object that we feel; neither is the same object perceived by the microscope, which was by the naked eye’ (DHP 245), he is generally happy to call a physical object a ‘collection’ (PHK 1), a ‘combination’ (PHK 4, DHP 175), or a ‘congeries’ (DHP 249) of sensible qualities or ideas. He also calls it a ‘complex or compounded’ idea (IN 9). As such, a physical object comprises ideas of several senses.1 Within each sense modality the ideas that constitute a physical object belong to numerous, highly complex, lawfully regulated sequences. Because the sensible ideas proper to each sense are heterogeneous, a Berkeleian physical object comprises qualities that share no common space and, thus, have no common spatial location. Tangible ideas, however, play a central role within such a complex and diverse combination: if one speaks of the whereabouts of a physical object, one must refer to a certain solid (itself complex) in tangible space. As several commentators have rightly observed, the sensible ideas that constitute a physical object are so numerous that during its lifetime a finite mind never immediately perceives more than a small part of some of them,2 and none at all of a great many others, even of those with which we are familiar. Of course, we may immediately perceive a greater proportion of the sensible ideas that constitute a cherry or a pea than of those that make up an oak tree, but a much smaller proportion of those that constitute Mt Rushmore.3 One should distinguish the question of the unity of a physical object from the question of its individuation. The question of individuation, as I understand it, is the following: Given all of the sensible ideas (of various senses) that one might immediately perceive, what is it about

The Problem of the Unity of a Physical Object in Berkeley

51

them that distinguishes them into numerically distinct physical objects, whether of the same or of different sorts? Why is there not just one combination, a unique physical object composed of all possible sensible ideas? The question of the unity of a physical object takes things conversely: Given certain sensible ideas, what is it about them in virtue of which they are combined so as to make one physical object of whatever sort, rather than two, or three, and so on? One might expect both questions to be discussed together.4 To do so, however, would exceed the limits of this paper, as would even a more detailed discussion of the question of unity. Let us therefore focus on only one aspect of the question of unity: Are the combinations that Berkeley calls physical objects made by finite minds or by God?5 Several authors believe that in some sense finite minds make physical objects. Of course, they do not hold that finite minds produce sensible ideas; these are caused by God. Nor do they say that the regularity, order, and lawfulness of the sequences of sensible ideas that constitute a physical object are produced by finite minds. Rather, they hold that finite minds make physical objects in the sense that, by means of one or more epistemic activities, they combine together the sensible ideas that constitute such objects. Of course, to the extent that the finite mind’s combining is based on experience, it is not done arbitrarily. Nevertheless, it is held that the combining is done by a finite mind, and the result of a finite mind’s (non-arbitrary) combining is a physical object. One reads, for instance, that ‘one of [Berkeley’s] strangest and most astounding principles, enunciated in the Essay on Vision and carried consistently throughout his works ... is the assertion that in some sense, human perceivers make objects.’6 As worded by a second critic, ‘semi-permanent items [‘things like coaches’] are fictions. These are constructed by us out of sense data, because otherwise “the endless number of confusion of names would render language impracticable.”’7 A third says: ‘So-called physical objects, he [Berkeley] tells us in [P 1], are “collections of ideas.” These collections are the product of human mental activity.’8 According to a fourth, ‘Berkeley seems always to have believed that the bundling of sense-specific ideas into individual substances [sic] ... is “the workmanship of the mind.”’9 A fifth has it that by this process of collecting we convert the proper objects of perception into ... common objects or, in other words, bodies ... Congeries, proper objects are presented or ‘given’ to us, but collections, common objects, are

52

Richard Glauser made, they are the products of the mind’s imaginative activities ... [The upshot is that] we arbitrarily create bodies by collecting together these proper objects by means of our (usually unconscious) activities ... we literally make ‘the furniture of the world.’10

A sixth says: ‘the mind ... not only receives but organizes ideas to form complex ordinary physical objects,’ the consequence being that ‘Berkeleian real objects are merely epistemic creations: they are things formed by judgment.’11 Although the authors have textual evidence to support their interpretation, the strongest being two passages in the Dialogues (245, 249), which will be discussed at some length in due course (section 4), I wish to challenge their reading. 1. Are Physical Objects Made by Finite Minds? Preliminary Remarks, and the Outline of an Alternative Reading Let us first explain a general reason for which their interpretation is difficult to accept. It seems to jeopardize needlessly Berkeley’s intention of making his immaterialist conception of bodies as compatible as possible with commonsense realism. Of course, as many critics have often noticed, Berkeley is in any case at quite a remove from commonsense realism in holding that physical objects exist only if perceived, and also for several other reasons. But he would be at a much further remove if he also held that individual physical objects are literally made by finite minds in the sense mentioned. Surely the claim that finite minds make cherries and oak trees by their epistemic activities is no part of the commonsense realism with which Berkeley wishes – as far as possible – to align immaterialism in such passages as the following: It will be urged that thus much at least is true, to wit, that we take away all corporeal substances. To this my answer is, that if the word substance be taken in the vulgar sense, for a combination of sensible qualities, such as extension, solidity, weight, and the like – this we cannot be accused of taking away. (PHK 37; my italics) Take away this material substance, about the identity whereof all the dispute is, and mean by body what every plain ordinary person means by that word, to wit, that which is immediately seen and felt, which is only a combination of sensible qualities, or ideas: and then their most unanswerable objections come to nothing. (PHK 95; my italics)

The Problem of the Unity of a Physical Object in Berkeley

53

philonous: With all my heart: retain the word matter, and apply it to the objects of sense, if you please; provided you do not attribute to them any subsistence distinct from their being perceived. I shall never quarrel with you for an expression ... hylas: Well but, Philonous, since I am content to give up the notion of an unthinking substance exterior to the mind, I think you ought not to deny me the privilege of using the word matter as I please, and annexing it to a collection of sensible qualities subsisting only in the mind ... (my italics) philonous: I do not pretend to be a setter-up of new notions. My endeavors tend only to unite and place in a clearer light that truth, which was before shared between the vulgar and the philosophers: the former being of opinion, that those things they immediately perceive are the real things; and the latter, that the things immediately perceived are ideas, which exist only in the mind. Which two notions put together, do in effect constitute the substance of what I advance. (DHP 261–2)

Berkeley says that the ‘vulgar sense’ of the expression ‘corporeal substance’ and ‘what every plain ordinary person means’ by the word ‘body’ is one and the same: ‘a combination of sensible qualities.’ He fully accepts this common usage. He goes as far as to say that one might even retain the word ‘matter’ in order to name ‘a collection of sensible qualities subsisting only in the mind.’ The only restriction Berkeley places on the use of such expressions is that one keep in mind that the combinations and their constituent qualities have no ‘subsistence distinct from their being perceived.’ He does not add a further restriction to the effect that such combinations are made by finite minds, which would be quite another thing. Yet, if he does think so, why does he not say so? One might reply that, according to Berkeley himself, a philosopher may ‘think with the learned, and speak with the vulgar’ (PHK 51). But the context of this often-quoted phrase indicates that discrepancy between learned thought and ordinary speech can be tolerated only when hardly avoidable and that what a philosopher does think must be made clear from the outset. To think with the learned and speak with the vulgar is not to engage in deceit. Furthermore, in the three texts just quoted, in particular the passage from the Dialogues, Berkeley is precisely trying to explain how learned, philosophical thought is to be accommodated to ordinary manners of speech. Yet he would be deceiving both the vulgar and the learned if, when writing them, he thought, without saying there, that physical objects are made by finite minds.

54

Richard Glauser

Of course, the question of just how far one can take Berkeley’s immaterialism to be compatible with commonsense realism is a controversial issue. Quite a few of his tenets – not only the esse est percipi principle – prevent him from fully endorsing commonsense realism. Furthermore, there has long been justified disagreement among scholars as to how many such tenets there are, and even as to what some of them are exactly.12 For all that, though, one should attempt to arrive at an interpretation that conforms as closely as possible to an author’s declared intentions. Hence, if it is possible to propose a reading of Berkeley that does not commit him to holding that finite minds make physical objects, it is to be preferred. I wish to show that it is possible. The reading to be defended has its own difficulties to face, the most important being the two passages from the Dialogues alluded to above. Now, there are passages that directly imply that combinations of sensible ideas are made by God, not by finite minds: Ideas are not anyhow and at random produced ... there are also several combinations of them, made in a very regular and artificial manner, which seem like so many instruments in the hand of Nature, that being hid as it were behind the scenes, have a secret operation in producing those appearances which are seen on the theatre of the world. (PHK 64; my italics) The reason why ideas are formed into machines, that is, artificial and regular combinations, is the same with that for combining letters into words. That a few original ideas may be made to signify a great number of effects and actions, it is necessary they be variously combined together: and to the end their use be permanent and universal, these combinations must be made by rule and with wise contrivance. (PHK 65; my italics)

In PHK 64 the combinations are said to be ‘like so many instruments in the hand of Nature.’ And there is no doubt the combinations are physical objects, for Berkeley calls them natural ‘machines.’ In PHK 65, the ‘machines’ are to words what their constituent ‘few original ideas’ are to letters. The combining of ‘letters’ (constituent ideas) into ‘words’ (combinations) is done ‘by rule and wise contrivance,’ the rules being those instituted by God and the ‘wise contrivance’ his own. Thus, the combining from which combinations result is the work of God, not of finite minds. In PHK 66 Berkeley goes on to say that the combinations (‘words’) are natural ‘signs instituted by the Author of Nature,’ the interpretation and understanding of which ‘ought to be the employ-

The Problem of the Unity of a Physical Object in Berkeley

55

ment of the natural philosopher.’ As the context makes clear, the kind of ‘sign to signified thing’ relations under discussion are those between what we take to be empirical causes and effects; they are relations holding between physical objects. Now, finite minds do not institute the language of the Author of Nature; this is done by God. Thus, they do not make natural signs; they gradually discover and interpret them. In passages such as these it should be clear that physical objects, ‘machines,’ are not made by finite minds.13 Other texts do not directly imply that physical objects are made by God but at least suggest that they are not made by finite minds: And as several of these [ideas actually imprinted on the senses] are observed to accompany each other, they come to be marked by one name, and so to be reputed as one thing. Thus, for example, a certain colour, taste, smell, figure and consistence having been observed to go together, are accounted one distinct thing, signified by the name apple. Other collections of ideas constitute a stone, a tree, a brook, and the like sensible things. (PHK 1; my italics) And it seems no less evident that the various sensations or ideas imprinted on the sense, however blended or combined together (that is, whatever objects they compose) cannot exist otherwise than in a mind perceiving them. (PHK 3; my italics) But, say you, it sounds very harsh to say we eat and drink ideas, and are clothed with ideas. I acknowledge it does so, the word idea not being used in common discourse to signify the several combinations of sensible qualities, which are called things. (PHK 38; my italics)

Passages such as these do not say in so many words that the combinations are made by God. On the other hand, their being made by finite minds seems to be ruled out. For, if some sensible qualities are ‘observed to go together’ by a finite mind, then the fact of their being combined in such a way as ‘to go together’ is not the result of a finite mind’s activity. In any case, given Berkeley’s declared intention of aligning his immaterialist conception of bodies as closely as possible with commonsense realism, and given, too, the passages from PHK 64–6 mentioned above, there is a strong presumption that the three texts just quoted have been carefully worded so as not to suggest what is not stated, that is, so as not to imply that physical bodies are made by finite minds. For the time

56

Richard Glauser

being, though, this is no more than a strong presumption. To go further we must examine the passages that seem to lend support to the alternative reading. Before doing so, it may be useful here to give a brief outline of the reading to be defended. A combination of sensible ideas that constitutes a physical object is made by God and gradually discovered by finite minds through observation and experience. Such a combination is nothing over and above a multiplicity of regularly ordered series of sense-specific ideas connected by certain relations. A combination of sensible ideas is made by God in two complementary senses. God produces the sensible ideas belonging to the combination, and He combines them together by the mere fact of His producing them along with the relations that hold between them. As a finite mind progressively discovers the relations holding between the sensible ideas, it gives their whole a name and considers it one thing. To consider a combination of sensible ideas as one thing is neither to confer unity upon it in a literal sense, nor to literally combine its constituent ideas together; it is merely to judge that it is one, for ‘in case every variation was thought sufficient to constitute a new kind or individual, the endless number or confusion of names would render language impracticable’ (DHP 245). On the other hand, there are combinations of ideas that are made by finite minds. But, strictly speaking, they are combinations of ideas of memory and imagination that represent sensible ideas proper. Such combinations are not physical objects but their representations in the imagination. So, although finite minds do not make physical objects (by means of their epistemic activities), they do make representations of them in the imagination, often through the process of mediate perception.14 A combination of ideas (or a complex idea) of the imagination is made by the finite mind in two complementary senses. Each idea of the imagination that belongs to the combination is caused, or produced, by the finite mind, and the activity of combining is performed by it as well. It would seem that there is no positive theory of the unity, as such, of a physical object in Berkeley. But there is a theory of the objective foundation upon which finite minds base the judgment that this or that multiplicity of sensible ideas constitutes one physical object, and upon which, too, finite minds produce combinations of ideas in the imagination that represent physical objects. The interpretation proposed is neutral as to whether physical objects can be both immediately and mediately perceived (as some commentators hold), or whether they can only be mediately perceived (as others believe).15

The Problem of the Unity of a Physical Object in Berkeley

57

It is important throughout this discussion to keep to Berkeley’s own conception of a sensible idea, namely an idea, or quality, that can be immediately perceived by sense, where ‘immediately’ excludes both an intervening idea and an inference.16 Thus, a sensible idea is not an idea of the imagination or of memory; their ideas are not perceivable by sense, although they have sensory content. The following section rehearses some points concerning the difference between ideas of sense and of imagination in Berkeley. Next, because Berkeley often connects (a) the claim that what we consider a unit when counting physical objects is a combination of ideas, with (b) the claim that unity and number are ‘creatures of the mind,’ section 3 briefly examines (b) and the reasons for which Berkeley holds it. This leads to the question of the sortal-dependence of unity and number as used to count physical objects and to a discussion of Berkeley’s claim that sorts and kinds are made by finite minds. Section 4 attempts to defend and to develop the interpretation outlined above and discusses the texts that seem to oppose it. 2. Ideas of Sense and Ideas of Imagination Ideas of sense are ‘more strong, lively, and distinct than those of the imagination; they likewise have a steadiness, order and coherence and are not excited at random, as those which are the effects of human wills often are’ (PHK 30). Sensible ideas as immediately perceived by a finite mind do not last very long: they are in ‘a continual succession ... some are anew excited, others are changed or totally disappear’ (PHK 26). According to Siris, they are ‘fluent and changing’ (Siris 292), ‘in a perpetual flux or succession, ever differing and various’ (Siris 347), ‘always flowing’ (Siris 349). Ideas of memory and imagination,17 on the other hand, must be related to mental dispositions in such a way that they can recur on occasion or can be recalled at will: ‘I find I can excite ideas in my mind at pleasure, and vary and shift the scene as oft as I think fit. It is no more than willing, and straightway this or that idea arises in my fancy’ (PHK 29). In other words, an idea of memory or imagination presupposes a dispositional stability that a sensible idea lacks. Whereas the latter is present to a finite mind only as long as it is immediately perceived, the former can be recurrently called to mind whenever useful. Ideas of the imagination are generally, though not always, copies of sensible ideas which they represent by means of some resemblance. Berkeley says as much in three entries of the Notebooks.18 The same

58

Richard Glauser

appears in PHK 33: whereas ‘ideas imprinted on the senses by the Author of Nature are called real things ... those excited in the imagination being less regular, vivid and constant, are more properly termed ideas, or images of things, which they copy and represent.’ Of course, resemblance varies in degrees; it can be more or less. For this reason Berkeley puts two restrictions on the claim that ideas of the imagination resemble those of sense. One restriction occurs in a letter of 1709 to Samuel Molyneux. Berkeley says that ideas of the imagination can represent sensible ideas without, strictly speaking, being images of them. This happens when sensible ideas are so numerous and complex that the imagination, due to its limited powers, is unable to copy them accurately (see W 8: 25). Another restriction appears in PHK 1: ideas of the imagination may be divided and compounded by that faculty to such an extent that the resulting ideas of imagination barely represent the original ideas of sense. But notice that if they come to ‘barely’ represent ideas of sense, it is presumably because, due to much fanciful compounding and dividing, the resulting ideas of the imagination come to lose much of their original resemblance: the lesser the resemblance, the lesser the representative function. Thus, Berkeley’s main position seems to be that, as long as sensible ideas are not too numerous to be accurately copied by the imagination, and before the imagination goes on to fancifully divide and compound its ideas at will, ideas of the imagination basically, and most often, represent sensible things by means of a resemblance. We have not yet finished with the first restriction that Berkeley places on the imagination’s representational capacity: the restriction that when sensible ideas are very numerous and complex the imagination, due to its limited powers, is unable to copy them accurately. A Berkeleian physical object, a combination of sensible ideas, is enormously complex. So, it is only natural to suppose that when a finite mind produces a combination of ideas of memory or imagination that represents a physical object, the resulting combination within the imagination does not contain an idea of imagination for each and every sensible idea included in the physical object. In most cases it will contain far less. This means that there is presumably some process of selection, based on experience, determining which ideas of sense, among all those that make up a physical object, are to be represented by a corresponding idea of the imagination within the combination that the mind produces. We will soon see that Berkeley clearly suggests as much. It is also important to note that Berkeley virtually acknowledges that, when writing, he occasionally blurs his own distinction between ideas

The Problem of the Unity of a Physical Object in Berkeley

59

of sense and of imagination, sometimes speaking of the latter as though they were really sensible. In an important letter of 1711 to Jean Leclerc, editor of the Bibliotheque choisie, Berkeley comments on the NTV and makes the point that ‘tangible’ can be taken either in a narrow or in a broad sense.19 The broad sense of the expression ‘tangible idea’ includes certain ideas formed by the mind: ‘ideas a mente ad similitudinem tactu perceptarum fictas,’ that is, ideas of the imagination ‘per operationem mentis formentur.’ So much might have been gathered from the NTV itself: ‘Note that when I speak of tangible ideas, I take the word idea for any the immediate object of sense or understanding, in which large signification it is commonly used by the moderns’ (NTV 45; my italics). Much the same point is made later, in TVV 51: ‘Figures and motions which cannot be actually felt by us, but only imagined, may nevertheless be esteemed tangible ideas, forasmuch as they are of the same kind with the objects of touch, and as the imagination drew them from that sense.’ The passages that seem to favour a reading according to which physical objects are made by finite minds often relate our activity of combining ideas to the claim that we arbitrarily consider a multiplicity of ideas as a unit when counting objects of a certain sort. So, before I directly address these passages, something must be said about unity and number in Berkeley. 3. Unity and Number, Creatures of the Mind, Sortal-Dependence Berkeley vigorously reacts against two claims made by Locke concerning unity and number. The first is that the idea of unity is received by all five senses and by reflection.20 The second is that unity and number are primary qualities.21 As regards Locke’s first claim, Berkeley rejects it by holding that unity and number are not ideas; nor are there any ideas of them. In PHK 13 he says that unity is an abstract idea, and PHK 120 makes it clear that it is therefore not an idea at all: Unity in abstract we have considered before in Sect. 13, from which and what has been said in the Introduction, it plainly follows there is not any such idea. But number being defined a collection of units, we may conclude that, if there be no such thing as unity or unit in abstract, there are no ideas of number denoted by the numerical names and figures.

Thus, if unity is not an idea, and if there is no idea of unity, it follows that given a physical object – a combination of sensible ideas – there is no idea of oneness belonging to the combination. This is one of the rea-

60

Richard Glauser

sons for which ‘number ... is nothing fixed and settled really existing in things themselves’ (NTV 109).22 As to Locke’s second claim – that unity and number are primary qualities – Berkeley argues in PHK 12 that one could prove that ‘number is entirely the creature of the mind even though the other qualities be allowed to exist without.’ In other words, even if the other qualities held by Locke to be primary qualities – ‘extension, figure, motion, rest, solidity or impenetrability’ (PHK 9) – were ‘allowed to exist without,’ that is, in matter, it would still be ‘evident’ that number is not one of them. On this precise point one can say that, if (per impossibile) matter did exist, Berkeley’s position would be closer to that of Descartes than to that of Locke. Descartes held that ‘we should ... think of [order and number] simply as modes under which we consider the things in question’ (PP I.55, 211). And that seems to be what Berkeley is saying in the first sentence of PHK 12. As to the expression ‘creatures of the mind,’ it seems to be a liberal translation of the Latin ‘entia rationis.’ Spinoza, for instance, held that number is no more than an ‘ens rationis,’ a ‘being of reason,’ a merely mental entity.23 As to the kind of ‘creature[s] of the mind’ that unity and number are, Berkeley seems to waver between three different positions. (1) In the Notebooks he says that number consists in relations: ‘Unite no simple Idea. I have no Idea meerly answering the word one. all Number consists in Relations’ (NB 545). This follows from Berkeley’s definition of number as ‘a collection of units.’ (2) In other entries of the Notebooks, in the Principles and in Alciphron, unity and numbers are said to be conventional signs.24 (3) In Siris Berkeley says that number is an act of the mind.25 The three statements are not necessarily incompatible. It is possible to take them as expressing three aspects of a coherent theory. Developing such a reading, however, is beyond the scope of this paper. One point should be stressed, though: as long as unity and number ‘in the application are referred to things,’ they have some objective foundation in reality inasmuch as ‘the signs, indeed, do in their use imply relations or proportions of things’ (Alc VII.12, 305). This is borne out, for example, by NTV 142, where Berkeley claims as an objective fact that a visible square has ‘four distinct equal parts corresponding to the four sides of the tangible square,’ and ‘four other distinct and equal parts whereby to denote the four equal angles of the tangible square.’ If it were not an objective fact, a visible square would not be ‘fitter than the visible circle to represent the tangible square’ (my italics).26 Thus, although there is no idea or quality of oneness in a combination of

The Problem of the Unity of a Physical Object in Berkeley

61

ideas, our considering the combination as one nevertheless rests on some objective foundation. This is important to keep in mind when reading the following passage: It ought to be considered that number (however some may reckon it amongst the primary qualities) is nothing fixed and settled, really existing in things themselves. It is entirely the creature of the mind, considering an idea by itself, or any combination of ideas to which it gives one name, and so makes it pass for an unit. According as the mind variously combines its ideas the unit varies: and as the unit so the number, which is only a collection of units, doth also vary. We call a window one, a chimney one, and yet a house in which there are many windows and many chimneys hath an equal right to be called one, and many houses go to the making of one city. In these and the like instances it is evident the unit constantly relates to the particular draughts the mind makes of its ideas, to which it affixes names, and wherein it includes more or less as best suits its own ends and purposes. Whatever, therefore, the mind considers as one, that is an unit. Every combination of ideas is considered as one thing by the mind, and in token thereof is marked by one name. Now, this naming and combining together of ideas is perfectly arbitrary, and done by the mind in such sort as experience shows it to be most convenient: without which our ideas had never been collected into such sundry distinct combinations as they now are. (NTV 109; my italics)

Even conventional physical objects such as a window, a chimney, and a house all have a right to be called one. Although the right is equally shared among them, there is something about them in virtue of which they have this right, for our considering them as one is based on experience. For instance, Berkeley does not say that the hypothetical combination of a window plus a chimney would have the same right to be called one, nor that the combination of a window plus a piece of floor would have the right. Why not? Presumably because there is some objective foundation, discovered by experience in certain sequences of sensible ideas originally belonging to different senses, in virtue of which they are recognized as, and hence, called ‘one,’ a foundation lacking in other hypothetical combinations. This should put us on guard when interpreting the phrase ‘this ... combining together of ideas is perfectly arbitrary.’ The phrase cannot mean ‘whichever way one fancies,’ because Berkeley immediately adds that the combining is ‘done by the mind in such sort as experience

62

Richard Glauser

shows it to be most convenient.’ In order to combine ideas into ‘one complex idea’ one must ‘have experienced which do regularly coexist and are proper to be bundled up together under one name’ (NTV 110; my italics). The arbitrariness in question, I suggest, has to do with the contingency of the combinations. If our experience were different – for instance, if God produced the sensible ideas we immediately perceive in a different order, with different relations between them, or if He produced different sensible ideas – then the combinations would be different. However, given the sensible ideas He does produce, and the order in which He produces them, there is some objective basis, discovered through observation and experience, entitling some combinations, rather than other hypothetical ones, to the ‘right’ to be called ‘one.’ There is no doubt, however, that in NTV 109 and elsewhere Berkeley says that some combinations are made by finite minds. Here are some other passages to the same effect: Number not without the mind in any thing, because tis the mind by considering things as one that makes complex ideas of ’em tis the mind combines into one, wch by otherwise considering its ideas might make a score of wt was but one just now. (NB 104) Complex Ideas are the Creatures of the Mind, hence may appear the Nature of Numbers, this to be deeply discuss’d. (NB 760) That number is entirely the creature of the mind, even though the other qualities be allowed to exist without, will be evident to whoever considers that the same thing bears a different denomination of number as the mind views it with different respects. Thus, the same extension is one, or three, or thirty-six, according as the mind considers it with reference to a yard, a foot, or an inch. Number is so visibly relative, and dependent on men’s understanding, that it is strange to think how any one should give it an absolute existence without the mind. We say one book, one page, one line; all these are equally units, though some contain several of the others. And in each instance it is plain, the unit relates to some particular combination of ideas arbitrarily put together by the mind. (PHK 12; my italics)

Passages such as these may suggest that Berkeley believes that finite minds make individual physical objects by combining their constituent ideas together. I believe, however, that that is not the point he is making. So, what is he saying? Let us work our way to an answer by first

The Problem of the Unity of a Physical Object in Berkeley

63

taking account of two things which Berkeley does not make explicit in passages such as these but which can justifiably be considered as presupposed when writing them. The first is that the sorts and kinds of physical objects, as distinguished from particulars, are in some sense made by finite minds: ‘Species of all sensible things made by the mind, This proved either by turning Men’s Eyes into magnifyers or diminishers’ (NB 271; cf. 288 and 289). The claim that sorts are made by finite minds is far from original; it had been made, for instance, by both Spinoza27 and Locke (cf. Essay III.iii.9–14, 412–17). Because Locke had forcefully asserted it, Berkeley does not think it necessary to repeat it at length in his published writings.28 The important point, though, is that by itself the claim that sorts of physical objects are made by finite minds does not imply that particulars are as well. It only implies that we might have made a different conceptual scheme of sorts and kinds and that, had we done so, particulars would have been classified differently. This, I believe, is clear enough in Locke. And there is nothing in Berkeley to show that he holds that because we make sorts, we therefore make particular physical objects. Surely the proof of our making sorts that he alludes to in NB 271 (‘turning Men’s Eyes into magnifyers or diminishers’) does not suffice to show that individual physical objects are made by finite minds. It only suggests that if our eyes were turned into ‘magnifyers or diminishers,’ the lawfully regulated sense-specific sequences of sensible qualities we would then immediately perceive would presumably be so different from the ones we are familiar with that we would observe and discover physical objects quite different from the ones we are acquainted with and that for this reason we would have a different conceptual scheme of classification. True, Berkeley might be prepared to say that because classifying particular physical objects under sorts depends on our being selectively attentive to certain resemblances, rather than to others, between such objects, we might have arrived at a different conceptual scheme of sorts without our senses being turned into ‘magnifyers or diminishers,’ but remaining in their present state. We might have done so, in sum, if we had been attentive to other resemblances between physical objects than the ones we do, in fact, consider salient. This being granted, two points must nevertheless be made in reply. First, as David Raynor aptly says, ‘relations within the stream of ideas undoubtedly make one way of carving up [our God-given sensory] intake more natural than another.’29 And this is only to be expected, for (A) given that God pro-

64

Richard Glauser

duces both (1) finite minds, with their needs for survival and wellbeing, ‘their own ends and purposes’ (NTV 109), and (2) ‘relations within the stream of ideas,’ and (B) given that He adapts the latter to the limited epistemic powers of the former, our present conceptual scheme based on what we in fact take to be salient resemblances between individual physical objects is presumably close enough to the one God can be taken to wish that we should arrive at. This can be shown by Berkeley’s wholeheartedly endorsing in its ‘most natural, obvious, and literal sense’ (DHP 253) the Mosaic account of Creation that took place when mankind did not exist: ‘Moses mentions the sun, moon, and stars, earth and sea, plants and animals; that all these do really exist, and were in the beginning created by God, I make no question’ (DHP 30).30 If our present conceptual scheme of sorts were not reasonably close to the one we can take God to have intended us to have, we probably would not know what sorts of things are referred to in the Mosaic account as created by Him. The second point to be made in reply is that the fact that we make sorts and kinds by selectively focusing on certain resemblances, rather than on others, between physical objects, and that we might therefore have produced a different scheme of classification, does not show that we make individual physical objects. All that it shows is that we would classify them differently if we were to focus on resemblances between them other than those that we have been led by experience to consider as salient. The second point that Berkeley seems to be presupposing in NTV 109 and PHK 12 is that number, when used to count, is sortal-dependent. It depends on the sort of thing to be counted. ‘How many are there?’ is an incomplete question which in contextually unclear circumstances of utterance prompts the reply ‘How many what are there?’ Because windows, chimneys, and houses – or lines, pages, and books – are different sorts of things, number, as used to count, varies according to the sort under consideration.31 If this is Berkeley’s position, as Flage correctly suggests,32 then he will hold that before one is able to determine the number of anything, the sort of thing to be counted must be settled. This means that we must have classified at least some individual objects into at least some sorts and kinds, and have given the individuals a common name before going on to count them. Berkeley, it seems, embraces this position in the Notebooks and perhaps in Alciphron: Children are unacquainted with Numbers till they have made some Progress in language. This could not be if they were suggested by all the senses. (NB 762)

The Problem of the Unity of a Physical Object in Berkeley

65

If we suppose rude mankind without the use of language, it may be presumed they would be ignorant of arithmetic. (Alc VII.12, 304)

The feature of language to which Berkeley is alluding here is the signification of common nouns which, when used to name physical objects, stand for whatever particular objects bear to each other a resemblance sufficient to justify their being called by the same name, that is, to be considered of the same sort: ‘a word becomes general by being made the sign, not of an abstract general idea but, of several particular ideas, any one of which it indifferently suggests to the mind’ (IN 11).33 These two points – (1) sorts and kinds are made by finite minds, (2) unity and number as used to count are sortal-dependent – shed some light on the passages quoted above and prompt the following line of reasoning. According to Berkeley unity and number as used to count are relative to combinations of ideas made by finite minds, to ‘draughts the mind makes of its ideas.’ But if unity and number are relative in the sense of being sortal-dependent, and if the sorts of physical objects are in some way made by finite minds, then the process by which we combine ideas into complex ones is surely a process by which we make sorts of physical objects. In other words, the complex ideas Berkeley discusses in the four passages quoted above (NB 104, 760; NTV 109; PHK 12) are (particular) ideas of sorts of things. This is confirmed by the fact that by far the greater part of Berkeley’s examples of names given to a combination of ideas are common nouns, not proper names.34 This is not to say that we do not sometimes give proper names to a combination of ideas; of course we do.35 But more often than not, the name is a common noun, for otherwise ‘the endless number or confusion of names would render language impracticable’ (DHP 245).36 What, then, is the complex idea that we make of a sort of thing? Intrinsically it is a particular idea, as all ideas are in Berkeley. It is a combination of particular, determinate ideas that becomes extrinsically, or functionally, general by being taken to stand for whatever other combinations of particular ideas are considered to resemble it to a sufficient degree for all of them to be classified under the same general name.37 Let us hereafter refer to such a complex particular idea as ‘general,’ or ‘sortal,’ in the sense just mentioned. There is nothing to prevent a complex sortal idea from being constituted by sensible ideas proper; that is, there is nothing to prevent one from taking a particular physical object itself as a sign representing other particulars of the same sort. However, in the four passages quoted above where the sortal idea is the result of

66

Richard Glauser

a finite mind’s activity of combining based on experience, the complex sortal idea is made of a selection of ideas of the imagination.38 4. Combinations of Ideas as Made by Finite Minds In order to defend this position, one must investigate both the experience on which the combining is based and the kind of combining that is performed. For instance, the experience cannot be merely the simultaneous and immediate sense perception of ideas belonging to different senses, as when one simultaneously and immediately perceives by sight, by touch, by smell, and by taste several sensible qualities that constitute an apple. Why not? Because mere simultaneity of a multiplicity of sensible ideas is an insufficient ground for considering them to be one physical object. One might just as well immediately perceive all of the following sensible ideas simultaneously: visible ideas of a table, visible ideas of a book lying on it, visible ideas of a window behind the book, visible ideas of a church seen through the window, tangible ideas of the wall one is leaning against and of the glass in one’s hand, the taste of sherry, and the smell of a nearby fireplace. Obviously, we do not consider all of these simultaneously and immediately perceived sensible ideas as one combination to which we give a sortal name. Why not? Because the ideas that are supposed to be combined into a complex idea and considered as one physical object, to which we give a sortal name, must have a ‘right’ to be so combined; they must be ‘proper to be bundled up together.’ The combining is supposed to be ‘done by the mind as experience shows it to be most convenient’ (NTV 109). Next, supposing that a finite mind immediately perceives several ideas that are ‘proper to be bundled up together,’ and granted that the finite mind has a power to combine ideas, is it plausible that the ideas it combines together are sensible ideas proper rather than a selection of ideas of the imagination? This does not seem plausible at all, for several reasons. First, if a physical object comprises an infinite quantity of sensible ideas, as suggested above, then a finite mind will not be able to combine them together. Only God can do that. Second, it is doubtful that immediately perceived ideas of sense last long enough for us to be able to combine them together. They are in ‘a continual succession ... some are anew excited, others are changed or totally disappear’ (PHK 26). Without the aid of memory and imagina-

The Problem of the Unity of a Physical Object in Berkeley

67

tion, how could one combine a sensible idea that one is immediately perceiving with others that one no longer immediately perceives? The first is present to sense when the others have already disappeared. The difficulty is overcome, however, if all of the ideas a finite mind combines together are copies or representations of sensible ideas proper, that is, if they are ideas of memory and imagination. Third, if the kind of complex idea that we form were made up of sensible ideas taken strictly, then such a complex idea would itself be a ‘fleeting’ part of a ‘continual succession.’ Its presence in a finite mind would last no longer than the time during which the mind simultaneously and immediately perceives its constituents, which would be a very short time indeed. Nor would it be an idea that one can call to mind at will, because sensible ideas cannot be called to mind at will; they are passively received in the order they are produced in us by God. But surely the combinations that we make and which, Berkeley says, are to serve as units to count other complex ideas of the same sorts are more stable than this. For, they must be such that they can be called to mind whenever we repeatedly need to count complex ideas of the same sorts given to sense, whether we count today, tomorrow, or the next day. The particular complex ideas formed by the finite mind must be somehow dispositionally stored so as to be called to mind – ‘excited in the imagination’ – whenever useful. The requirement can be satisfied if such combinations are made of ideas of memory and imagination, not if they are made of ‘fleeting’ ideas of sense. Fourth, the activity of combining, according to Berkeley, is selective. In the passage quoted above from NTV 109 he says that ‘the unit constantly relates to the particular draughts the mind makes of its ideas, to which it affixes names, and wherein it includes more or less as best suits its own ends and purposes’ (my italics). Now, if the ‘particular draughts’ were really individual physical objects, it would be absurd to say that they are made through a process of selection whereby the mind ‘includes more or less as best suits its own ends and purposes.’ For then, it would seem, the resulting combination would not be a physical object but only part of one, and furthermore only that part which happens to be of special interest to the finite mind. However, the claim that the activity of combining is selective and dependent on the mind’s ‘own ends and purposes’ makes good sense if the combinations are produced by the imagination with its ideas. For we have seen that the imagination is often unable to copy all of the sensible ideas the mind has previously received and that constitute a physical object. So, it is

68

Richard Glauser

only reasonable to suppose that, because of this natural handicap, the mind selects – according to ‘its own ends and purposes’ – the sensible ideas that it retains copies of in the imagination and that only these copies are combined into complex ideas. Furthermore, if the resulting combinations are made by the imagination by a process of selection based on experience, and if – as we have seen – they are sortal ideas, then they can be roughly construed as a Berkeleian equivalent to Lockean nominal essences. If so, there is no mystery as to how or why ‘Species of all sensible things [are] made by the mind’ (NB 271), even though – as in Locke – the finite mind does not make physical objects themselves. Fifth, the only way in which sensible ideas as such are actually given to a finite mind is in immediate perception. Yet, Berkeley never includes an activity of combining in immediate sense perception. When he does speak of some sort of activity of combining, he relates this to the imagination and its own ideas. So, for the reasons adduced, let us take it as plausible that whatever combinations are made by finite minds are combinations of ideas of imagination. It is important to note, however, that Berkeley mentions two quite different kinds of activities of combining ideas of the imagination. The first is when he says that ‘I have a faculty of imagining, or representing to myself the ideas of those particular things I have perceived and of variously compounding and dividing them’ (IN 10; my italics). The same activity is mentioned in PHK 1, where he speaks of the imagination as ‘either compounding, dividing, or barely representing those [ideas] originally perceived.’ But this sort of compounding is fanciful and done at will. It is not performed ‘in such sort as experience shows it to be most convenient,’ for the combinations he mentions are fictitious entities: ‘a man with two heads or the upper parts of a man joined to the body of a horse’ (IN 10). In NB 753, the ‘compositions’ are those of ‘a Blue Horse or Chimera.’ The other sort of combining that takes place in a finite mind is the psychological, associative linking of ideas originally received by one sense to ideas originally received by another in such a way that the former come to suggest the latter in mediate perception, as explained in the NTV, the TVV, and Alciphron. The finite mind gradually learns by experience of some regular, yet arbitrary, correlations between visible ideas and certain tangible ideas of sense. Due to the mind’s experience, its repeated observation of the regular correlations, the immediate perception of some visible ideas of sense comes to suggest ideas of the imagination that represent possible tangible ideas of sense.39 The psy-

The Problem of the Unity of a Physical Object in Berkeley

69

chological, associative connections by which the ideas of one sense come to suggest – and to enable mediate perception of – ideas originally belonging to another sense can well be called a ‘combining’ of ideas together. It is not the effect of will but of custom.40 Let us now give a slightly more accurate picture of the mind’s combining ideas through suggestion and mediate perception, by making four points. (1) The regular and arbitrary correlations of visible and tangible ideas proper are not produced by the finite mind; Berkeley says that they are observed by it (cf. NTV 25, 65, 110; TVV 39, 45). This is only to be expected. Since God produces all the ideas of sense, any regular correlations between visible and tangible ideas of sense are produced by Him as well. These correlations are presumably natural relations of signification instituted by God, by which certain visible ideas naturally signify certain possible tangible ideas within the Language of the Author of Nature. According to Berkeley, we can ‘have a notion of relations between things or ideas, which relations are distinct from the ideas or things related, inasmuch as the latter may be perceived by us without our perceiving the former’ (PHK 89; my italics). For instance, a blind person suddenly made to see will immediately perceive certain visible ideas, but the person does not know which tangible ideas to expect because, due to lack of the proper experience, the person has no notion of the natural relations of signification by which the immediately perceived visible ideas naturally signify certain possible tangible ideas of sense. Hence, we can take it that the objective foundation on which our activity of combining ideas originally belonging to various senses is based are such inter-modal correlations, natural relations of signification, instituted by God. They are not made by us and can only be discovered through experience.41 There is more to be said about the objective foundation constituted by relations of signification. Berkeley distinguishes between several kinds of relations of signification. Let us focus on two of them. First are the inter-modal correlations just mentioned, which are described in the NTV, the TVV, and Alciphron. These relations hold between visible and tangible ideas in such a way that both sorts of ideas connected by them are generally to be considered part and parcel of one and the same physical object. Second, in the Principles and later works Berkeley says that what we mistakenly take to be empirical causal relations between ideas are in fact relations of signification (cf., for instance, PHK 65–6, 108). One of the differences between the two kinds of relations of signification is that we are not supposed to consider what we (mistakenly) call an empirical cause and an empirical effect to be part and parcel of

70

Richard Glauser

one and the same physical object, because cause and effect are distinct. Now, in PHK 65 an example Berkeley gives of what might (inappropriately) be called a ‘causal’ relation of signification is that between a sound and ‘this or that motion or collision of the ambient bodies.’ So, although he considers visible and tangible ideas, along with tastes and smells, to be constituents of physical objects (cf. PHK 1), he would not rank sounds among them.42 They are empirical ‘effects,’ rather than constituents, of the ‘ambient bodies’ said to ‘cause’ them. What this shows, however, is that without notions of both kinds of relations of signification, we would not know which ideas to combine, and which not to. In order to do so we must have acquired, by observation and experience, notions of both kinds of relations. Thus, the objective foundation on which our activity of combining ideas originally belonging to various senses is based not only consists in relations of signification of the sort described in the NTV, the TVV, and Alciphron, but also in ‘causal’ relations of signification. This should come as no surprise. The distinguishing feature between ideas of sense and imagination is not only that the former are more lively than the latter, but especially that the former have ‘steadiness, order, and coherence’; that is, they are produced by God according to ‘Laws of Nature’ (PHK 30), what we call empirical ‘causal’ laws. (2) Although in the NTV and the TVV Berkeley limits his theory of suggestion and mediate perception to ideas of sight and touch, in Alciphron he expands his account so as to include smells, tastes, and natural sounds among natural signs ‘that inform us of other qualities to which they have neither likeness nor necessary connexion’ (Alc IV.12, 157).43 The addition made in Alciphron broadens the scope of Berkeley’s account to the extent that at least some clusters of ideas of each one of the five senses are natural signs of at least some ideas of any one or more of the other senses. (3) However, the ideas suggested in mediate perception are not sensible ideas proper but ideas of the imagination that represent possible ideas of sense: they are ‘suggested to the imagination, whose objects they are, and which alone perceives them’ (TVV 39; my italics; see also TVV 9–10 and DHP 204). This is why the associative combinations made by the mind are combinations of ideas of the imagination. (4) It is important to distinguish perhaps more sharply than Berkeley himself does between, on one hand, the natural relations of signification instituted by God between sensible ideas and, on the other hand, the associative connections produced by the finite mind between ideas of the imagination.44 The reason for this is that Berkeley’s account of

The Problem of the Unity of a Physical Object in Berkeley

71

perception, just as any other, requires a distinction between veridical and non-veridical perception. According to the Dialogues, Berkeley’s way of making the distinction seems to be that, since immediate sense perception is infallible, non-veridical perception can only be mediate.45 The point is made concerning an apparently bent oar: ‘But his mistake lies not in what he perceives immediately and at present (it being a manifest contradiction to suppose he should err in respect of that) but in the wrong judgment he makes concerning the ideas he apprehends to be connected with those immediately perceived: or concerning the ideas, that, from what he perceives at present, he imagines would be perceived in other circumstances’ (DHP 238; my italics). This means that an idea suggested to the imagination may not adequately represent the possible sensible idea one takes it to represent. Because of this, one should distinguish between (1) the combinations of ideas of the imagination that finite minds make by psychological, associative relations, and (2) the combinations of sensible ideas that God produces by instituting natural relations of signification between the components produced by Him. Not only are the former mere mental representations of the latter, but – because mediate perception is fallible – to some extent at least they may be inadequate representations. Let us now return to the passages quoted above (NB 104, 760; NTV 109; PHK 12) that seem to favour the position I am opposing. We have seen that there is ample reason to believe that the ‘draughts the mind makes of its ideas,’ in NTV 109, are sortal combinations of ideas of the imagination. This reading can be extended to PHK 12 because of the close thematic similarities of the two passages. As to NB 104, it, too, can (but might not) be read this way. Berkeley says: ‘tis the mind by considering things as one that makes complex ideas of ’em tis the mind combines into one’ (my italics). The ‘things’ considered as one are indeed physical objects made of sensible ideas proper, but the complex ideas that result from the mind’s combining are not. I suggest they are ideas of them; that is to say, the combinations that we make are ideas of the imagination that represent the sensible things we consider as one. The same reading can be extended to NB 760 because of its proximity with NB 753, which is about the ‘Composition of Ideas’ made by the imagination, and with NB 775: ‘By Idea I mean any sensible or imaginable thing.’ Let us also look at some related passages in Siris: The person or mind of all created beings seemeth alone indivisible, and to partake most of unity. But sensible things are rather considered as one than

72

Richard Glauser truly so, they being in a perpetual flux or succession, ever differing and various. (Siris 347; my italics) In things sensible and imaginable, as such, there seems to be no unity, nothing can be called one prior to all act of the mind; since they, being in themselves aggregates, consisting of parts or compounded of elements, are in effect many. (Siris 355; my italics) Thus much upon the whole may be concluded, that, distinct from the mind and her operations, there is in created beings neither unit nor number. (Siris 357)

Berkeley is saying that physical objects have no real unity since, contrary to God, they are not indivisible, but ‘ever differing and various’ and ‘many.’ This confirms that there is no positive theory of the unity of a physical object in Berkeley, although there is a theory of the objective foundation upon which we base our act of considering a physical object as one, and upon which, too, we combine ideas of the imagination in order to represent it and other objects of the same sort.46 What, then, are physical objects ‘in themselves’? Berkeley says that ‘in themselves’ they are ‘aggregates,’ ‘compounded’ entities. Now, if taken in themselves they are ‘compounded,’ then the compounding is no result of the finite mind. It must be God’s doing, as we have seen in PHK 64–5. According to Siris 347 the finite mind considers ‘sensible things’ as one. No mention is made of an activity of making physical objects by our combining sensible ideas together. The discussion pursued to this point has assumed that only two alternatives are to be envisaged: either physical objects are produced by God or they are made by the epistemic activities of finite minds. A third possibility – that physical objects are made by God in conjunction with finite minds – would make physical objects the result of a collaborative joint venture. One should resist this reading as well. Apart from the fact that Berkeley says nothing of the sort, it contradicts what he does say. If a physical object were the result of the joint, collaborative activities of God and finite minds, it would be a collection of ideas comprising both sensible ideas and ideas of the imagination. Berkeley’s repeated claim is that a physical object is a collection of sensible ideas as distinguished from ideas of the imagination; he does not say that it is a collection of sensible ideas and ideas of the imagination mixed together. For all that, though, it must be admitted that the interpretation so far

The Problem of the Unity of a Physical Object in Berkeley

73

proposed has its own difficulties to face. It cannot be easily reconciled with a literal reading of two passages of the Dialogues mentioned above: philonous. But in case every variation was thought sufficient to constitute a new kind or individual, the endless number or confusion of names would render language impracticable. Therefore to avoid this ... men combine together several ideas, apprehended by diverse senses, or by the same sense at different times, or in different circumstances, but observed however to have some connexion in Nature, either with respect to coexistence or succession; all which they refer to one name and consider one thing. (DHP 245; my italics) philonous. A cherry, I say is nothing but a congeries of sensible impressions, or ideas perceived by various senses: which ideas are united into one thing (or have one name given them) by the mind; because they are observed to attend each other. (DHP 249; my italics)

In the first passage Berkeley should have written ‘originally apprehended by diverse senses, or by the same sense.’ The adverb ‘originally’ would have implied that the ideas combined together are no longer apprehended by sense but are now copies in the imagination. In the second passage Berkeley should have said that the sensible ideas belonging to the congeries are considered one thing by the mind, or judged to constitute one thing, not literally that they are ‘united into one thing ... by the mind.’ Alternatively, he should have said that the ideas ‘united into one thing’ are ideas of the imagination and that they form a complex particular idea of a sort of thing we name ‘a cherry.’ Can these difficulties be overcome? To a certain extent they can. The difficulty in the first passage disappears if Berkeley is using ‘ideas apprehended by sense’ to mean ‘ideas with sensory content’ (as suggested by Marc Hight in this volume), since ideas of the imagination have sensory content. And the difficulty encountered in the second passage vanishes if Berkeley is using the expression ‘united into one thing’ to mean no more than ‘having one name given them,’ as the text might be taken to suggest. Both interpretive manoeuvres, however, are admittedly a bit strained. After all, if Berkeley is using ‘ideas apprehended by sense’ to mean ‘ideas with sensory content,’ thus including ideas of the imagination, that will not conceal the fact that he is nevertheless somewhat blurring

74

Richard Glauser

the difference between ideas of sense proper and ideas of the imagination. As to the second passage it is not obvious how the meaning of ‘united into one thing’ can be semantically reduced to that of ‘having one name given to them.’ Surely uniting (literally) a manifold and giving a name to the combination are two distinct operations however closely connected or interdependent they might be. In sum, it is preferable to acknowledge outright that Berkeley is unwittingly blurring the distinction between ideas of sense and imagination in both passages. Yet, this is not an unreasonable assumption for the following three reasons. First, we have seen that on several occasions Berkeley himself virtually acknowledges that he sometimes overlooks the distinction, as, for example, in his letter of 1711 to Leclerc quoted above. And it is easy to do so since ideas of the imagination have sensory content. Second, there is a plausible reason for which he might unintentionally neglect the distinction in both passages. The difference between ideas of sense and imagination, although important, is comparatively less fundamental than the theoretical distinction between perceivable qualities and the purely hypothetical unperceivable qualities that would belong to material substances if per impossibile any existed. Now, one can easily come to set aside a distinction of some importance within one’s own philosophy when focusing on a much more fundamental distinction lying between one’s philosophy and that of one’s opponents. This, I submit, is what takes place in both passages. Their respective contexts show that the distinction Berkeley is really concerned with is that between perceivable qualities and the hypothetical, unperceivable qualities of matter. In the context of the first passage Philonous contrasts perceivable qualities with ‘supposed originals [that] are in themselves unknown’ and describes Hylas as holding a position leading to scepticism with regard to ‘certain absolutely existing unperceived substances’ (DHP 246). In the context of the second passage Philonous opposes his conception of a cherry as a congeries of perceivable qualities to ‘an unknown nature distinct from all those sensible qualities ... something distinct from its being perceived’ (DHP 249). In sum, the distinction Berkeley is concentrating on in both passages is between (1) perceivable qualities, whether ideas of sense or of imagination, and (2) hypothetical unperceivable qualities of matter. Given that ideas of sense and of imagination are both perceivable, they are on the same side of the great divide. In such a context, ‘sensible’ seems to mean ‘perceivable,’ or ‘having sensory content.’ This is why Berkeley expresses himself without stressing the difference between the two kinds of ideas.

The Problem of the Unity of a Physical Object in Berkeley

75

Third, in any case the two passages just cannot be taken literally. Remember that the complexity of a Berkeleian physical object is such that we seldom ever immediately perceive more than a small part of the sensible ideas that constitute one. Now, in both passages from the Dialogues quoted above Berkeley says that the ideas combined together are ideas that have been perceived by us. This raises two problems we have already encountered. (A) The sensible ideas we have immediately perceived no longer exist in our minds; only copies of them subsist in memory and imagination. So how could one possibly combine together ideas one no longer has? (B) What of the enormous amount of possible ideas of sense we have not immediately perceived and that nevertheless belong to the same physical object as do some of those that we have perceived? Surely, if we made a physical object by combining sensible ideas together, we would have to combine the sensible ideas as yet unperceived by us with those that we have perceived; otherwise the result would not be a physical object, but a small part of one. But how could one possibly combine together ideas one has never had with those one no longer has? Both difficulties are overcome, however, if Berkeley is taken to mean the following. On the basis of our past experience of relations of signification, (1) we consider as one physical object a multiplicity of sensible ideas comprising ideas of sense that we remember, and possible ideas of sense that we have not immediately perceived but that we imagine we would immediately perceive under certain conditions; and (2) the ideas we do combine together are ideas of the imagination that represent a selection – a selection made ‘as best suits [our] own ends and purposes’ – of some past and some possible sensible ideas, so as to produce a complex idea of a sort of physical object. If this reading is on the right track it is mistaken to say that in virtue of their epistemic activities finite minds, either by themselves or in conjunction with God, make physical objects, although, of course, they make representations of them in the imagination (largely through mediate perception based on experience). It enjoys consistency with Michael Ayers’s statement, in this volume, that ‘the ontological status of sensible things is for Berkeley a fundamental metaphysical fact, a fact about the structure of reality, so to speak.’ Notes 1 ‘Several’ rather than ‘all five’ because we will see that it is doubtful that Berkeley includes sounds among the qualities that constitute the physical

76

2 3

4

5

6 7 8 9 10

11 12

Richard Glauser objects that are said to ‘produce’ them. In this essay the expression ‘physical object’ is used merely as a determinable ranging over things such as cherries, coaches, apples, churches, books, tulips, oak trees, planets, and so on, all of which are immaterial, mind-dependent non-substances in Berkeley. The expression carries no materialist, mind-independency, or substantialist assumptions. ‘Commonsense object’ could be a synonym in this context. Cf., for instance, George Pitcher, Berkeley (London: Routledge & Kegan Paul, 1977), 164ff. I am assuming, like many others, that physical objects in Berkeley are collections (or combinations, or congeries, etc.) of sensible ideas. In this volume Marc Hight denies that Berkeley holds this position. The textual evidence in favour of the traditional reading is too abundant and explicit to warrant an interpretation that refuses to take Berkeley’s often-repeated claim at face value, and I am unconvinced by Hight’s arguments to the contrary. NTV 110 shows that Berkeley is aware of the difference between individuation and unity. The question of individuation has been discussed by Daniel Flage in ‘Berkeley, Individuation, and Physical Objects,’ in Individuation and Identity in Early Modern Philosophy, ed. Kenneth F. Barber and Jorge J.E. Gracia (Albany: State University of New York Press, 1994), 133–54. Might one not consider a third possibility – that physical objects are made conjunctively both by God and by the epistemic activities of finite minds? The hypothesis will be briefly considered, and discarded, in section 4. Paul J. Olscamp, The Moral Philosophy of George Berkeley (The Hague: Nijhoff, 1970), 26. Ian C. Tipton, Berkeley: The Philosophy of Immaterialism (London: Methuen, 1974), 210. George Pitcher, ‘Berkeley on the Perception of Objects,’ Journal of the History of Philosophy 24 (1986): 101. David Raynor, ‘Berkeley’s Ontology,’ Dialogue 26 (1987): 619. Robert G. Muehlmann, Berkeley’s Ontology (Indianapolis: Hackett, 1992), 213, 215, 222. Cf. also 221: ‘houses, mountains, and rivers ... are collections. Strictly speaking, collections are not immediately perceived, they are made. And we are able to make them only after accumulating masses of immediate perceptions.’ Muehlmann’s distinction between ‘proper objects’ and ‘common objects’ is explained in his chapter 7; his ‘common object’ (cf. 213) is what I call a physical object. Flage, ‘Berkeley, Individuation,’ 142 and 147. Among many other contributions on this issue, cf. George Pappas, Berkeley’s Thought (Ithaca, NY: Cornell University Press, 2000), chs. 7 and 8.

The Problem of the Unity of a Physical Object in Berkeley

77

13 Note that God does not only combine ideas of several senses in order to form physical objects; he also combines natural, visual signs: ‘you cannot deny that the great Mover and Author of Nature constantly explaineth Himself to the eyes of men by the sensible intervention of arbitrary signs, which have no similitude or connexion with the things signified; so as, by compounding and disposing them, to suggest and exhibit an endless variety of objects, differing in nature, time, and place; thereby informing and directing men how to act with respect to things distant and future, as well as near and present’ (Alc IV.12, 157; my italics). 14 One aspect of Marc Hight’s paper in this volume concurs with the view defended here inasmuch as we agree that (at the level of metaphysics) bodies are collections of sensible ideas, and that such collections are the work of God, not of finite minds’ epistemic activities. On the other hand I disagree with Hight’s statement that (at an epistemic level) what a finite mind takes to be a physical (commonsense) object is an idea of the imagination. Berkeley is quite clear that ideas of the imagination are produced by finite minds. Therefore, it would follow from Hight’s position that what a finite mind takes to be a physical (commonsense) object is made by the finite mind, in the exact sense in which all ideas of the imagination are caused or produced by finite minds, and this seems unacceptable. Suppose I am perceiving a bird that I take to be a sparrow. If the bird I take to be a sparrow is a sparrow, the sparrow is a collection of sensible ideas made by God, not by myself. And if that which I take to be a sparrow is not a sparrow, but a robin, then the robin, too, is a (different) collection of sensible ideas made by God, not by myself. Of course, however, it should go without saying that, whether or not the perceived bird is a sparrow, when I take it to be a sparrow I have an idea of the imagination which I do produce or cause and which represents, correctly or not, a sparrow. This (partly) accounts for misperception at the level of mediate perception in Berkeley, as discussed in my ‘La structure de la perception médiate dans la théorie berkeleyenne de la vision,’ in Berkeley: Langage de la perception et art de voir (Paris: Presses Universitaires de France, 2003), 103–33. Hallucinations, too, can be explained in Berkeley by extending this account. 15 On this last issue, cf. Pappas, Berkeley’s Thought, ch. 6. 16 According to Berkeley, to perceive x immediately by sense is (1) to perceive x by sense without the means of (the perception of) an intermediate idea; and (2) to perceive x by sense without making an inference. The two senses are distinguished in the Dialogues: ‘Are those things only perceived by the senses which are perceived immediately? Or may those things properly said to be sensible, which are perceived mediately, or not without the intervention

78

17 18

19

20

21

Richard Glauser of others? ... the senses perceive nothing which they do not perceive immediately: for they make no inferences’ (DHP 174; my italics). Berkeley assimilates ideas of memory and ideas of imagination; cf. PHK 1. ‘Properly speaking Idea is the picture of the Imagination’s making[;] this is the likeness of & refer’d to the real Idea or (if you will) thing’ (NB 657a); ‘Would they [Descartes & Malebranche] have it that the Ideas of imagination are images of & proceed from the Ideas of Sense, this is true but cannot be their meaning’ (NB 818); ‘Ideas of Sense are the Real things or Archetypes. Ideas of Imagination ... are copies, images of these’ (NB 823). ‘Concerning the Object of Geometry, it must be noted that the word tangible admits of a double sense. In the strict sense it designates only what can be stroked and touched by the hands. In a wider sense, it embraces not only things that can be perceived in one way or another (ut Quies & Inane quae non sunt palpabiles) but also ideas fashioned by the mind on the basis of things perceived by touch (ideas a mente ad similitudinem tactu perceptarum fictas). So, when I assert that “the Object of Geometry is tangible Extension,” this must be understood in the latter sense. Assuredly, figure, angle, line, point, (all objects) which Geometers contemplate in their mind, although they themselves are not objects of sensory perception, are nevertheless referred to touch, to which, in some way, they owe their origin, when by an operation of the mind they are formed on the pattern of Ideas that were originally imprinted by means of that sense (per operationem mentis formentur ad speciem Idearum Tactui primitus impressarum).’ The quotation is from the final version of the letter, the one Leclerc actually received, published in Pierre Bellemare and David Raynor, ‘Berkeley’s Letters to Le Clerc (1711),’ Hermathena 146 (1989): 12–13. A draft of the letter is in W 8: 50. The idea of unity is ‘suggested to the Understanding, by every Object without, and every Idea within ... Whatever we consider as one thing, whether a real Being, or Idea, suggests to the Understanding, the Idea of Unity’ (Essay II.vii.7, 131). ‘Amongst all the Ideas we have, as there is none suggested to the Mind by more ways, so there is none more simple, than that of Unity, or One: it has no shadow of Variety or Composition in it: every Object our Senses are employed about; every Idea in our Understandings; every Thought of our Minds brings this Idea along with it. And therefore it is the most intimate to our Thoughts, as well as it is, in its Agreement to all other things, the most universal Idea we have. For Number applies it self to Men, Angels, Actions, Thoughts, every thing that either doth exist, or can be imagined’ (Essay II.xvi.1, 205). Number is ‘utterly inseparable from the Body, in what estate soever it be’ (Essay II.viii.9, 134; cf. II.viii.12–13, 136). Number is ‘really in them [Fire or

The Problem of the Unity of a Physical Object in Berkeley

22

23 24 25 26 27 28

29 30

31 32 33

79

Snow] whether any ones Senses perceive them or no: and therefore may be called real Qualities, because they really exist in those Bodies’ (Essay II.viii.17, 137–8; cf. II.viii.22, 140 and II.viii.23, 140–1). As with the other primary qualities, number belongs to the solid parts of a body ‘whether we perceive them or no’ (Essay II.viii.23, 140). Berkeley might well accept the following reasoning. All ideas are either sensible ideas or ideas of the imagination. All sensible ideas are sensible qualities, and vice versa. All ideas of sensible ideas are ideas (of the imagination) of sensible qualities, and vice versa. Unity is not a sensible quality; hence it is not an idea, nor is there an idea of it. Furthermore, we know that a finite substance is one, and that God is one, because we know that they are indivisible. Yet they are not ideas, they are not combinations of ideas, and they are not represented by ideas. Hence, our knowing that they are one requires no idea of unity or oneness. Spinoza, Letter XII, in The Collected Works of Spinoza, ed. Edwin Curley (Princeton: Princeton University Press, 1985), 1:203. Cf. for example NB 714, 763, 766–8; PHK 120–2; Alc VII.5, 293 and VII.12– 14, 304–8. Siris 288: ‘Number is no object of sense: it is an act of the mind. The same thing in a different conception is one or many.’ For more on this point, see the essay in this collection by Ralph Schumacher. Spinoza, Appendix Containing Metaphysical Thoughts, in Collected Works, 301. ‘Treating of Matter I had better say the proportion & Beauty of Things than their species (wch Locke hath proved already) are the Workmanship of the Mind’ (NB 836). Raynor, ‘Berkeley’s Ontology,’ 619. ‘I say farther, in case we conceive the Creation, as we should at this time a parcel of plants or vegetables of all sorts, produced by an invisible power’ (DHP 252; my italics). ‘When Moses speaks of herbs, earth, water, &c. as having been created by God; think you not the sensible things, commonly signified by those words, are suggested to every unphilosophical reader?’ (DHP 255). Cf. Berkeley to Percival, W 8: 37–8. The same position has been defended more recently by Peter Geach, Mental Acts (London: Routledge and Kegan Paul, 1957), 28. Cf. Flage, ‘Berkeley, Individuation,’ 136–7. Notice, too, that in NB 762 Berkeley implies that the sortal-dependence of unity and number counts against Locke’s claim that an idea of unity is received by sense. Berkeley’s argument seems to be that if ideas of unity and number were suggested by the senses, then a small child or ‘rude man-

80

34 35 36 37

38

Richard Glauser kind’ would be able to count before having made any ‘Progress in language,’ that is, before learning any common nouns and being able to classify some particulars into sorts; however, given the sortal-dependence of unity and number, one does not learn to count before making ‘Progress in language’; therefore, it is not true that ideas of unity and number are suggested by our senses. In NB 746 Berkeley wishes to attain the same conclusion without basing it on the sortal-dependence of unity and number: ‘Will any man say that Brutes have the ideas, unity & Existence? I believe not. yet if they are suggested by all the ways of sensation, tis strange they should want them.’ In addition to NTV 109 and PHK 12, cf., for instance, PHK 1 and DHP 249. Cf. IN 9: ‘Peter, James, and John.’ Cf. Locke, Essay III.iii.2, 409. ‘By observing how ideas become general, we may the better judge how words are made so. And here it is to be noted that I do not deny absolutely there are general ideas, but only that there are any abstract general ideas ... I believe we shall acknowledge, that an idea, which considered in itself is particular, becomes general, by being made to represent or stand for all other particular ideas of the same sort ... And as that particular line becomes general, by being made a sign, so the name line which taken absolutely is particular, by being a sign is made general. And as the former owes its generality, not to its being the sign of an abstract or general line, but of all particular right lines that may possibly exist, so the latter must be thought to derive its generality from the same cause, namely, the various particular lines which it indifferently denotes’ (IN 12); ‘universality, so far as I can comprehend, not consisting in the absolute, positive nature or conception of anything, but in the relation it bears to the particulars signified or represented by it: by virtue whereof it is that things, names, or notions, being in their own nature particular, are rendered universal’ (IN 15). ‘In truth, there is no such thing as one precise and definite signification annexed to any general name, they all signifying indifferently a great number of particular ideas’ (IN 18). As mentioned above, the reading advocated is neutral as to whether physical bodies in Berkeley may be immediately perceived as well as mediately perceived, or whether they may only be mediately perceived. It should also be added here that it carries no direct implications relative to the question whether the perception of a physical object in Berkeley is, or is not, necessarily sortal-dependent. For a good analysis of the question of sortaldependency, and sortal-independency, of perception, cf. Michael Ayers, Locke, vol. 1, Epistemology (London: Routledge, 1991), 173ff. Cf. also idem,

The Problem of the Unity of a Physical Object in Berkeley

39

40

41

42 43 44 45 46

81

‘Individuals without Sortals,’ Canadian Journal of Philosophy 4 (1974): 113–48; and idem, ‘Is “Physical Object” a Sortal Concept? A Reply to Xu,’ Mind and Language 12 (1997): 393–405. When saying that ideas of sense are ‘possible,’ I mean ‘possible relative to a finite mind that is not immediately perceiving them at a certain time t1’; they are sensible ideas that one would immediately perceive in such and such empirical circumstances. These ideas (or their archetypes) are nevertheless actual in God’s mind at that time tl (cf. PHK 90 and DHP 230–1). That Berkeley does hold that suggestion and mediate perception are related to an activity of combining ideas is borne out by NTV 110, although the combinations mentioned here are only made up of visible ideas. For more on this point, see Genevieve Migely’s discussion of Berkeley’s treatment of the active and passive aspects of mind in her essay in this collection. On this point, cf. Tipton, Berkeley, 195 and 371. Although Euphranor says that they are natural signs, he does not grant that they constitute natural languages, as the ideas of sight do. This is explained at greater length in my ‘La structure de la perception médiate.’ Cf. also NB 846, NTV 45, TVV 25; and Alc IV.12, 157–8. On this point, cf. Tipton, Berkeley, 211. The unity Berkeley calls ‘true’ or ‘real’ in these passages from Siris is indivisibility. The fact that a physical object has no unity in the sense of indivisibility is compatible with the claim that the sensible ideas constituting a physical object share ‘a right to be called one’; it only means that the objective foundation, the ‘right to be called one,’ is not indivisibility.

Why My Chair Is Not Merely a Congeries: Berkeley and the Single-Idea Thesis m a rc a . h i gh t

A centrepiece of Berkeley’s immaterialism is his treatment of ordinary objects (which I shall call ‘commonsense objects’).1 Unfortunately, understanding this crucial area of his thought has been clouded by the dubious assumption that the only non-phenomenalist reading of Berkeley is the view that commonsense objects are straightforwardly nothing more than collections of ideas.2 My intent here is to demonstrate that there are textual as well as philosophical reasons for believing that Berkeley holds a slightly more sophisticated view. From the perspective of finite minds, commonsense objects are single ideas associated with collections of sensory ideas. Metaphysically, commonsense objects are collections, but when we recognize Berkeley’s inclusion of an explicitly distinct epistemic element, a superior theory emerges. The word ‘chair,’ for instance, names a single idea that is in turn associated with a collection of sensory ideas. The word we use to name a putatively macro object names the single idea and only indirectly the set of the sensory ideas with which the single ideas are associated. In our ordinary lives single ideas serve as epistemic unifiers of diverse possible sensory experiences. In this epistemic sense, commonsense objects are single ideas. In metaphysical reality, commonsense objects are collections associated with these single ideas. I will not argue here against phenomenalist interpretations3 except insofar as to note that my arguments are incompatible with them. 1. Collections and Names Berkeley appears to hold that the only things we perceive are ideas.4 He also vociferously claims that his theory preserves the intuitions of ordi-

Berkeley and the Single-Idea Thesis

83

nary people (‘the vulgar’). One might then expect him to reconcile his immaterialism with ordinary beliefs. The interpretation that strikes many as an obvious candidate to satisfy vulgar intuitions about sensible objects is the collections view (hereafter ‘CV’). Commonsense objects are collections of immediately perceived sensory ideas. Berkeley’s first discussion of ordinary objects seems to lend initial support to this reading: Smelling furnishes me with odours; the palate with tastes, and hearing conveys sounds to the mind in all their variety of tone and composition. And as several of these are observed to accompany each other, they come to be marked by one name, and so to be reputed as one thing. Thus, for example, a certain colour, taste, smell, figure and consistency having been observed to go together, are all accounted one distinct thing, signified by the name apple. (PHK 1)

The traditional advocate of the collections view thus interprets Berkeley as arguing that perceiving sensory ideas is just to perceive the object itself. A typical argument schema would run this way: (1) Commonsense objects are collections of sensible ideas. (2) Sensible ideas and collections of sensible ideas are immediately perceivable. (3) Thus, commonsense objects are immediately perceived. The point is that when one perceives some object, that it is perceived as something (an apple, for example) is not an additional fact. Perceiving the red shape on the table is perceiving the apple, and there is no conceptual gap to be bridged from one to the other. George Pappas, a recent proponent of the collections view, argues in precisely this way: For instance, if person S immediately perceives a ‘collection’ of ideas O, and this collection of ideas is identical to a physical object R, then S will immediately perceive R. And if S immediately perceives a cluster of ideas O, and the ideas in the cluster are constituents of the physical object R, then S will also immediately perceive R.5

Pappas holds that commonsense objects are non-ideas that are also immediately perceived. I will engage this view subsequently in my essay. What I want to underscore for the moment is that on CV read-

84

Marc A. Hight

ings, perceiving a group of sensory ideas and perceiving the object are the same perceptual process. Now Berkeley certainly seems to write as if he espouses the collections view on several occasions. Committing him to this view, however, runs afoul of at least two difficulties. First, a careful reading of the passages where he mentions collections reveals that they do not necessarily commit him to such a view. Second, there are significant tensions between CV interpretations and what Berkeley says about ideas and how we come to perceive sensible things. Let us turn first to the texts themselves and convince ourselves that Berkeley was up to more than a simplistic collections view. Consider PHK 1 again. Although he clearly calls objects collections, in what sense does he do so? He notes that a group of ideas attend one another. Put together, they are labelled a distinct object and ‘signified by the name apple’ (my italics). Analysis suggests that Berkeley’s apple involves a single idea whose name is associated with (i.e., ranges over) a group of sensory ideas. A cautious study of Berkeley reveals that this is not just a random slip of the pen. Virtually all of the passages explicitly dealing with collections of ideas invoke the commonsense object as a single idea, under either names or words. Consider the following prominent passages: philonous: Therefore to avoid this as well as other inconveniences which are obvious upon a little thought, men combine together several ideas, apprehended by divers senses, or by the same sense at different times, or in different circumstances, but observed however to have some connection in nature, either with respect to coexistence or succession; all which they refer to one name, and consider as one thing. (DHP 245; my italics) philonous: Since it is not a being distinct from sensations; a cherry, I say, is nothing but a congeries of sensible impressions, or ideas perceived by various senses: which ideas are united into one thing (or have one name given them) by the mind; because they are observed to attend each other ... But if by the word cherry you mean an unknown nature distinct from its being perceived, then indeed I own, neither you nor I, nor anyone else can be sure it exists. (DHP 249; my italics)

As it turns out, Berkeley rarely speaks of commonsense objects as collections of ideas independently of some reference to a unifying name.6 Thus, for us, the single idea is the commonsense object (we consider that

Berkeley and the Single-Idea Thesis

85

idea or name to be the object), even though in metaphysical strictness the story is more complicated. Even in those instances where Berkeley is not directly concerned with collections of ideas, he nonetheless remains consistent in his attaching names to objects as signifiers and single ideas: [Those] things which pass for abstract truths and theorems concerning numbers, are, in reality, conversant about no object distinct from particular numerable things, except only names and characters; which originally came to be considered, on no other account but their being signs, or capable to represent conveniently whatever particular things men had need to compute. (PHK 122)

The context here is his attack on abstract ideas, but it is important to note how names function even here with numbers. A single idea, a name, represents a plurality of particular things. Berkeley consistently uses names as single ideas that range over groups of particulars. A word of warning is required when we examine all of Berkeley’s writings. He has a tendency to treat words in two overlapping senses. Often words (and names) are signifiers; they signify ideas. However, he also uses them as bearers of meaning and sometimes as both. The potential for confusion does not pose a problem for this analysis, for at least on one occasion Berkeley is clear about the position he wants to stake as his own. philonous: You indeed, who by snow and fire mean certain external, unperceived, unperceiving substances, are in the right to deny whiteness or heat to be affections inherent in them. But I, who understand by those words the things I see and feel, am obligated to think like other folks. (DHP 230)

Hylas has his words mean something (in addition, perhaps, to their functioning as signifiers), while Philonous uses words to represent sensory perceptions. ‘Snow,’ for instance, does not mean having certain perceptions (of coldness and whiteness etc.); rather, Berkeley understands by the word ‘snow’ a set of sensory ideas with which the name is associated. This is a subtle point for which Berkeley deserves some praise. Many commonsense objects are cold and white; thus merely having those sensations is not in itself sufficient to perceive snow as opposed to something else. This explains why Berkeley uses the word

86

Marc A. Hight

‘name’ so frequently: he is careful not to confuse signifiers with meaning-bearers.7 Evidence of the subtlety of the point may also be found in the frequent assertion that Berkeleian sensible objects are ‘constructed’ or ‘made’ by the mind. The word ‘snow’ stands for a collection of sensory perceptions but does not mean that collection. Since Berkeley notes that sensory ideas are volitionally independent of finite minds, it would be difficult to reconcile the claim that commonsense objects are collections with the assertion that they are essentially contingent mental constructs. The idea we use as a name is selected by us; what sensory ideas are proper members of the collection associated with the single idea is determined by God.8 With the textual evidence before us it becomes apparent that when Berkeley speaks of the status of commonsense things, he is not directly identifying them with collections. Instead, some unifying single idea, represented by a name, intervenes. Thus, what is strongly suggested by our first careful pass at the texts is that Berkeley’s considered view is that commonsense objects involve single ideas and not simply collections of sensory ones. Before we turn and attempt to construct a positive theory of how this works, it is worthwhile to stop and reflect briefly on some of the wellknown difficulties that accompany collections-view readings of Berkeley. What I would like to emphasize, however, is that the problem is not merely that the collections view commits him to untenable philosophical positions.9 Rather, the difficulty stems from the fact that, assuming Berkeley holds this view, the problems that arise are both obvious and ignored by Berkeley. This suggests that if Berkeley had another option that neatly avoided these concerns, he likely adopted it. At a minimum it is a useful philosophical exercise in charity to see whether such an option can preserve (more of) his system without undue violence to its spirit. One obvious problem with the collections reading concerns the conditions required for successfully perceiving some sensible object. Assume that person P perceives sensible object O. X, Y, and Z are member (sensory) ideas in the collection that constitutes O at that time. How many of the member ideas must P perceive to genuinely perceive O? Too few and we have no way to distinguish between putatively distinct sensible objects. Too many and one can reasonably argue that we never really perceive sensible objects at all. I do not believe that this problem is insurmountable; but what strikes me as significant is that Berkeley at no time squarely addresses this problem. One would expect him to say something about the membership conditions of collections if he adopted

Berkeley and the Single-Idea Thesis

87

the view. After all, the worry is sufficiently obvious that even beginning philosophy students frequently pick up on it. A second problem concerns the stability of the collections. If the collections view were correct we should expect that the member ideas in a collection ought to remain relatively stable and continuous over time. If not, we face two immediate worries: (1) If the membership of a collection is not stable, how do we identify and then re-identify the collection over time? It is not the focus of this discussion to answer this problem. I raise this as a concern because it is a deep and difficult issue, perhaps best solved by avoiding it altogether. And (2) how could one distinguish between collections if the membership is in constant flux? If I taste a cherry, then I should have that sensation as long as I think I am tasting a particular cherry (as opposed to some other object or nothing at all). We would expect this for several reasons. Since my perceiving the cherry is nothing more than my having certain sensory perceptions, when I cease to have those sensations I cease perceiving the cherry. Thus, if the ideas were to come and go during the time that I was chewing on my cherry, it would seem to follow that the cherry would be ‘blinking’ in and out of existence inside my mouth. Given that no two distinct sensory perceptions, strictly speaking, are of the same object, it would be different cherries that I would be tasting had I numerically different sensory perceptions. The result is a position with serious consequences for Berkeley’s system. We not only appear to perceive collections of ideas; we also appear to have the power to call them to mind at various times in the future and to count them (both at a time and over time). If commonsense objects are merely collections and nothing else, Berkeley needs to provide an account for how we can meaningfully accomplish tasks like counting over time and re-identifying things. Unless these collections are at least relatively stable with respect to the minds that perceive them, there is no clear way that Berkeley can account for these actions we perform.10 2. The Single-Idea Thesis The position I want to ascribe to Berkeley is the one we have already seen is strongly suggested by his treatment of commonsense objects. Recall that an apple, for instance, is a cluster of sensory ideas that have been observed to go together and have been signified by the name ‘apple’ (PHK 1). Drawing on passages like these, we arrive at the following thesis:

88

Marc A. Hight

Single-Idea Thesis (SIT): Finite minds consider commonsense objects to be single ideas (of the imagination) represented by names that signify diverse collections of sensible ideas. An apple, then, is a single idea suggested by certain particular sensory ideas. That is, the name ‘apple’ attaches to a single idea that signifies a collection of apple-like sensory impressions. We discover that these sensory ideas attend one another, so that by experience we find upon having certain apple-like visual perceptions we are apt to have certain apple-like tactile perceptions, and so forth. Berkeley thus invokes an important distinction between the commonsense object per se and the object as we perceive it qua object. In our ordinary lives, we unify the sensory perceptions we have over time into a single idea that we consider as the object. We are, in truth and strictness, mistaken about what we say, but as I shall argue subsequently, Berkeley makes a clean distinction between what is said about the world and what is true about the world. What then is this single idea? What kind of idea is it? The single ideas to which we attach names are sensory ideas of the imagination.11 Berkeley makes a distinction between ideas of sense (as distinct from sensory ideas, which concerns the content of the ideas) and ideas of imagination (PHK 30, cf. NB 582). The former come to us involuntarily from without the mind, while the latter are voluntarily generated by the mind and are under the control of our will. So whereas we cannot control the ideas of sense we receive from the sensible world, we can conjure up all sorts of images at will. We perceive one or more sensory ideas and slide insensibly to a single idea of the imagination (Berkeley says the latter are ‘suggested’ by the former). This single idea, in turn, is associated with a collection of sensory ideas. Thus, upon having a particular tactile sensation, I slide to the single idea of the imagination that names the object, which in turn leads me to expect other members of the collection with which it is associated. Ideas which are observed to be connected with other ideas come to be considered as signs, by means whereof things not actually perceived by sense are signified or suggested to the imagination, whose objects they are, and which alone perceives them. And as sounds suggest other things, so characters suggest those sounds; and, in general, all signs suggest the things signified, there being no idea which may not offer to the mind another idea which hath been frequently joined with it. (TVV 39)

Berkeley and the Single-Idea Thesis

89

Although this passage does not explicitly discuss commonsense objects, it does lay out the mechanism and how it works. It even goes so far as to explicitly note that some objects are perceived only by the imagination. We have, at this early stage in our discussion, at least an initially plausible reading. A number of minor worries and/or consequences must be addressed. One initial worry a critic might raise concerns the nature of ideas of imagination. Berkeley introduces the distinction between ideas of the imagination and those of sense in the Principles by appealing to levels of vivacity. Ideas of sense are more ‘strong, lively, and distinct’ (PHK 30). If this were the crucial distinction then one might wonder at the selection of ideas of the imagination to be commonsense objects. But this is only one axis of difference. Another difference lies in their dependence or independence from the will. The dependence distinction is the crucial one for our purposes. Ideas of the imagination can be conjured at will. This allows us to attach such an idea with a collection of sensory ideas by convention – just as Berkeley indicates is actually the case. Note, for instance, the similarity here with how he claims how general ideas function. A particular idea (image) becomes general by how it is used and not by being abstract. The details of my sensory idea (not the idea of sense) may potentially differ from yours even though our respective single ideas range over the same collection (and hence in every relevant respect are the same commonsense object). Thus the features of commonsense objects are not arbitrary even if the single ideas (names) we attach to similar collections might be.12 More significantly, making objects single ideas of the imagination allows us to conjure up objects even when we are not confronted via sensory perception with members of that object’s collection. Another superficial worry is the ‘real sun’ passage from the Principles. There Berkeley writes: These [ideas of sense] are said to have more reality in them than [ideas of the imagination]: by which is meant that they are more affecting, orderly, and distinct, and that they are not fictions of the mind perceiving them. And in this sense, the sun that I see by day is the real sun, and that which I imagine by night is the idea of the former. (PHK 36)

One might think that this passage does not cohere with the single-idea thesis. How are we to maintain Berkeley’s distinction between real and fictive objects if we take ordinary objects to be ideas of the imagination?

90

Marc A. Hight

Berkeley has already provided the answer. There is actually no difference in the reality of the idea itself – Berkeley needs that point to defend his anti-scepticism – but rather in the place the idea has in the order of ideas we perceive. After all, our remembered idea of the sun must have similar content in order for it to be about the sun at all. The remembered idea at night is less real because what drives one to have that idea is not an immediate sensory idea. It is thus already detached to a certain degree from the order of ideas of sense. Berkeley’s account of the distinction between the real and imaginary does not actually impact on whether the ideas perceived are of the imagination or not. It so happens that our ideas of sense are more orderly and regular, which is why we take them to be ‘real’ in the first place. But that implies nothing about the nature of idea qua of the imagination or qua of sense. One consequence of my reading deserves mention.13 It follows on my account that any seriously pre-linguistic mind does not in fact perceive commonsense objects. Infants at birth perceive colours and shapes, textures and sounds, but not tables and chairs. They must learn to make the associations between collections of sensory ideas and single ideas of the imagination that signify them. I find this, however, entirely Berkeleian. We learn the language of the Author of Nature like we learn any other normal language. The resultant picture preserves Berkeley’s core metaphysics, avoids the well-known problems with the collections view, and independently has greater explanatory power. When two persons perceive a commonsense object, they do not have to perceive the same (numerically or qualitatively) ideas of sense so long as the ideas they do perceive are all a part of the same collection. The actual content of the idea of the imagination (the single idea) might well vary between distinct individuals, but that matters not at all. The idea serves simply to signify the content of the collection. As a result, whether your particular image matches mine qualitatively is irrelevant so long as the ideas we have are associated more or less with the same collection of sensory ideas.14 The same analysis explains how Berkeley can address objections concerning how one identifies and re-identifies commonsense objects. When we perceive one or more sensory ideas, we have learned to associate certain groupings with single ideas (what we think of as unified objects). An object is (re-)identified when we attach the same single idea to another grouping of sensory ideas we perceive. Sometimes we err; but so long as the Author of Nature makes the groupings well ordered, we will have a generally reliable mechanism for picking out what we consider to be objects, even if the actual collections are different or in flux.

Berkeley and the Single-Idea Thesis

91

I do wish to be up-front about the deeper problems with which this interpretation has to deal. There are, so far as I can see, two principal hurdles. First, if SIT is correct, then Berkeley must directly challenge some plain views that the vulgar hold, despite his loud noises to the contrary. My strategy here will be to simply deny that Berkeley firmly adheres to the beliefs of the vulgar. His commitment to ordinary intuitions is largely a rhetorical device. When he has on his careful philosophy cap, it is just not possible to construe his words as conforming to commonsense intuitions. This is of particular importance when applied to the issue of commonsense objects, and I am not the first to notice this.15 In any event, the collections view suffers from conflicts with vulgar intuitions as well, precluding this issue from being truly decisive. The second problem is both more serious and more interesting. On a number of occasions Berkeley seems to indicate that commonsense objects are immediately perceived. A number of scholars have recently made much of this.16 In one sense, the single-idea thesis denies that this is strictly true. Metaphysically, ordinary objects are not immediately perceived by the senses but instead are suggested subsequently by some sensory ideas that are not themselves the proper objects of the imagination. But I deny that there is really a problem here. I agree that Berkeley takes commonsense objects to be perceived immediately; the evidence is significant and compelling, and Berkeley’s system seems to require it. The solution here lies in a deeper analysis of immediate and proper perception (beyond the proliferation of distinctions already invoked with respect to it). I argue that there are two kinds of immediacy that are relevant; I call these perceptual and process immediacy.17 Although we perceive all of our ideas immediately in the perceptual sense, frequently we also perceive ideas process mediately. This insight, when coupled with a careful application of Berkeley’s use of the concept of proper perception, will ultimately provide additional justification for the single-idea thesis. I will engage each of these problems in turn. If a satisfactory accounting can be provided for each, then, given the other virtues of this reading, we will be able to conclude at a minimum the plausibility of the addition to the collections reading that commonsense objects are considered by finite minds to be single ideas. 3. The True and the Said As regards the first of our two hurdles, Berkeley does make it a point to argue that his philosophy ‘vindicates common sense’ (DHP 244). Yet it

92

Marc A. Hight

is important to note that he does draw a firm distinction between what is (philosophically) correct and what is held to be true by the vulgar, as in one of his most famous lines, ‘Think with the learned, and speak with the vulgar’ (PHK 51). In an obvious sense, Berkeley is challenging at least one common view right off – that ordinary things have a material existence independent of our perceiving them. One should be suspicious, then, when Berkeley goes on to say that he sides in all things with the mob. It is just not true. In fact, there is reason to believe that Berkeley knows it is not true. The most prominent divergence from ordinary intuitions is Berkeley’s claim that we do not see the same thing we feel (see DHP 245). Here he does not even attempt to argue that this coheres with the plain person’s conception of the world. Instead, in those moments in which he is engaged in defending his philosophical views, Berkeley takes the importantly weaker tack of defending how the vulgar speak about the world, not how they conceive it to be. In the same passage where he speaks with the vulgar, it is significant that he restricts his claim as to how far his views match those of ordinary folk. ‘A little reflection on what is said here will make it manifest, that the common use of language would receive no manner of alteration or disturbance from the admission of our tenets’ (PHK 51). He makes no mention of whether what they say is true. When responding to Hylas’s question about whether two persons see the same object, Berkeley (through Philonous) answers by reconciling his view with how ordinary people use the word ‘same’: philonous: If the term same be taken in the vulgar acceptation, it is certain (and not at all repugnant to the principles I maintain) that different persons may perceive the same thing; or the same thing or idea exist in different minds. Words are of arbitrary imposition; and since men are used to apply the word same where no distinction or variety is perceived, and I do not pretend to alter their perceptions, it follows, that as men have said before, several saw the same thing, so they may upon like occasions still continue to use the same phrase, without any deviation either from propriety of language, or the truth of things. (DHP 247)

Worries about this response aside, Berkeley says that his theory is acceptable because it does not alter how ordinary people speak. In this case, we can say that two people see the same thing without violating Berkeley’s theory since it has a mechanism to allow for, strictly speaking, false utterances.

Berkeley and the Single-Idea Thesis

93

In general, Berkeley is far more circumspect in his siding with the vulgar than many believe. He is, however, adamant about two closely related issues: the mob is correct in trusting their senses and in rejecting scepticism (these are not quite the same, since scepticism can be generated for reasons other than those owing to doubts about perception). Outside of these two claims, Berkeley is less wedded to his vulgar allies. Most of the specific claims Berkeley makes with respect to the vulgar involve trusting the senses. He remarks in the Notebooks that ‘we must with the Mob place certainty in the senses’ (NB 740), and he sums it up well with Philonous: ‘In short you [Hylas – a materialist] do not trust your senses, I do’ (DHP 245). Closely allied with this strain is Berkeley’s desire to fight scepticism. Here the real villain is materialism, and Berkeley is using the mob as a club both to combat materialists and to buttress his own theory. When Berkeley is trying to ‘vindicate commonsense,’ it is most often in the context of how materialism commits people to scepticism whereas immaterialism does not (see DHP 229–30). Berkeley is selective in his invoking of the mob, which strongly suggests that he is consciously using it as a ploy and not as a guiding principle for his mature theory. Before we turn to the second problem, however, it is enlightening to note that the collections view also forces Berkeley to abandon ordinary intuitions. If the collections view is his position, then to perceive some commonsense object is to perceive some sensory idea (or ideas) in that object’s collection. But this entails that I can perceive an object via some of its ideas and not others. Ideas not perceived do not exist. Thus, sensible objects would be able to exist, as it were, incompletely. I can perceive the top of a table without its legs actually existing. It is precisely these sorts of problems that have driven some interpreters to accept a phenomenological reading of Berkeley. Other similar minor problems plague the collections view, but it is sufficient to note here that the collections view has nothing over the single-idea thesis with respect to preserving the intuitions of ordinary people. In fact, there is good reason to suppose that SIT is superior in this respect. When we speak about ordinary objects in common parlance we typically do so without immediately appealing to their sensory properties. When I mention the book on my desk I might subsequently take note of its size or colour, but I usually do so only after invoking the idea of the book. The single-idea thesis allows this to be an accurate description of how we engage the world. Although we must initially learn to match collections of sensory ideas with single ideas of the imagination that signify those collections, once the correspondences have been

94

Marc A. Hight

established we can simply invoke the single idea of the object and then slide to the members of its associated collection. I do not think of a blue, rectangular, yellowed sensation which is ipso facto my book; I think of my book and then I remember that it is blue and rectangular and has yellowed pages. Berkeley can more honestly preserve this way of speaking if he endorses something like the single-idea thesis. In any event, whether or not SIT conforms completely to vulgar opinions about objects is not a good test as to whether it is an acceptable idealist interpretation of Berkeley. Rivals to the present account fail to uphold commonsense, and we have good reason to think that Berkeley was not, in his careful philosophy, focused on remaining consistent with all of our ordinary intuitions. 4. Immediate Perception The more intriguing obstacle to the acceptance of the single-idea thesis concerns the nature of immediate perception and what sorts of things are actually so perceived. We are presented in the texts with an initial tension. On the one hand, (1) Berkeley seems to hold that only ideas are immediately perceived. Physical objects are mediately perceived via these sensory impressions. On the other hand, (2) there are numerous places in the texts where Berkeley seems to explicitly say that commonsense objects are nonetheless immediately perceived. The single-idea thesis requires that there be some kind of mediacy between our sensory ideas and the single idea that signifies the collection. The single idea is distinct from the member ideas of the collection; thus some sort of inference or suggestion must be taking place. I must explain how the singleidea theorist can hold both that ordinary macro-objects are immediately perceived and that only ideas are immediately perceived. At first glance the resolution seems relatively obvious. Berkeley claims that only ideas are immediately perceived in a sense that straightforwardly excludes ordinary macro-objects like tables and chairs. Philonous is quite clear: ‘For whatever is immediately perceived is an idea’ (DHP 202). And this is not an isolated point. philonous: This point then is agreed between us, that sensible things are those only which are immediately perceived by sense. You will further inform me, whether we immediately perceive by sight anything beside light, and colours, and figures: or by hearing, anything but sounds: by the palate, anything besides taste: by the smell, beside odors: or by touch, more than tangible qualities. (DHP 175)

Berkeley and the Single-Idea Thesis

95

Although we might be misled by his initial claim that ‘sensible things’ are immediately perceived (which might include commonsense objects), he corrects any possible misunderstanding in the remainder of the passage. According to Berkeley here, I do not immediately see a chair; I see light and colours from which I infer the existence of a chair. The same point is made at the start of the Principles, where commonsense objects are again apparently excluded from the list of things immediately perceived. More importantly, Berkeley frequently speaks as if the perception of putatively physical objects involves some sort of inference from sensory ideas. Consider first Philonous’s response to Hylas’s question about someone perceiving an oar in the water which appears bent: philonous: But his mistake lies not in what he perceives immediately and at present, but in the wrong judgment he makes concerning the ideas he apprehends to be connected with those immediately perceived: or concerning the ideas that, from what he perceives at present, he imagines would be perceived in other circumstances. (DHP 238)

If an error is made in thinking that the oar is really bent, the mistake lies in connecting the sensory perception of a bent figure with the objectidea of a bent oar instead of with a straight one. Note that Berkeley explicitly draws a link between ideas and ideas immediately perceived. He is not merely talking about how commonsense objects appear in differing circumstances, as is indicated by the careful disjunction he creates between the two cases. Thus, Berkeley is simply admonishing us to be careful about the commonsense objects we infer from certain sensory perceptions – a common warning even in materialist theories. An even more lucid passage comes earlier in the Dialogues: philonous: For instance, when I hear a coach drive along the streets, immediately I perceive only the sound; but from the experience I have had that such a sound is connected with a coach, I am said to hear the coach. It is nevertheless evident, that in truth and strictness, nothing can be heard but sound: and the coach is not then properly perceived by sense, but suggested from experience. (DHP 204)

While ‘strictly speaking,’ Berkeley is clear that an inference (or a suggestive slide) is made from sensory experience to the idea of an object (the coach). This is not an isolated passage. Berkeley makes essentially the same point in the Alciphron:

96

Marc A. Hight alciphron: Do we not, strictly speaking, perceive by sight such things as trees, houses, men, rivers, and the like? euphranor: We do indeed, perceive or apprehend those things by the faculty of sight. But will it follow from thence that they are the proper and immediately objects of sight? (Alc IV.10, 154) alciphron: I see, therefore, in strict philosophic truth, that rock only in the same sense that I may be said to hear it, when the word rock is pronounced. euphranor: In the very same. (Alc IV.11, 155)

We do perceive commonsense objects by our senses, but strictly speaking only in a derivative manner. If all of the texts were like these, there would be no deep interpretative problem. Commonsense objects would be the mediate objects of perception (we perceive the single idea we consider to be the object after ‘sliding’ to it from a sensory experience), and the single-idea thesis would be a straightforward improvement over the collections view. But Berkeley often makes remarks that strongly suggest that commonsense objects are themselves directly perceived. Pappas persuasively presents this case in his recent book. He cites texts like the following: ‘But to fix on some particular thing; is it not a sufficient evidence to me of the existence of this glove, that I see it, and feel it, and wear it?’ (DHP 224). The pronoun ‘it’ cannot reasonably be taken to refer to anything other than the glove – a common, ordinary object. Part of the allure of these passages is that they do seem to more strongly cohere with Berkeley’s protestations about siding with our ordinary ways of speaking. Lastly, whether the premises considered, it be not the wisest way to follow Nature, trust your senses, and laying aside all anxious thought about unknown natures or substances, admit with the vulgar for real things, which are perceived by the senses? (DHP 246)

Here it must be admitted that Berkeley does not say that we immediately perceive these objects by the senses. But he does elsewhere. Consider the two following passages, also invoked by Pappas in support of his position that commonsense objects are immediately perceived: But, say you, it sounds very harsh to say we eat and drink ideas, and are clothed with ideas. I acknowledge it does so, the word idea not being used in common discourse to signify the several combinations of sensible qual-

Berkeley and the Single-Idea Thesis

97

ities which are called things: and it is certain that any expression which varies from the familiar use of language will seem harsh and ridiculous. But this doth not concern the truth of the proposition, which in other words is not more than to say, we are fed and clothed with those things which we perceive immediately by the senses. (PHK 38) Wood, stones, fire, water, flesh, iron, and the like things, which I name and discourse of, are things that I know. And I should have not known them, but that I perceived them by my senses; and things perceived by the senses are immediately perceived. (DHP 230)

Like Pappas, I grant that these passages straightforwardly seem to say that we really do immediately perceive commonsense objects. The trouble now, of course, is to reconcile the conflicting passages. Traditionally there have been two ways to do this. We might seek either (1) to deny that Berkeley is speaking carefully when he says that we perceive ordinary objects or (2) to build a case such that Berkeley, strictly speaking, believes that we immediately see more than just sensory ideas. George Pitcher defends the first view; George Pappas has recently endorsed the second. In accordance with an explanation I first encountered in Pitcher’s work, one might read Berkeley here as drawing a line between his careful philosophy and his appeals to ordinary intuitions when we talk about the world. That is, Berkeley did not really mean to imply that we immediately perceive tables, chairs, wood, and stones. He was, as it were, speaking with the vulgar. Pitcher fastens on Berkeley’s tendency to qualify his discussions of his careful philosophy with phrases like ‘in truth and strictness.’ A careful review of the coach passage and others like it reveals that when Berkeley denies that we immediately perceive things like coaches, he invokes a higher philosophical standard. As he notes, it is evident, ‘in truth and strictness,’ that nothing can be heard but sound. In the initial passages cited by Pappas, no such qualifiers occur. This strongly suggests that we have been misled by Berkeley’s rhetorical strategy. There is nothing wrong with saying – in everyday speech – that we immediately see chairs and so on. But that does not make such claims true. Pitcher’s analysis, however, has not dissuaded others from pursuing the second course. Pappas, following a line of analysis first advanced by Kenneth Winkler, has argued that these passages contain yet another important qualifier. The key is not the ‘truth and strictness’

98

Marc A. Hight

clause but rather the claim that objects like coaches are not the proper and immediate objects of the senses.18 In analysing these same passages where Berkeley makes claims about the strict truth of matters, they draw a distinction between immediate objects proper to a single sense and immediate objects common to multiple sense modalities. This is the same distinction invoked by Aristotle in De Anima, and in the relevant texts cited by Pitcher Berkeley appears to be calling on this very distinction. Consider two passages we have already examined: the coach passage and the one from Alciphron where Euphranor notes, ‘We do indeed, perceive or apprehend those things by the faculty of sight. But will it follow from thence that they are the proper and immediate objects of sight?’ Things like coaches are immediate objects common to several senses (sight and hearing), whereas things like sounds are immediate objects proper to a single sense modality. Euphranor is making the same point with respect to trees, houses, men, and rivers. According to Winkler and Pappas, the ‘in truth and strictness’ is a reference to this distinction and not the mediacy of the perception generally.19 When he says that ‘strictly speaking’ nothing can be heard but sound, he intends that what is immediately and properly perceived by one modality is not immediately perceived by another. This claim is technically consistent with commonsense objects being immediately perceived. The distinction is thus between objects immediately and properly perceived as opposed to objects merely immediately perceived. I think there can be little doubt that Pappas and Winkler are right on this score. In The Theory of Vision Vindicated and Explained Berkeley makes his intentions clear: By a sensible object I understand that which is properly perceived by sense. Things properly perceived by sense are immediately perceived. Besides things properly and immediately perceived by any sense, there may be also other things suggested to the mind by means of those proper and immediate objects. Which things so suggested are not objects of that sense, being in truth only objects of the imagination, and originally belonging to some other sense or faculty. Thus, sounds are the proper objects of hearing, being properly and immediately perceived by that, and by no other sense. But, by the mediation of sounds or words all other things may be suggested to the mind, and yet things so suggested are not thought the object of hearing. (TVV 9)

Here the distinction is laid out in its entirety. Strictly speaking, sensory

Berkeley and the Single-Idea Thesis

99

objects are immediately perceived only by the sense modality to which they are proper. Although this does leave open the possibility that there are objects that are immediately perceived, it does not establish that said items can be non-ideas. Pappas’s position is stronger than the mere denial of Pitcher’s claim; he seeks to positively establish that things other than (proper) ideas can be immediately perceived: Here an equally important but distinct thesis is explored: that if something is immediately perceived, then that thing is an idea. Contrary to a common and traditional interpretation, I argue that Berkeley also did not accept this latter thesis ... Berkeley held that, in addition to ideas, ordinary physical objects – what he called ‘sensible objects’ or ‘sensible things’ – are immediately perceived.20

What is distinctive about this view is that Pappas takes physical objects to be non-ideas that are immediately perceived. This seems directly at odds with Berkeley, as when he says ‘I take the word idea for any immediate object of sense or understanding’ (NTV 45). Anything that is an immediate object (not merely a proper one) is automatically an idea. Some textual wrangling might follow, but I am content to let this worry slide. A larger problem looms: the distinction Pappas uses against Pitcher also seems to apply against his own analysis. In the passages where Berkeley invokes proper and immediate perception, he does so always with respect to a particular sense modality. The coach is not immediately perceived by hearing (and Berkeley emphasizes the word ‘hearing’) because coaches are not the proper objects of hearing. A glance back at the other relevant passages reveals the same pattern with respect to sight. Recall Euphranor’s words quoted above: ‘We do indeed, perceive or apprehend those things by the faculty of sight. But will it follow from thence that they are the proper and immediate objects of sight?’ We may mediately perceive by hearing something proper to sight (as when we conjure up a visual image of a bell when we hear a ringing sound), but as it turns out all immediate perception is proper to some sense modality or faculty. The importance of the assertion that all immediate perception is proper to something does not surface until one remembers that Berkeley adheres to the heterogeneity thesis. There are no common sensibles. As Berkeley phrases it, there is no ‘such thing as one idea or kind of idea common to both senses’ (PHK 127). So unless there is some sense modality specific to ordinary objects, they cannot possibly be immedi-

100 Marc A. Hight

ately perceived. Pitcher’s insightful analysis now returns with a vengeance – was Berkeley just engaging in more rhetoric? Pappas can produce all of the passages he likes that imply that commonsense objects are immediately perceived, but unless he can explain how this works without contradicting the seriously held heterogeneity thesis, his argument is in jeopardy. Even if physical objects are not ideas, they must be proper to some sense modality or faculty. Berkeley tells us what objects are proper to the traditional senses – light, colour (and perhaps shape) for sight, sound for hearing, and so on – and commonsense objects are nowhere to be found in these lists. Pappas provides many texts where Berkeley seems to say that commonsense objects are immediately perceived, but he does not provide the mechanism for how this works. At best he gives us a quick passing appeal to the collections view. Perceiving an ordinary object is to perceive some of the members of its collection of sensory ideas: In all then, based on these many passages, we have ample support for the view that Berkeley holds that physical objects are immediately perceived. Of course, when one immediately perceives an object one must also immediately perceive one or more of its sensible qualities; so, these, too, will count as entities that are immediately perceived.21

We do have ample support for the claim that commonsense objects are immediately perceived, and Pappas’s textual analysis is first-rate. What we lack is an explanation of how this can be, given our tension and the rest of Berkeley’s philosophical system. Given that there is a serious initial philosophical difficulty with this position, it would be unwise to accept his view absent some persuasive story about how the perception of physical objects can be reconciled with these concerns. 5. All Is Not Lost Fortunately all is not lost, for the single-idea thesis has the resources to address this problem. In one respect the single-idea thesis provides an easy solution to the original tension with which we started. Only ideas are immediately perceived. What we take to be commonsense objects can be immediately perceived because they are also ideas. What remains, of course, is to explain exactly how this works within Berkeley’s system. The key lies in separating immediate perception from the process of perception. It is convenient to refer to this as the difference between

Berkeley and the Single-Idea Thesis

101

perceptual immediacy and process immediacy. In one sense saying that there are two kinds of immediacy here is a bit misleading. Rather, there is one genuine sense of immediacy (perceptual) and a different aspect of perception that is easy to conflate with immediate perception. I am happy to follow Pappas’s analysis of what constitutes immediate perception.22 If we omit some complicating details that are not relevant at the moment, a particular act of perception is immediate when there is no third thing present during an act of perception. An act is otherwise perceptually mediate. Thus, for example, my perceiving the martial virtues of Caesar from reading a book is perceptually mediate. But we (including Berkeley) also use the word ‘immediate’ as an indicator of time. Something that happens immediately happens right now or at this instant. Perception, like any act of the mind, can occur over time. The mind’s engagement with an idea of sense is not only immediate in the sense of not involving some tertium quid; it is also temporally immediate. It happens, as it were, in an instant. But we do not perceive every idea of sense all at once. We perceive them in an order over time. Sometimes we perceive some ideas of sense at one time because we perceived a distinct idea at an earlier time. Had we not perceived the earlier idea we would not have perceived the latter. In this way one might say that the latter idea is process mediate even though it is immediately perceived. What I wish to suggest is that our perception of commonsense objects works in exactly this way. We immediately perceive the single ideas that signify various collections of sensory ideas – there is no third thing in virtue of which it is perceived at the time in which it is perceived – but there is a process involved in so doing. It is still possible to immediately perceive the idea without the suggestive sensation, so the idea is not being perceived mediately through the sensation. The story is fairly simple. We learn to associate a particular idea of the imagination with a collection of sensory ideas.23 We give that idea a name which signifies that collection. Thereafter, the mind ‘slides’ insensibly over time (no matter how short a span) from the perception of a member of the collection to the single idea (which in turn signifies the entire collection). Naturally, this does not deny that other inferences may be made to things perceived mediately, just as Berkeley indicates. This distinction allows us to reconcile those passages where Berkeley seems to say that we immediately perceive commonsense objects with those where he explicitly claims that we only immediately perceive ideas, without forcing us to accept the implausible collections view. Furthermore, it supports the single-idea thesis generally.

102 Marc A. Hight

It might be objected that Berkeley blocks this view. Philonous agrees with Hylas when the latter proclaims that ‘[the senses] make no inferences’ (DHP 174). That is certainly right. No inference is being made. We attach single ideas to collections (and vice versa) by constant conjunction from experience. Eventually this connection becomes so familiar that we insensibly slide from one to the other. Berkeley himself explicitly describes this process and separates suggestion from judgment or inference: alciphron: You would have us think, then, that light, shades, and colours, variously combined, answer to the several articulations of sound in language; and that, by means thereof, all sorts of objects are suggested to the mind through the eye, in the same manner as they are suggested by words or sounds through the ear, that is, neither from necessary deduction to the judgment, nor from similitude to the fancy, but purely and solely from experience, custom, and habit. (Alc IV.10, 154)24

Euphranor affirms this, adding only that he obliges Alciphron to submit to nothing more than the force of truth. The truth is that we are driven to our ideas of objects by the habit and custom of constant conjunction. This is not an inference but a ‘suggestion’ and so does not technically require an act of judgment. Importantly, Berkeley even claims (at the end of NTV 77) that suggestions can occur immediately (in the temporal sense), as with the perception of distance: ‘I say they [ideas that suggest distance] do not first suggest distance, and then leave the mind from thence to infer or compute magnitude, but suggest magnitude as immediately and directly as they suggest distance.’25 The last piece of the puzzle is the problem with which I confronted Pappas’s view. It would seem that all ideas perceived immediately are proper to a sense modality or faculty. That is exactly right. Berkeley tells us that some ideas are the objects of the imagination and not of any sense modality: The peculiar objects of each sense, although they are truly or strictly perceived by that sense alone, may yet be suggested to the imagination by some other sense. The objects therefore of all the senses may become objects of the imagination, which faculty represents all sensible things. (TVV 10)

The proper objects of the senses become the secondary objects of the imagination. That is, a particular colour is the secondary object of the

Berkeley and the Single-Idea Thesis

103

imagination. But the idea of a book, which is secondary to sight, is the proper object of the imagination: euphranor: Thus, for example, in reading we run over the characters with the slightest regard, and pass on to the meaning. Hence it is frequent for men to say, they see words, and notions, and things in reading of a book; whereas in strictness they see only the characters which suggest words, notions, and things. And, by parity of reason, may we not suppose that men, not resting in, but overlooking the immediate and proper objects of sight, as in their own nature of small moment, carry their attention onward to the very thing signified, and talk as if they saw the secondary objects? which, in truth and strictness, are not seen, but only suggested and apprehended by means of the proper objects of sight, which alone are seen. (Alc IV.12, 156)

Notice the careful distinctions that are made here. We immediately and properly see a certain bit of light and colour and then insensibly slide to the commonsense object, which is secondary to that sense. The idea of the imagination – the single idea of the object – is not properly seen at all: What we immediately and properly perceive by sight is its primary object, light and colours. What is suggested or perceived by mediation therefore, are tangible ideas which may be considered as secondary and improper objects of sight. (TVV 42)

The idea is secondary to the sense but proper to the imagination. Berkeley might not say this as perspicuously as one might like, but his analysis is clear enough: Ideas which are observed to be connected with other ideas come to be considered as signs, by means whereof things not actually perceived by sense are signified or suggested to the imagination, whose objects they are, and which alone perceives them. (TVV 39; my italics)

The ideas that serve as signs are perceived by the imagination alone. That is, they are proper to the imagination. The analysis is not difficult to extend generally to Berkeley’s system, and I am at a loss as to how this could work any other way. The idea of a colour or a sound is not proper to the imagination; hence something else is proper to it. Berkeley tells us that the imagination can conjure up ideas itself and these ideas are not

104 Marc A. Hight

perceived by the senses (PHK 28–30). Imagining is a species of perceiving or, yet more broadly, of thinking. I thus independently defend this move on the grounds of charity. It is both consonant with the spirit of Berkeley’s overall project and an emendation that improves its strength. These objects of the imagination are certainly strange. They depend on the sensory perceptions we have but are not proper to any sense modality (cf. NB 582). They are connected by custom and habit with our sensory perceptions but are not necessarily constructed by judgment from them. I do not pretend to have a robust theory about the nature of these ideas, nor do I suppose Berkeley had a clear story to tell. But the position is consistent with the rest of his metaphysics and neatly avoids the problems we have been discussing. The single-idea thesis ties all of these elements together in a way that is strikingly consistent with the texts. Consider one extended bit of text from the New Theory. Berkeley tells us that the only immediate objects of sight are light and colour (NTV 129). Then he invokes the strictness language to fix the proper objects of a sense modality: ‘I am not able to attain so great a nicety of abstraction: in a strict sense, I see nothing but light and colours’ (NTV 130). Nothing else is immediately perceived by sight. And then Berkeley goes on to invoke the single-idea thesis in the same paragraph: It must be owned that by the mediation of light and colours other far different ideas are suggested to my mind: but so they are by hearing, which beside sounds which are peculiar to that sense, doth by their mediation suggest not only space, figure, and motion, but also all other ideas whatsoever that can be signified by words. (NTV 130)

The last phrase is delightful. Tables and chairs, as ideas, can be suggested by the mediation of ideas of sense. These, in turn, are signified by words that range over the collection of ideas associated with that single idea of the imagination. Thus, even though metaphysically all that exists to constitute commonsense objects are collections of ideas (and Berkeley admits this), we (finite minds) use single ideas as commonsense objects to provide a coherent unity we can utilize to manipulate, and function in, the world. Conclusion With our obstacles safely behind us, we can now conclude that the single-idea thesis is a plausible interpretation of Berkeley’s account of

Berkeley and the Single-Idea Thesis

105

commonsense objects. The chairs and tables that we perceive are considered by us to be single ideas that signify collections of sensory ideas. When I say that I now perceive a book, the word ‘book’ refers to a single idea that functions as a signifier. Reading Berkeley this way enables us to neatly avoid some of the philosophical and textual problems of the traditional collections view. To perceive a book is not ‘just’ to perceive some of the members of its collection. Perceiving some of the members of the collection leads us (by habit learned by prior constant conjunctions) to have an idea of the imagination associated with the entire collection. This view allows us to explain error: some sensory ideas (or their qualitative duplicates) are members of more than one collection, and that might prompt us to slide to the ‘wrong’ idea.26 Worries about speaking with the vulgar either do not apply specially to this theory or, as is more likely, do not apply at all given that Berkeley separates speaking with the vulgar from what is true. Our concerns about the nature of proper and immediate perception ultimately turn out to provide support for the single-idea thesis, given that only this hypothesis can cleanly reconcile Berkeley’s diverse philosophical claims. It is now no longer sufficient to dismiss idealist interpretations of Berkeley’s views on perceptions by simply attacking the weak collections view. This, in turn, reopens the possibility that Berkeley’s overall theory has more to be said for it than has previously been thought. Although this is not the place to pursue such inquiries, a more savvy understanding of Berkeley concerning commonsense objects reveals the potential for uncovering other insights formerly hidden by the simple collections view. Notes I am indebted to many of the participants at the Berkeley Conference at Texas A&M University for their critical discussion of this paper. In particular, Richard Glauser, Ian Tipton, and Steve Daniel gave generously of their time to pursue these and other themes in Berkeley scholarship. 1 I avoid using the term ‘physical object’ because one might be tempted, inadvertently, to make materialist assumptions about their nature in Berkeley’s ontology. The OED, for instance, first associates ‘physical’ with material, a connection Berkeley would not allow. I do not dispute that others may use the word ‘physical’ innocently as a sortal term. 2 This is the standard interpretation of Berkeley. As just a small sample, see

106 Marc A. Hight

3 4

5

6 7

8 9

10 11

12 13

A.C. Grayling, Berkeley (La Salle, IL: Open Court, 1986), 53, 63; I.C. Tipton, Berkeley: The Philosophy of Immaterialism (London: Methuen, 1974), 185; J.O. Urmson, Berkeley (New York: Oxford University Press, 1982), 41; and G.J. Warnock, Berkeley (Notre Dame, IN: University of Notre Dame Press, 1953), 130. In this collection, see the essay by Richard Glauser. For a defence of a phenomenalist reading of Berkeley, see Kenneth Winkler, Berkeley: An Interpretation (Oxford: Clarendon Press, 1989), esp. 191–203. Recent commentators have argued against this claim, and I will engage that issue subsequently. George Pappas in particular denies that only ideas are immediately perceived; see his Berkeley’s Thought (Ithaca, NY: Cornell University Press, 2000), especially 172–8. Pappas, Berkeley’s Thought, 12. In his essay in the current volume, Pappas also argues that representative theories of perception need not always require an inference (cf. especially his section 5). Cf. NTV 96, 97, 106, and especially 109 for representative examples. One passage might mislead here. Compare his NB 763: ‘Numbers are nothing but Names, meer Words,’ and Berkeley to Molyneux, 8 December 1709: ‘Truth on’t is Numbers are nothing but Names’ (W 8: 25). The former is the corrected version of the Notebooks, which is much less suggestive than what appears in the Luce and Jessop edition: ‘Numbers are nothing but Names, never Words.’ But this correction does not weigh against my claims here, since Berkeley does not believe that there is any thing ‘a number’ to be signified. Douglas Jesseph has fairly clearly made this point already; cf. his Berkeley’s Philosophy of Mathematics (Chicago: University of Chicago Press, 1993), 107–11. See Glauser’s ‘The Problem of the Unity of a Physical Object in Berkeley’ in this volume for an excellent analysis engaging the constructionist position. Many such difficulties have been adequately explored elsewhere. See Jonathan Bennett, Learning from Six Philosophers, 2 vols. (New York: Clarendon Press, 2001), 2: secs. 172–7, 226–7. Cf. Glauser’s essay in this collection. Although I agree in the main with Glauser’s analysis on the nature of Berkeleian commonsense objects, I take my thesis to be an extension and improvement to his view, which unhappily concludes that Berkeley is ‘unwittingly blurring the distinction between ideas of sense and imagination.’ If we note the distinction between the single ideas that name objects and the collections with which they (the single ideas) are associated, I believe we may avoid attributing this confusion to Berkeley. I thus hope to confirm and reinforce Glauser’s analysis. My thanks to Doug Jesseph for raising this point in private conversation.

Berkeley and the Single-Idea Thesis

107

14 An additional potential complication might need to be sorted out at some point. Following Locke, are these single ideas of the imagination simple or complex? Fortunately my view can remain neutral with respect to these options (I believe a case can be made for both alternatives). A detailed analysis of the consequences of each reading lies outside the scope of this discussion. 15 George Pitcher is the first (of whom I know) to make a sustained argument against the collections view based on dividing Berkeley’s apparent endorsement of it against his explicit denials of the same thesis. My reasons and ultimate position are distinct from his, although our views are broadly consonant and I owe no small bit of inspiration to his work. See George Pitcher, Berkeley (Boston: Routledge, 1977), 99–100. 16 Pappas is one, cited earlier (see also the Pappas essay in this collection). Winkler (Berkeley, 149–60) is another. 17 I owe the label ‘process mediacy’ to Lex Newman, whose suggestion greatly clarified this position. 18 Winkler, Berkeley, 155; Pappas, Berkeley’s Thought, 180–2. 19 See Winkler, Berkeley, 154–61, for an extended analysis of these passages. 20 Pappas, Berkeley’s Thought, 172–3. 21 Ibid., 176. 22 Ibid., ch. 6, esp. 159. 23 On this point, see Genevieve Migely’s essay in this collection. 24 Cf. TVV 42: ‘To perceive is one thing; to judge is another. So likewise, to be suggested is one thing, and to be inferred another.’ 25 The rest of the paragraph provides yet more evidence of my point concerning suggestion and constant conjunction that becomes habitual movement among ideas. 26 I am not implying that we ever perceive incorrectly, only that we might make errors in judgments about what ideas might attend to others. One error of this sort is to associate a sensory idea with an inappropriate single idea (name). Thus I might see an image which suggests a chair to me when in fact (revealed later after I have had additional sensory experiences) the commonsense object is a picture.

Berkeley on Visible Figure and Extension ral p h s c h u mach e r

1. Introduction: The Question of the Status of Visible Figure and Extension The question of how sensory cognition is supposed to be directed to objects in the perceiver’s environment is a central issue of philosophical theories of perception. According to Berkeley, it is the awareness of sensory ideas that makes possible cognizance of real objects. In the following I examine how, according to Berkeley’s theory, visual ideas are supposed to contribute to the mind’s directing awareness to the geometrical properties of physical objects. The claim that the direct objects of sight and touch are entirely different is the central thesis of Berkeley’s criticism of the doctrine of common sensibles.1 Accordingly, only light and colours are immediate objects of sight. He articulates this view, among other places, towards the end of his New Theory of Vision: All that is properly perceived by the visive faculty amounts to no more than colours, with their variations and different proportions of light and shade ... planes are no more the immediate object of sight than solids. What we strictly see are not solids, nor yet planes variously coloured: they are only diversity of colours. (NTV 156, 158)

Two- and three-dimensional shapes, in contrast, are supposed to be exclusive objects of touch, which are the proper objects of geometry. The view that there are no common objects of direct visual and tactile perception is the theoretical background of Berkeley’s negative answer to

Berkeley on Visible Figure and Extension

109

the Molyneux question (NTV 132–6): The newly sighted man is not able to tell at once which of the objects seen by him is the cube, and which the globe, because what he directly perceives by sight are only light and colours but neither two- nor three-dimensional geometrical shapes. At several places, however, Berkeley not only talks of light and colours as direct objects of sight but also mentions what he calls ‘visible figure’ and ‘visible extension,’ which are supposed to be entirely different from tangible shapes (NTV 43–5, 49, 130, 133, 137, 140, 151–9). On his view, visible figure and extension are marks of tangible shapes. These marks are supposed to direct our visual awareness to tangible shapes. Accordingly, we indirectly see tangible shapes by directly seeing the visible marks that stand for them. Considering Berkeley’s repeated claim that only light and colours are direct objects of sight, the following question arises: How do we have to understand the relationship between light and colours, on the one hand, and visible figure and extension, on the other? Is it possible to reconcile his claim that only light and colours are immediate objects of sight with his remarks about visible figure and extension? 2. The Role of Visible Figure and Extension as Immediate Objects of Sight The passages in which he talks about these visible shapes cannot be neglected because the claim that visible figure and extension are immediate objects of sight is a central assumption of the following two arguments. First, this assumption plays an important role for his criticism of the distinction between primary and secondary qualities. This criticism is primarily based on the rejection of abstract ideas.2 In his New Theory of Vision Berkeley argues that visible figure and extension cannot be different from certain patterns of light and colours because we are unable to form abstract ideas of visible figure and extension (NTV 43, 130). Consequently, visible figure and extension are nothing but patterns of light and colours: And as for figure and extension, I leave it to anyone that shall calmly attend to his own clear and distinct ideas to decide whether he had any idea intromitted immediately and properly by sight save only light and colours: or whether it be possible for him to frame in his mind a distinct abstract idea of visible extension or figure exclusive of all colour: and on

110 Ralph Schumacher the other hand, whether he can conceive colour without visible extension? For my own part, I must confess I am not able to attain so great a nicety of abstraction: in a strict sense, I see nothing but light and colours. (NTV 130)

A further consequence that is not explicitly mentioned in this passage is that figure and extension cannot be common objects of sight and touch because there can be no common abstract ideas of figure and extension that represent objects of sight as well as objects of touch. In PHK 10 Berkeley also argues that it is impossible to conceive visible shapes without colour. From this he draws the conclusion that both colours and visible shapes can exist only ‘in the mind.’ Accordingly, at least some of the so-called primary qualities cannot exist without the mind. Yet it is important to note that this argument, contrary to what Berkeley thinks, is in fact restricted to visible shapes because it is based on considerations which at best show that we are unable to abstract visible figure and extension from colour. Since, on Berkeley’s account, only tangible shapes are to be regarded as proper objects of geometry, his argument – by his own theoretical standards – does not show that twoor three-dimensional geometrical shapes can exist only in the mind! For this reason, it does not affect the geometrical properties which, for instance, John Locke discusses under the name of primary qualities. Second, the assumption that visible shapes are direct objects of sight plays an important role for Berkeley’s explanation of misperception. In the Third Dialogue he analyses the visual misperception of a straight stick – that looks broken if held in water – as broken, as the result of the wrong association of the direct visual perception of a broken shape with the expectation of a tactile perception of something broken: Thus in the case of the oar, what he immediately perceives by sight is certainly crooked; and so far he is in the right. But if he thence conclude, that upon taking the oar out of the water he shall perceive the same crookedness; or that it would affect his touch, as crooked things are wont to do: in that he is mistaken. (DHP 238)

Since this analysis explains misperception in terms of wrong inferences about the connections between the different objects of direct sensory perception, it presupposes that visible shapes are direct objects of sight.

Berkeley on Visible Figure and Extension

111

3. Visible Figure and Extension as Patterns of Light and Colours? In view of the importance of the claim that visible figure and extension are directly perceivable by sight, one has to look for an interpretation that reconciles this claim with Berkeley’s view that light and colours are the only direct objects of sight. In the following I am going to examine the interpretation that visible figure and extension are nothing but patterns of light and colours which are regarded as visible shapes, because they are marks of tangible shapes. Although this interpretation seems to be quite plausible at first sight, it faces two serious problems. The first problem is related to Berkeley’s explanation of misperception. According to his approach, we can directly perceive by sight something as broken without this visible figure being a mark of a broken tactile shape. On the one hand, this is advantageous, because it enables his theory to account for the phenomenology of optical illusions: Straight sticks always look broken if held in water, since our knowledge that the stick actually is straight has no influence on its appearance. But on the other hand, this description implies that something can be a visible shape of something broken although it might be associated with the tangible shape of something straight. For this reason, being a certain visible shape must be entirely independent of standing for a certain tangible shape. Consequently, Berkeley’s considerations on misperception support the view that visible shapes are not just combinations of light and colours which are regarded as marks of tangible shapes. The second and more important problem is a consequence of Berkeley’s conception of visible shapes as marks of tangible shapes. On his view, the visible shapes used by us as marks of tangible shapes constitute a ‘universal language’ (NTV 147) or a ‘language of nature’ (NTV 140). However, this does not mean that there is a necessary connection between visible and tangible shapes that constitutes the representational relation holding between them. In contrast, Berkeley emphasizes repeatedly that there is no such necessary relation and that visible shapes do not represent tangible shapes by virtue of their intrinsic properties. There is nothing in their nature that constitutes their representational character. Just like Locke’s simple sensory ideas, visible shapes do not have intrinsic intentionality.3 All Berkeley intends to say by maintaining that visible shapes constitute a ‘universal language’ or a ‘language of nature’ is that independently of human perceivers there exists some

112 Ralph Schumacher

co-variation between certain visible and tangible shapes. Therefore, our interpretation of certain combinations of colours as visible shapes that signify tangible shapes has to be based on the observation of this kind of co-variation. We come to regard certain patterns of light and colours as visible shapes that are marks of certain tangible shapes as the result of the repeated experience of their co-variation: Having of a long time experienced certain ideas, perceivable by touch, as distance, tangible figure, and solidity, to have been connected with certain ideas of sight, I do upon perceiving these ideas of sight forthwith conclude what tangible ideas are, by the wonted ordinary course of Nature like to follow. (NTV 45)

Hence, we have to acquire empirical knowledge about the co-variation of certain patterns of light and colours, on the one hand, and certain tangible shapes, on the other, in order to regard the former as visible shapes that stand for geometrical properties of physical objects. As Berkeley says in the Principles: By a connexion taught us by experience, they [visible ideas] come to signify and suggest them [tangible ideas] to us, after the same manner that words of any language suggest the ideas they are made to stand for. (PHK 43)

It is this kind of empirical knowledge that distinguishes us from the Molyneux man. The man born blind and made to see ‘would not consider the ideas of sight with reference to, or as having any connexion with, the ideas of touch’ (NTV 79). Since he lacks the knowledge of which combinations of colours stand for which tangible shapes, these visual ideas cannot direct his perceptual awareness to geometrical properties of physical objects. Thus, he is unable to tell at once which of the objects seen by him is the cube, and which the globe. As a consequence of this conception, not only does the visual perception of tangible shapes mediated by visible marks depend on empirical knowledge, but also the recognition of combinations of colours as visible shapes, because a certain combination of colours only becomes a visible shape by being associated with a certain tangible shape. Visual awareness of tangible shapes thus involves ideas of two different kinds: firstorder visual ideas; and second-order ideas that refer to these sensory ideas, representing them as standing in relations of co-variation to tangible shapes and making us aware of these tangible shapes. There-

Berkeley on Visible Figure and Extension

113

fore, both kinds of visual perception have to be regarded as cases of indirect perception because Berkeley characterizes indirect perception in terms of inferences and associations based on experience. Visible figure and extension cannot be direct objects of sight, because as objects of immediate awareness visual ideas are bereft of any representational character. 4. The Indirect Perception of Visible Figure and Extension One might object that Berkeley implicitly distinguishes between the following two kinds of perceptual directness:4 (1) Sensory perceptions are direct, if no cognitive intermediaries are involved. (2) Sensory perceptions are direct, if no inferences or associations are involved.5 Accordingly, although the visual perception of tangible shapes is indirect in both senses, because it is mediated by visible marks that have to be regarded as marks of tangible shapes, the visual perception of visual shapes, in contrast, seems to be indirect merely in the second sense, since we need no marks to see visible figure and extension. Therefore, there still seems to be a way of qualifying the visual perception of visible figure and extension as direct. However, since cognitive intermediaries intervene not in a spatial but rather in an inferential sense, this objection is not convincing. They are not spatially located between the perceiver and the object. Instead, we have to regard these intermediaries as starting points for inferences that enable us to perceive other objects for which these intermediaries stand. For instance, let us take Berkeley’s example of the statue of Caesar (DHP 203–4). This statue is not spatially located between the perceiver and Caesar himself but rather in the inferential sense that it is the basis of our judgments about Caesar. Since we have to recognize that certain patterns of light and colours co-vary with certain tangible shapes in order to perceive these patterns as visible figure and extension, cognitive intermediaries in this inferential sense are also involved in the case of visible shapes. In this case the relations of co-variation are the cognitive intermediaries. Therefore, it is also true of the visual perception of visible figure and extension that it is indirect in both of the above-mentioned senses.

114 Ralph Schumacher

Thus, we face the following dilemma: Either visible shapes are contingent marks of tangible shapes, in which case they cannot be direct objects of sight, or they are immediate objects of visual perception, in which case they cannot be contingent marks of tangible shapes that direct our visual awareness to the geometrical properties of physical objects. 5. Berkeley’s and Locke’s Difficulties with the Concept of ‘Seeing As’ Obviously, there is a conflict internal to Berkeley’s thought: On the one hand, he wants to maintain that – contrary to the position he ascribes to Locke – light and colours are exclusive objects of immediate visual perception, because this position is essential for his criticism of the doctrine of common sensibles. On the other hand, he regards visible figure and extension as useful to account for the fact that there is a way in which we do – although only indirectly – perceive geometrical shapes by sight. These visible shapes have to be contingent marks of tangible shapes in order to be compatible with his negative answer to the Molyneux question. Otherwise, if he had assigned intrinsic intentionality to them, the newly sighted man would be able to tell at once which of the objects seen by him is the cube, and which the globe. But the main problem is that Berkeley – just like Locke – fails to recognize that ideas that are not intrinsically intentional need to be regarded as representations in order to be able to direct our awareness to the objects represented by them. This is why he overlooks how the cognitive function which he ascribes to visible figure and extension – namely to direct our visual awareness to tangible shapes – is not compatible with regarding visible figure and extension as immediate objects of sight. Consequently, to ensure that visible figure and extension are really suited to fulfill their assigned cognitive function, it would be necessary to describe them rather as indirect objects of sight. This requires that it must be possible to see patterns of light and colours as visible shapes. Yet this concept of seeing as creates a serious problem for Berkeley’s theory of sensory perception, because his theory does not provide the conceptual resources to account for the influence of judgments on the content of sensory perceptions.6 To illustrate this problem, it is helpful to consider Locke’s example of the visual perception of the three-dimensional shape of a globe of uniform colour.7 On his account, the first stage consists of the visual per-

Berkeley on Visible Figure and Extension

115

ception of ‘a flat circle variously shadowed’ or of a ‘variety of shadow and colour.’ In the second stage we learn to associate this pattern of different colours with a certain tangible shape that co-varies with this pattern. Finally, in the third stage, as the result of a habit established by the repeated experience of the co-variation between our visual and tactile ideas, the pattern of light and colours directs our visual awareness to the object’s geometrical shape: So that from that, which truly is variety of shadow or colour, collecting the Figure, it [the judgment] makes it [the variety of shadow or colour] pass for a mark of Figure, and frames to itself the perception of a convex Figure, and an uniform Colour; when the Idea we receive from thence, is only a Plain variously colour’d, as is evident in Painting. (Essay II.ix.8, 145)

Hence, Locke’s explanation of the influence of judgments on the contents of our visual perceptions of three-dimensional shapes implies that none of our visual sensations initially marks the shapes and colours of the objects of sight. Instead, we have to learn to see the real qualities of things. Locke’s attempt to account for the influence of judgments on perceptual content gives rise to two problems. The first problem is related to his conception of the reality of simple ideas, according to which all simple sensory ideas are real, because they all stand for real qualities or powers of things which give rise to them. If what we immediately perceive by sight is not a three-dimensional globe of uniform colour, but rather a round and two-dimensional surface with patterns of different colours, then our visual ideas of colours cannot be real, because they correspond only to the apparent, but not the real, colours of the globe. Consequently, contrary to Locke’s thesis, it is not true of all simple sensory ideas that they are real. The only way to solve this problem would be to qualify certain perceptual conditions as ‘normal’ and to restrict the definition of the reality of simple ideas in such a way that an idea is real if and only if it signifies its ‘normal’ or ‘right’ cause. The second problem concerns the implication that we have to learn to see the real qualities of things. According to a recent paper of Martha Bolton, this threatens Locke’s empiricism because it tacitly requires the mind to have other ideas than the simple sensory ones – namely, second-order ideas – that are supposed to establish the marking function of the simple ideas.8 The similarity of Locke’s description of the visual perception of a

116 Ralph Schumacher

globe of uniform colour to Berkeley’s own theory is remarkable, since Berkeley would agree with Locke on all three stages. The question is now whether Berkeley also gets into similar problems about the explanation of the influence of judgments on the contents of our experiences. Is he able to offer a consistent description of the transformation of the visual perception of a pattern of different colours into the visual perception of a globe of uniform colour? An important requirement for this task is that his theory must be able to account for the fact that the sensory ideas initially perceived by sight cannot be regarded as the globe’s real properties. To answer this question, let us focus on his considerations concerning the misperception of a straight stick in the Third Dialogue (DHP 238). Does Berkeley distinguish between real and merely apparent properties of physical objects? I think his considerations about misperception support that the answer is ‘no.’ Although the inference from the broken visible shape to the broken tangible shape is actually wrong, the relevant reason for why this inference is incorrect is not that the stick in the water does not have the geometrical property of being broken. Instead, the relevant reason is that under these specific perceptual conditions it is simply wrong to predict that something that now looks broken will also feel broken. As I understand Berkeley, it is his aim to reduce our ordinary talk about real and apparent properties to propositions about correct and incorrect inferences concerning the connections between different perceptions. Accordingly, the distinction between real and apparent properties does not describe a metaphysical difference. Now, is it possible, according to Berkeley, that qualities directly perceived by us are not real properties of physical objects? Since there is no metaphysical difference between real and apparent properties, I think Berkeley would answer that this question does not make sense if we understand it as a philosophical question. All we can say is that under different perceptual conditions we might perceive different qualities. Since we do not have non-arbitrary criteria at our disposal which allow us to qualify some perceptual conditions as ‘normal’ conditions, all qualities immediately perceived by our senses are on a par. Therefore, we cannot be mistaken about them. All possible mistakes are mistakes about the relations between the objects of direct sensory perception: But his mistake lies not in what he perceives immediately and at present (it being a manifest contradiction to suppose he should err in respect of that) but in the wrong judgment he makes concerning the ideas he apprehends

Berkeley on Visible Figure and Extension

117

to be connected with those immediately perceived: or concerning the ideas that, from what he perceives at present, he imagines would be perceived in other circumstances. (DHP 238)

Hence, sensory ideas immediately perceived by sight cannot be regarded as merely apparent qualities. Berkeley therefore is not in the position to maintain that a globe that really is of uniform colour in direct sensory perception merely appears to us as a pattern of different colours. But this is just a precondition for saying that we perceive a pattern of different colours as the visible shape of a globe of uniform colour. For this reason, Berkeley’s theory of sensory perception does not provide the conceptual resources to account for the influence of judgments on perceptual content. Consequently, we cannot even regard visible figure and extension as indirect objects of sight. It is interesting to note, by the way, that both Locke and Berkeley have difficulties accounting for the influence of judgments on perceptual content for the same reason, namely, that within their theories it is not possible to regard the qualities immediately perceived by our senses as merely apparent properties. In both theories this is due to the fact that the epistemological status of sensory ideas is qualified in a particular way which is supposed to guarantee that there can be no doubts about their reality. 6. Concluding Remarks To sum up, we come to the following result. First, Berkeley’s analysis of the misperception of a straight stick that looks broken supports the interpretation that being a certain visible shape must be entirely independent of standing for a certain tangible shape. Second, we have seen that there is no place for visible shapes as contingent marks of tangible shapes within Berkeley’s theory – neither as direct nor as indirect objects of sight. Therefore, both lines of thought contradict the initially suggested interpretation that visible figure and extension are nothing but patterns of light and colours that are regarded as marks of tangible shapes. Since this interpretation was offered to reconcile conflicting elements of Berkeley’s theory of visual perception, we come to the conclusion that he is not able to explain how visual ideas contribute to the mind’s directing awareness to the geometrical properties of physical objects. The best option to avoid this problem within a Berkeleian model of

118 Ralph Schumacher

sensory perception would be to describe a visible shape as a distinct kind of direct object of sight that intrinsically represents a tangible shape. On the one hand, this implies that one has to revise the negative answer to the Molyneux question, because – according to this suggestion – perceivers do not have to acquire empirical knowledge in order to be able to direct their visual awareness to the geometric properties of physical objects. But on the other hand, since it is still possible to regard visible and tangible shapes as different entities, it is also possible to maintain the criticism of the doctrine of common sensibles. Thus, Berkeley should have distinguished more carefully between the following two claims: (1) The immediate objects of sight and touch are entirely different. (2) To direct our visual awareness to the geometric properties of physical objects, it is necessary to acquire knowledge about the co-variation of certain visible and tangible ideas. I think within a Berkeleian model of sensory perception it is possible to maintain proposition (1) without proposition (2) and to replace (2) by the following proposition (3): (3) Since visible figure and extension are intrinsic signs of tangible shapes, empirical knowledge is not required to direct our visual awareness to the geometric properties of physical objects. Hence, the problems internal to Berkeley’s theory of sensory perception lead finally to the interesting question of whether it is plausible from a systematic point of view to claim that even though the immediate objects of sight and touch are different, no empirical information is needed to direct our visual awareness to the geometric properties of physical objects. But this is an empirical question which can only be answered on the basis of psychological research. Notes 1 For overviews of theories of visual perception in the seventeenth and eighteenth centuries, see Margaret Atherton, Berkeley’s Revolution in Vision (Ithaca, NY: Cornell University Press, 1990), part 1; Gary Hatfield, The Natural and the Normative: Theories of Spatial Perception from Kant to Helmholtz (Cambridge, MA: Cambridge University Press, 1990), ch. 2, part 2; Martha Brandt Bolton,

Berkeley on Visible Figure and Extension

2

3

4

5

6

119

‘The Real Molyneux Question and the Basis of Locke’s Answer,’ Locke’s Philosophy: Content and Context, ed. G.A.J. Rogers (Oxford: Oxford University Press, 1994), 83–94; and Lorne Falkenstein, ‘Intuition and Construction in Berkeley’s Account of Visual Space,’ Journal of the History of Philosophy 32 (1994): 63–84. For Berkeley’s critique of the doctrine of common sensibles, see Margaret Dauler Wilson, ‘The Issue of “Common Sensibles” in Berkeley’s New Theory of Vision,’ in her Ideas and Mechanism: Essays on Early Modern Philosophy (Princeton: Princeton University Press, 1999), 257–75. For a clear analysis of Berkeley’s arguments against abstract ideas, see George Pappas, Berkeley’s Thought (Ithaca, NY: Cornell University Press, 2000), ch. 3. This interpretation of Locke’s theory of sensory ideas is convincingly defended by Michael Ayers in his comprehensive study of Locke’s philosophy, Locke: Epistemology and Ontology, 2 vols. in 1 (London: Routledge, 1991), 1:38–42. This distinction between different kinds of perceptual directness is already discussed by Kenneth Winkler, Berkeley: An Interpretation (Oxford: Clarendon Press, 1989), 149–50. According to Marc Hight’s interpretation, Berkeley holds the view that the perception of things is always indirect in this second sense. This interpretation is based on Hight’s ‘single-idea thesis’ (described in his essay elsewhere in this collection). He ascribes to Berkeley the position that commonsense objects are single ideas of the imagination represented by names that signify diverse collections of sensible ideas. It is important to note that this kind of epistemic perception is of an entirely different type from the kind of perception discussed by Pappas (Berkeley’s Thought, 12) and Hight (see essay in this volume). The epistemic perception of an object under a certain description as, for example, an apple implies that the perceiver thereby forms the belief that there is an apple in front of him. However, this belief is not implied in the case of the non-epistemic perception of, for instance, an apple by perceiving a collection of sensory ideas which actually constitute that apple. In other words, one can see an apple without having the belief that there is an apple. It is perfectly possible to say of a person that she perceives an apple (e.g., by seeing a certain collection of sensory ideas) without ascribing the belief to her that there is an apple in front of her. Therefore, when Pappas correctly claims that, according to Berkeley, we see things by seeing certain combinations of sensory ideas, he is talking about non-epistemic perception – and not about the epistemic perception of a combination of sensory ideas under a certain description. This interpretation is also defended by Winkler (Berkeley, 154–60).

120 Ralph Schumacher 7 For a detailed analysis of Locke’s theory of spatial perception see Laura Berchielli, ‘Color, Space, and Figure in Locke: An Interpretation of the Molyneux Problem,’ Journal of the History of Philosophy 40 (2002): 47–65; and Ralph Schumacher, ‘What Are the Direct Objects of Sight? Locke on the Molyneux Question,’ Locke Studies 3 (2003): 41–61. 8 Martha Brandt Bolton, ‘Locke on Sensory Representation,’ in Perception and Reality: From Descartes to the Present, ed. Ralph Schumacher (Paderborn: Mentis-Verlag, 2004), 146–67.

Perceiving and Berkeley’s Theory of Substance p hi l l i p d. c u m mi ns

It is not difficult to prove that Berkeley explicitly affirmed mental substances in A Treatise concerning the Principles of Human Knowledge and Three Dialogues between Hylas and Philonous. Confirming texts abound. Berkeley affirmed mental substance as a distinct and unique kind of entity. He held minds (souls, spirits, perceivers) exist and insisted that they are completely different from sensibles. The latter are either sensible qualities or sensible objects – collections or congeries of sensible qualities.1 Sensibles are what we sense when we sense and what we imagine when we imagine; they are perceived by minds, which on occasion also bring them into existence by acts of will. Unlike sensibles, minds are held to be substances – simple, indivisible, thinking substances. Nonetheless, a number of well-respected Berkeley scholars have expressed doubts about his commitment to mental substance or to the substance tradition.2 They, of course, acknowledge the textual evidence but find in Berkeley no intelligible or consistent account of mental substance or, more generally, no conceptual place for the category of substance. In what follows I shall not directly address these authors. Instead, I shall attempt to show Berkeley had an idiosyncratic but intelligible theory of substance. Beginning with a puzzling passage from the Principles, I shall describe Berkeley’s conception of mental substance and explain why he claimed minds are the only substances. Thereafter, I shall answer some of the main arguments which purport to show that Berkeley could not consistently affirm mental substance.3 1. A Problem Text: Three Questions Answered One of the most important and yet most puzzling passages among Berkeley’s philosophical writings is PHK 7. It is highly important because

122 Phillip D. Cummins

it is where Berkeley first states and defends his fundamental ontological thesis, S, minds and only minds are substances.4 In what follows I hope to show why it is puzzling. PHK 7 begins, ‘From what has been said, it follows, there is not any other substance than spirit, or that which perceives.’ In context this assertion implies (1) No non-minds are substances, and (2) All minds are substances.5 They allegedly follow from ‘what has been said.’ Before assessing this claim, one must address a scope problem regarding (1). In PHK 1 and 2 Berkeley distinguished between sensibles and minds and implied that nothing is both a sensible and a mind. A sensible is an idea both in the sense that it is an immediate object of sensory awareness and in the sense that for it to be is to be perceived.6 The first meaning is introduced in PHK 1, the second at the end of PHK 2. Since sensibles are definitely non-minds, thesis (1) implies that no sensibles are substances. However, ‘non-mind’ is not equivalent to ‘sensible.’ Since a sensible is what is perceived and a mind is what perceives, the distinction between them allows for a third kind of entity, a non-sensible non-mind, something which neither perceives nor is perceived. Without implying their existence, (1) denies any nonsensible non-mind is a substance. Thus it is not unreasonable to take Berkeley to be claiming that from what has been said he has established three claims: (1a) No sensible is a substance; (1b) No non-sensible nonmind is a substance; and (2) All minds are substances. Having fixed Berkeley’s ontological thesis, let us return to the first sentence of PHK 7, ‘From what has been said, it follows, there is not any other substance than spirit, or that which perceives.’ From this one might infer that Berkeley is drawing a summarizing conclusion from previously stated arguments. But whether one examines all the sections of the Principles which precede this one or, more reasonably, just the preceding six sections, one discovers this is not the case. There are no such explicit arguments. Moreover, until PHK 7 there is no reference whatsoever to non-sensible non-minds or to substance. Therefore, any arguments for (1a), (1b), and (2) in PHK 1 through 6 were implicit. This being so, one expects Berkeley to make those arguments at least partially explicit in the continuation of PHK 7. Minimally, this would involve specifying what it is to be a substance, why minds qualify, and why non-minds do not. His initial words, ‘But for a fuller proof of this, let it be considered,’ suggest this approach, but what follows disappoints. Here is PHK 7 in its entirety: From what has been said, it follows, there is not any other substance than spirit, or that which perceives. But for the fuller proof of this point, let it be

Perceiving and Berkeley’s Theory of Substance

123

considered, the sensible qualities are colour, figure, motion, smell, taste, and such like, that is, the ideas perceived by sense. Now for an idea to exist in an unperceiving thing, is a manifest contradiction; for to have an idea is all one as to perceive: that therefore wherein colour, figure, and the like qualities exist, must perceive them: hence it is clear there can be no unthinking substance or substratum of those ideas.

Notice that Berkeley neither defines ‘substance’ nor indicates informally what is required to be one. Moreover, he offers no explicit reason why minds are substances. Indeed, he offers no reason why sensibles cannot be substances. The ‘fuller proof’ turns out to be Berkeley’s attempt to disprove the existence of a special kind of non-sensible nonmind – substance as unthinking substratum of sensible qualities.7 In other words, it is a partial argument for (1b), but since non-sensible non-minds were not even mentioned in sections 1 through 6, it is difficult to see how Berkeley could be making explicit in PHK 7 a prior argument about them. One’s puzzlement should deepen when the argument is closely examined. It begins with the words, ‘let it be considered, the sensible qualities are colour, figure, motion, smell, taste, and such like, that is, the ideas perceived by sense.’ Given the arguments of PHK 3 through 6, Berkeley surely means that the qualities just specified are ideas in the second sense of ‘idea.’ Hence we can reformulate his initial claim as (a) sensible qualities cannot exist unperceived. Berkeley concludes by denying the existence of an unperceived and unperceiving substance. His explicit conclusion is ‘there can be no unthinking substance or substratum of those ideas.’ Were there such a substratum, sensible qualities would exist in it, but Berkeley insists (b) sensible qualities cannot exist in unperceiving substance. It is far from obvious, however, why one cannot affirm (a) but deny (b). It seems possible to conceive a situation in which one or more sensible qualities exist in an unperceived and unperceiving substance, thus violating (b), yet, as is required by (a), are perceived by some mind distinct from that substance. Berkeley rejects this scenario as impossible, writing, ‘Now for an idea to exist in an unperceiving thing, is a manifest contradiction; for to have an idea is all one as to perceive: that therefore wherein colour, figure, and the like qualities exist, must perceive them.’ If we replace ‘idea’ with ‘sensible quality,’ it becomes apparent that Berkeley’s crucial premise is (h), to have a sensible quality is all one as to perceive it. Its role in securing (b) is best understood by considering again the scenario according to which an unperceived and unperceiving substance has a

124 Phillip D. Cummins

quality which must be perceived to exist and which, accordingly, is perceived by a mind distinct from the substance having the quality. If, as is required by (h), to have a quality is to perceive it, the substance having the quality must perceive, contrary to the scenario’s stipulation that it is an unperceiving substance. Put differently, the supposed substratum both perceives – given (h), in virtue of having the quality – and is an unperceiving thing (by hypothesis), so is a manifest impossibility. Although the argument of PHK 7 is made stronger, perhaps even valid, with the addition of (h), puzzlement remains. Why should one accept (h)? Why is having a perceiver-dependent quality nothing more than perceiving it? Why not distinguish them? Obviously, Berkeley made no such distinction. But why not? The explicit argument thus raises the third of three pressing questions about PHK 7. The first is why does he give no argument for (1a) – that is, why does Berkeley write as if he has already established that no sensible objects are substances? The second is why does he offer no argument for (2) – that is, why does he write as if he has already established that spirits, that is, perceivers, are substances? Finally, to repeat, how can he justify his fundamental premise, (h), that to have a sensible quality just is to perceive it? Answering these questions provides an account of how Berkeley defended (1a), (1b), and (2), and thus his key claim, (S), minds and only minds are substances. The remainder of this section will be devoted to this task. I begin with Berkeley’s implicit argument for (1a), no sensible is a substance. The Principles (Part I) begins with a list of the objects of human knowledge. Characterizing the first such objects, ideas of sense, amounts to offering a partial enumeration of the different kinds of sensible qualities and claiming that sensible objects are combinations of them that are provided with names and regarded as individual things. In PHK 2 minds are introduced as a distinct set of beings. What are they? Perceivers. In the final sentence of PHK 2 Berkeley asserts that the existence of sensibles consists in being perceived. For them to be is to be perceived. What does this imply for sensibles? For one thing, that their existence presupposes perceiving and something that perceives them. Clearly, for Berkeley, no sensible quality or combination of sensible qualities can perceive. Consequently, none can exist unless something else, a mind or minds, does what is required for them to exist. The ground of a sensible’s existence lies outside itself. It is thus a dependent entity. The dependence concerns existence but is not causal. One might call it ontological dependence.

Perceiving and Berkeley’s Theory of Substance

125

Consider next two principles for determining whether or not an entity is a substance. The first is that if an entity is an ontologically dependent entity, then it is not a substance. The second is that if an entity is not an ontologically dependent entity, but some other entity depends upon it, then it is a substance. One could plausibly claim that given the first substance principle and the arguments of PHK 3 through 6, it follows that sensibles are not substances. They have a dependent existence that is distinct from causal dependence. What about spirits? Notice, first, that in PHK 7 ‘spirit’ is defined or explicated as that which perceives. Consequently, S amounts to Only perceivers are substances. Now, according to Berkeley, for a spirit, that is, a perceiver, to be is to perceive. That is, for things which perceive, existence consists in perceiving. Perceiving is thus considered essential to being a spirit. The occurrence of a perceiving guarantees the existence of a spirit. Note, further, that perceiving is something a perceiver does. Unlike the case of sensibles, the state that constitutes or grounds the existence of a mind is something it has or does, not a state bestowed on it in virtue of what some other thing has or does. A mind, then, is not ontologically dependent. Further, by perceiving, a mind also does what is required for a sensible’s existence. Sensibles are ontologically dependent entities and minds are the entities upon which they depend. Because the latter satisfy the second principle, they are substances. This, too, can plausibly be regarded as a conclusion implicit in PHK 3 through 6. Turning next to the PHK 7 argument for (b), we recall that a sensible is either a sensible quality or a collection of sensible qualities. As was noted in the last paragraph, for a sensible quality to be is to be perceived, and being perceived requires a perceiver distinct from the entity perceived. An existing sensible quality thus requires a mind, that is, a perceiver, to ground its existence. My suggestion is that given the perceiver dependence of sensible qualities, Berkeley translates the traditional principle that qualities cannot exist without a substance in which to exist into a principle regarding perception. Sensible qualities are dubbed ‘ideas’ in part because for them to be is to be perceived. For the same reason, I suggest, Berkeley asserts (h), that to have a sensible quality just is to perceive it. Berkeley’s statement in PHK 7, ‘to have an idea is all one as to perceive,’ indicates that perceiving is meant to give content to the supposed relationship of having or supporting. Given this position, a quality perceived by a mind requires no other substance besides mind to have or support it. As for an unperceiving substance having, that is, perceiving, qualities, we now should be able to appreci-

126 Phillip D. Cummins

ate Berkeley’s contention that that is a manifest contradiction. Consequently, unless one can provide a new and intelligible sense for ‘having a quality’ or ‘supporting a quality,’ a project Berkeley dismisses in PHK 16, 17, and 49, one cannot easily accept esse est percipi and the primacy of perceiving over being perceived and, nonetheless, reject Berkeley’s PHK 7 claim that there cannot be an additional unperceiving substance which supports sensible qualities. 2. Berkeley’s Conception of Substance One useful way to explicate and assess Berkeley’s pivotal claim that for sensibles to be is to be perceived is to consider the various reasons he offers for accepting it. These range from announcements in the Principles that it is a self-evident truth discovered by reflecting on the meaning of ‘exist,’ through repeated attempts in the Dialogues to equate sensibles with sensations or to infer perception dependence from perceptual variation. Another approach, followed here, is to examine the inferences it generates. First comes the conclusion that the existence of a sensible, that is, its being perceived, presupposes perceiving. Perceiving in turn requires a perceiver – something to do the perceiving. Nothing can fill the role, perceived, unless something fills the role, perceiver. Hence, the existence of a sensible implies the existence of a perceiver. Of course, were a single entity able to be both a perceived item and that which perceives or a component of that which perceives, the requirement that perceiving requires both a perceived and a perceiver could be satisfied without supposing two genuinely distinct entities, each filling one and only one role. In both the Principles and the Dialogues, Berkeley regards perceiving and being perceived as not just different but as contraries. He never permits the possibility that a sensible quality can perceive itself or some other sensible quality.8 Furthermore, no combination of sensible qualities can perceive itself, any other sensible quality, or any combination of sensible qualities. In short, the requirement for a perceiver of an existing sensible cannot be satisfied solely with reference to sensibles. An entity of a different kind is needed. Thus, the existence of a sensible, that is, its being perceived, implies the existence of a mind, a perceiver different in both number and kind from the sensible. Through its perceiving, a mind is implied by the existence of a sensible. Let us next consider how this can be combined with a commitment to substance. By ‘substance’ I mean, minimally, an individual which has strict unity at a time and strict identity over time and secures unity and

Perceiving and Berkeley’s Theory of Substance

127

identity for its properties or actions. A substance at a time is one thing, a single individual, in a strict sense. It unifies a class of properties or actions by uniting each to itself, so that the individual and its states at a given time constitute one thing in a derivative but genuine sense. In the language of C.D. Broad, unity of system presupposes unity of centre. Over time a substance remains one and the same thing in a strict sense; by its strict identity and the derivative unity it provides for its states, it secures genuine but derivative identity over time for the system comprising it and its states even when those states change over time. To be a substance, however, is not merely to be an enduring individual. It is also to be an ontologically primary thing, something upon which other, subordinate, entities, depend for their existence. On this construal to be a substance is to be a non-causal support or ground for the existence of a dependent entity. My interpretive hypothesis is that Berkeley provided content or determinate meaning to ‘support,’ as applied to substances and sensible objects, in terms of perceiving, an activity which could plausibly be said to yield a contrast between a subordinate entity, the sensible whose existence consists in and depends upon being perceived, and a superior entity, the mind, whose awareness or perceiving is an indispensable condition for the existence of a sensible. The perceiver, as non-causal ground for the existence of an ontologically subordinate sensible, fills the role of substance. This not only secures the claim that minds are substances, but also and more importantly, totally precludes non-perceiving substance. It does so by making perceiving the experiential meaning of ‘supports,’ the relation with reference to which substance had been defined. As in the case of ‘exists,’ which can have more than one concrete (i.e., non-abstract) meaning but must have a concrete meaning appropriate to the kind of entity of which it is affirmed or denied, so in the case of substance, Berkeley did not absolutely equate substance with perceiver but inculcated the doctrine that relative to sensibles, substance can only be made intelligible in terms of perceiving and perceivers, since only with reference to them could the support function and the ontological dependence of sensible qualities and objects be made intelligible. Unlike Hume, who questioned the doctrine of substance in part because ontological dependence seemed to him unintelligible or vacuous (see especially THN I.iv.6, 232–4), Berkeley retained and employed the contrast between subordinate and superior entities embedded in the substance tradition. However, I contend, he rethought ontological

128 Phillip D. Cummins

dependence. Within the substance tradition a quality existed in and was dependent upon the substance of which its name was predicated, so that to be for qualities was to inhere in (be supported by) a substance. Berkeley rejected the relation of inherence, at least for sensibles and perhaps entirely, and used perceiving to redefine dependence on a substance. This use of the substance tradition is made especially clear in PHK 91, where he wrote: It were a mistake to think, that what is here said derogates in the least from the reality of things. It is acknowledged on the received principles, that extension, motion, and in a word all sensible qualities, have need of a support, as not being able to subsist by themselves. But the objects perceived by sense, are allowed to be nothing but combinations of those qualities, and consequently cannot subsist by themselves. Thus far it is agreed on all hands. So that in denying the things perceived by sense, an existence independent of a substance or support wherein they may exist, we detract nothing from the received opinion of their reality, and are guilty of no innovation in that respect.

In the passage here quoted Berkeley claims orthodoxy for his position that a quality is a subordinate or dependent entity. He claims innovation or special insight only in discovering the true nature of the dependency and identifying the substance a quality requires. PHK 91 continues: All the difference is, that according to us the unthinking beings perceived by sense, have no existence distinct from being perceived, and cannot therefore exist in any other substance than those unextended, indivisible substances, or spirits, which act, and think, and perceive them.9

Note in this passage the ‘therefore’ which marks the transition from Berkeley’s statement of the genuine dependence of sensible qualities to his specification of the kind of substance required by that dependence. It shows, I think, the linkage Berkeley found or inserted between esse est percipi and the doctrine of substance. Earlier, in PHK 73, the position of PHK 91 had been expressed in such a way that the traditional substance doctrine was completely captured by esse est percipi and the deduction of a perceiver upon which the sensible quality depends. In the process of uncovering and invalidating the motive which supposedly induced commitment to material substance, Berkeley writes:

Perceiving and Berkeley’s Theory of Substance

129

First therefore, it was thought that colour, figure, motion, and the rest of the sensible qualities or accidents, did really exist without the mind; and for this reason, it seemed needful to suppose some unthinking substratum or substance wherein they did exist, since they could not be conceived to exist by themselves. Afterwards, in process of time, men being convinced that colours, sound, and the rest of the sensible secondary qualities had no existence without the mind, they stripped this substratum or material substance of those qualities, leaving only the primary ones, figure, motion, and such like, which they still conceived to exist without the mind, and consequently to stand in need of a material support. But it having been shewn, that none, even of these, can exist otherwise than in a spirit or mind which perceives them, it follows that we have no longer any reason to suppose the being of matter. Nay, that it is utterly impossible there should be any such thing, so long as that word is taken to denote an unthinking substratum of qualities or accidents, wherein they exist without the mind. (PHK 73)

It is important for Berkeley’s argument that matter is taken to signify ‘unthinking substratum.’ Matter, meaning material substance, is impossible because being an unthinking thing precludes perceiving sensible qualities, the only way in which those qualities can be supported, given their newly revealed nature. Incapable of supporting qualities in the only way allowed as intelligible, an insentient thing cannot be a substance. It cannot perform the substance function. Notice, too, that throughout the passage the subordinate status of sensible qualities is not challenged. Instead, Berkeley claims that human beings, philosophers presumably, aware of the dependence of qualities, but ignorant of the true ground of their existence and, thus, of their dependence, have posited a vaguely conceived type of entity, a substratum, to function as that ground. When the genuine ground of the ontological dependence of sensible qualities is understood – when, that is, ‘existing’ is found to mean ‘being perceived’ and ‘existing in’ is found to mean ‘being perceived by’ for each and every kind of quality, not just secondary qualities – one realizes that the only legitimate or intelligible sense of ‘substance’ is ‘that which perceives,’ in which case no unperceiving thing can be a substance. A familiar passage in the Dialogues reinforces these points while also proving Berkeley denied a sensible can perceive and affirmed mental substance. In the Third Dialogue, trying to find inconsistency in Philonous’s position, Hylas states the following objection:

130 Phillip D. Cummins Notwithstanding all you have said, to me it seems, that according to your own way of thinking, and in consequence of your own principles, it should follow that you are only a system of floating ideas, without any substance to support them. Words are not to be used without a meaning. And as there is no more meaning in spiritual substance than in material substance, the one is to be exploded as well as the other. (DHP 233)10

Hylas claims that Philonous holds that the word ‘substance’ lacks meaning and so is not to be used; but after dismissing material substance (presumably because ‘material substance’ is meaningless), he then inconsistently affirms spiritual substance. Note well the choice not made in Berkeley’s response. He does not have Philonous first assign a seemingly neutral characterization to substance and then provide reasons why minds and only minds actually fit that characterization. Instead he writes: How often must I repeat, that I know or am conscious of my own being; and that I myself am not my ideas, but somewhat else, a thinking active principle that perceives, knows, wills, and operates about ideas. (DHP 233)

Berkeley through Philonous claims to know or be conscious of the self and claims, further, that the self of which one is conscious is not only numerically distinct from the ideas one perceives, but also a thinking active principle. An argument for the second claim follows. In presenting it Berkeley employs his key inference that since a sensible cannot perceive, its existence requires a perceiver in addition to itself, an entity which differs from it in both number and kind. Philonous continues: I know that I, one and the same self, perceive both colours and sounds: that a colour cannot perceive a sound, nor a sound a colour: that I am therefore one individual principle, distinct from colour and sound; and, for the same reason, from all other sensible things and inert ideas. (DHP 233–4)

Berkeley here argues that no sensible quality or combination of sensible qualities can fill the role of perceiver, which must therefore be an individual principle distinct from the colour, the sound, and all other sensible qualities and things.11 Philonous’s answer to Hylas concludes by addressing the original

Perceiving and Berkeley’s Theory of Substance

131

charge that for Berkeley ‘substance’ has no more meaning when applied to spirit than when applied to bodies. It reads: Farther, I know what I mean, when I affirm that there is a spiritual substance or support of ideas, that is, that a spirit knows and perceives ideas. But I do not know what is meant, when it is said, that an unperceiving substance hath inherent in it and supports either ideas or the archetypes of ideas. There is therefore upon the whole no parity of case between spirit and matter. (DHP 234)

Substance here is conceptualized and validated by Berkeley with reference to perceiving, that is, with reference to mind, that which perceives. There are intelligible notions of supporting and of substance in the case of minds and its objects. Supporting is perceiving. Substance is that which perceives. In Philonous’s overall argument for spiritual substance, self-as-substance is construed as self-as-perceiver. In contrast, material substance is defined as an unperceiving substance which supports sensible objects or the sensible qualities they comprise; so defined, it is a contradiction in terms, since ‘support’ means ‘perceives.’ 3. Substance, Autonomy, and Superordination A misunderstanding must now be prevented. To articulate its status as ontologically primary, one could stipulate that a substance is that which can exist alone. Call this the autonomy conception of substance. According to it, a substance is of a superior ontological category inasmuch as it is an autonomous being, a thing which can exist in the absence of any other entity, save perhaps God.12 In contrast, modes or accidents cannot exist alone; they exist in or are conceived through something else. An example may illustrate this conception of substance. One might deny that someone’s smile on some occasion is a substance on the ground that one can only conceive of a smile as an arrangement of the features of a face. Similarly, a face is not a substance inasmuch as it is conceived of as an arrangement of the features of the front surface of a head. In contrast, a head or, perhaps, the body of which it is the head is a substance. Another example: a cube can be conceived as an object without being conceived as a state or modification of some further thing; a surface of the cube is conceived with reference to the cube. Berkeley seems to reject this sense of the autonomy of substance

132 Phillip D. Cummins

explicitly in the Principles. Responding in PHK 67 to an attempt to introduce matter or material substance as a bare, negatively conceived substratum, he writes, ‘It seems no less absurd to suppose a substance without accidents, than it is to suppose accidents without a substance.’ This is as it should be, given his position that perceivers are substances and non-perceivers are not. Consider the existence of perceivers. Just as he held a sensible’s existing is merely its being perceived, so Berkeley held that for minds esse est percipere. The existence of a mind consists in perceiving. Because of the nature of perceiving, conceiving a universe containing perceivers and their perceivings but absolutely nothing else is a conceptual impossibility. Perceiving requires an object perceived.13 In other words, perceiving and being perceived are correlatives, and correlativity cuts both ways. For there to be a perceived there must also be a perceiver; equally, however, for there to be a perceiver there must also be a perceived.14 Can a mind play both roles and thereby exist alone? No. Berkeley’s thesis that no single entity can be both perceived and perceiver, either simultaneously or successively, also cuts both ways. ‘No sensible is a perceiver’ implies and is implied by ‘No perceiver is a sensible.’ Hence, a perceiver requires a sensible object distinct from itself, just as a sensible object requires a perceiver distinct from itself.15 Once again there appears to be reciprocity between perceiver and perceived. Neither can exist alone. Hence, on the autonomy conception, neither can be a substance. It is my contention that Berkeley had available to him and utilized an alternative criterion for being a substance. Its central conception is ontological dependence, which is a non-causal dependence that subordinates one entity or class of entities, the dependent, to a second, that upon which the former depends for its existence. Like causal dependence, ontological dependence concerns existence. If the existence of an entity of one type presupposes something done by an entity of a second kind, but not conversely, then the former is ontologically dependent upon the latter. If, further, the latter is not similarly dependent upon an entity of a third kind, it is a substance. A substance may be said to support the dependent entity because it or something it does is required for the existence of the latter. Even if neither the dependent entity nor the supporting entity ever occurs alone, the latter is a substance.16 Though this superordination conception of substance is logically distinct from the autonomy conception, it is very abstract, and its central element, ontological dependence, might seem quite vague. As I interpret him, Berkeley made it determinate with reference to perception

Perceiving and Berkeley’s Theory of Substance

133

and its objects, claiming that sensibles are ontologically dependent entities, whereas perceivers are substances. Since, for him, the existence of a sensible is its being perceived, which presupposes perceiving and a perceiver, and since no sensible can perceive, the existence of a sensible is a dependent existence. In contrast, the existence of a mind is defined in terms of perceiving, a state or activity of its own. Its existence is thus not other-dependent even though its perceiving must always have an object, and so it cannot be conceived to exist alone. The ground of its existence is its own perceiving. In sum, for Berkeley, the required noncausal dependence is perception dependence, supporting is perceiving, and to be a substance is to be a mind. A somewhat different but related point concerns singular statements about perception. When ‘perceives’ is combined with a pair of singular terms to make a complete assertion – for example, Mary perceives a tree – what is implied about the two relata differs from many other singular relational statements. ‘Perceives’ and similar terms such as ‘sees,’ ‘hears,’ and ‘smells’ imply non-interchangeable roles for their two relata. That which perceives is an agent or doer; that which is perceived is not. A contrasting case may make this clearer. If a piece of paper, M, touches a second, N, then, equally, N touches M. M and N contribute equally to the relationship and have interchangeable roles. It would, perhaps, be more perspicuous to say M and N are touching. In contrast, if P perceives O, P and O do not have interchangeable roles. Furthermore and more importantly, P, as perceiver, does something. Hence it is not just an element in a state of affairs; it is an agent in at least a minimal sense. To the other, O, something is done. It is an object of perception in virtue of what P does. If one wants to portray perceiving as a two-term relation, so that the perceived and the perceiver are correlatives, one can still insist that only one, the perceiving subject, is active. Because perceiving by its nature takes an object, there is another relatum, the perceived, but doing or activity belongs only to the perceiver. Hence, it is the dominant partner which supports the dependent sensible object. It is, therefore, the substance. This contention about ontological dependence receives analogical support from the case of causation.17 Cause and effect are correlatives. Just as there cannot be an effect without a cause, so there cannot be a cause without an effect.18 Nevertheless, effect and cause are not on a par; they are not interchangeable roles. We insist effects depend upon and are subordinated to their causes. We deny the converse. Why? Because the cause is thought to bring about the effect. Hence, the cause is the dominant

134 Phillip D. Cummins

partner, the primary entity in the relationship. I am not maintaining that Berkeley regards perceiving and causing as identical or strictly parallel; I am only insisting that Berkeley wants us to recognize for perceiving and perceivers a set of implications familiar from causing and causes. Perceiving is primary, though correlative, and minds are ontologically dominant, though not ontologically autonomous. My argument has been that despite the correlativity of perceiving and being perceived, Berkeley had two ways to establish an order of dependence and thus provide a basis for construing supporting as perceiving and holding that to be a substance is to be a mind. First, the existence of the perceived is other-implying; the existence of a mind is based on its own state, its perceiving. Moreover, whereas being perceived is not doing, perceiving is. 4. Perception, Inherence, and Predication The position I am ascribing to Berkeley is that he provided content to the substance doctrine in terms of esse est percipi and the primacy of perceiving over being perceived. This position should not be identified with that ascribed to him by the so-called inherence interpretation.19 Even though both construe Berkeley as a substance philosopher, they differ fundamentally. On my interpretation, the subordination of sensibles to minds which generates an asymmetry regarding substance is based on esse est percipi. A sensible must be perceived to exist. Consequently, its existing in a substance just is its being perceived by an entity distinct from itself, which by perceiving it does what is required for its existence. On the inherence interpretation, in contrast, a subordinate being’s existing or inhering in a substance is taken as basic, so that a sensible’s dependence on its perceiver is understood as its having to inhere in a substance, specifically, a mental substance. To say a sensible quality must be perceived to exist is to say that (i) a quality, which cannot exist by itself, can only exist as a quality of a substance, and (ii) minds, which perceive them, are the only substances. Hence esse est percipi requires the premise that there are no substances besides minds, whereas on my account the former is employed as a premise to secure the latter. The inherence account also seems to portray a mind’s perceiving a sensible quality as that quality’s inhering in the mind, so that a mind’s perceiving a sensible object is the inhering in it of the sensible qualities which taken together constitute the sensible object.20 My interpretation involves no such claim.

Perceiving and Berkeley’s Theory of Substance

135

I believe Berkeley realized that these two positions terminologically resemble each other closely enough for an unwary reader to conflate them. To demonstrate the difference between them he stated an objection based on an apparent consequence of the inherence theory, then explained why it did not apply to his position. Here is the beginning of PHK 49: It may be objected, that if extension and figure exist only in the mind, it follows that the mind is extended and figured; since extension is a mode or attribute, which (to speak with the Schools) is predicated of the subject in which it exists.

After announcing in PHK 2 that the existence of an idea consists in being perceived, Berkeley in PHK 3 had asserted of sensible objects, unthinking things, ‘Their esse is percipi, nor is it possible they should have any existence, out of the minds or thinking things which perceive them.’ From this point forward Berkeley commonly employed the contrast pairs ‘in/out’ and ‘within/without,’ claiming repeatedly that sensible objects and the qualities of which they are composed exist in or within the mind and cannot exist out of or without the mind. Because of the frequent use of these terms, especially ‘in the mind,’ the objection of PHK 49 is not unprepared. It also has considerable plausibility when directed against the inherence interpretation. Recall that on it, a mind’s perceiving a sensible is interpreted as the sensible’s inhering in the mind, that is, being a mode or quality of the mind. Since on the traditional doctrine of substance, a quality, that is, the name of a quality, is predicated of the substance in which it inheres, the inherence interpretation of perceiving might be thought to require Berkeley to hold that the mind (the soul) is extended. This would be a very embarrassing result, committed as he was to proving the natural immortality of the soul on the basis of its being unextended (see, for example, PHK 141).21 Berkeley emphatically rejects predicating perceived qualities of their perceivers, that is, treating them as states of their perceivers. In so doing, he provides textual evidence sufficient for rejecting the inherence interpretation of esse est percipi. Far from perceiving being explicated as inherence, Berkeley insists that the only genuine meaning of ‘in the mind’ is ‘perceived by the mind.’ Here is his answer to the objection of PHK 49: I answer, those qualities are in the mind only as they are perceived by it,

136 Phillip D. Cummins that is, not by way of mode or attribute, but only by way of idea; and it no more follows, that the soul or mind is extended because extension exists in it alone, than it does that it is red or blue, because those colours are on all hands acknowledged to exist in it, and no where else.

The crucial words are ‘qualities are in the mind only as they are perceived by it.’ This response would be patently ineffective, not to say disingenuous, were, first, being perceived itself construed as inhering in and, second, being predicated of tied to inhering in. To save Berkeley from this result, the inherence interpretation breaks the second of these links and divorces predication from inherence. This move invites the criticism that, so divorced, inherence is inexplicable. In the substance tradition inherence was introduced as the ontological relationship underlying predication and thus signified by singular propositions such as ‘Socrates is short.’ Once the divorce occurs inherence can no longer be so explicated. Of course, one could interpret inhering as perceiving, but then one cannot retain the first link, that is, explicate perceiving as inhering, without circularity.22 Consequently, I submit, we should take Berkeley to be explicating ‘being in the mind,’ that is, inhering, in terms of ‘being perceived by the mind.’ Since a quality is in the mind only by way of idea, that is, only as being perceived by that mind, there is no basis for predicating it of the mind as there would be were being perceived interpreted as inhering in. Berkeley’s response to the objection of PHK 49 thus provides strong, perhaps conclusive, evidence that perceiving is being used to provide content to the substance doctrine rather than being explicated in terms of an already intelligible inherence model. In the Third Dialogue Philonous summarizes his position for the judgment of those who possess ‘plain common sense, without the prejudices of a learned education.’ The passage reinforces my interpretation of the import of PHK 49: That there is no substance wherein ideas can exist beside spirit, is to me evident. And that the objects immediately perceived are ideas, is on all hands agreed. And that sensible qualities are objects immediately perceived, no one can deny. It is therefore evident there can be no substratum of those qualities but spirit, in which they exist, not by way of mode or property, but as a thing perceived in that which perceives it. I deny therefore that there is any unthinking substratum of the objects of sense, and in that acceptation that there is any material substance. (DHP 237)23

Perceiving and Berkeley’s Theory of Substance

137

Shortly thereafter, Philonous clarifies further what is meant and not meant by ‘in the mind,’ writing: Look you, Hylas, when I speak of objects as existing in the mind or as being imprinted on the senses; I would not be understood in the gross literal sense, as when bodies are said to exist in a place, or a seal to make an impression upon wax. My meaning is only that the mind comprehends or perceives them; and that it is affected from without, or by some being distinct from itself. (DHP 250)

It is plain, then, that in Dialogues as well as the Principles, to be in the mind is not to inhere in the mind; it is to be perceived by it. As might be expected, the issue of inherence and predication is more complex. The continuation of PHK 49, to which I now turn, raises an important question concerning Berkeley’s position on predication and its implications regarding substance. It reads: As to what philosophers say of subject and mode, that seems very groundless and unintelligible. For instance, in this proposition, a die is hard, extended and square, they will have it that the word die denotes a subject or substance, distinct from the hardness, extension and figure, which are predicated of it, and in which they exist. This I cannot comprehend: to me a die seems to be nothing distinct from those things which are termed its modes or accidents. And to say a die is hard, extended and square, is not to attribute those qualities to a subject distinct from and supporting them, but only an explication of the meaning of the word die.

This passage contrasts the account routinely offered by philosophers of the logical structure of ‘A die is hard, extended, and square’ and similar propositions, with the supposedly correct, that is, Berkeleian, account. It might be thought to raise the following problem for my interpretation. On the basis of the expression ‘As to what philosophers say of subject and mode; that seems very groundless and unintelligible,’ one might argue that Berkeley intends to provide a completely general argument regarding how one should analyse all singular propositions, an account that eliminates any logical subject corresponding to the grammatical subject. This supports the further claim that Berkeley implicitly, but completely, eliminates the category of substance. If the grammatical subject of a proposition is just a place marker for the several predicates used to explicate it, it does not denote a substance, spir-

138 Phillip D. Cummins

itual or material, in addition to the qualities signified by the several adjectives predicated of the grammatical subject. From this line of thought one might be tempted to conclude that Berkeley implicitly reveals in PHK 49 that spiritual substance is to be rejected along with material substance because he explicitly replaces the substance analysis of predication with a bundle-of-qualities analysis. In relation to predication, the category of substance offers unity by providing a common entity which has qualities whose names are predicated of it. Berkeley’s PHK 49 account of singular referring expressions is supposed to eliminate substance by replacing unity of centre with unity of system or merely conventional unity. Hence, substance has no role to play in Berkeley’s system. Nevertheless, he continued to employ the term as though it had some theoretical meaning. He was therefore inconsistent in affirming spiritual substance. So goes the argument. This objection gains its force by generalizing from Berkeley’s comments on singular statements about sensible objects to all singular statements, including statements about minds. However, one need not so generalize. There are two less radical, equally text-compatible alternatives regarding the implications of Berkeley’s account of the proper analysis of ‘A die is hard, extended and square.’ One is that it concerns only singular propositions about sensible objects. It is not intended to be a completely general claim about the logical structure or ontological implications of all singular propositions employing the ‘is’ of predication. On this alternative nothing said about ‘A die is hard, extended and square’ applies to sentences like ‘I am sad’ or ‘I am thinking.’ It might be argued in support of this position that by PHK 49 Berkeley had already denied sensible objects are substances and proclaimed the impossibility of unperceived material substances, so that it was to be expected that he would reject a substance/mode analysis for singular propositions about such things as dies, tables, apples, and the like. In contrast, mental or spiritual substance, as the individual principle which perceives, thinks, and wills, had already been endorsed, so that in the case of propositions about minds there is something to fill the role of logical subject, a perceiving substance. The other alternative, which I prefer and shall use later for purposes of illustration, concedes that Berkeley’s comments were intended to apply to all singular propositions which purport to employ an ‘is’ of predication. On this interpretation, rather than signifying a genuine distinction between subject and mode, all such propositions appropriately and perspicuously merely use the predicate expression to expli-

Perceiving and Berkeley’s Theory of Substance

139

cate the subject term, which therefore does not signify a substance distinct from the qualities signified by the predicates which perform the explication. Notice, however, that this generalized analysis of ‘is’ predications is far less unfriendly to the category of substance than might be supposed. Consider my interpretation, according to which Berkeleian minds, that is, Berkeleian substances, are active beings, doers. They perceive, they think, they believe, they will. Hence, Berkeley could argue, with no inconsistency, that the grammatical form which truly exhibits the ontological structure of facts regarding minds, mental substances, is subject–verb–object rather than subject–copula–adjective. The perspicuous grammatical form for propositions about minds is illustrated by ‘He hears a sound’ and ‘I love Jerry’ rather than ‘Leon is sad’ and ‘I am angry.’ On this view, sentences of the form, S is P, subject–copula–adjective, either properly explicate the subject term, reflecting a whole/part structure of sensible objects and their qualities, or have a misleading grammatical form and, thus, should be translated into the subject–verb– object form so as to perspicuously represent a state of affairs involving an active substance, its acts, and their objects. It might be objected against the second alternative that ontological analysis based on the deep features of propositions, as opposed to their surface grammar, must always be topic-neutral. Ontological structure, it may be said, is to be revealed by syntax, and syntax is prior to or at least independent of semantics. Consequently, propositions of the form, S is P, have the same ontological analysis, whether they are used to make statements about sensible objects or statements about minds. If this is true, it was not a truth recognized by Berkeley.24 For him the consequences of what he took to be the radical heterogeneity of sensibles and minds extend to the analysis of abstract terms, even logical terms. PHK 142 may say it best: ‘Spirits and ideas are things so wholly different, that when we say, they exist, they are known, or the like, these words must not be thought to signify any thing common to both natures. There is nothing alike or common in them.’25 Berkeley early on applied this position to existence, arguing that it means perceived when applied to sensible objects and perceives when applied to minds. In PHK 142 he used it to conclude that ‘our souls are not to be known in the same manner as senseless inactive objects, or by way of idea.’ Throughout this essay I have presented evidence that Berkeley denied the topic-neutrality of the ontological concept of ‘substance.’ Had he regarded ‘substance’ to be intelligible when applied to sensible

140 Phillip D. Cummins

objects, there is no reason to believe he would have held it to apply in the very same sense to minds. In fact, he regarded its application to sensibles as unintelligible or contradictory. Nevertheless, there is irrefutable textual evidence that he endorsed the intelligibility of ‘substance’ as applied to minds and indeed held S, minds and only minds are substances. My conclusion is that Berkeley may well have envisioned a differential analysis of singular propositions, depending upon whether they concerned sensibles or perceivers, even if doing so would now be universally and correctly regarded as philosophically indefensible. Since Berkeley’s statements about minds and their activities are not numerous, it may not be possible to decide between the two alternatives just stated. This makes no difference for the question at hand, since, as has been shown, neither alternative supports the contention that Berkeley abandoned or exploded the category of substance through his analysis of ‘A die is hard, extended, and square,’ and both are at least as plausible as the interpretation according to which he did.26 5. Perceiving and Willing: Two Forms of Doing In presenting an interpretation of Berkeley that makes perceiving the basis for his affirmation of mental substance and denial of material substance, I have emphasized that for him perceiving is what a mind does and have suggested that the perspicuous linguistic representation of a perceptual state of affairs is subject–verb–object, not subject–copula–predicate. My present task is to distinguish this kind of doing from another Berkeley insisted upon in arguing that minds alone are causes. Willing (choosing) is also doing – volitional doing; the perspicuous way to represent it linguistically is also subject–verb–object. As indicated earlier, Berkeley doubly elevated conscious beings in his immaterialism, committing himself to two parallel claims: S, minds and only minds are substances, and C, minds and only minds are causes. Crudely put, Berkeley’s argument for C is that only what is active can be a cause and only minds are active. Volitional activity is emphasized to secure the point that minds are active.27 Willing thus may seem to be as important in Berkeley’s philosophy as perceiving. In fact, however, the explicit Principles argument for C uses S as a premise, so it might be said to make perceiving-as-doing an indispensable part of the argument for C. In PHK 25 Berkeley argues that ideas, that is, sensible qualities or combinations of sensible qualities, cannot be causes, because they are inactive, and a cause must be active. In PHK 26 he adds:

Perceiving and Berkeley’s Theory of Substance

141

We perceive a continual succession of ideas, some are anew excited, others are changed or totally disappear. There is therefore some cause of these ideas whereon they depend, and which produces and changes them. That this cause cannot be any quality or idea or combination of ideas, is clear from the preceding section. It must therefore be a substance; but it has been shewn that there is no corporeal or material substance: it remains that the cause of ideas is an incorporeal active substance or spirit.

Note that, after denying in PHK 25 that ideas can cause other ideas, Berkeley here insists, first, that a cause of an idea must be a substance, and, second, that there are no material substances. He concludes that minds, active beings and the only other kind of substances, are causes. S is thus employed in the argument for C. This wrinkle goes unexplained. Did Berkeley think that to be a mental substance is to satisfy the activity requirement? If so, perhaps he did so because he had already implicitly conceived of the support relation as perceiving and regarded perceiving as doing, so that one condition of being a mental substance, perceiving, guaranteed its being active, at least in a minimal sense.28 This reading is supported by what Berkeley says first in PHK 27: ‘A spirit is one simple, undivided, active being: as it perceives ideas, it is called the understanding, and as it produces or otherwise operates about them, it is called the will.’ Even though willing, unlike perceiving, is tied directly to causing and operating on ideas, both are cited in order to expand the claim that a spirit is an active being. Moreover, both are next held to be imperceptible and unrepresentable by ideas on the ground that passive ideas cannot represent activity. It must be acknowledged that in his early published writings Berkeley did not work out with care and in detail exactly how willing and perceiving relate to and differ from each other and why they should both be classified as doings. Still, there is no good reason to equate willing and doing, that is, to contrast perceiving to doing rather than see it as a type of minimal doing. Similarly, there are many questions which can be asked regarding Berkeley’s views on the relationships between perceiving and willing. They include: Does perceiving presuppose antecedent willing by the perceiver or some other cause? Does willing include or presuppose perceiving? Does willing issue in sensibles or in perceivings? Answering some of these questions requires noting a distinction between two kinds of perceiving, sensing and thinking, a distinction I have pretty much ignored. Again, however, answering these questions is not required in order to defend my interpretation. What does need to be considered is

142 Phillip D. Cummins

whether willing, robust doing, is a constituent of perceiving and, if not, whether perceiving can still be considered doing. These questions are occasioned by a passage in the Dialogues of direct relevance to my interpretation. There, in the person of Philonous, Berkeley maintains that in sensing, properly so-called, the perceiver does not will its object of sense. This claim occurs in the First Dialogue as part of a complex attempt to reduce to absurdity a distinction Hylas invokes while trying to retract his earlier concession that no sensible object can exist unsensed (DHP 194–7). To justify his retraction Hylas suggests that one must distinguish in every perception between a sensation and its object. The former is an action, which belongs to the mind and so cannot occur in unperceiving things, whereas the remainder is the object sensed, which can exist unperceived or in an unperceiving thing. In keeping with earlier agreements, the object is characterized as one or more sensible qualities. While purportedly eliciting information on Hylas’s new distinction, Philonous develops a principle to which Hylas uncritically assents: philonous: And this action cannot exist in, or belong to any unthinking thing; but whatever beside is implied in a perception, may. hylas: That is my meaning. philonous: So that if there was a perception without any act of the mind, it were possible such a perception should exist in an unthinking substance. hylas: I grant it. But it is impossible there should be such a perception. (DHP 195–6)

Hylas and Philonous here take it as a given that a perception cannot occur in an unthinking or unperceiving substance. Prodded by Philonous, Hylas admits that for him this is only because a perception includes an act or action of the mind. Hence if it were shown that a perception includes no act, a perception would on his view be capable of existing in an unperceiving substance. Next, Philonous asks Hylas when the mind is said to be active; the latter responds, ‘When it produces, puts an end to, or changes anything,’ and then he agrees further that a mind can produce, discontinue, or change anything only by an act of will. Armed with these concessions Philonous next forces Hylas to acknowledge that since in perceiving, properly so-called, there is no act of will, there is no causation and hence no activity. For example, in smelling and seeing there is no act of will, hence no activity, even though

Perceiving and Berkeley’s Theory of Substance

143

there may be an act of will involved in moving an object near one’s nose or in opening one’s eyes. When Hylas accepts this position, Philonous invokes the principle agreed upon earlier and reduces Hylas to absurdity. He states: Since therefore you are in the very perception of light and colours altogether passive, what is become of that action you were speaking of, as an ingredient in every sensation? And doth it not follow from your own concessions, that the perception of light and colours, including no action in it, may exist in an unperceiving substance? And is not this a plain contradiction? (DHP 197)

Hylas replies, ‘I know not what to think of it.’ The contradiction being alleged is the existence in an unperceiving substance of the perception of light and colour. Hylas must allow this because he has already maintained that everything in a perception besides the act (the missing act of volition) can exist in an unperceiving substance. So much for the passage. Now for the objection. It might be held that since, on my account of Berkeley’s commitment to substance, perceiving-as-doing is crucial for establishing the dependence of the sensed and the primacy or substancehood of the perceiver, Philonous’s denial of Hylas’s act/object distinction discredits or at least severely weakens that account. There is an act only if there is volition. But in perception there is no volition; consequently, in perception there is no act. Since there is no act, perceiving cannot be a doing, and so the substance requirement, the primacy of the mind over the sensible, cannot be established. This objection fails. In the first place, it equates causing with acting and acting with doing. By doing so, it ignores the possibility that doing might be of two different kinds, one being perceiving and related cognitive states, assignable to the understanding, the other being volitions, assignable to the will. Moreover, it ignores several salient features of the relevant texts. Notice, first, that Philonous’s reductio ad absurdum of Hylas’s position turns on two propositions. One is that the act which secures the mind-dependence of perception, but requires an object distinct and independent from itself, is a volition which produces, eliminates, or changes something.29 The other is that, unlike the various actions which may accompany or prepare for it, sensing proper is nonvolitional. The first is asserted by Hylas, the second by Philonous. Once Hylas accepts the second, Philonous’s denial of volitional perceiving,

144 Phillip D. Cummins

he is quickly reduced to absurdity. Nothing in this exchange, however, implies that Philonous accepts Hylas’s identification of acting with willing.30 In my opinion the critical reader’s quite proper doubt that Berkeley has effectively answered the act/object objection to esse est percipi is founded on an unwillingness to accept the identification of acting with willing. Hence, it is natural for the reluctant reader to assume that because of his commitment to esse est percipi Berkeley would have insisted upon the identification. Surely, though, this last assumption is unfounded. Berkeley could also have insisted that perceiving is doing, distinguished between perceiving and causing, and held that only the latter involves an act/object distinction strong enough to secure the perception independence of the object of the action.31 In short, Berkeley maintained that causing is willing and willing is doing, but these positions do not imply either that only causing is doing or that only willing is doing. Consequently, his holding them does not imply that he denied perceiving is doing. Another textual point is that throughout Philonous’s response to Hylas’s act/object proposal, there is nothing to suggest that Berkeley’s point is that in perceiving by sight or smell there is no sensing, no doing, in addition to the odours smelled and the lights or colours seen. On the contrary, in denying that perceiving, properly so-called, includes volition, Philonous uses verbs of action, ‘seeing’ and ‘smelling,’ to illustrate and validate his stance. He asks, for example, ‘Tell me now, whether seeing consists in perceiving light and colours, or in opening and turning the eyes?’ Again, the absurdity attributed to Hylas is that, on his principles, because it lacks a volitional component, perceiving light and colour, for example, can occur in an unperceiving substance. Since he is in the process of retracting esse est percipi for sensible qualities, Hylas would hardly admit absurdity if his position implied only the presence of light and colour in an unperceiving thing. That is, after all, the position he is trying to secure. The reductio ad absurdum would thus lose its force unless Philonous is committed to and commits Hylas to a sensory process, a perceiving, which cannot be done by an insentient thing. There is thus nothing in the argument which shows Berkeley denied perceiving is doing. Its point is not that no distinction can be drawn, for example, between seeing a colour and the colour seen; it is, rather, that no distinction can be drawn which implies the colour’s capacity to exist unseen. It is true, admittedly, that in this exchange with Hylas, Philonous states that in sensing the mind is passive, but it should be recalled that

Perceiving and Berkeley’s Theory of Substance

145

this is stated against the background of Hylas’s volitional criteria for mental activity. ‘Passive’ is also tied to ‘passion’ as in an expression Berkeley sometimes employs, ‘passions or sensations of the soul’ (cf. DHP 197, NTV 41). To say colours or sounds are sensations in or passions of the soul is to affirm with emphasis that they cannot exist unperceived. It may also provide a sense in which sensing is passive. As he did in the Principles, Berkeley argues in the Dialogues that the object sensed must have a cause and the cause is a mind distinct from its created perceiver. Consequently, the latter in perceiving is acted upon by something outside itself. If we consider the cause to be the agent, the perceiver can be deemed the patient and its perceptions assigned the status of passions. To accept this way of expressing that in sensing the mind does not determine its own objects is not, however, to deny that hearing, for example, is something a mind does or that there is a minimal distinction between the hearing and the sound heard. Finally, there is also undeniable evidence that Berkeley recognized non-volitional mental doing in addition to and independent of choosing.32 Mediate perceiving provides a relevant example.33 On Berkeley’s account of mediate perception, when upon immediately seeing a colour, I mediately perceive something else, I am doing something, something more than what occurs when I merely see the colour. The something more, the interpretive element, involves thought, even belief, but to think or believe is not to will. Mediate perceiving is not itself a volitional state nor the effect of conscious choosing. Equally, it is not the result of a process of reasoning, which could be said to involve choice. Berkeley is insistent that mediate seeing, though something the mind does, is usually or always a product of psychological conditioning or associating. It indicates an active being, but not choice or volition. One need only recall Berkeley’s treatment of seeing distance, magnitude, and situation in An Essay towards a New Theory of Vision to see the relevance and strength of this claim.34 Without identifying it with mediate perception or equating their degrees of activity, I want to suggest that in immediate perception there is also doing without volition. Having blocked the objection stemming from Philonous’s rejection of Hylas’s distinction between act and object, let me add that Berkeley may well have had philosophical commitments which led him to downplay perceiving as doing as well as commitments which led him to emphasize it. As has been suggested by many philosophers and scholars before me, Berkeley’s immaterialism may well harbour deep contrary tendencies. To be specific without arguing these claims in the

146 Phillip D. Cummins

detail they require, I want to suggest that to secure his two main metaphysical tenets, S and C, Berkeley needed a strong idealistic position regarding sensibles. To secure it he characterized sensible qualities as sensations or as inseparable elements within sensations. His paradigmatic sensations were feelings of pleasure or pain. To emphasize the impossibility of unperceived sensibles, Berkeley utilized the language of sensations and portrayed the distinction between the feeling and the felt as exceedingly thin. However, when he wished to claim that unlike its rivals, his theory of perception and its objects secured immediate knowledge of bodies and saved common sense, he set aside the language of sensations and wrote, instead, of sensible objects and qualities or even of bodies (apples, tables) and their qualities (red, sweet), thereby strengthening and widening the distinction between perceiving and the perceived. Many find the gap between these rhetorical stances unbridgeable. Equally, Berkeley’s treatment of mental substance may have been affected by his commitment to the mind-dependence of the objects of sense.35 In the argument for mental substance, as I have reconstructed it, he begins with esse est percipi, then uses the being perceived/perceiving correlation to arrive at a distinct entity which perceives and thus fills the dominant role required in the substance tradition. However, when, in attempting to secure esse est percipi, his idealistic starting point, Berkeley insists that colour is a sensation or passion of the soul, the distinction between perceiving and object perceived is concealed and the perceiver as doer seems to fade from view. To note this tension, however, is not to deny mind as doer and, thus, substance in Berkeley’s account of sensory perception and defence of immaterialism. It is merely to recognize the logical and rhetorical distance between these passages and the vigorous affirmation of mental substance in PHK 7. Notes 1 I shall use ‘sensible’ as a blanket noun covering both sensible qualities and sensible objects, when the distinction between the two makes no difference. When it is important to differentiate between them, I shall use the more precise terms. 2 Purported problems about mind in Berkeley’s philosophy are of two main kinds. There are, first, concerns about whether and how mind can be known. Since I will not be addressing this concern, the vast literature on minds, ideas, and notions will be ignored in this paper. For an attempt to

Perceiving and Berkeley’s Theory of Substance

3

4 5

6 7

8

147

lay those concerns to rest, see Phillip D. Cummins, ‘Hylas’ Parity Arguments,’ in Berkeley: Critical and Interpretive Essays, ed. Colin M. Turbayne (Minneapolis: University of Minnesota Press, 1982), 283–94. There are also, however, concerns about the ontological status of mind as mental substance. It is thought, for example, that Berkeley’s views on language or his anti-abstractionism destroyed the foundations of the traditional doctrine of substance. Some examples: Colin M. Turbayne, ‘Berkeley’s Two Concepts of Mind,’ Philosophy and Phenomenological Research 20 (1959): 85–92, and 22 (1962): 383–6; and Robert Muehlmann, Berkeley’s Ontology (Indianapolis: Hackett Publishing, 1992), ch. 6, ‘Mental Congeries and Mental Substances.’ For two interesting attempts to understand what Berkeley means by mental substance, see Harry M. Bracken, ‘Substance in Berkeley,’ in New Studies in Berkeley’s Philosophy, ed. Warren E. Steinkraus (New York: Holt, Rinehart and Winston, 1966), 85–97; and Colin M. Turbayne, ‘Lending a Hand to Philonous: The Berkeley, Plato, Aristotle Connection,’ in Berkeley: Critical and Interpretive Essays, ed. Turbayne, 295–310. It should be noted at the outset that others have done crucial work on the place of mind in Berkeley’s philosophy. For example, Margaret Atherton has shown just how weak much of the evidence for Berkeley’s alleged inconsistency regarding mind really is in her ‘The Coherence of Berkeley’s Theory of Mind,’ Philosophy and Phenomenological Research 43 (1983): 389–99. On the other hand, Charles McCracken has demonstrated how Berkeley agonized regarding mind in his notebooks in his ‘Berkeley’s Notion of Spirit,’ History of European Ideas 7 (1986): 597–602, reprinted in The Empiricists: Critical Essays on Locke, Berkeley, and Hume, ed. Margaret Atherton (Lanham, MD: Rowman and Littlefield, 1999), 145–52. Berkeley has two fundamental theses: the second is C, minds and only minds are causes. It will be discussed in section 5 of this essay. A logical purist might insist that what Berkeley wrote in PHK 7 implies only (1); perhaps, but in context and in line with what he later claimed, for example, in PHK 26 and 36, it makes more sense to regard Berkeley in PHK 7 as asserting both (1) and (2). No explicit definition of ‘idea’ is provided at the beginning of PHK. Those provided here are contextual. In PHK 8 Berkeley rejects a different kind of non-sensible non-mind, unperceived material objects which are pictured and thus represented by ideas. Subsequently, in PHK 25, unperceived unperceiving causes of ideas are denied. In the Principles Berkeley simply takes the point for granted; in the Dialogues he offers an explicit argument which we shall soon be examining.

148 Phillip D. Cummins 9 Why did Berkeley think he was entitled to the conclusion that perceiving substances cannot be extended and extended substance cannot perceive? One possibility is that he endorsed the argument, sometimes called the Achilles argument, that thinking and extension are fundamentally incompatible. Another possibility is that, since he thought extension must be perceived to exist, he concluded that the only relation extension could have to a mind is being perceived by it. He might also have thought that were a thing which perceives extended, it too would have to be perceived to exist, in which case it would not be an autonomous entity and, thus, not a substance. On the Achilles argument, see Lorne Falkenstein, ‘Hume and Reid on the Simplicity of the Soul,’ Hume Studies 21 (1995): 25–45. 10 The pattern of argument being employed is analysed in Cummins, ‘Hylas’ Parity Arguments.’ 11 It is one individual principle because, presumably, the colour and sound are objects of the same perceptual consciousness. There is no question-begging here on behalf of perceiving substance since even on a bundle theory the colour and sound would be regarded as components of one and the same bundle. This is not to say there is an easy way to justify the assignment of both the colour and the sound to a single perceiver beyond saying they are both objects I perceive, which admittedly seems circular. 12 Descartes (PP I.51, 210) seems to employ both the autonomy conception of substance and what I shall call the superordination conception. He writes first, ‘By substance we can understand nothing other than a thing which exists in such a way as to depend on no other thing for its existence.’ Later in the same article he writes, ‘In the case of created things, some are of such a nature that they cannot exist without other things, while some need only the ordinary concurrence of God in order to exist. We make this distinction by calling the latter “substances” and the former “qualities” or “attributes” of those substances.’ Descartes may have thought that the two conceptions are equivalent. I shall argue that according to Berkeley, perceivers, unlike sensibles, are not dependent entities even though strictly speaking they cannot exist in the absence of sensibles. 13 This is so even for misperceiving, as it is usually understood, since there is misperceiving precisely because there is a lack of fit between what is perceived (something distinct from the perceiver) and the world. Berkeley, of course, would want to distinguish between immediate and mediate perception and deny immediate misperception. 14 For more on this point, see Genevieve Migely’s essay in this collection. 15 See PHK 135–48 for an extended defence of the claim that no perceiver is an object of sense. However, Berkeley does allow that a perceiver can be conscious of itself.

Perceiving and Berkeley’s Theory of Substance

149

16 Clearly, what is non-causally dependent upon another entity for its existence cannot exist alone and so cannot qualify as a substance on the autonomy conception of substance. The important point is that the superior entity, upon which something else depends for its existence, but which does not so depend upon any other thing, may for all that not be capable of existing alone. Put another way, failing the autonomy test does not entail failing the ontological dependence test. 17 The analogy between causal dependence and perception dependence being developed in the text should not lead one to conclude there is no distinction between them. In the substance tradition, the dependence of mode or accident on substance was recognized as a form of ontological dependence distinct from causal dependence. 18 See Hume, THN I.iii.3, 82, where he notes, ‘effect is a relative term of which cause is the correlative.’ Hume famously proceeded to argue that the correlativity of effect and cause, which implies that every effect must have a cause and every cause an effect, does not imply that whatever begins to exist must have a cause. 19 The paper which initiated discussion of the inherence interpretation is Edwin B. Allaire’s ‘Berkeley’s Idealism,’ Theoria 29 (1963): 229–44. In his introduction to Berkeley’s Metaphysics: Structural, Interpretive, and Critical Essays, ed. Robert G. Muehlmann (University Park: Pennsylvania State University Press, 1995), 1–4, Muehlmann provides a partial bibliography of papers in which the inherence interpretation is developed or assessed. 20 The present sketch of the position attributed to Berkeley by the inherence interpretation is just that, a mere sketch, and so suggests many questions which it cannot answer. One important one is: Does the inherence relation as such constitute perceiving, or does the nature of the substance in which a quality inheres determine whether the substance perceives the quality? On the first alternative a substance is a perceiver just because a quality inheres in it; on the second a quality is perceived because the substance in which it inheres is a mind. 21 For an account of the historical context of Berkeley’s commitment to the natural immortality of the soul, see Harry M. Bracken, ‘Berkeley on the Immortality of the Soul,’ Modern Schoolman 37 (1960): 77–94 and 197– 212. 22 The present essay defends an explication that is an alternative to this circular view. 23 The first sentence of the passage is puzzling. It seems to anticipate the conclusion reached shortly afterwards: ‘It is therefore evident there can be no substratum of those qualities but spirit.’ However, the second sentence, which clearly states a premise in the argument for that conclusion, begins

150 Phillip D. Cummins

24 25

26

27

with ‘And’ and thus seems to make the first sentence a premise as well, in which case it would be a premise in its own proof. On how this relates to claims about other issues regarding spiritual matters (e.g., religious mysteries), see Roomet Jakapi’s essay in this collection. See also PHK 89, where Berkeley writes, ‘Thing or being is the most general name of all, it comprehends under it two kinds entirely distinct and heterogeneous, and which have nothing common but the name, to wit, spirits and ideas.’ This might be the appropriate place to consider briefly a second objection to Berkeley’s employment of the category of substance. It stems from his use of arguments from perceptual variation in the Dialogues for the conclusion that sensible qualities and objects cannot exist unperceived. The objection is this: whether or not those arguments succeed, substance cannot be invoked to secure derivative unity at a time and derivative identity over time, because for someone like Berkeley, perceptual variation implies that what is immediately perceived is what it is perceived to be. Consider the case of simultaneously seeing and touching an oar in water. When Hylas questions Philonous about it and similar cases, the latter insists that strictly speaking one does not see and touch the same thing. He rejects, that is, a single substance that has and thereby unifies the properties we immediately perceive. This is done to avoid the absurdity of the oar being both bent and straight at the same time (i.e., simultaneously having two contrary qualities). This was the type of absurdity Hylas was forced to acknowledge in the first dialogue regarding heat and cold and other contrary qualities. The first step, then, of the objection is that Berkeley rejected substance in connection with sensible objects to preserve the veridicality of immediate perception in the face of perceptual variation. The second step is that he must reject substance regarding minds for the same reasons. Must the second step be taken? Granting that self-awareness for Berkeley is always veridical, need one grant that he did acknowledge or should have acknowledged an analogue of perceptual variation for self-awareness? Perhaps, but only after the objector establishes that there are genuine cases in which one seems to be in two or more contrary cognitive or affective states simultaneously. Since perceptual variation seems always to involve conditions unique to sensory perception, such as perspectival differences, we should not expect such cases any time soon. PHK 28 begins, ‘I find I can excite ideas in my mind at pleasure, and vary and shift the scene as oft as I think fit. It is no more than willing, and straightway this or that idea arises in my fancy; and by the same power it is

Perceiving and Berkeley’s Theory of Substance

28

29

30

31

32

151

obliterated, and makes way for another. This making and unmaking of ideas doth very properly denominate the mind active.’ Each perceiving of a mind is a state of affairs, so a sequence of perceivings provides change, the mind being that which endures through change. Perceiving can thus allow for a distinction between an enduring agent and its momentary states. Causation, however, requires a more robust sense of agency than is exhibited by perceiving as doing. It is not obvious why an object of volition as such is capable of mind-independent existence, as the argument seems to assume. If I choose to imagine an object as it would appear if seen from some unusual perspective, my imagining, which presumably results from my willing, is hardly mindindependent. The same might be said of the appearance. One should keep in mind a similar dialectical situation. Near the beginning of the First Dialogue, Hylas and Philonous agree in stipulating that a sceptic is one who denies the reality of sensible things (DHP 173–4). Hylas next defines the reality of a sensible thing as its being capable of existing unperceived (DHP 175). Philonous proceeds to convince Hylas that no sensible thing is capable of so existing (DHP 175–94). After Hylas at the beginning of the Second Dialogue confesses to scepticism and claims Philonous must join him, the latter points out that he did not accept Hylas’s definition of reality and so escapes the charge (DHP 211–12). I am suggesting the act/ object case is analogous. Hylas equates acting (doing) with willing and so must accept the absurd consequence that a perception, since it does not include an action on his equation of acting and willing, is capable of existing in a non-sentient substance. Must not Philonous share the absurdity? Not if he rejects the equation, which he does. Must he, too, reject the act/ object distinction? It is not obvious that he must. If one maintains that Berkeley must deny perceiving is doing because affirming it requires an act–object distinction and such a distinction implies the perception independence of the object, one is arguing from a Humean perspective there is no reason to believe Berkeley shared. For Hume, if one entity is not a proper part of a second, then the existence of the latter is independent of that of the former. Why assume Berkeley would agree? The distinction between doing and volitional acting also can be illustrated using bodily analogies. My breathing is something I do, but not a volitiondirected action. My heart’s beating is something I do or it does, not an intentional action I or it performs. So, too, I am suggesting, sensing for Berkeley is something a mind does, but not a volition-directed action of the mind. The present bodily examples cannot be used as models to explain perceiving as doing. A heart’s beating may take adverbial modifiers, e.g.,

152 Phillip D. Cummins rapidly, but it has no object. In contrast, with perceiving, it is always appropriate to ask, what is being perceived? This point may caution us against uncritical acceptance of the so-called adverbial interpretation of Berkeleian perceiving. 33 For more about mediate perception, see the essays in this collection by Richard Glauser and Marc Hight. 34 See the essay by Ralph Schumacher in this collection. 35 One of the best treatments of this issue is to be found in S.A. Grave’s ‘The Mind and Its Ideas: Some Problems in the Interpretation of Berkeley,’ Australasian Journal of Philosophy 42 (1964): 199–210.

Berkeley’s Actively Passive Mind gen e v i e ve m i g e ly

Berkeley defines the mind as an active substance or being: ‘a spirit is one, simple, undivided, active being: as it perceives ideas, it is called the understanding, and as it produces or otherwise operates about ideas, it is called the will’ (PHK 27). The will is the activity of producing and operating about ideas and the understanding is the activity of perceiving ideas. As will and understanding, then, the mind is activity. Even though this activity is distinct from its objects and effects (viz., ideas and volitions), we can still say that ideas are both produced and perceived by the mind and volitions are acts of the mind. This way of speaking about the mind, its activity, and its products makes it difficult to understand how Berkeley can claim emphatically that the mind is entirely active while also claiming that the finite mind in perception is entirely passive. It seems, at first glance, that there is an apparent inconsistency in the very nature of a finite substance. However, I will offer an interpretation of passivity that preserves the active nature of the mind and demonstrate that, rather than being inconsistent, Berkeley is securing his commonsense realism without abandoning his dualism of minds and ideas. 1. Passivity and Activity in Perception Berkeley repeatedly asserts that the soul is a simple, active being whose existence consists in perceiving ideas, but certainly not in being perceived (PHK 2, 7, 27, 71, 89, 141; and DHP 231, 233–4). In these passages, he sets up his dualism of spirits and ideas. These are two different ways of being: the mind is ‘wholly active,’ while ideas are ‘altogether passive and inert.’ Confusion enters into this sharp distinction in regards to perception. In the Dialogues, Philonous admits to Hylas that he is ‘alto-

154 Genevieve Migely

gether passive’ in perception as there is ‘no action in it’ (DHP 196–7). Numerous commentators1 argue that Berkeley cannot have his entirely active mind if it is also a passive mind.2 Berkeley is here charged with inconsistency since the always-active mind is at the same time passive, ‘involving no mental activity.’3 This position also blurs the distinction Berkeley establishes between minds and ideas. If mind is passive, it is similar in nature to ideas. George Pitcher is entirely comfortable with this ontological blurring for he finds the mind to be only ‘weakly distinct’ from ideas in the sense that ideas qualify the mind. Pitcher understands the Berkeleian soul as ‘not altogether active, for perceiving ideas is a passive state of the mind.’4 George Stack cautions that such a view would lead Berkeley to a Humean analysis of mind in which the mind is simply a bundle of perceptions or ideas.5 Beyond the inconsistency and the fuzzy ontological lines between things, there is a further concern in regard to the structure of the mind itself. Charles McCracken notices that a mind that has opposing functions (active and passive) defies the concept of a unified whole: ‘The mind’s essence could, of course, be held to reside in both active volition and passive perception, but these very different states must then be reconciled with the mind’s unity and simplicity.’6 At stake are Berkeley’s dualism and his mental ontology. Berkeley states that the finite soul is passive in exactly two discussions: the tulip-smelling discussion and a letter to Samuel Johnson.7 There are some elements of sense perception, such as bodily motion, that clearly involve activity of the mind: ‘The mind therefore is to be accounted active in its perceptions, so far forth as volition is included in them. In plucking this flower, I am active, because I do it by the motion of my hand, which was consequent upon my volition’ (DHP 196). However, the smell of the flower is a different matter: ‘But I do not find my will concerned any farther. Whatever more there is, as that I perceive such a particular smell or any smell at all, this is independent of my will, and therein I am altogether passive’ (DHP 196). We are active in that we may will that we smell, but we cannot determine by an act of will what we smell. The mind is ‘in these respects altogether passive’ (DHP 196). The only other text in which Berkeley states that the finite soul is passive is in response to a question raised by Johnson in a previous letter. Johnson writes on 5 February 1730: ‘There is certainly something passive in our souls, we are purely passive in the reception of our ideas’ (PW 351). In his letter of 24 March 1730, Berkeley replies: ‘That the soul of man is passive as well as active, I make no doubt’ (PW 354). The numbering in the letters of the questions and replies indicates that

Berkeley’s Actively Passive Mind

155

Berkeley agrees with Johnson that in the reception of our ideas (viz., in sense perception), the mind is passive. That is, passivity in the soul occurs only during sense perception.8 However, this passivity in sense perception does not occur in all forms of sense perception. Berkeley distinguishes between two types of perception: immediate perception and mediate perception.9 Only light, colours, and figures are immediately perceived by sight; only sounds by hearing; only tastes by the palate; only odours by smell; and only tangible qualities by touch (DHP 175). The classification and recognition of a collection of these sensations as an object seems to be mediate perception.10 ‘A cherry, I say, is nothing but a congeries of sensible impressions, or ideas perceived by various senses: which ideas are united into one thing (or have one name given them) by the mind’ (DHP 249). The will is involved in mediate perception since the mind is clearly operating about the ideas it has and is producing other ideas from those ideas perceived. The distinction between immediate and mediate perception is evident in his discussion of distance: ‘It is shewn that distance or outness is neither immediately of it self perceived by sight ... but that it is only suggested to our thoughts, by certain visible ideas and sensations attending vision’ (PHK 43). It is by means of another idea that the idea of distance is discovered, not by an idea of immediate perception. This type of perception, in which the mind is performing operations and producing ideas different from the ideas perceived in immediate perception, involves volition.11 It is an active process of relating and categorizing ideas from our sensory experience.12 Berkeley can only be referring to immediate perception when he considers the mind to be utterly passive. In his discussion of the tulip, he states, ‘You are in the very perception of light and colours altogether passive ... the perception of light and colours, including no action in it’ (DHP 197). So, it is in the bare reception of ideas of sense that the mind is utterly passive, not in its process of operating about those ideas of sense. Berkeley is very specific about why he considers the mind passive in immediate sense perception. In the case in question, ‘activity’ means involving an act of will. ‘Passive’ here means independent of the will in the sense of involuntary. What is involuntary in immediate perception he makes clear: we are not the cause or origin of our immediate perceptions of sense (PHK 29, 32, 44, 56, 90; DHP 214–15, 242). What it is that I see, feel, taste, smell, or hear, and whether or not I sense it, is not up to me. That I sense it is up to me in that I can close my eyes or not inhale, but its content is not determined by me:

156 Genevieve Migely But whatever power I may have over my own thoughts, I find the ideas actually perceived by sense have not a like dependence on my will. When in broad daylight I open my eyes, it is not in my power to choose whether I shall see or no, or to determine what particular objects shall present themselves to my view; and so likewise as to the hearing and other senses, the ideas imprinted on them are not creatures of my will. There is therefore some other will or spirit that produces them. (PHK 29)

In order to secure his commonsense realism, it is necessary that objects of immediate sense perception not be caused by finite wills. If they were, his idealism would be reduced to a mere fiction of each finite mind. Yet, it is a part of every man’s experience that sensations are received involuntarily. What this means is that we are not the cause or origin of those sensations we perceive. It is in a very strict sense then that Berkeley considers the mind passive: only in immediate sense perception which involves no act of our will to produce the content of that sensation.13 This then appears to be the main difference between immediate and mediate perception: one involves no activity of the mind, whereas the other does. However, it is not that simple. Ideas of sense are considered real things because of their constant order. That regularity, however, is again not of our own doing. ‘Now the set rules or established methods, wherein the mind we depend on excites in us the ideas of sense, are called the Laws of Nature: and these we learn by experience, which teaches us that such and such ideas are attended with such and such other ideas, in the ordinary course of things’ (PHK 30). In many of our mediate perceptions, we again do not determine by an act of our will which ideas go with which.14 This is something we learn by experience of always seeing certain ideas together. When our mind is affected by one such idea, we will immediately think of the other idea with which it has been regularly experienced.15 This judgment is a result of repeated experience of certain ideas. It is obvious there is mental activity involved in mediate perception since the mind is making inferences, even if those judgments are involuntary in the sense that we do not always decide which ideas are so related.16 We may not always have a choice about which ideas we relate together, but we are the cause of that relation of ideas. In this sense, mediate perception is certainly not passive, precisely because it involves mental operations.17 Rather than restricting ‘active’ to the production of all ideas, Berkeley uses ‘active’ to describe the mind’s pro-

Berkeley’s Actively Passive Mind

157

duction of ideas concerning the ideas the mind did not necessarily produce itself.18 What Berkeley means by ‘activity’ then is not simply the ability to choose what we think about, since nature often chooses for us, but more importantly, that we are the ‘cause of’ the ideas we form based upon sensory ideas. As ‘that active principle of motion and change of ideas,’ the mind is still active in its operations regardless of whether it is the cause of the ideas involved in these processes (PHK 27). What needs to be examined then is whether there is activity in immediate perception.19 Berkeley finds an intimate relation between volition and perception: It seems there can be no perception, no Idea without Will, being there are no Ideas so indifferent but one had rather Have them than annihilation, or annihilation than them. or if there be such a equall Balance there must be equal mixture of pleasure & pain to Cause it. there being No Ideas perfectly void of all pain & uneasiness But what are preferable to annihilation. (NB 833)

Hedonic sensation is part of all perceptions, as heat, tastes, smells, and sounds are nothing but a particular pleasure or pain (DHP 177–80). Pleasure or pain comes with desire or aversion (volitions).20 Further, there is always assent involved in order to register the perception. ‘To be sure or certain of what we do not actually perceive (I say perceive not imagine) We must not be altogether Passive, there must be a disposition to act, there must be assent, which is active, nay what do I talk There must be Actual Volition’ (NB 777). This assent is present at all times for Berkeley: ‘While I exist or have any Idea I am eternally, constantly willing, my acquiescing in the present state is willing’ (NB 791). Assenting is an act of the mind (Alc VII.3, 288). Therefore, immediate perception is active since it involves activity in the evaluation and confirmation of sensory ideas. Accordingly, Berkeley employs ‘passivity’ in immediate perception in a very narrow sense. Minds are ‘passive’ in that finite perceivers are not the origin or cause of the ideas of sense.21 Although an act of the finite will does not produce ideas of immediate perception, the finite will is involved in other actions in immediate sense perception, such as desire, aversion, and assent. The will is active in producing mental activities and ideas about ideas of sense.22 Perception, then, does not violate the nature of the finite, nor infinite, mind. The being of a substance is determined by its activity: ‘Substance

158 Genevieve Migely

of a spirit is that it acts, causes, wills, operates’ (NB 829). Though the active, finite mind is passive in perception, the infinite mind is never passive. As an imperfect being, the finite mind perceives ideas of nature by sense. We are affected by ideas of sense. However, a perfect being cannot be ‘affected’ by anything. Berkeley emphatically denies that the infinite perceiver perceives as we do: That God knows or understands all things, and that He knows among other things what pain is, even every sort of painful sensation, and what it is for His creatures to suffer pain, I make no question. But that God, though He knows and sometimes causes painful sensations in us, can Himself suffer pain, I positively deny. (DHP 240)

Finite minds are connected to the perception of a body, so we are affected by sense. God simply perceives ideas without being affected by them. ‘God May comprehend all Ideas even the Ideas which are painfull & unpleasant without being in any degree pained thereby. Thus we our selves can imagine the pain of a burn etc without any misery or uneasiness at all’ (NB 675). God understands what pain is without ever having to be in pain. Finite minds are considered passive in immediate perception because we are not the cause of the content of ideas of sense. God, however, determines the content of every idea of sense. As the cause of ideas of sense, God produces ideas but is never involuntarily affected by them. He is therefore ‘impassive’ (DHP 214). 2. The Interdependence of Mind and Ideas Though perception is an activity, calling the mind ‘passive’ still lends itself to the view mentioned earlier that Berkeley has not made a sharp enough distinction between minds and ideas. Ian Tipton finds that Berkeley’s stance on perception ‘made it difficult to see how the mind conceived of as that which perceives, or passively receives, can be thought of as distinct from ideas perceived.’23 There are even passages where Berkeley seems to be equating mind and ideas (NB 585, 609, 655; PHK 5, 98).24 For example, shortly after presenting his official doctrine of a distinct mind at PHK 2, Berkeley comments in PHK 5 that no distinction can be made between a perception and an idea: ‘The things we see and feel, what are they but so many sensations, notions, ideas or impressions on the sense; and is it possible to separate, even in thought, any of these from perception? For my part I might as easily divide a

Berkeley’s Actively Passive Mind

159

thing from itself.’ Initially, the mind and ideas are distinct, but then they seem to be identical. While arguing that the mind is always thinking, Berkeley admits that he cannot conceive of the mind without thought: ‘Whoever shall go about to divide in his thoughts, or abstract the existence of a spirit from its cogitation, will, I believe, find it no easy task’ (PHK 98). The very being of a mind is its ‘thoughts.’ Berkeley appears to be undermining his own dualism in these passages. Mind and ideas cannot be both distinct and the same. However, this alleged inconsistency is based upon a process/product ambiguity. When Berkeley claims that the perception and the idea are inseparable, ‘perception’ may refer to either the process of perception or the object of perception. If he means the object of perception, there is nothing contentious here. PHK 5 is in the middle of his argument for idealism. He is attempting to prove that external sensible qualities are ideas, that is, things perceived or perceptions. The abstraction he finds as illegitimate is between the existence of sensible objects and ‘their being perceived’ (PHK 5). It is impossible to separate an idea from a perception because an idea is what is perceived, or a perception. That this is the proper referent for ‘perception’ can be seen in the subsequent passage, where Berkeley reiterates his point: ‘to be convinced of which, the reader need only reflect and try to separate in his own thoughts the being of a sensible thing from its being perceived’ (PHK 6). The existence of a sensible object is to be perceived, that is, to be an object of perception. Berkeley is using ‘perception’ here to designate the object of perception rather than the activity of perception. This same ambiguity between the process and the object of the process is also present in PHK 98. The term ‘cogitation’ refers to both the act and the object of thought. Berkeley uses it both ways: ‘Idea is the object or Subject of thought; that I think on whatever it be, I call Idea. thought it self, or Thinking is no Idea tis an act i.e. volition i.e. as contradistinguish’d to effect, the Will’ (NB 808). In PHK 98, if ‘cogitation’ means the activity of thinking, then it is no surprise that Berkeley would make this statement. To be a spirit is to be a thinking, perceiving thing. That is what a spirit is by definition; it is an active being. To remove the activity from a soul is to take away its very essence: ‘such is the nature of spirit, or that which acts’ (PHK 27).25 Further evidence for this reading is Berkeley’s agenda in this passage: it is to prove that the soul always thinks. In other words, it is to prove that the soul is always performing the activity of thinking. It is about the action of thinking, not about the objects of those actions. In both passages, then, if we rely

160 Genevieve Migely

on a charitable reading of the terms, there is no need to reach the conclusion that Berkeley is anywhere equating the mind with ideas. Passivity in perception does not open the door for a reading in which the mind is identical to ideas. However, let us say he is referring to the activity perceiving in PHK 5 rather than to the object of the activity – what S.A. Grave and Robert Muehlmann propose Berkeley is doing. He would then be making the claim that perceiving is inseparable from perception. Yet, this does not entail the reduction of the activity into its objects. Inseparability does not logically entail identity. For example, the second floor is inseparable from the first floor: there would be no second floor without a first floor. Though the second floor is inseparable from the first floor, that does not mean it is identical, and reducible, to the first floor. For Berkeley, there is a necessary epistemic relation between the act of perceiving and the object perceived. As A.A. Luce explains, ‘such a subject [mind] as he describes must have its object; such a perceiving being must have something to perceive; such an active being must have that on which to act; its sea must find a shore.’26 You cannot have one without the other. And this is perfectly in keeping with Berkeley’s anti-abstractionism. If it is impossible for two things to exist separately, it is likewise impossible to conceive of them separately. Berkeley denies ‘that [he] can abstract one from another, or conceive separately, those qualities which it is impossible should exist so separated’ (IN 10). The activity of perceiving cannot occur without the object of perception and vice versa. Therefore, the activity and the object are inseparable. But further proof is required to make the additional claim made by Grave and Muehlmann that they are therefore identical. If Berkeley is claiming perceiving is incapable of being distinguished from ideas, he does not mean that they are identical; rather, he simply means they are incapable of separate existence. Similarly, in PHK 98, even if Berkeley is referring to the objects of thoughts rather than to the activity of thought, there is still no identity claim present. On this reading, Berkeley would be stating that it is a difficult task to conceive of the existence of a spirit apart from its thoughts. It would involve an illicit abstraction for him. Again, the perceptual relation between perceiving and perceptions or between thinking and thoughts is a necessary epistemic relation. One must be thinking about something. The nature of a thought is being thought about. It does not qualify as a thought without being thought of. The thinking soul is also intimately tied to its thoughts. If the soul is not thinking about any-

Berkeley’s Actively Passive Mind

161

thing, it is not thinking, it is not active, and therefore does not exist. Though this statement reveals the nature of the dependency between the soul and its ideas, it does not make the further claim that the soul is identical to its ideas. This relation that Berkeley sets up between the mind and its ideas has unique ontological and existential implications, different from the traditional substance analysis in which substances are independent existents and ideas are modes dependent upon that substance.27 The most important feature of Berkeley’s ontology, that the mind and its ideas are ontologically distinct, can be expressed in the following: Ontological Distinctness Principle (ODIS): Things are ontologically distinct when they are two different types of beings due to their two completely different natures. Ideas are passive beings while minds are active beings. The way in which minds exist is completely different from the way in which ideas exist. Berkeley argues that ‘it is therefore necessary, in order to prevent equivocation and confounding natures perfectly disagreeing and unlike, that we distinguish between spirit and idea’ (PHK 139). The mind is an active thing. ‘Substance of a Spirit is that it acts, causes, wills, operates’ (NB 829). Ideas, however, are passive, inert things. They are ‘visibly inactive, there is nothing of power or agency in them ... the very being of an idea implies passiveness and inertness in it, insomuch that it is impossible for an idea to do any thing, or strictly speaking, to be the cause of a thing’ (PHK 25). This is why for Berkeley there is, strictly speaking, no idea of spirit, since something inactive cannot represent or resemble something active. He argues that ‘the words will, soul, spirit, do not stand for different ideas, or in truth, for any idea at all, but for something which is very different from ideas, and which being an agent cannot be like unto, or represented by, any idea whatsoever’ (PHK 27). In insisting that the mind is ‘a thing entirely distinct from them [ideas]’ (PHK 2), Berkeley reveals his commitment to the ontological distinction principle. Unlike the traditional substance analysis, this ontological distinction does not have existential import. According to Berkeley, mind and ideas are existentially inseparable: Existential Inseparability Principle (EXIN): Two things are considered to be existentially inseparable when they are unable to exist apart

162 Genevieve Migely

from each other. Separate existence of either being results in the annihilation of that being. First, ideas depend ontologically upon minds for their existence. Ideas are what the mind is immediately aware of. Ideas then must be thought of or perceived in order to exist. Ideas, if they exist, must exist in a mind. As Berkeley explains: So long as they [ideas] are not actually perceived by me, or do not exist in my mind or that of any other created spirit, they must either have no existence at all, or else subsist in the mind of some eternal spirit: it being perfectly unintelligible and involving all the absurdity of abstraction, to attribute to any single part of them an existence independent of spirit. (PHK 6)

Ideas, then, are incapable of existing apart from some mind. It may be the infinite mind or a finite mind that perceives them. But if an idea is not perceived by a mind, it does not exist. This is Berkeley’s idealism. Second, minds depend epistemically on ideas for their activity. Berkeley continues the Cartesian view that the mind is always thinking: ‘To say the mind exists without thinking is a Contradiction, nonsense, nothing’ (NB 652). Therefore, the mind must always be thinking about something – that is, the mind must always have ideas: ‘certainly if there were no sensible ideas there could be no soul, no perception, remembrance, love, fear etc. no faculty could be exerted’ (NB 478). For Berkeley, mental activity does not require any one particular idea or set of ideas, but simply that the activity is directed: ‘Some Ideas or other I must have so long as I exist or Will. But no Idea or sort of Ideas is essential’ (NB 842). Thinking, then, entails thought. If a mind is not thinking any thoughts, it is not an active thing and so does not exist. The mind existing is the mind being active about something, and so, the mind existing is the mind having ideas. The very substance of a mind is that it is a thinking, perceiving thing: ‘a soul or spirit is an active being, whose existence consists not in being perceived, but in perceiving ideas and thinking’ (PHK 139). Therefore, a mind must have thoughts to be active, which is the essence of the soul’s existence.28 A mind is also then incapable of existing without any ideas. Minds and ideas are ontologically distinct (ODIS), yet existentially inseparable (EXIN). It is simply false to say that Berkeley ‘uses “substance” in one of its classic senses, to mean a being that is conceivable as existing separately.’29

Berkeley’s Actively Passive Mind

163

Now, ideas are the only ontologically distinct feature of Berkeley’s mental architecture. The will and the understanding are the mind and represent different functions of the active system. The acts of the will are volitions and the objects of the understanding are ideas. The substantial mind is an integrated, unified system of activity in which the will and the understanding operate. These operations produce volitions and ideas. As Berkeley notes, ‘it seems to me that Will & understanding Volitions & ideas cannot be severed, that either cannot be possibly without the other’ (NB 841). None of these elements is capable of separate existence. They all depend, ontologically and epistemically, upon one another. However, there are still distinctions to be made – even between the mind and volitions. Muehlmann suggests that since it is the very nature of a substance to be active, there may be no real distinction between the substance and its acts or volitions. They in fact could be one and the same.30 A substance then would not be a thing which acts, but simply actions, namely, volitions. This suggestion is appealing for Muehlmann, as it avoids a real substance account of mind which he finds in violation of Berkeley’s anti-abstractionism. However, Berkeley makes it clear that the will and the understanding are not reducible to their effects and objects: ‘I must not Mention the Understanding as a faculty or part of the Mind, I must include Understanding & Will etc. in the word Spirit by which I mean all that is active. I must not say the Understanding differs not from the particular ideas, or the Will from particular volitions’ (NB 848). The understanding, as the mind, is not the same as ideas. Furthermore, the will, as the mind, is not the same as volitions. Berkeley is explicit that ‘the Spirit, the Mind, is neither a Volition nor an idea’ (NB 849). It is a system of activity with interrelated, interdependent items. Even though the items are distinct, they are nonetheless existentially inseparable (EXIN). To be clear, by system I am not referring to the view that individual parts are combined to create a whole. This is not like a mechanical item such as a watch that is composed of different parts which function together to create the system ‘watch.’ True, the watch is only a watch when it is functioning with all the parts working together and it is ‘telling time.’ However, parts of the watch still retain their identity as things apart from the system. The cog is still a cog even if it is removed from the watch. The system of the mind is not like that. The mental items are incapable of being what they are removed from the system. Ideas cease to be ideas if they are not perceived. Volitions cease to be

164 Genevieve Migely

acts of the will if there is no will. The will and the understanding are not parts of the mind; they constitute the mind itself. They are not parts of the system; they are the system. Thus, by system I am not referring to a collection of things that are part of a whole to serve some function. Rather, by system, I mean one simple, undivided, substantial unity of activity. This interpretation constitutes a radical departure from the common view of Berkeley as continuing the Cartesian and Lockean tradition of substance as an independent existent with ideas as modes inhering in it. In this sense I adopt a view similar to that found in the work of Stephen Daniel, who emphasizes a different influence on Berkeley. Daniel argues, quite convincingly, that Berkeley is following the lead of Ramus’s Stoic logic in his account here. He advances a semantic analysis of the relationship between the mind and its ideas. The mind, according to this view, ‘is essentially linguistic and rhetorical,’ where ‘the meaning or logismos of each thing is defined by its function in a structure of relations that constitutes logic.’31 This logic is the ‘semantic ontology’ by which one can understand the mind’s relation to ideas.32 In this discourse, the identity of any thing ‘consists in its affective place in the discourse.’33 Ideas become intelligible, that is to say exist, only when perceived by a mind. In turn, a mind is what it is in its identification of ideas. Daniel claims that ‘mind is the principle or activity whereby ideas are identified and subsequently thought in relation to one another.’34 However, the mind remains distinct from ideas. In fact, Daniel employs the Stoic distinction here between subsistence and existence to demonstrate the different ways in which minds and ideas can be said to exist, or be identified. Existence is being perceived. Since minds cannot be the object of perception, they cannot be said technically to ‘exist’ in the same sense that ideas exist. Daniel explains that ‘the act of perceiving identifies a thing and constitutes the existence of the idea. This activity is obviously not the same as the things it produces, so it is technically incorrect to say that spirits exist; rather, they subsist.’35 Though minds and ideas are different ways of being, they are inseparable. Unlike the Cartesian and Lockean version of immaterial substance, the mind is not an independently existing entity. Instead, ‘the subsistence of minds or souls is nothing other than the existence of ideas.’36 Such an account intimately ties minds and ideas while maintaining the distinction between them. ‘To say that an idea exists means nothing more than that it is the object of mind, and to refer to the subsistence or being of a mind is to refer to the existence of its ideas. There-

Berkeley’s Actively Passive Mind

165

fore, to think that minds exist in the same way that ideas exist is to think that minds are things like ideas; and that is something Berkeley repeatedly cautions against.’37 This use of Ramist logic renders Berkeley consistent in his claims that minds and ideas are distinct and yet dependent upon each other. Although I continue to employ, like Berkeley does, the Cartesian terminology of substance, Daniel and I arrive at virtually the same basic conclusion regarding the Berkeleian mind: it is activity. Activity, not independent existence, is what defines substance for Berkeley. The mind may be inseparable from ideas and volitions, but it still remains distinct from them. Nor can mind or spirit be a volition, for as Daniel puts it, ‘mind or spirit is the will to differentiation by means of which ideas are identified.’38 Where I retain the ontological account of substance, Daniel provides a semantic account of substance: ‘This use of the word substance to refer to the topic of discussion – as, for example, when we ask, what is the substance of your objection? – is precisely what Berkeley has in mind when he applies the term to God and finite minds.’39 According to Daniel, in the discourse that comprises reality, immaterial substance is the activity by which ideas are identified, differentiated, and related to each other. Berkeley’s substance, on both of our views, is activity. The activities of will and understanding that comprise the mind are distinct from and yet inseparable from volitions and ideas. My endeavour here has been to show that most critics of Berkeley take him to be saying something he is not. Ultimately, Berkeley’s active mind is not challenged by his position that the mind is passive in immediate sense perception. In order for the criticism to hold that Berkeley offers an inconsistent view of the mind here, he must have a conception of the soul as passive in the sense of utterly inert and inactive. Only this use of ‘passive’ would violate the notion that the mind is always active. However, Berkeley does not regard the soul at any time as simply absorbing sensory information. The mind is not a sponge. Instead, when the mind perceives immediate sensory ideas, the mind is at the same time operating upon those ideas. As shown above, ideas are attended by volition involved in hedonic sensation. There are not two different ideas here, one the idea of sense and the other the idea of pleasure or pain. The idea of sense is the idea of pleasure or pain. Again, a great heat is a great pain; it is one and the same idea. The corresponding desire or aversion and assent to these ideas are all part of the package of perception (NB 692). It is not the next step or phase in the mental pro-

166 Genevieve Migely

cess. It is one activity. In perception, the mind is not at first passive and then active. In the very reception of ideas, the soul is operating about these ideas in its evaluation and confirmation of them, forming its own ideas about the ideas it has received. Additionally, for Berkeley, the mind is always thinking. He calls the mind ‘passive’ in immediate sense perception to show the contrast between ideas we do and do not cause. It is crucial to make this distinction in order to safeguard our commonsense realism that we are not the cause of the physical world. McCracken is entirely right in his assertion that ‘perceiving something, in an important sense, is not an action at all’ insofar as perceptions are understood in terms of their cause.40 However, this assumes that ‘action’ means only ‘the cause of ideas of sense.’ What is missing in this interpretation is Berkeley’s view that the mind is actively causing processes and ideas about ideas of sense. Further, this view of the mind does not violate its simplicity and unity. Perceiving and willing are both actions of the same mind. Since passivity in immediate sense perception still involves activity, the unified mind is never in opposite states of activity and inactivity.41 The apparent inconsistency dissipates when one realizes that Berkeley never considers the mind to be passive in the same sense as ideas. In fact, Berkeley never uses the terms ‘inert’ or ‘inactive’ to describe the mind. These terms are reserved exclusively for ideas. The mind is passive in the sense that it is not the cause of the ideas it is thinking about, but it is never passive in the sense that it is not causing anything. The mind is active in perception in the very act of perceiving as well as the cause of operations and the cause of ideas that are the result of those operations. The nature of mind is consistent in itself as well as with Berkeley’s dualism. Berkeley did not slip up; passivity in perception secures his commonsense realism and maintains his ontology of active substance. Notes 1 See Charles McCracken, ‘Berkeley’s Notion of Spirit,’ History of European Ideas 7 (1986): 602; Robert Muehlmann, Berkeley’s Ontology (Indianapolis: Hackett, 1992), 172–83; George Stack, ‘Berkeley’s Concept of Existence,’ Modern Schoolman 53 (1976): 283–9; Tom Stoneham, Berkeley’s World: An Examination of the Three Dialogues (Oxford: Oxford University Press, 2002), 81; George Pitcher, Berkeley (London: Routledge and Kegan Paul, 1977), 204; and I.C. Tipton, Berkeley: The Philosophy of Immaterialism (London: Methuen, 1974), 265.

Berkeley’s Actively Passive Mind

167

2 Stack finds Berkeley’s reliance on perceiving to demonstrate the soul’s active nature misguided: ‘Berkeley cannot consistently maintain that the active nature or “structure” of the mind is manifested in the process of perception as he has described it. Thus, when Berkeley does say that the existence of an active self consists in “perceiving ideas,” this is unconvincing in light of his depiction of perception as a form of passivity. In order to accept such a claim, we would have to assert the paradoxical view that the existence of an active self is manifested in passivity!’ See Stack, ‘Berkeley’s Concept of Existence,’ 283. 3 Stoneham, Berkeley’s World, 81. 4 Pitcher, Berkeley, 204. 5 Stack, ‘Berkeley’s Concept of Existence,’ 289. 6 Charles McCracken, ‘Berkeley’s Cartesian Concept of Mind: The Return through Malebranche and Locke to Descartes,’ Monist 71 (1988): 609. 7 As Bertil Belfrage argues in this volume, Berkeley’s view on passivity in perception may have changed during the composition of his Notebooks. The issue I am now addressing is how passivity affects his developed view of mind. 8 Kenneth Winkler tries to save Berkeley from this discrepancy by viewing the tulip-smelling argument as simply a tactic that Berkeley is employing in argumentation. The passage in question occurs in the Dialogues where Berkeley has been known to go along with opposing views to prove the point that even if he gives them what they say they want, they are still left with his immaterialism. There is some evidence that Winkler may be on to something, since Berkeley does show at the end of the argument that even if perception is passive, it still cannot exist in an unperceiving substance. However, Berkeley also mentions the mind’s passivity in perception in the letter to Johnson. In this letter, he is not using any form of argumentation to prove his point. See Kenneth Winkler, Berkeley: An Interpretation (Oxford: Clarendon Press, 1989), 7–9. 9 ‘In reading a book, what I immediately perceive are the letters, but mediately, or by means of these, are suggested to my mind the notions of God, virtue, truth, etc.’ (DHP 174). 10 It is, however, ambiguous what Berkeley means by ‘object’ since at times he uses it for an immediate perception and at other times, like here, he refers to it as a collection of such perceptions. On this issue, see the essays by Richard Glauser and Marc Hight in this volume. 11 ‘When I perceive a great number of intermediate objects such as fields, houses and rivers ... I thence form a judgment or conclusion that the object I see beyond them is at a great distance’ (NTV 3); ‘When upon perception of

168 Genevieve Migely

12 13

14

15

16

17 18 19

an idea I range it under this or that sort, it is because it is perceived after the same manner or because it has a likeness or conformity with, or affects me in the same ways as, the ideas of the sort I rank it under’ (NTV 128); ‘Number is entirely a creature of the mind’ (NTV 109; also PHK 12) (all emphases are mine). He gives an example of how a coach is suggested by a sound in DHP 204. For Berkeley, you perceive one sense, and other senses are suggested to your mind from that. These quotes demonstrate how mediate perception is an activity of the mind involving volition. There is a further issue about whether or not these relations are naturally or necessarily given to us. I shall take this up in a future work. Human agents can act in the natural world by causing the presentation of certain ideas of sense. In the tulip-smelling example, Philonous causes the bodily activity (reaching with his arm, plucking the flower with his hand, drawing the hand holding the flower near his nose, and inhaling through his nose) which allows him to experience that particular idea of sense (flower smell). Philonous does not cause the smell of the flower. However, he does cause the voluntary action that allows him to experience the smell of the flower. Berkeley’s definition of the soul as ‘that active principle of motion and change of ideas’ conveys this notion that finite activity in the natural world is limited to the rearrangement and presentation of ideas of sense rather than the production of ideas of sense themselves (PHK 27). I say ‘many’ rather than ‘all’ in order to show that we can choose which ideas go with which other ideas. This must be the case, or else we could never be mistaken in making judgments. However, Berkeley wants to claim at the same time that we are often led to certain inferences based upon repeated experience of always seeing such ideas constantly conjoined. ‘There has grown a habitual or customary connexion between those two sorts of ideas, so that the mind no sooner perceives the sensations arising from the different turn it gives the eyes ... but it withal perceives the different idea of distance which was wont to be connected with that sensation; just as upon hearing a certain sound, the idea is immediately suggested to the understanding which custom had united with it’ (NTV 17). I say ‘not always decide’ because there are clearly cases where we do decide and we decide incorrectly. Otherwise, we could never be mistaken since God would simply connect the dots for us all the time. In his essay in this collection, Belfrage claims that this is a late development in Berkeley’s Notebooks. ‘its [the mind’s] acts about ideas’ (PHK 142; my emphasis). Stack resigns himself to a ‘partial’ resolution to the issue here since we are

Berkeley’s Actively Passive Mind

20 21

22

23 24

169

not passive in all types of perception – only in immediate sense perception. However, Berkeley would still be considered inconsistent even if passivity only occurred in some, not all, perception. ‘in proportion to the Pleasure & pain Ideas are attended with desire aversion & other actions which include volition’ (NB 692). ‘Again, the things perceived by sense may be termed external, with regard to their origin, in that they are not generated from within, by the mind itself, but imprinted by a spirit distinct from that which perceives them’ (PHK 90). The other indication that Berkeley evidently allows for activity in immediate sense perception is his basic tenet that the understanding is perceiving: ‘as it perceives ideas, it is called the understanding’ (PHK 27). It is an elemental feature of Berkeley’s idealism and immaterialism that ideas can only exist in a mind. He finds it a strange opinion that sensible things have any existence apart ‘from their being perceived by the understanding’ (PHK 4). This perception by the understanding is the same thing as having ideas, which is itself a form of thinking. As he states, ‘the things by me perceived are known by the understanding, and produced by the will, of an infinite spirit’ (DHP 215). Both the ‘having’ and the ‘knowing’ ideas are clearly an activity of the mind since it involves thinking. ‘Thoughts do most properly signify or are mostly taken for the interior operations of the mind, wherein the mind is active’ (NB 286). The mind is considered entirely active for Berkeley because it is always thinking, even though it may not always be causing the ideas about which it thinks. Berkeley is perfectly consistent in stating that the mind is ‘wholly active’ since he contends that the mind is ‘always thinking’ (PHK 98). Unless Berkeley is taken to be stating a contradiction every time he states that the soul is a ‘perceiving, active, thinking being,’ then it is understood that Berkeley takes perceiving as an activity rather than as being inoperative and inert (PHK 2, 27, 91, 136–9; DHP 231, 233–4, 239). But, in fact, Berkeley does consider the understanding a power of the mind. ‘All things by us conceivable are 1st thoughts 2dly powers to receive thoughts, 3dly powers to cause thoughts’ (NB 228). He maintains this view even in his published works. ‘Do but leave out the power of willing, thinking, and perceiving ideas, and there remains nothing else wherein the idea can be like a spirit ... If therefore it is impossible that any degree of those powers should be represented in an idea, it is evident that there can be no idea of a spirit’ (PHK 138). Again, perception is an activity of the understanding. Tipton, Berkeley, 265. Grave and Muehlmann interpret these passages as identifying the mind

170 Genevieve Migely

25 26 27 28

29 30 31 32 33 34

35 36 37 38

39 40 41

with its ideas. See S.A. Grave, ‘The Mind and Its Ideas: Some Problems in the Interpretation of Berkeley,’ in Locke and Berkeley: A Collection of Critical Essays, ed. C.B. Martin and David Armstrong (Garden City, NY: Anchor Books, 1968), 298; and Robert Muehlmann, Berkeley’s Ontology (Indianapolis: Hackett, 1992), 172–8. ‘A soul or spirit is an active being, whose existence consists not in being perceived, but in perceiving ideas and thinking’ (PHK 139). A.A. Luce, Berkeley’s Immaterialism (London: Thomas Nelson, 1945), 50. For more on this point, see Phillip Cummins’s essay in this collection. Berkeley explains that ‘we have an intuitive Knowledge of the Existence of other things besides our selves & even praecedaneous to the Knowledge of our own Existence. in that we must have Ideas or else we cannot think’ (NB 547). See, for example, Robert A. Adams, ed., Three Dialogues between Hylas and Philonous (Indianapolis: Hackett, 1979), xx. Muehlmann, Berkeley’s Ontology, 178. Stephen Daniel, ‘The Ramist Context of Berkeley’s Philosophy,’ British Journal for the History of Philosophy 9 (2001): 491. Stephen Daniel, ‘Edwards, Berkeley, and Ramist Logic,’ Idealistic Studies 31 (2001): 56. Daniel, ‘Edwards, Berkeley,’ 63. Stephen Daniel, ‘Berkeley’s Stoic Notion of Spiritual Substance,’ in New Interpretations of Berkeley’s Thought, ed. Stephen H. Daniel (Amherst, NY: Humanity Books, 2007). Stephen Daniel, ‘Berkeley, Suárez, and the Esse-Existere Distinction,’ American Catholic Philosophical Quarterly 74 (2000): 633. Daniel, ‘Berkeley, Suárez,’ 623. Stephen Daniel, ‘Berkeley’s Christian Neoplatonism, Archetypes and Divine Ideas,’ Journal of the History of Philosophy 39 (2001): 249. Daniel, ‘Edwards, Berkeley,’ 64. Marc Hight and Walter Ott, ‘The New Berkeley,’ Canadian Journal of Philosophy 34 (2004): 1–24, interpret Daniel’s position as a version of the bundle theory, stating that minds are simply ‘collections of ordered activities’ (2). Basically, they view Daniel as claiming that the mind is volitions. However, this is obviously a mistaken reading of Daniel’s position. Daniel does not hold that the mind is a volition or even a collection of volitions. Stephen Daniel, ‘Berkeley’s Pantheistic Discourse,’ International Journal for Philosophy of Religion 49 (2001): 184. McCracken, ‘Berkeley’s Notion of Spirit,’ 599. Tipton also argues that sense perception is ‘one manifestation of the activity

Berkeley’s Actively Passive Mind

171

of that perceiving thing which is essentially active’ (Tipton, Berkeley, 269). Although he claims that sense perception is activity, he does not offer a complete explanation of what occurs in immediate sense perception to demonstrate the activity.

Berkeley’s Four Concepts of the Soul (1707–1709) b e rt i l be l f rage

‘The most important source for our knowledge of Berkeley’s thinking about spirit’ is, according to Ian Tipton,1 a manuscript often referred to as Notebook A.2 But the difficulty is that Berkeley presents very different views in this manuscript. Early on he develops a bundle theory: + Mind is a congeries of Perceptions. Take away Perceptions & you take away the Mind put the Perceptions & you put the mind. (NB 580)

But later on (NB 847), as well as in the Principles, ‘mind,’ ‘soul,’ or ‘spirit’ is defined as: one simple, undivided, active being: as it perceives ideas, it is called the understanding, and as it produces or otherwise operates about them, it is called the will. (PHK 27)

In the former case, ‘the mind’ (or ‘soul’ as in NB 577) is defined in terms of perceptions or ideas and is no separate being distinct from these ideas. But in the latter case, a spirit is a separate being very different from ideas; it ‘perceives’ and ‘operates about them.’ In this essay, I shall reconstruct the steps and developments from the early notes in 1707 to the view that he was to publish in the Principles.3 1. Berkeley’s Early Principles Berkeley opens Notebook A with a few assumptions which I shall refer to as his Early Principles. One of these is the assumption that

Berkeley’s Four Concepts of the Soul

173

(P 1) ‘All our knowlege & contemplation is confin’d barely to our own Ideas.’ (NB 606) To this he adds the semantic principle: (P 2) ‘No word to be used without an idea.’ (NB 422) and the decision to confine himself to the cognitive field of discourse as defined in (P 1): (P 3) ‘No reasoning about things whereof we have no idea.’ (NB 421) He uses these principles to define the meaning of crucial terms, to reject non-descriptive terms and propositions as nonsensical, and to mark the boundary of cognitive discourse. Another important assumption is that (G) God is the only active being in the world. When we read about ‘a Being wch wills wn the Effect follows the volition,’ then this is God, who produces ‘perceptions in us’ (NB 499): we are totally passive in thinking and perceiving. This passivity thesis plays an important part in the first three-fifths of Notebook A.4 2. A Conventional Concept of the Soul In the opening pages of Notebook A, Berkeley is reluctant to use his critical weapons against the concept of mind, but the tension between the Early Principles and a conventional concept of the soul comes to the surface at NB 576, where he writes: S We think we know not the Soul because we have no imaginable or sensible idea annex’d to that Sound. This the Effect of prejudice.

In this note, he uses the same words as he normally uses when rejecting empty talk as mere nonsense: ‘We have no imaginable or sensible idea annex’d to’ the term ‘soul,’ he says. Up to this point the conclusion has been: ‘Therefore absurd to talk or make propositions about it’ (NB 417). And even now, his immediate reaction is to follow his Early Principles.5 This is the first clash on the concept of the soul.

174 Bertil Belfrage

3. The Bundle Theory of the Soul After NB 576, Berkeley turns over the page and starts in note NB 577 to develop a bundle theory in accordance with his Early Principles. As the received view is that the religious Berkeley could not have defended the same view as the atheist David Hume was later to publish, it should be noted that, contrary to Hume’s theory in the Treatise (I.iv.6, 252), Berkeley’s version is embedded in a religious context. According to (G), God causes all our ideas; we exist no longer than it pleases God to produce ideas in our understanding. I shall return in a minute to this consequence of Berkeley’s theory (the intermittent existence of the soul) as well as to his early rejection of thinking substances; but I open with his definitions of ‘mind,’ ‘soul,’ ‘understanding,’ ‘volition,’ ‘will,’ and ‘person.’ ‘Mind,’ ‘soul,’ and ‘understanding’ are defined in this simple and straightforward way: + The very existence of Ideas constitutes the soul. (NB 577) + Mind is a congeries of Perceptions. Take away Perceptions & you take away the Mind put the Perceptions & you put the mind. (NB 580) The Understanding [is] not distinct from particular perceptions or ideas. (NB 614)

These three terms are all defined as a set of particular ideas or perceptions. It is true, in the early, uncritical part of Notebook A (when he refuses to apply his Early Principles to the concept of the soul), Berkeley distinguishes between ideas and a separate active being which ‘thinks’ (NB 437). But now, after the first clash, he explicitly rejects this opinion: + Say you the Mind is not the Perceptions, but that thing wch perceives. I answer you are abused by the words that & thing these are vague empty words without a meaning. (NB 581)

He adds this more elaborate argument against the mind as a separate being which perceives ideas: + Consult, ransack yr Understanding wt find you there besides several

Berkeley’s Four Concepts of the Soul

175

perceptions or thoughts. Wt mean you by the word mind you must mean something that you perceive or yt you do not perceive. a thing not perceived is a contradiction. to mean (also) a thing you do not perceive is a contradiction. We are in all this matter strangely abused by words. (NB 579)

This argument against a perceiving soul was later on, in the Principles, developed into an attack on the concept of matter. If we perceive only ideas, and if thinking about something entails being aware of an idea, then it is a contradiction to say, ‘I am thinking about, or perceive something, i.e. an idea, which is not an idea.’ And if a proposition is meaningful only if it expresses something imaginable, then, given that ‘our contemplation is confined to our own ideas,’ it is contradictory to say, ‘I contemplate what this proposition is about, i.e. an idea, but it is no idea.’ Again, the first time he uses this argument is here, not against matter, but against the conception of the soul as a separate being. The next concept of importance for the bundle theory is ‘volition.’ In the early part of Notebook A Berkeley refers so frequently to Locke’s Essay, that one can almost take the entries – particularly the series of notes on will and volitions – for commentaries in the margin of the Essay. We find, for instance, several comments on Locke’s view that the ‘motive to change,’ or to move some part of one’s body knowingly or willingly, ‘is always some uneasiness’ (Essay II.xxi.29, 249). God, Locke says, ‘has put into Man the uneasiness of hunger and thirst, and other natural desires ... to move and determine their Wills, for the preservation of themselves, and the continuation of their Species’ (Essay II.xxi.34, 252; cf. II.xxi.5, 236; II.xxi.15, 241; and II.xxi.28, 248). Berkeley’s first comment is: + Will not rightly distinguish’d from Desire by Locke. it seeming to superadd nothing to the Idea of an Action but the Uneasiness for it’s absence or non Existence. (NB 598)

Suppose, to take Locke’s example, that the action is to eat and that the uneasiness is to feel hungry. We can be uneasy or feel hungry without eating, Berkeley argues, just as we can think of eating without doing it, and we can be both hungry and vividly thinking about eating without doing so. The conclusion is that not even ‘uneasiness & Idea together’ is a sufficient condition for voluntary action (NB 611–13). But he accepts that ‘to have a desire for doing something’ means to have an idea of this

176 Bertil Belfrage

action together with a sensation of uneasiness for not doing it. Desire is therefore connected with pleasure as aversion is to pain, he argues. And the degree of desire and aversion can be measured ‘in proportion to the Pleasure & pain Ideas are attended with’ (NB 692). But as not even ‘the greatest uneasiness,’ and thus not even the greatest desire or aversion, is a sufficient condition for us to act (NB 624), he prefers to speak about a certain God-given disposition: S Complacency seems rather to determine or precede or coincide wth & constitute the Essence of Volition than uneasiness. (NB 630)

Instead of Locke’s ‘uneasiness,’ Berkeley takes complacency to be the motor which makes us move to enjoy pleasure and avoid pain. It constitutes ‘the Essence of Volition’ by determining the chain of ideas and actions in an actor’s way towards satisfaction.6 This view on volitions is most fully presented in a sermon preached in January 1708. Here, ‘volition’ includes (i) ideas of things which are attended with desire or aversion, (ii) ideas of means to reach or avoid them, and (iii) (ideas of) the actions actually performed in a person’s progress towards complacency (W 7: 13). For example, I feel hungry, look at a piece of bread, and become uneasy for not eating it; next, the God-given inclination to reach complacency makes me take the bread, eat it, and become satisfied. We can describe each single element in this process, but we cannot isolate one of the elements and refer to it as a volition; and we cannot freeze the process and picture it at one point of time (NB 643, 644, 657, 665, 706). Volitions are closely connected to the concept of will: ‘The Will [is] not distinct from Particular volitions’ (NB 615). Just as ‘mind,’ ‘soul,’ and ‘understanding’ mean a set of particular ideas or perceptions, ‘the will’ is a set of particular volitions. This does not mean that ‘the will’ refers to a real essence: it is merely ‘a Word’ or a nominal unity; it does not ‘make one Will’ (NB 714). This reflects a passage in that part of the Essay where Locke (like Hobbes) speaks about a ‘way of talking’ about will and understanding which has ‘produced great confusion’: If it be reasonable to suppose and talk of Faculties, as distinct Beings, that can act, (as we do, when we say the Will orders, and the Will is free) ... we may as properly say, that ’tis the singing Faculty sings, and the dancing Faculty dances; as that the Will chuses, or that the Understanding conceives (Essay II.xxi.17, 242)

Berkeley’s Four Concepts of the Soul

177

Just as no ‘dancing Faculty dances,’ there is no will that wills, according to the early Berkeley; to accept a substantivist concept of ‘will’ is to be ‘cheated by those general terms’: S If you ask wt thing it is that wills. I answer if you mean Idea by the Word thing or any thing like any Idea, then I say tis no thing at all that wills. This how extravagant soever it may seem yet is a certain truth. we are cheated by those general terms, thing, is &c. (NB 658) +S Again if by is you mean is perceived or do’s perceive. I say nothing wch is perceived or does perceive Wills. (NB 659)

Thus, by a line of argument parallel with the one he uses to reject mind, soul, or understanding as separate beings, he concludes that there is ‘no thing at all that wills’ (except God). There is only a succession of ideas and a series of particular volitions caused by God (NB 606, 579, 581, 587, 614). ‘Person,’ finally, is the most general of the technical terms that Berkeley uses in this context. When he comments on potential criticisms by ‘the Church or Church-men,’ he provides us with this information about his early concept of ‘person’ (a technical term in Anglican liturgy): S The Concrete of the Will & understanding I must Call Mind not person, lest offence be given, there being but one volition acknowleged to be God. Mem: Carefully to omit defining of Person, or making much mention of it. (NB 713) S You ask do these volitions make one Will. wt you ask is meerly about a Word. Unite being no more. (NB 714)

The concept of a person is identified with will and understanding, just as he says in this note: S Doctrine of identity best explain’d by Taking the Will for Volitions, the Understanding for ideas. (NB 681, cf. NB 614, 615, 689)

Thus, there are two sequences of events: (i) perceptions or ideas and (ii) volitions. ‘Understanding’ refers to the former and ‘will’ to the latter – provided that we do not take the understanding to be a separate being distinct from a succession of ideas (NB 579, 587, 614) or forget that ‘The

178 Bertil Belfrage

Will [is] not distinct from Particular volitions’ (NB 615), only ‘a Word’ or a nominal unity. That is to say, the term ‘person’ is used for an unbroken chain of volitions and ideas succeeding each other without (conscious) interruption. One of the most controversial consequences of Berkeley’s definition of ‘mind,’ ‘soul,’ ‘understanding,’ ‘volition,’ ‘will,’ and ‘person’ is the intermittent existence of the soul (will or person). He formulates it in a seemingly contradictory way. First he says, ‘the mind always & constantly thinks’; but then he adds that, sometimes, the mind does not think. That is when ‘there is no time no succession of ideas’; and, when this happens, ‘the mind exists not.’ Thus, ‘In Sleep & trances the mind exists not there is no time no succession of Ideas’ (NB 577, 460, 488, 590, 647, 651). But what does Berkeley mean in this context when he says that the soul always thinks? If he has in mind the eternal existence of the soul, then it should mean: (1) If X is a soul at one point of time, then it is always the case at any future point of time, say t, that X is thinking at t. But this does not allow that, sometimes, the soul does not think or that there is a future point of time when ‘the mind exists not.’ When we read that the proper measure of time is a ‘succession of ideas’ (NB 460, 488, 590, 647), and Berkeley maintains that ‘Each Person’s time being measured to him by his own Ideas’ (NB 590), then this opens the possibility that (2) If X is a mind, then it is always the case, at any point of X’s private time, that X is thinking at that very point of time. But from this solipsistic interpretation of ‘always thinking’ we cannot conclude that a person P does not exist at t, if not thinking at t, because t ranges over P’s private time continuum, and when there is no succession of P-ideas, then there is no P-time. As far as P’s private time is concerned, therefore, there is no point at which P’s ‘mind exists not.’ In Berkeley’s terms: T No broken intervals of death or Annihilation Those Intervals are nothing. Each Person’s time being measured to him by his own Ideas (NB 590; my emphasis)

Berkeley’s Four Concepts of the Soul

179

Just as waking up sometimes appears to follow the next moment to falling asleep, a meaningful way of formulating the question of immortality or resurrection is to ask: will there be any succession of ideas after the last one I had, or will have, at the moment of death? But this is not to say that the soul is ‘always thinking’ in the sense suggested in (1).7 Although there are no gaps in one person’s private time, we can very well observe how other persons are unconscious and thus unthinking at times, as ‘in sleep and trances.’ To ask ‘Does that person’s soul exist?’ would then be the same as asking ‘Is it likely that there are any ideas succeeding each other in that person’s mind?’ This is a plain empiricist approach in accordance with the bundle theory. Although the non-existence of thinking substances is a flat contradiction of what Berkeley was later to publish, it is a direct consequence of the Early Principles. And, as previously noted, he actually rejects the existence of a separate ‘thing wch perceives’ (NB 581). ‘Say you,’ he says, ‘there must be a thinking substance. Something unknown wch perceives & supports & ties together Ideas’ (NB 637). But according to (P 1), only a confused person or a liar could claim to imagine that such a thing as a thinking substance exists: ‘a Man can never be brought to Imagine any thing should exist whereof he has no Idea’ (NB 639).8 And, according to (P 2), we are using ‘empty words without a meaning’ if we speak about thinking substances that we have no idea of (NB 581, 639). His decision in (P 3), moreover, to confine himself to things that we do have an idea of effectively excludes any comment on thinking substances. As we cannot form a ‘mental proposition’ about such a thing, the mystery of immortality concerns things above our knowledge, and in matters of revelation no more than ‘an Humble Implicit faith becomes us’ (NB 720).9 But in the early part of Notebook A, he goes one step further in that he rejects thinking substances not only as being ‘unknown’ but even as being a useless conception (NB 637–9) – an argument which he later used against the concept of material substance.10 4. Abandoning the Early Principles So long as Berkeley argues from his Early Principles, his view on ‘ideas’ is simple: (1) ‘A person P has an idea of X’ means that P is aware of an image-picture of X.

180 Bertil Belfrage

But according to his New Theory of Vision, we can, for example, have an idea of, or see, invisible passions in a person’s mind by observing the colour of his face (NTV 9). In this example, (2) P has an idea of (the invisible) X by means of (the visible) Y. Several requirements have to be satisfied before we can see things which are (strictly speaking) invisible. I shall mention only two of them. The first is that ‘we see shame or anger in the looks of a man ... for no other reason, than barely because [certain colours and alterations of countenance] have been observed to accompany [shame or anger]’ – ‘without which experience, we should no more have taken blushing for a sign of shame, than of gladness’ (NTV 65). To ‘see shame’ or ‘have an idea of’ a passion, therefore, does not mean to perceive it by sense but to infer it by a primitive form of induction. The second requirement, based on these observed regularities, is that a psychological sign/signified relation has been established in a person’s mind between X and Y in (2). Once established, ‘no sooner shall he behold that colour to arise in the face of another, but it brings into his mind the idea of that passion which has been observed to accompany it’ (NTV 23). Provided that both requirements have been satisfied, a condensed way of formulating one aspect of the new sense in which Berkeley uses such terms as ‘to perceive’ or ‘to have an idea of’ something is this: (3) P can ‘imagine,’ ‘perceive,’ or ‘have an idea of’ (an invisible) X by means of (the visible) Y, if P has learned to take Y as ‘a sign of’ X. Or, in Berkeley’s terms: Ideas which are observed to be connected with other ideas come to be considered as signs, by means whereof things not actually perceived by sense are signified or suggested to the imagination, whose objects they are, and which alone perceives them. (TVV 39; my emphasis; see also NTV 138–48)

Although Berkeley develops this theme in the New Theory of Vision and thus in Notebook B (the notebook on vision), there is a hint of it even in Notebook A:11 S To be sure or certain of wt we do not actually perceive (I say perceive not imagine) We must not be altogether Passive, there must be a disposition

Berkeley’s Four Concepts of the Soul

181

to act, there must be assent, wch is active, nay wt do I talk There must be Actual Volition. (NB 777)

The ‘nay wt do I talk’ indicates a state of searching, and the development recorded in this part of the notebook is dramatic. He explicitly turns his back on his Early Principles in a note where he says: It seems to me that we have no certainty about Ideas but onely about Words. (NB 731)

It is true, he later noted, that ‘this seems wrong’ (NB 731a), but the reason he gives for this sudden change is nonetheless interesting: Tis improper to say I am certain I see, I feel &c. there are no Mental propositions form’d answering to these Words & in simple perception tis allowed by all there is no affirmation or negation & consequently no certainty. (NB 731)

He distinguishes between ‘simple perception’ (captured in statements such as ‘I see,’ ‘I feel’), and another kind of perception which answers to a ‘mental proposition’ such as ‘the Horse is White.’ As we cannot express certainty about a proposition ‘without affirmation or negation,’ we cannot believe that something is true or false without exerting an attitude towards a ‘mental proposition.’ This opinion, which Berkeley styles ‘my Doctrine of Certainty,’ also includes ‘a disposition to act’: if we believe that a proposition is true, then we will act, or be prepared to act, accordingly. And considering that what makes us act is our enjoyment of pleasure and avoidance of pain, and that no ideas are perfectly void of all pleasure, pain, or uneasiness, then ‘it seems,’ he argues, ‘there can be no perception, no Idea without Will’ (in a new sense of the term).12 Even to form a mental proposition, such as the imaginary ‘the horse is blue,’ is to exert our faculty of ‘Composition of Ideas’ which entails activity (NB 731, 777, 809, 753, 776, 833). The result is a new sense of ‘will’ and ‘understanding.’ So long as perceiving meant a passive reception of God-given ‘pictures,’ to perceive was perfectly unproblematic. But his inquiries into the psychology of perception puts this issue into the melting pot. He goes from denying ‘certainty about Ideas’ (NB 731) to the view that the understanding is infallible (NB 816) and ends up saying that ‘Understanding is in some sort an Action’ (NB 820–1, 833), not a set of passively received image-

182 Bertil Belfrage

pictures. When he still took the will for ‘particular volitions’ (NB 615), he denied that ‘the will wills’; but now he says: ‘I am of another Mind.’ At this later stage, the will is, or performs, acts (NB 795–8). Earlier, ‘The very existence of Ideas constitutes the soul,’ but now ‘our Ideas are distinct from the Mind’ (NB 847). Thus, at this later stage, he explicitly rejects the definitions of ‘understanding’ and ‘will’ that he defended in the first three-fifths of Notebook A: ‘I must not say that the Understanding differs not from particular Ideas, or the Will from particular Volitions’ (NB 848). The bundle theory is even rejected as a ‘Dangerous’ view: ‘that were the Way to prove spirits are Nothing’ (NB 788, 808, 872; cf. 577). In NB 806, moreover, he criticizes Hobbes for taking ‘those things for nothing wch are not Ideas’ – which he himself had done in the early part of Notebook A. 5. The Soul as ‘Pure Act’ ‘Thinking’ no longer refers to a succession of ideas; it is ‘an act,’ and according to the next phase of Berkeley’s development, ‘this Act is the Spirit’ (my emphasis). He refers to it as ‘pure act’ (purus actus), but, unhappy about this scholastic term, he then prefers ‘pure Spirit or active Being’ (NB 829, 867, 870), which is ‘neither a Volition nor an Idea’ in the old sense of these terms (NB 841, 848–9). At this stage, he summarizes his view as follows: S Substance of a Spirit is that it acts, causes, wills, operates, or if you please (to avoid the quibble yt may be made on ye word it) to act, cause, will, operate & it’s substance is not knowable not being an Idea. (NB 829)

It is misleading, he says, to use the term ‘it’ for the soul in phrases such as ‘It acts,’ ‘It wills,’ and so on: ‘spirit’ means ‘to act,’ ‘cause,’ ‘will.’ Just as we can explain and predict natural events by the law of gravitation, we can predict how a perceiver will judge about size or distance in a given situation by the laws established in empirical psychology. But it is as mistaken to hypostatize spirit and take it as an ‘it’ as to take ‘gravity’ to denote an actor and attribute to ‘it’ such qualities as ‘is red’ or ‘is moving.’13 In the literature on Berkeley, Geneviève Brykman has placed great emphasis on this psychological aspect and identifies Berkeley’s soul with the imagination.14 Stephen Daniel focuses on the soul’s activity of

Berkeley’s Four Concepts of the Soul

183

perceiving and willing: ‘Just as the existence of ideas consists in their being perceived,’ Daniel argues, ‘so also my own existence as a mind consists in perceiving ideas.’15 On these accounts, the controversial consequence is that the soul has an intermittent existence: without ideas, no imagination; and without mental acts, no soul. I agree, Berkeley did affirm the soul’s intermittence not only in the early part of Notebook A but even as late as in the Principles Manuscript (a manuscript of the end of the Principles), where he actually maintains that the soul ‘exists no longer than it thinks.’16 According to this manuscript, there are two kinds of beings besides God: perceiving spirits and perceived ‘sensible things.’ ‘Spirit’ means a succession of acts; that is, (1) If there is ‘spirit’ (not a spirit), then there is activity. Or, in a formulation that emphasizes intermittence: (1´) If there is no activity, then there is no ‘spirit.’ Again, according to this manuscript, the soul ‘exists no longer than it thinks’; there is not ‘a 3rd Sort of Being which exists tho it neither wills nor perceives nor is perceived’ (except God). ‘Sensible things’ exist no longer than they are perceived by a mind, and souls no longer than they think or perceive. The subsistence of the soul, finally, is explained by an element of mysticism: when the soul does not exist, it subsists in the divine mind.17 But this is not the view that he published. 6. A Metaphysical Concept of the Soul The intermittent existence of the soul creates a tension between it and Berkeley’s new concept of ‘sensible things.’ In the Principles Manuscript, just as in PHK 89, ‘sensible things’ are ‘dependent beings, which subsist not of themselves, but are supported by or exist in minds’ or souls – which themselves have an intermittent existence according to the manuscript. This tension comes to the surface in the very last sentence of the Principles Manuscript, which ends as follows, in mid-sentence: ... the Existence of other spirits is capable of being made known, only by the effects their Operations

He never added the verb (‘have,’ ‘produce,’ ‘cause,’ or whatever term

184 Bertil Belfrage

he may have chosen). Contrary to (1), he now speaks about spirits as separate beings. He mentions ‘their Operations,’ and adds this sufficient condition of their existence: (2) If there is activity, then there is a spirit. At this stage, he has abandoned not only the Early Principles but also the thesis (G), that God is the only active being in the world.18 Thence arose the metaphysical concept of the human soul that he included in PHK 27: (3) ‘A spirit is one simple, undivided, active being.’ Here, the soul is undivided, no longer a succession either of ideas or of acts; it is a separate ‘active being’ or thinking substance. This is the end of the development of Berkeley’s concept of the soul, as we can follow it in the manuscripts he wrote between 1707 and 1709. Notes I wish to express my gratitude to the Hultengrens Fond, Lund University, for a grant that made it possible to finish this essay. I am also grateful to Wolfgang Breidert, Geneviève Brykman, and Stephen Daniel for constructive discussions. 1 Ian Tipton, Berkeley. The Philosophy of Immaterialism (London: Methuen, 1974), 260–1. 2 A.C. Fraser found two manuscripts, Notebook A and Notebook B, which he published in 1871 as Berkeley’s Commonplace Book. Later editors followed Fraser in publishing these two manuscripts as one, but they differ from Fraser in editing B before A. This explains why the first note in A is numbered 400 by A.A. Luce, whose numbering I follow. He renamed them Philosophical Commentaries in 1944. 3 The traditional question (which I shall not ask) is: Which one is Berkeley’s ‘real’ concept of mind (presuming that he never changed)? Most scholars argue that the ‘real’ view is the one that he published in the Principles. See for instance A.A. Luce, The Dialectic of Immaterialism (London: Hodder and Stoughton, 1963), 24ff, 173. C.M. Turbayne takes the contrary view in ‘Berkeley’s Two Concepts of Mind,’ in Berkeley’s Principles of Human Knowledge: Critical Studies, ed. Gale W. Engle and Gabriele Taylor (Belmont, CA: Wadsworth, 1968), 24–33. Turbayne argues that the bundle theory is Berkeley’s

Berkeley’s Four Concepts of the Soul

4

5

6

7

8 9 10

11

185

‘real’ view, although he did not publish it in the Principles but kept it secret in his private notebook. The received view is that Berkeley continuously held the passivity thesis, according to which, in the act or activity of perceiving, we are entirely passive. What is given in vision is then like pictures on a camera screen. Some scholars take these ‘pictures’ to represent things in God’s mind; others take them to represent physical objects which Berkeley identifies with tactual objects. See C.M. Turbayne, ‘Commentary,’ in George Berkeley, Works on Vision, ed. C.M. Turbayne (Indianapolis: Bobbs-Merrill, 1963), vii–xlv. Cf. Genevieve Migley’s contribution to the present volume. Berkeley moves towards a bundle theory very early on in Notebook A, as when he says: ‘If there were no sensible ideas there could be no soul, no perception, remembrance, love, fear &c no faculty could be exerted’ (NB 478). But we do not have an explicit formulation of the tension between the Early Principles and a conventional concept of the soul until NB 576. Later on Berkeley develops this theme into an argument from design for the existence of God. Thus, in PHK 146, he asks the reader to consider ‘the exact harmony and correspondence of the whole, but above all, the never enough admired laws of pain and pleasure, and the instincts or natural inclinations, appetites, and passions of animals.’ David Berman argues in George Berkeley: Idealism and the Man (Oxford: Clarendon Press, 1994), 65, that the solipsistic concept of time ‘supplied him [Berkeley] with an argument for immortality. In short, if there is no public time, then there would seem to be no time in which I could cease to exist.’ But the solipsistic approach only helps us to formulate the question of immortality; the crucial link, the Resurrection, remains a mystery. For an analysis of Berkeley’s solipsist concept of time, see Geneviève Brykman, Berkeley et le voile des mots (Paris: Vrin, 1993), 93–106, 282–5. God’s existence is, however, a presumption of Berkeley’s entire metaphysical system. On this theme, see my ‘The Theological Positivism of George Berkeley (1707–8)’ (forthcoming). To accept the existence of material objects void of causal power, he says in PHK 19, ‘is to suppose, without any reason at all, that God has created innumerable beings that are entirely useless, and serve to no manner of purpose.’ For a more detailed account of the complex structure of Berkeley’s ‘ideas,’ see my ‘The Scientific Background of George Berkeley’s Idealism,’ in Eriugena, Berkeley and the Idealist Tradition, ed. Stephen Gersh and Dermot Moran (Notre Dame, IN: University of Notre Dame Press, 2006), 213–16.

186 Bertil Belfrage 12 In NTV 59 Berkeley notes, ‘We regard the objects that environ us, in proportion as they are adapted to benefit or injure our own bodies, and thereby produce in our minds the sensations of pleasure or pain.’ See my ‘The Constructivism of Berkeley’s New Theory of Vision,’ in Minds, Ideas, and Objects: Essays on the Theory of Representation in Modern Philosophy, ed. Phillip D. Cummins and Guenter Zoeller (Atascadero, CA: Ridgeview, 1992), 176. 13 It is true, inspired writers speak about immortal souls. But this concerns things which are ‘not knowable’: we cannot affirm or deny anything about things outside the scope of human knowledge. 14 See Brykman, Berkeley, 97–8, 150–2, 176–9, 258–9, 262–4. 15 Stephen H. Daniel, ‘Berkeley’s Christian Neoplatonism, Archetypes, and Divine Ideas,’ Journal of the History of Philosophy 39 (2001), 239–58, especially 244. He adds these more precise details: ‘A mind exists (or more precisely, subsists) only to the extent that it has or perceives ideas. Those ideas are the content of the mind’s activity. Without such content, there is no mental activity, there is no mind. Mind cannot subsist apart from thinking because mind is defined as volitionally conditioned perception. A mind is not some thing or substance that exists apart from having such perceptions’ (245). Daniel develops this interpretation further in ‘Berkeley’s Stoic Notion of Spiritual Substance,’ in New Interpretations of Berkeley’s Thought, ed. Stephen H. Daniel (Amherst, NY: Humanity Books, 2007). 16 This manuscript is now in the British Library, catalogued as Add. MS. 39304. See in particular folios 49–50. Unfortunately, the editors of Berkeley’s Works (A.A. Luce and T.E. Jessop) omitted this and other passages from their edition. A new edition of this manuscript is in preparation. 17 Folio 50 recto, lines 5–6, 11–17 (first stratum), cf. PHK 66, 149. 18 In the first of the Three Dialogues, Berkeley asks: When is the human mind active in perceiving, and when is it passive? (DHP 195–7) ‘The mind,’ he answers, ‘is to be accounted active in its perceptions, so far forth as volition is included in them’ (DHP 196; my italics). As we passively receive such disparate raw data as plain sensations of light-and-colour, smells, tastes, sounds, and so on, it is God who produces them in us (cf. PHK 29). ‘You are then in these respects altogether passive’ (DHP 196; my italics), we read. From the observation that we are passive in some respects, the hasty conclusion is that we are passive in all respects, as we are according to the early passivity thesis. But already when we regard something as ‘one thing’ – as when ‘a certain colour, taste, smell, figure, and consistence having been observed to go together, are accounted one distinct thing, signified by the name apple’ (PHK 1) – then, I argue, this complex unity is created by us,

Berkeley’s Four Concepts of the Soul

187

according to the New Theory of Vision. It is the result of a ‘combining together of ideas,’ which is ‘done by the mind in such sort as experience shows it to be most convenient: without which our ideas had never been collected into such sundry distinct combinations as they now are’ (NTV 109; cf. NTV 59, 147, 153–4, 157–9; PHK 12).

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism ro o me t jak a pi

There is a tradition of interpreting Berkeley’s views on language and meaning in the light of modern linguistics and philosophy of language. Commentators have ascribed, for example, instrumentalism regarding scientific and religious language1 and ethical emotivism2 to him. In this essay, I am not going to offer another explanation of Berkeley’s views on language, meaning, or signs. What I hope to do is to cast some light on the religious context in which one specific account or doctrine – that of the ‘other ends’ or ‘other uses’ of language besides ‘the communicating of ideas’ – occasionally occurs. Berkeley famously introduces it in the Introduction to A Treatise concerning the Principles of Human Knowledge: The communicating of ideas marked by words is not the chief and only end of language, as is commonly supposed. There are other ends, as the raising of some passion, the exciting to, or deterring from an action, the putting the mind in some particular disposition; to which the former is in many cases barely subservient, and sometimes entirely omitted, when these can be obtained without it, as I think doth not infrequently happen in the familiar use of language. (IN 20)3

To make this doctrine intelligible to a modern reader, some scholars have explained it in terms of non-cognitive, emotive, or purely pragmatic – as opposed to cognitive, informative, descriptive – uses of language.4 It seems that some of the most interesting points Berkeley makes about the ‘other ends/uses’ of language appear in a context typically unfamiliar to a contemporary philosopher. That is, in a number of relevant cases, he speaks of the use and significance of terms and propositions5 related to the Christian mysteries such as the heavenly rewards, the Holy

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism

189

Trinity, and the grace of God.6 His defence of the (belief in) mysteries in Alciphron VII also partly relies on the theory of word usage. As for the ‘non-cognitivist’ readings, I will not examine them in detail here.7 However, I will point out a few significant features apparently shared by them. For one thing, these interpretations seem to suggest that, for Berkeley, such terms as ‘grace’ or ‘incarnation,’ when used in certain religious contexts, do not stand for, or (to use an anachronistic term) refer to, anything. Accordingly, these readings imply, the relevant propositions in which such terms occur (1) are neither true nor false; (2) do not provide information; (3) are cognitively meaningless; and (4) are typically written or spoken merely to provoke emotions, attitudes, or actions. I want to argue that – as far as terms and propositions regarding the Christian mysteries are concerned – Berkeley saw things differently. To see this clearly, one should consider what he meant by ‘mysteries’ in these contexts. This is no place to examine the historical, linguistic, and theological background of the term ‘mystery,’ its etymology, uses in ancient Greek religion and in the New Testament, and so forth. At the same time, a relevant definition to be found in The Oxford English Dictionary is apposite. Thus, theologically speaking, mystery is a religious truth known only from divine revelation; usually, a doctrine of the faith involving difficulties which human reason is incapable of solving.8

The definition accords well with what we find in Berkeley. I think we can discern at least four meanings of the term in his works. This is not to say, however, that he drew explicit separation lines between them. First, mystery is a religious truth. For such a usage, we can turn to a note Berkeley composed for a sermon preached in 1729: After all – what mysterious truths come by Xt? St Mat. 13.11 to you it is given to know the mysteries of heaven. Wt are these important principal truths? True God revealed. our own state by original sin & our own corruption. The cure thereof. life & immortality with the way to come at it. our eyes open’d as to sin & duty, hell & heaven &c, Rom 15. 29 25. Xt king wt his kingdom. Priest wt his sacrifice. Prophet wt his doctrine. Love of God & neighbour Summary thereof. other doctrines try’d by this touchstone. (W 7: 84)9

190 Roomet Jakapi

In this record, the ‘principal truths’ or foundations of the Christian religion, including the existence of God and immortality of the soul, are termed ‘the mysteries of heaven.’ Second, mystery is a religious doctrine (also an ‘article of faith’). Thus in Alciphron Berkeley discusses the ‘doctrine of grace’ to be found in the New Testament (Alc VII.4, 289–91) and ‘the doctrine of the Trinity ... revealed in Holy Scripture’ (Alc VII.8, 296–8). Such doctrines are ranked among revelations made by God (Alc VI.10, 239–40; VI.12, 242; VII.8, 296–8) as well as ‘articles of (our Christian) faith’ (Alc VII.4, 290; VII.15, 308). Berkeley takes ‘the explication of mysteries in divinity’ to be an ‘attempt as fruitless as the pursuit of the philosopher’s stone in chemistry or the perpetual motion in mechanics’ (Alc VII.9, 301). In sum, a mystery is primarily a religious truth or doctrine that cannot be discovered or (fully) explained by human reason. It is only via divine revelation that human beings know it. Third, in entry 720 of the Notebooks Berkeley identifies ‘holy mysteries’ with ‘propositions about things out of our reach’ (my emphasis). This is not particularly surprising. On a traditional account, the divine truths or doctrines are delivered in, or revealed by means of, certain propositions to be found in the Bible. One and the same doctrine can be ‘contained’ in a number of propositions, both scriptural and theological. Accordingly, in Berkeley’s opinion, the particular ‘wording of a mystery’ is not a substantial issue (Alc VII.10, 302). I take it that Berkeley always held the relevant propositions in the Bible to be true. For, in sharp contrast to typical examples of non-cognitive or other uses of language, as examined in contemporary theories of meaning, in the case under observation, God, not man, is the speaker. He speaks by the mediation of divinely inspired writers (see Alc VI.6– 10, 227–40). And for Berkeley, of course, God does not lie or speak nonsense. In other words, the propositions in question are true simply because they come from God and because they are to be found in the Scripture, the written Word of God.10 This is an orthodox theological position that Berkeley does not seem to question in any of his writings at any stage of his philosophical development.11 The truth of the biblical propositions regarding mysteries is presupposed in his philosophical reflections on their meaningfulness. Finally, there appears to be one more meaning of the term ‘mystery’ in Berkeley’s works. That is, there are the hidden supernatural realities themselves that the doctrines, truths, or propositions concern or ‘are about.’ As far as we know, Berkeley was no Unitarian or free-thinker. I

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism

191

suppose that he sincerely believed ‘that the Father, the Son, and the Holy Ghost, are God, and that there is but one God’ (Alc VII.8, 297) and that he believed in the doctrine of the Trinity, variously formulated.12 He likewise seems to have endorsed the doctrine of the general resurrection and the belief that the resurrection will indeed take place (Alc VI.10–11, 239–41). Two passages from Alciphron should make this point clear: describing the use and effects of the Christian faith, Berkeley says of the evil intentions of a freethinker: Whereas that very man, do but produce in him a sincere belief of a Future State, although it be a mystery, although it be what eye hath not seen, nor ear heard, nor hath it entered into the heart of man to conceive, he shall, nevertheless, by virtue of such belief, be withheld from executing his wicked project: and that for reasons which all men can comprehend, though nobody can be the object of them. (Alc VII.10, 303)

And concerning the mysteries in general he observes: Why men that are so easily and so often gravelled in common points, in things natural and visible, should yet be so sharp-sighted and dogmatical about the invisible world and its mysteries is to me a point utterly unaccountable by all the rules of logic and good sense. (Alc VI.30, 280)

In the first passage cited, Berkeley refers to rewards in afterlife. The same Pauline description of heaven13 is to be found in the sermon ‘On Immortality’ (W 7: 12), the Manuscript Introduction (fol. 22), and the essay ‘The Future State’ in the Guardian (W 7: 184). For Berkeley, of course, the heavenly rewards14 are there, although, unlike St Paul, we have not perceived them (yet). The second passage includes a distinction between two realms, natural and supernatural. Such mysteries as the future rewards and punishments, the expected general resurrection, and the Holy Trinity, are not parts of the natural, perceivable order of things. Rather, they belong to the supernatural realm. They are ‘invisible,’ at least for human beings living in the natural world. Put differently, we have no (clear and distinct) ideas of them. It seems that all aspects of the Christian mysteries considered above fit together perfectly in Berkeley’s theological position. There are propositions, in the Bible, by means of which certain divine truths or doctrines have been revealed to mankind. These propositions express or

192 Roomet Jakapi

reveal what God wants humans to know ‘about the invisible world and its mysteries.’ Several matters of fact concerning the supernatural realm have thus been revealed to us. This holds in spite of the fact that we cannot have or form any (‘clear and distinct,’ or ‘abstract,’ ‘precise,’ ‘positive’) ideas of the supernatural. Berkeley’s theological stance on the mysteries, I think, did not change (significantly) over the years. His philosophical conception of meaning, on the other hand, developed during the time of his writing the Notebooks and continued thereafter. The Notebooks (NB 720), the Manuscript Introduction (fols. 22–4, 27), the sermon ‘On Immortality,’ and Alciphron Dialogue VII, all reflect a tension between Berkeley’s theory of ideas (including the early conception of meaning: ‘all significant words stand for ideas,’ NB 378) and his religious commitment to the belief in mysterious truths revealed in the Bible. One way to highlight this tension is to recognize how the ‘content’ of the revelation concerning the mysteries, or the relevant information provided through the revelation, cannot be ‘cashed out’ in terms of (clear and distinct) ideas. Thus, if the meaning of the terms and propositions regarding the mysteries is to be explained solely in terms of such ideas, these terms and propositions turn out to be meaningless. Berkeley repeatedly tries to address and solve this difficulty, insisting in the end on the practical uses of signs and the good influence such propositions have on believers. One might think that the development of his views on language and meaning was motivated in part by his religious position. However, the textual evidence does not suggest that his shifts in philosophical views on meaning were preceded, accompanied, or followed by respective changes in his religious notion of the mysteries. Neither is there a reason to suppose that philosophical reflections on the relevant passages (e.g., St Paul’s cryptic description of heaven) made him think that those passages are non-descriptive, non-informative, or cognitively meaningless. He tried to defend the Holy Book against attacks of free-thinkers who indeed may have regarded the biblical descriptions of heavenly mysteries as non-cognitive nonsense. But it seems inappropriate to attribute such a position to Berkeley. I suggest, then, that Berkeley’s religious beliefs and theological assumptions regarding the mysteries were in a sense fundamental or prior to his philosophical explanations of the meaning of the terms and propositions in question. To be sure, there are some places where his philosophical argumentation is based on the account of the ‘other ends/uses’ of language. And there are other places where he indicates

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism

193

that one can have some ‘distant representation’ of the supernatural realm by means of metaphors (W 7: 12, 183–4, 223; Alc VI.11, 240–1). So, as even his contemporaries were quick to point out, Berkeley’s arguments are not unproblematic.15 Nonetheless, as I indicated earlier, the truth of the biblical propositions regarding the mysteries is presupposed in his discussions of their meaningfulness. It is also presupposed that these propositions reveal something important about such supernatural realities as the grace of God, original sin, and a future state. The latter are not reduced to images in human minds or explained away by means of semantic principles. To better understand Berkeley’s position, it is useful to consider Peter Browne’s contemporary account of divine analogy with respect to the Christian mysteries. In his Things Divine and Supernatural, conceived by Analogy with Things Natural and Human (1733), Browne (in contrast to Berkeley) offers a detailed account of how supernatural ‘things’ or ‘objects,’ variously called ‘objects of another world,’ ‘heavenly objects themselves,’ ‘divine things as they are in themselves,’ and ‘supernatural realities,’ are represented to human minds (see TDS 173, 175, 210, 228). He argues that supernatural ‘things’ or ‘objects’ as they are ‘in their own nature’ are ‘utterly unknown and incomprehensible to us,’16 and are therefore ‘conceived by the help of objects natural and human’ (TDS 168). That is, we have no ideas or conceptions of the supernatural realities as they are ‘in themselves,’ but we have ideas and conceptions of earthly things which represent those realities by means of analogy. Browne distinguishes two parts in every mystery: (1) the supernatural object as it is ‘in itself’ or ‘in its own nature,’ and (2) the revealed doctrine (which is ‘delivered’ in propositions) concerning that object. The divine object, as it is in itself, is incomprehensible to us, but the doctrine and the relevant propositions are easily understood, as they relate to ideas and conceptions of familiar things (TDS 168). For example, Browne proposes that the mystery that The Logos is the only-begotten Son of God has two parts: (1) ‘the divine supernatural derivation or generation of the Word from his divine Father as it is in it self,’ of which we have no perception or idea, and (2) ‘that analogous representation of this divine derivation under that of an only-begotten Son; whereof the conception is merely natural, and expressed in terms of common speech.’ Thus the supernatural divine derivation or generation is represented to us by means of a conception of ‘natural/human generation’ (TDS 223–6). According to Browne, God himself has chosen certain natural things or objects by means of which he has revealed the divine things to us.

194 Roomet Jakapi

The relation between those divine things and the natural things in question is not an arbitrary one. There is a real analogy between them, even though one part of the relation lies beyond our understanding.17 Berkeley does not share such a view regarding analogy and meaning, but that does not prevent him from agreeing with Browne that (1) hidden supernatural realities exist; (2) several truths or doctrines concerning those realities have been revealed in the Scripture; and (3) these truths or doctrines are expressed in propositions. The core of Berkeley’s defence of the belief in the Christian mysteries (developed specifically in Alciphron VII) lies in his account of the usefulness of such belief, which, in turn, is related to the doctrine of ‘other uses’ of language. As I have discussed this topic in detail elsewhere,18 I will confine myself to, and briefly comment on, one example only. About the belief in the Trinity, Berkeley writes the following: alciphron: Fear not: by all the rules of right reason, it is absolutely impossible that any mystery, and least of all the Trinity, should really be the object of man’s faith. euphranor: I do not wonder you thought so, as long as you maintained that no man could assent to a proposition without perceiving or framing in his mind distinct ideas marked by the terms of it. But, although terms are signs, yet having granted that those signs may be significant, though they should not suggest ideas represented by them, provided they serve to regulate and influence our wills, passions, or conduct, you have consequently granted that the mind of man may assent to propositions containing such terms, when it is so directed or affected by them, notwithstanding it should not perceive distinct ideas marked by those terms. Whence it seems to follow that a man may believe the doctrine of the Trinity, if he finds it revealed in Holy Scripture that the Father, the Son, and the Holy Ghost, are God, and that there is but one God, although he doth not frame in his mind any abstract or distinct ideas of trinity, substance, or personality; provided that this doctrine of a Creator, Redeemer, and Sanctifier makes proper impressions on his mind, producing therein love, hope, gratitude, and obedience, and thereby becomes a lively operative principle, influencing his life and actions, agreeably to that notion of saving faith which is required in a Christian. (Alc VII.8, 296–7)

The idea that propositions enunciating the doctrine of the Trinity have a good influence on those who assent to them is perfectly consistent with the view that these propositions are true and reveal something

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism

195

important about God and His relation to us. The passage is meant to promote and assure belief in the Trinity and to defend this belief by means of the semantic doctrine in question. No doubt is cast on the existence of the Trinity, authority of the Bible, or authenticity of the revelation. The above considerations, I believe, make it quite difficult to accommodate biblical descriptions of mysteries (as Berkeley saw them) to a framework of non-cognitive uses of language. In the present case, the relevant propositions in the Bible are taken to be true and informative. Even though not written by the immediate ‘hand’ of God, they do express (reveal) what God himself rather than any human being wanted to express. It might be objected that Berkeley’s reflections on the uses of language (or of signs in general) can and should be regarded separately from his religious beliefs. The philosophical content of his remarks on the relevant biblical propositions, then, is not to be confused with, or muddled by, his religious position on the matter. Thus, it might be said, the philosopher Berkeley viewed propositions concerning mysteries as cognitively meaningless utterances (or at least he did so at a certain stage of his philosophical development). As a Christian and cleric, he may have believed that nonsense, but this is irrelevant to the good philosophical point he made. This reasoning, even if it seems correct and desirable from a ‘pure’ philosophical perspective, is misleading. We are dealing here with specifically religious propositions uttered in a specific religious context. In the present case, the strict separation of the philosophical or semantic aspect from the religious is unnatural and may lead to inappropriate conclusions both about Berkeley’s account of meaning and about his religious position. For example, if his ‘theory of emotive meaning’ were to be applied to biblical expressions such as ‘the dead shall be raised incorruptible’ (1 Corinthians 15:52), someone might conclude (erroneously) that Berkeley did not really believe in the resurrection. From Berkeley’s perspective, the Christian mysteries are said to imply something mysterious, something hidden from the eyes of the mortal, something inexplicable by any philosophical means (including the vocabulary of cognitivism/non-cognitivism). Accordingly, the nature of the Trinity, original sin, divine grace, and so on remains hidden from us. The relevant passages in Scripture give us some hint of the supernatural, but they are not supposed to be (fully) explained. The best and most we can do is to believe these saving doctrines and thus believe the relevant

196 Roomet Jakapi

propositions to be true. Berkeley’s main effort in Alciphron VII and NB 720 is thus aimed more to opening up the possibility for reasonable assent to the basic doctrines of Christianity rather than providing a comprehensive account of the meaning, purpose, and nature of the talk of mysteries.19 Berkeley’s general ‘doctrine of signs’ is no doubt complex, and perhaps it would be useful to study all its manifestations in his works in terms of syntax, semantics, and pragmatics. Also, much remains to be studied in the historical background of Berkeley’s ‘doctrine of signs.’ However, neither has been my aim. My point is simply that, while interpreting passages related to the mysteries, we should take Berkeley’s religious commitments seriously.20 In this context, propositions in the Bible ‘containing’ various divine truths cannot be taken for (noninformative) sentences uttered by human beings in ordinary circumstances. Berkeley held those propositions to be quite specific, indeed, unique, as to their (supernatural) nature and origin. Notes 1 See Melvin W. Beal, ‘Berkeley’s Linguistic Criterion,’ The Personalist 52 (1971): 499–514, especially 510ff. 2 See Stephen R.L. Clark, ‘God-Appointed Berkeley and the General Good,’ in Essays on Berkeley: A Tercentennial Celebration, ed. John Foster and Howard Robinson (Oxford: Clarendon, 1985), 240, 244. 3 Cf. Manuscript Introduction, fol. 25, in George Berkeley’s Manuscript Introduction, ed. Bertil Belfrage (Oxford: Doxa, 1987). See also Alc VII.5, 292; and Alc VII.14, 307. 4 See, for example, David Berman, ‘Cognitive Theology and Emotive Mysteries in Berkeley’s Alciphron,’ Proceedings of the Royal Irish Academy 81 (C) (1981): 219–29; idem, George Berkeley: Idealism and the Man (Oxford: Clarendon, 1994), 11–17, 144–63; Bertil Belfrage, ‘Berkeley’s Theory of Emotive Meaning (1708),’ History of European Ideas 7 (1986): 643–9; and Belfrage’s editorial introduction to George Berkeley’s Manuscript Introduction. It has to be noted that Belfrage’s interpretation is confined to the early development of Berkeley’s conception of meaning. Berman’s and Belfrage’s accounts of that development diverge in several respects, but both scholars explain it in terms of cognitivism and non-cognitivism. They also share the view that, in the years 1707–8, Berkeley’s conception of meaning developed from a strict cognitivist position (as represented in the paper ‘Of Infinites,’ W 4: 235–8) to a theory of emotive meaning (as brought forth in the Manuscript Introduction).

Christian Mysteries and Berkeley’s Alleged Non-Cognitivism

197

5 I take it that by ‘propositions’ he usually means declarative sentences. 6 See the Manuscript Introduction, fols. 22–4, and Alc VII.7–10, 295–303. 7 For critical examinations of those interpretations, see Roomet Jakapi, ‘Emotive Meaning and Christian Mysteries in Berkeley’s Alciphron,’ British Journal for the History of Philosophy 10 (2002): 401–11; idem, ‘Entry 720 of Berkeley’s Philosophical Commentaries and “Non-Cognitive” Propositions in Scripture,’ Archiv für Geschichte der Philosophie 85 (2003): 86–90; and Kenneth Williford, ‘Berkeley’s Theory of Operative Language in the Manuscript Introduction,’ British Journal for the History of Philosophy 11 (2003): 271–301. 8 Oxford English Dictionary, 2nd ed. (Oxford: Clarendon, 1991), 10:173. 9 See also George Berkeley, A Discourse addressed to Magistrates (1738), in W 6: 201–22, esp. 219. 10 Berkeley acknowledged problems related to the transcription and translation of the texts included in the Bible. Transcribers and translators could have made mistakes. This, however, concerns circumstantials, not essentials of the divine revelation (see Alc VI.7, 229–33). 11 I disagree with Bertil Belfrage’s view that, for example, in NB 720, the propositions concerning mysteries are deemed neither true nor false. See Belfrage’s essay in this collection. 12 Thus in the Creeds and the Thirty-Nine Articles of Religion, art. 1. 13 ‘But as it is written, Eye hath not seen, nor ear heard, neither have entered into the heart of man, the things which God hath prepared for them that love him’ (1 Corinthians 2:9). See The Bible: Authorized King James Version (Oxford: Oxford University Press, 1997). 14 He also speaks, for example, of the ‘pleasures of Heaven,’ ‘coelestial joys’ (W 7: 13), ‘Good Things’ in heaven, ‘pleasures of the Saints in Heaven,’ and ‘Joys of the blessed’ (Manuscript Introduction, fols. 22, 23). 15 For a contemporary criticism of Alciphron IV and VII, see Peter Browne’s Things Divine and Supernatural, conceived by Analogy with Things Natural and Human (1733; repr. Bristol: Thoemmes, 1990), ch. 8. For discussions of the Browne–Berkeley relation, see Paul J. Olscamp, The Moral Philosophy of George Berkeley (The Hague: Martinus Nijhoff, 1970), ch. 9; Arthur R. Winnett, Peter Browne. Provost, Bishop, Metaphysician (London: SPCK, 1974), ch. 11; James O’Higgins, ‘Browne and King, Collins and Berkeley: Agnosticism or Anthropomorphism?’ Journal of Theological Studies 27 (1976): 88–112; and Berman, Berkeley, 140–52. 16 Or, ‘utterly unknown and imperceptible to us’ (TDS 238). 17 According to Browne (TDS, ch. 1), such analogy amounts to real resemblance and correspondence, whereas metaphors used to describe divine

198 Roomet Jakapi things (e.g., God’s ‘hand’ or ‘eye’) are arbitrary, mere products of imagination and figures of speech. 18 See my ‘Emotive Meaning,’ 406–11. 19 For a detailed argument to that effect, see my ‘Faith, Truth, Revelation and Meaning in Berkeley’s Defense of the Christian Religion (in Alciphron),’ The Modern Schoolman 80 (2002): 23–34, esp. 29–33. 20 The same can be said about interpreting Berkeley’s discussions of morality in Alciphron and elsewhere. See Laurent Jaffro’s essay in this volume.

Berkeley’s Criticism of Shaftesbury’s Moral Theory in Alciphron III l au r e n t jaf fro

My aim is to expound Berkeley’s argument against Shaftesbury’s ethical views, concentrating on the Third Dialogue of Alciphron. I will address one limited question: To what extent is Berkeley’s point correct? How far is his reading of Shaftesbury pertinent? I would also like to tackle a second question: Does Berkeley’s criticism of Shaftesbury rely on specifically Berkeleian premises? Or, to put it in Aristotelian terms, is this refutation dialectical? How far does it draw on Berkeley’s own basic conceptions of the nature of morality? One might recognize in these two questions the two sides which should be distinguished in every philosophical criticism: First, does the criticism inform us correctly of the views that are held by the criticized? Second, does the criticism express the critic’s own views on the subject? Our attention must be drawn to the fact that in order to be fair and relevant a criticism must fulfill the first criterion, not necessarily the second. A criticism cannot be fair without being well informed, but it can be well informed without being grounded on the critic’s principles. It is also to be noted that some historians of philosophy consider that the criticism one philosopher can make about another is in most cases a dialogue of the deaf. Such a statement amounts to saying that if a criticism fulfills the second criterion (if it expresses the critic’s own views), then it should fail to fulfill the first (it will not inform us correctly of the views of the criticized). We may suspect that, should such an inference be true, there would not be any genuine discussion among philosophers. These methodological observations are intended only to make my point clear. In this essay, I maintain that Berkeley’s criticism of Shaftesbury is fair and relevant and is not a dialogue of the deaf. As a further preliminary, it should be recalled that Alciphron was

200 Laurent Jaffro

written against the free-thinkers and in defence of Christianity. Today free-thinking and religious scepticism still smack of heresy, not to mention complete atheism. But even though free-thinkers in the British Enlightenment may have been heretical on the subject of religion, the majority of them were at the same time orthodox in the matter of morality. They considered Christianity as a tale, but they did not attack virtue or benevolence. They were even happy with the Gospel when viewed as a recommendation of justice and charity. Most of them rejected the supernatural in the name of the natural, namely of the good nature of humanity and society. Only a few, and in particular Bernard Mandeville, opposed divine commands with social conventions instead of nature. Mandeville, who is present in Alciphron as Lysicles – the main character of the Second Dialogue, an advocate of vice – extended to morality and to Shaftesbury’s ‘natural sense of right and wrong’ the libertine criticism of the supernatural. In his view, morality as an autonomous realm was just another tale. When he rejected the enthusiasm for virtue as a new illusion, Mandeville seemed to stick to Hobbesian conventionalism; he might as well be regarded as a pessimistic disciple of Augustine. But this is not my point here. Rather, I want to focus on how Berkeley makes Lysicles interrupt Alciphron’s Shaftesburian speech in the Third Dialogue. The moral sense is ‘at bottom mere bubble and pretence’ because the qualities (in particular the alleged beauty of virtue) which it is supposed to appreciate are ‘things outward, relative, and superficial’ (Alc III.12, 131). This suggests that Berkeley was able to make use of Mandeville’s relativist arguments against Shaftesbury without subscribing to Mandevillian views. Putting Mandeville aside as the exception that proves the rule, the most usual strategy of British free-thinking against Christianity was to promote the autonomy of morality, that is, to assert that moral values are at the same time independent of religion and sufficient to organize the life of man. Shaftesbury embodied this mainstream insofar as he was responsible for promoting a moral sense which was antecedent to any sense of religion. This is precisely what Berkeley denounces at the beginning of his Theory of Vision Vindicated and Explained, referring to Shaftesbury’s Characteristics of Men, Manners, Opinions, Times (1711): All that is said of a vital principle of order, harmony, and proportion; all that is said of the natural decorum and fitness of things; all that is said of taste and enthusiasm, may well consist and be supposed, without a grain

Berkeley’s Criticism of Shaftesbury’s Moral Theory

201

even of natural religion, without any notion of law or duty, any belief of a lord or judge, or any religious sense of a God; the contemplation of the mind upon the ideas of beauty, and virtue, and order, and fitness, being one thing, and a sense of religion another. (TVV 4)

It is no exaggeration to say that Alciphron was written against the primacy of morality, as is evident in the defiant declaration of Alciphron himself in the Third Dialogue: The less religious the more virtuous. For virtue of the high and disinterested kind no man is so well qualified as an infidel; it being a mean and selfish thing to be virtuous through fear or hope. The notion of a Providence, and future state of rewards and punishments, may indeed tempt or scare men of abject spirit into practices contrary to the natural bent of their souls, but will never produce a true and genuine virtue. (Alc III.3, 116)

This is not only a revival of the paradox of the virtuous atheist, but, as Alciphron terms it, the ‘basis’ for a ‘scheme of duty.’ Bearing this in mind, we should renounce all thought of piecing back together Berkeley’s moral philosophy.1 In Berkeley’s opinion, the moral philosophers are the minute ones, who believe in the existence of a moral and social realm, as if the order of morality were autonomous and did not depend on the divine command. On the contrary, Christianity saves us the trouble of developing a so-called moral philosophy insofar as it provides us with a religion, in which there is everything necessary not only to our salvation hereafter but also to the conduct of our life here-below. Thus it should be no surprise that Berkeleian ethics could not be expressed as a ‘moral philosophy.’2 To put it in Berkeley’s terminology, a conscience exempts us from any moral sense. In the first version of Shaftesbury’s Inquiry concerning Virtue (1699), the moral sense was referred to only as a ‘natural sense of right and wrong.’3 The stress must be put on ‘natural’ as well as on ‘sense.’ On the contrary, Berkeley’s conscience is a supernatural sense of right and wrong. Actually it is not a ‘sense’ at all. One might consider that this is just a play upon words. Does not Berkeley admit a moral sensibility, or an ability to discern moral properties, insofar as he allows a conscience? But a careful reading shows that Berkeley refuses to describe this ability as a ‘sense.’ He makes use of the term ‘conscience’ as if it were the antonym of ‘moral sense’ or of any ‘sense’ whatever.

202 Laurent Jaffro

For instance, Euphranor declares, ‘In earnest, can any ecstasy be higher, any rapture more affecting, than that which springs from the love of God and man, from a conscience void of offence, and an inward discharge of duty, with the secret delight, trust, and hope that attend it?’ Alciphron replies: ‘O Euphranor, we votaries of truth do not envy but pity the groundless joys and mistaken hopes of a Christian. And, as for conscience and rational pleasure, how can we allow a conscience without allowing a vindictive Providence?’ (Alc III.7, 122). Throughout Berkeley’s writings, ‘conscience’ is constantly associated with ‘religion’ and deliberately opposed to moral sense (or to a sense of honour, a sense of morals). Berkeley carefully sticks to this terminological distinction, which, as far as I know, is his own. In Shaftesbury and Francis Hutcheson (and later in Joseph Butler and Lord Kames), ‘conscience’ and ‘sense,’ whether moral or natural, are synonymous; and the same holds true for many writers who do not belong to the moral sense school.4 When Euphranor declares in the First Dialogue, ‘Conscience always supposeth the being of a God’ (Alc I.12, 52), he simply gives the correct Berkeleian definition of this term. This terminology should be kept in mind.5 There are two different expositions of Shaftesbury’s ethical theory in Alciphron III. In Alciphron’s speech and throughout his discussion with Crito and Euphranor, from section 3 onwards, there is a good deal from Shaftesbury. Let us call this the first exposition. But there is a second exposition: Shaftesbury is specifically represented by Cratylus, whose writings are quoted by Crito and Alciphron in sections 13 and 15. The first exposition is a broad outline of Shaftesbury’s conceptions in his Inquiry concerning Virtue. This outline is correct when it puts the focus on the notion of ‘relish,’ ‘taste,’ or ‘sense,’ which must be seen from two angles: first, as the ability in a rational creature to discern beauty and goodness of an action or of an agent (this is moral sense as judgment), and second, as a practical principle that governs the action and motivates the agent’s self-control (this is moral sense as motivation). This conflation between judgment and motivation is typical of Shaftesbury. I quote Alciphron’s speech: To relish this kind of beauty there must be a delicate and fine taste; but where there is this natural taste, nothing further is wanting, either as a principle to convince, or as a motive to induce men to the love of virtue. And more or less there is of this taste or sense in every creature that has reason. (Alc III.3, 117)

Berkeley’s Criticism of Shaftesbury’s Moral Theory

203

The account of Shaftesbury’s ethical theory in this section 3 could have been even more satisfactory if it had distinguished more explicitly between what Shaftesbury himself calls goodness and virtue. Goodness, according to Shaftesbury, refers to actions motivated by first-order affections (especially but not exclusively benevolent affections or ‘natural affections’); virtue refers to actions motivated by second-order affections, that is, by a rational appraisal of the first-order affections. Actions of the first kind may be termed good or bad, depending on whether they promote the well-being of the species and the universe in general. Only good actions of the second kind deserve to be termed virtuous, insofar as they do not originate directly in an affective motivation but in the critical judgment which the rational agent makes on affective motivation: In a creature capable of forming general notions of things, not only the outward beings which offer themselves to the sense are the objects of the affections, but the very actions themselves, and the affections of pity, kindness, gratitude and their contraries, being brought into the mind by reflection, become objects. So that, by means of this reflected sense, there arises another kind of affection towards those very affections themselves, which have been already felt, and are now become the subject of a new liking or dislike.6

One might recognize here the Stoic distinction between, on the one hand, the simple or animal use of representations (in which our conduct or desire is immediately determined by representations that function as stimuli) and, on the other hand, the critical or rational use of representations (in which representations are tested and checked). In any event, Alciphron sticks to Shaftesburian orthodoxy in the continuation of the passage I have just quoted, at least if it is carefully read: All rational beings are by nature social. They are drawn one towards another by natural affections. They unite and incorporate into families, clubs, parties, and commonwealths by mutual sympathy. As, by means of the sensitive soul, our several distinct parts and members do consent towards the animal functions, and are connected in one whole; even so, the several parts of these rational systems or bodies politic, by virtue of this moral or interior sense, are held together, have a fellow feeling, do succour and protect each other, and jointly co-operate towards the same end. Hence that joy in society, that propension towards doing good to our kind, that gratulation and delight in beholding the virtuous deeds of other men, or in reflecting on our own. (Alc III.3, 117)

204 Laurent Jaffro

Alciphron does not sacrifice the rational dimension of the moral sense in his definition of ‘affection,’ and in this he is faithful to Shaftesbury’s frequent use of the term ‘affection’ as a synonym for ‘disposition.’ This is why there are affective affections (emotions, passions, inclinations, etc. – the Latin affectus, in Greek pathos) and non-affective affections (the Latin affectio, in Greek skhesis, which Shaftesbury sometimes translates ‘disposition’ or ‘relation’).7 Contemporary commentators often forget this ancient meaning of ‘affection,’ but Berkeley was well aware that Shaftesbury used it to identify the moral sense with a rational disposition. Berkeley could have just denied that reason, insofar as it is a ‘discursive faculty,’ should be regarded as a ‘sense’ or ‘taste.’ But he chose to put into Euphranor’s mouth a slightly different criticism. Merging ‘sense’ and ‘passion,’ which is not faithful to Alciphron’s argument, Euphranor points out that it seems ‘a very uncertain guide in morals, for a man to follow his passion or inward feeling’ and suggests that this rule would ‘infallibly lead different men different ways, according to the prevalency of this or that appetite or passion’ (Alc III.5, 120). If my understanding of this passage is correct, Euphranor claims that a Shaftesburian moral sense would be subjective – and as such arbitrary – insofar as it is identical with ‘passion or inward feeling.’ Here Euphranor reuses the argument which Berkeley had employed in his Sermon on Passive Obedience (1712): Tenderness and benevolence of temper are often motives to the best and greatest actions; but we must not make them the sole rule of our actions: they are passions rooted in our nature, and, like all other passions, must be restrained and kept under, otherwise they may possibly betray us into as great enormities as any other unbridled lust. Nay, they are more dangerous than other passions, insomuch as they are more plausible, and apt to dazzle and corrupt the mind with the appearance of goodness and generosity. (W 6: 23)

We may well admit that passion as such is not a reliable basis for morality. But what has this to do with the moral sense issue? Euphranor could have better objected to Alciphron – who did not conflate sense with passion – that the moral sense as a subjective disposition does not guarantee the universality of moral rules, even if we recognize that it is quite different from the whim of passion. If the moral sense theory does not give a satisfactory account of the universality of moral rules, it is not

Berkeley’s Criticism of Shaftesbury’s Moral Theory

205

because such a theory would merge conscience and passion, but just because it exempts the moral agent from following a rule. According to such a view, the moral action does not consist in following a rule but in following the dictates of a moral taste. How the same moral taste, as a subjective ability to appraise the value of an action, can be found in a number of individuals is far more complicated to explain than just to say that the same moral rule is to be followed by different people, whether they appreciate it or not.8 In short, Alciphron’s first exposition of Shaftesbury is fair enough, but Euphranor’s criticism is quite cavalier. Anyway, this is not Euphranor’s last word in his criticism of moral sense philosophy; nor is Euphranor the sole speaker on that topic, as we will see in a moment. Euphranor has another argument which does not presuppose any conflation of sense with passion. He makes this new point when he insists that even if we have the experience of some immediacy in moral appreciation, this does not justify the appeal to a moral sense. Alciphron has put stress on the fact that we do not deliberate, compute, premeditate, or even argue in the matter of morality. He considers that this rules out any explanation that would have recourse to interest or deliberation and that this legitimates the Shaftesburian hypothesis. ‘How can we account for this but by a moral sense, which, left to itself, has as quick and true a perception of the beauty and deformity of human actions as the eye has of colours?’ Euphranor is right when he replies that such a hypothesis is superfluous: ‘May not this be sufficiently accounted for by conscience [my emphasis], affection, passion, education, reason, custom, religion; which principles and habits, for aught I know, may be what you metaphorically call a moral sense?’ (Alc III.6, 121). We suspect here that, according to Berkeley, moral conduct is not something that has to be appreciated but is rather something that first has to be taught and learned. The point against the moral sense is that it does not explain how morality can be enforced; it even seems to exempt us from the necessity of any enforcement of morality. For Berkeley, the heart is not that which gives an account of moral duties but that in which moral duties are to be inculcated. Here it is obvious that Berkeley’s criticism of Shaftesbury on morality is but a part of a wider conflict regarding education. Let us turn now to the second exposition, which is indirect. Cratylus’s opinions as a writer are reported by Alciphron. This is a very different method. The first exposition was a broad yet fair summary of Shaftesbury’s ethical theories as if they were known only from hearsay; the

206 Laurent Jaffro

second is a precise account of Shaftesbury’s argument in his Characteristics, and especially of the essay Soliloquy or Advice to an Author. It is to be noted that Alciphron’s first exposition draws mainly on Shaftesbury’s Inquiry concerning Virtue, with some materials from Hutcheson’s Inquiry concerning the Original of our Ideas of Virtue or Moral Good (1725).9 The examination of the Soliloquy, even if it is brief, testifies to a further reading of Shaftesbury. Here is the big question: Why two different expositions? Why did Berkeley have two goes at it? My answer is: because Berkeley was well aware that Shaftesbury had changed his mind concerning the moral sense and that in the Soliloquy he had substantially modified the views he had expressed formerly in his Inquiry concerning Virtue. If my understanding is correct, the twofold exposition of Shaftesbury in the Third Dialogue is intended to suggest that Berkeley has a thorough knowledge of what he is criticizing and that he knows that Shaftesbury, at the beginning of his philosophical career, began with a conception that regarded the sense of right and wrong as analogous to our sensory organs (and this view was shared by Hutcheson, who reformulated it in a Lockean vocabulary, namely the discussion of moral properties as qualities, i.e., powers in the objects to produce sensations in us). As Alciphron puts it in the first exposition, thanks to our moral sense we have ‘a perception of the beauty and deformity of human actions as the eye has of colours’ (Alc III.6, 121). But Shaftesbury, in his Soliloquy, had developed a quite different conception, according to which the moral sense, as a moral taste, has to be cultivated and trained. The reality of moral values is not diminished. The new point is that moral values can be correctly apprehended only through a process of self-improvement, ‘a regimen and discipline of the fancies,’10 and this is the price to be paid for reaching the height of disinterestedness and moral finesse. The moral sense is thus not ready-made equipment but that which we should cultivate within ourselves. Crito’s description of Cratylus is correct when it stresses the connection between ‘taste’ and ‘politeness’: It is true the main scope of all his writings (as he himself tells us) was to assert the reality of a beauty and charm in moral as well as in natural subjects; to demonstrate a taste which he thinks more effectual than principle; to recommend morals on the same foot with manners; and so to advance philosophy on the very foundation of what is called agreeable and polite ... For the sake therefore of the better sort, he has, in great goodness and wisdom, thought of something else, to wit, a taste or relish: this, he assures us, is at least what will influence; since, according to him, whoever has any

Berkeley’s Criticism of Shaftesbury’s Moral Theory

207

impression of gentility (as he calls it) or politeness, is so acquainted with the decorum and grace of things as to be readily transported with the contemplation thereof. (Alc III.13, 133)

In the same section, Crito makes a shrewd observation on the consequences of the perfectionist premises that are typical of Shaftesbury’s philosophy. The mind that is entirely taken up with Stoic enthusiasm ‘will ever be its own object, and contemplate mankind in its own mirror’(Alc III.6, 132). Crito is not wrong at all in suggesting that for Shaftesbury mankind is not the totality of all men and women but some heroic ideal that only the happy few can fulfill. Crito gives us to understand that a perfectionist viewpoint leads Shaftesbury to prioritize conversation with oneself to the detriment of conversation with others. The perfectionist definition of humanity entails the primacy of conversation with oneself. If humanity is not a collection but a perfection, then one needs to take care of one’s real self in order to raise it up to the ideal of a ‘better self.’ This is why the method of Shaftesburian perfectionism is soliloquy or ‘self-discourse,’ ‘self-practice,’ ‘home-regimen,’ which corresponds to the Stoic intelligent use or control of representations. The real self has to converse with the inward daemon. This impartial spectator is the embodiment of accomplished humanity. In his essay Soliloquy or Advice to an Author, Shaftesbury had formulated the theory of this ‘self-discourse’ in the following way: By a certain powerful figure of inward rhetoric, the mind apostrophizes its own fancies, raises them in their proper shapes and personages and addresses them familiarly, without the least ceremony or respect. By this means it will soon happen that two formed parties will erect themselves within. For the imaginations or fancies being thus roundly treated are forced to declare themselves and take party.11

I place particular emphasis on Crito’s perspicacity. When Berkeley makes him declare that the Stoic contemplates mankind in his own mirror, this remark is directed against the connection between moral perfectionism and self-cultivation. In short, Berkeley makes use of two arguments against Shaftesbury’s moral sense: (1) In the first exposition, the moral sense as a natural subjective disposition – a faculty for discerning the qualities that motivate actions, analogous to a sensory organ – is accused of being arbitrary and in

208 Laurent Jaffro

any case superfluous. Shaftesbury was aware that the best way to escape this objection was to move to another conception which would require that the sense be trained. (2) In the second exposition, the moral sense as a cultivated and artful taste, analogous to those dispositions that we reach only through good breeding, is accused of being elitist insofar as it supposes a personal achievement. As Crito puts it, ‘Cratylus, having talked himself, or imagined that he had talked himself, into a Stoical enthusiasm about the beauty of virtue, did, under the pretence of making men heroically virtuous, endeavour to destroy the means of making them reasonably and humanly so’ (Alc III.13, 132). Indeed this is a satire, but it is not a caricature. It is difficult to deny that the claim that the moral taste should be refined, or that something more than our ordinary equipment is required to appreciate the beauty of virtue for its own sake, rests upon the assumption that our sensitivity to morality is a matter of refined education and some kind of connoisseurship. It is to be noted that the first argument, namely the claim that the moral sense is arbitrary and in any case superfluous, foreruns Jeremy Bentham’s charge against moral-sense theories in his Principles of Morals and Legislation.12 The second argument, namely, that the moral sense as a cultivated taste might be unattainable to the average human being, replaces the question of education at the centre of the controversy. The point is whether education should be regarded as an aristocratic selfimprovement or as a collective training. Is it the artistic formation of an outstanding personality or the regulation of habits through the discipline of pleasures and pains? At the very end of Alciphron, Berkeley proclaims to be a defender of Plato’s and Aristotle’s political education against the Stoic ideal of heroic self-achievement. Moral dispositions are not the individual perfections that I should cultivate within myself and by myself; on the contrary, as common habits, they are first enforced through laws. I would not say with David Berman that Berkeley did not ‘shine’ in the Third Dialogue, nor, with Bonamy Dobrée, that ‘he never really gives his minute philosophers their due.’13 I subscribe to the opinion expressed by Stephen Clark: ‘More detailed and careful study of the texts under review would show Berkeley to have allowed his opponents some excellently contrived speeches, and to have been very much

Berkeley’s Criticism of Shaftesbury’s Moral Theory

209

to the point in his mockery.’14 My account of Berkeley’s criticism of the moral sense also rectifies two claims made in 1970 by Paul Olscamp:15 (1) that ‘Berkeley and Shaftesbury actually agreed with each other more than they disagreed.’ On the contrary, I maintain that Berkeley was well aware of the existence of two different stages in Shaftesbury’s thought or, at least, two different conceptions of the moral sense in the latter’s work, both of which he completely rejected with adequate arguments. (2) that the Third Dialogue of Alciphron expresses a ‘systematic moral philosophy.’ My suggestion is rather that if we put aside the arguments which Berkeley uses against his enemies and if we want a clear expression of Berkeley’s own views on morality, we should stick to Euphranor’s concise remarks on the nature of conscience as that which supposes the being of a God. This was enough for Berkeley. He was pushed to elaborate on the subject of morality only within the context of a refutation. One may easily accept this latter statement only insofar as it applies to substantive ethical claims about what is right or wrong. In that sense, there is no doubt Berkeley considers that morality consists in nothing else than following the Gospel. But if by ‘moral philosophy’ we mean an account of the justification of those ethical claims, an inquiry into the nature of rules – in a word moral theory as roughly distinguished from substantive normative assumptions – then there is no reason why Berkeley should be shy about it. In fact, some commentators claim to find a full-blown theoretical account of moral rules in Passive Obedience (sec. 8), where Berkeley carefully distinguishes between two ‘ways’ or ‘methods’ of achieving the ‘well-being of mankind’: Either first, without the injunction of any certain universal rules of morality, only by obliging every one upon each particular occasion, to consult the public good, and always to do that which to him shall seem, in the present time and circumstances, most to conduce to it. Or, secondly, by enjoining the observation of some determinate, established laws, which, if universally practised, have, from the nature of things, an essential fitness to procure the well-being of mankind. (W 6: 621)

On that basis, one may be tempted to ascribe to Berkeley a rule-consequentialist moral theory, according to which the moral quality of an act

210 Laurent Jaffro

does not depend on the consequences of the act itself but rather on the consequences of the rule or the set of rules (‘some determinate, established laws’) to which that act conforms.16 However, I would maintain (1) that Berkeley’s argument against the view that morality should be submitted to subjective evaluation and left to private determination is identical (as has already been suggested) with his first argument against the moral sense theory in Alciphron III; and (2) that the recourse to utility does not entail that Berkeley subscribes to a consequentialist account of morality, nor even that he looks beyond the Gospel for a criterion of morality. In Alciphron V, Crito admits that ‘it is certainly right to judge of principles from their effects,’ but he immediately adds that ‘we must know them to be effects of those principles’ (Alc V.9, 182). In a sense, Christianity is useful to mankind. We can draw on the positive effects of Christian principles on societies to justify our claim that we should follow Christian principles. Yet this does not imply that Christianity is true because it is useful. On the contrary, it is useful because it is true.17 George Pitcher is correct when, citing Passive Obedience (sec. 31), he says that the view of Berkeley as a rule utilitarian needs to be qualified since the philosopher does not claim that moral rules should be obeyed because their observance increases the general happiness: For Berkeley ... it is their being God’s commands to us that makes them the set of rules we ought to obey; our obligation to obey them stems not, then, from the fact that their general observance leads to more human happiness, but directly from the fact that God has willed them: ‘nothing is a law merely because it conduces to the public good, but because it is decreed by the will of God, which alone can give the sanction of a law of nature to any precept.’18

My point goes beyond the fact that Berkeley has a divine-command account of the source of normativity, since Berkeley’s God is the source not only of obligation but also of utility. The promotion of the wellbeing of mankind is an effect of Christian principles; it does not prove that Christian principles are true only insofar as they are favourable but that there must be some truth in Christian principles which produces such an effect. If utility were the criterion, then since pagan religions also sometimes have utility value, we should equally consider being pagan. In fact, utility is not a criterion, but rather a sign of the truth of practical principles. Therefore we should express the matter the other

Berkeley’s Criticism of Shaftesbury’s Moral Theory

211

way around: if pagan religions are also sometimes useful, they must contain some principles of such a nature as to produce these effects, that is, Christian principles. This is the line of argument found in Alciphron V.10: We do not deny but there was something useful in the old religions of Rome and Greece, and some other Pagan countries. On the contrary, we freely own they produced some good effects on the people. But then these good effects were owing to the truths contained in those false religions: the truer therefore, the more useful. I believe you will find it a hard matter to produce any useful truth, any moral precept, any salutary principle or notion in any Gentile system, either of religion or philosophy, which is not comprehended in the Christian, and either enforced by stronger motives, or supported by better authority, or carried to a higher point of perfection. (Alc V.10, 183)

In short, even though Berkeley gives a seemingly rule-consequentialist account of morality in some passages of Alciphron and in Passive Obedience, his discussion about the origin of moral rules shows that his point is not that a law is divine and should be obeyed because it is useful to mankind, but that it is useful to mankind because it is the law of God. Here again Berkeleian moral theory amounts to his strict notion of what ‘conscience’ is and of what, or rather whom, it supposes. Notes 1 The quest for Berkeley’s ‘moral philosophy’ is as old as Berkeleian studies. See the collection of republished papers, Money, Obedience, and Affection: Essays on Berkeley’s Moral and Political Thought, ed. Stephen R.L. Clark (New York: Garland, 1989). 2 See George Pitcher, Berkeley (London: Routledge & Kegan, 1977), 228: ‘It is surprising that Berkeley does not devote more of his energies to moral philosophy ... Of course he has views about these matters, but it cannot be said that he anywhere provides, or tries to provide, adequate backing for them.’ 3 The Inquiry was rewritten and republished in 1711 as a part of Shaftesbury’s famous collection of essays. The phrase ‘moral sense’ appears only in this second corrected edition: An Inquiry Concerning Virtue, I.iii.2, in Characteristics of Men, Manners, Opinions, Times, 3 vols. (London: John Darby, 1711), 2: 46: ‘For notwithstanding a man may through custom, or by licentiousness of practice, favoured by atheism, come in time to lose much of his

212 Laurent Jaffro

4

5

6

7

8

9

natural moral sense, yet it does not seem that atheism should of itself be the cause of any estimation or valuing of anything as fair, noble, and deserving, which was the contrary.’ See Isabel Rivers, Reason, Grace, and Sentiment, vol. 2, Shaftesbury to Hume (Cambridge: Cambridge University Press, 2000), 124. Thomas Burnet, who restored this terminology against Locke, used ‘sense’ as synonymous with ‘natural conscience.’ See his Third Remarks upon An Essay concerning Humane Understanding (London: M. Wooton, 1699), 7–8, reprinted in Remarks on John Locke, ed. George Watson (Doncaster: Brynmill Press, 1989), 63: ‘I understand by natural conscience a natural sagacity to distinguish moral good and evil, or a different perception and sense of them, with a different affection of the mind arising from it; and this so immediate as to prevent and anticipate all external law, and all ratiocination.’ In his introduction to Money, Obedience, and Affection, Clark remarks that ‘some elements of Berkeleyan ethics almost amount to making a dictionary of words’ (xxvi). An Inquiry concerning Virtue, I.ii.3, in Characteristics, 2:28; see also Inquiry, I.ii.4, in Characteristics, 2:36: ‘As to those creatures who are only capable of being moved by sensible objects, they are accordingly good or vicious, as the sensible affections stand with them. It is otherwise in creatures capable of framing rational objects of moral good. For in one of this kind, should the sensible affections stand ever so much amiss, yet if they prevail not, because of those other rational affections spoken of, it is evident, the temper still holds good in the main; and the person is with justice esteemed virtuous by all men.’ See my ‘La question du sens moral et le lexique stoïcien,’ in Shaftesbury. Philosophie et politesse, ed. Fabienne Brugère and Michel Malherbe (Paris: Honoré Champion, 2000), 61–8. A similar argument is employed in Passive Obedience (sec. 21), even if in a different context: ‘to suppose all actions lawful which are unattended with those starts of nature would prove of the last dangerous consequence to virtue and morality’ (W 6: 28). J.O. Urmson claims that Alciphron III draws mainly on Hutcheson. This appreciation is the result of Urmson’s focus on the question of beauty in the first exposition. He also affirms that ‘it is surely wrong for Jessop to claim, as he does in his edition of the Alciphron, that Alciphron either faithfully presents or is meant faithfully to present the ideas of Shaftesbury, though Shaftesbury, clearly referred to under the name of Cratylus in the dialogue, is represented as being of the same school of thought.’ See J.O. Urmson, ‘Berkeley on Beauty’ (1985), in Berkeley: Alciphron in Focus, ed. David Ber-

Berkeley’s Criticism of Shaftesbury’s Moral Theory

10 11

12

13

14 15 16

17

18

213

man (London: Routledge, 1993), 179–84. Against Urmson, I have shown that, even in the first part of the third Dialogue, Berkeley’s exposition of Shaftesbury is quite faithful. Soliloquy or Advice to an Author, I.ii, in Characteristics, 1:186. Characteristics, 1: 188. Berkeley had a perfect understanding of the importance of self-discourse for Shaftesbury. He goes back to the same topic in the Fifth Dialogue, where Euphranor mocks Shaftesbury’s free verse. Alciphron reformulates Cratylus’s doctrine: ‘You must know this great man has (to use his own words) revealed a grand arcanum to the world, having instructed mankind in what he calls mirror-writing, self-discoursing practice, and author practice’ (Alc III.22, 200). ‘One man says ... he has a thing made on purpose to tell him what is right and what is wrong; and that is called a “moral sense”; and then he goes to work at his ease, and says, such a thing is right, and such a thing is wrong – why? “Because my moral sense tells me it is.”’ An Introduction to the Principles of Morals and Legislation, ed. J.H. Burns and H.L.A. Hart (London: The Athlone Press, 1970), 26. David Berman, George Berkeley: Idealism and the Man (Oxford: Clarendon Press, 1994), 164; Bonamy Dobrée, ‘Berkeley as a Man of Letters,’ Hermathena 82 (1953), quoted by Ian Tipton, Berkeley: The Philosophy of Immaterialism (London: Methuen, 1974), 9–10. Clark, introduction to Money, Obedience, and Affection, xxi. Paul J. Olscamp, The Moral Philosophy of George Berkeley (Dordrecht: Martinus Nijhoff, 1970), 154; see also 171. It is a commonplace that Berkeley was responsible for the first presentation of rule consequentialism. See, for instance, Brad Hooker, ‘Rule Consequentialism,’ in The Blackwell Guide to Ethical Theory, ed. Hugh LaFollette (Oxford: Blackwell, 2000), 183. Crito is well aware that his recourse to utility theory sounds Mandevillian: ‘We should endeavour to act like Lysicles upon another occasion, taking into our view the sum of things, and considering principles as branched forth into consequences to the utmost extent we are able.’ Yet Mandeville would not approve of the qualification that follows: ‘Now, the Christian religion, considered as a fountain of light, and joy, and peace, as a source of faith, and hope, and charity (and that it is so will be evident to whoever takes his notion of it from the Gospel), must needs be a principle of happiness and virtue’ (Alc V.4, 178). Pitcher, Berkeley, 239.

Berkeley Poetized w o l f g a ng b re i d e rt

1. Philosophy in Poems: Biography or Philosophy? We should not think of Berkeley as a poet. We know that Johann Gottfried Herder dismissed him as ‘the good bishop Berkeley, who was not a poet.’1 According to A.A. Luce, Berkeley’s ‘America or the Muse’s Refuge’ (1726) was the ‘only ... serious poem’ known to have been written by him (W 7: 369), and in Luce’s appraisal, ‘some of the lines, not all, are admirable’ (W 7: 270). ‘The lines,’ Luce concludes, ‘are the work of a seer, a vates sacer.’ He attests to the poem’s ‘insight, foresight, [and] prediction’ and characterizes the famous passage ‘Westward the course of empire takes its way’ as ‘haunted ... simple, marching, and true.’ Yet Luce concedes that Berkeley’s poem had left no noticeable trace within his lifetime. In any event, I shall not deal with how Berkeley influenced poets and artists.2 Instead, I am concerned with how he fares in regard to the ways in which poems exploit philosophy. Often poets are more concerned with the biography of a philosopher than with his philosophy. Poems about Empedocles, for instance, say more about his death and sandals than about his doctrine of the four elements, because many poets are drawn more to the notion of the unappreciated genius than to a scientific theory of natural elements and forces. There are many poems about Spinoza polishing lenses. Spinoza gets pigeonholed as suffering from persecution – a Jew rejected by other Jews. Such poetry cannot be said to revolve around his philosophy. But the opposite is true for Berkeley. Most poetry about him focuses on his immaterialistic philosophy and neglects his biography.

Berkeley Poetized

215

2. Examples In one of his satirical poems Alexander Pope, a biting poet and Berkeley’s good friend, praised the bishop’s character, neither mentioning details of his biography nor referring to his work: Ev’n in a Bishop I can spy Desert; Secker is decent, Rundel has a Heart, Manners with Candour are to Benson giv’n, To Berkley, ev’ry Virtue under Heav’n.3

The famous distichs written jointly by Johann Wolfgang von Goethe and Friedrich Schiller, published under the title ‘Xenien’ (1788–1805), contain a special group of rhymes called ‘The Philosophers.’ Only a few of the couplets’ titles actually carry the names of the philosophers involved, but it is not difficult to identify whom is meant. We find Descartes’s ‘cogito’ as well as Spinoza’s doctrine that ‘all is one.’ These are followed by the pair of lines: I say just the opposite. Nothing exists but myself! All other things ascend within me as a bubble.4

The editors’ note to this claim suggests Berkeley as its subject. But if the allusion is to Berkeley, it is mistaken, since the passage lacks any mention of God; Berkeley never taught that nothing exists besides the self. I disagree, then, with the interpretation that Goethe and Schiller here refer to Berkeley, particularly since they never mention him elsewhere in their works.5 Lord Byron’s poetry is a different matter. Don Juan (Canto XI, stanzas 1 and 2) explicitly mentions Berkeley. Luce even suggests that Berkeley’s PHK 77 ‘may have been the source of Byron’s witticism’ in that segment of the poem (W 2: 74): When Bishop Berkeley said ‘there was no matter,’ And proved it – ’twas no matter what he said: They say his system ’tis in vain to batter, Too subtle for the airiest human head; And yet who can believe it! I would shatter Gladly all matters, down to stone or lead,

216 Wolfgang Breidert Or adamant, to find the World a spirit, And wear my head, denying that I wear it. What a sublime discovery ’twas to make the Universe universal Egotism! That all’s ideal – all ourselves: I’ll stake the World (be it what you will) that that’s no Schism. Oh, Doubt! – if thou be’st Doubt, for which some take thee, But which, I doubt extremely – thou sole prism Of the Truth’s rays, spoil not my draught of spirit! Heaven’s brandy, – though our brain can hardly bear it.6

Irish poet and Nobel Prize winner (1923) William Butler Yeats (1865– 1939) names Berkeley in two poems of his collection The Winding Stair and Other Poems (1933). ‘Blood and the Moon’ contains two stanzas with a key to interpretation. Yeats, living some years in a secluded tower with his spiritist wife, employed the tower as a symbol indicating a traditional conceptual framework: I declare this tower is my symbol; I declare This winding, gyring, spiring treadmill of a stair is my ancestral stair; That Goldsmith and the Dean, Berkeley and Burke have traveled there. ... And God-appointed Berkeley that proved all things a dream, That this pragmatical, preposterous pig of a world, its farrow that so solid seem, Must vanish on the instant if the mind but change its theme; ... Saeva Indignatio and the labourer’s hire, The strength that gives our blood and state magnanimity of its own desire; Everything that is not God consumed with intellectual fire.7

This poem attributes disdain of the world and dreamy idealism to Berkeley. In another poem, ‘Seven Sages,’ Yeats tries to involve Berkeley in his own poetic anti-intellectualism.8 The text takes the form of a dialogue. Seven wise men represent historical persons, among whom we find Edmund Burke and Oliver Goldsmith. The bishop of Cloyne is associated with music and tar-water, but there is no mention of idealism, abstraction, or any other Berkeleian theme. Instead, we find a political remark:

Berkeley Poetized

217

Goldsmith and Burke, Swift and the Bishop of Cloyne All hated Whiggery ...

But the main claim made by the seventh sage is that these four scholars were educated not in schools but on ‘the roads’ by ‘mimicking what they heard’: They understood that wisdom comes of beggary.

Now, that may be a surprising description of the bishop of Cloyne for many a Berkeleian scholar, but it may refer to Berkeley’s pleading for common sense. Some poetry poses as didactic material, a sort of history of philosophy in verse. In our times the French poet René François Armand Sully Prudhomme (1839–1907) gains less acclaim than he deserves. In 1901 he won the first Nobel Prize for literature, but today his name is virtually unknown.9 Prudhomme was well versed in philosophy. He wrote a great poem in alexandrines on Le Bonheur (Happiness). Besides a dedication and foreword (‘Au Lecteur’), the poem has three parts: I. ‘Les ivresses’ (Ecstasies), II. ‘La pensée’ (Thought), and III. ‘Le suprème essor’ (The Highest Elevation). Part II is of particular interest, since it includes the sections ‘La philosophie antique’ and ‘La philosophie moderne,’ containing as it were a history of philosophy in verse. In the second section, Sully Prudhomme objects to Plato, Plotinus, and medieval thought and introduces Descartes, Malebranche, and the French moralists (Bossuet, Fénelon, and also Pascal). We next encounter Spinoza and Leibniz. Any contemporary book on the history of philosophy would follow this with John Locke, but Prudhomme prefers to focus on Doctor Faust next. Faust ‘consults those whose solemn eyes cast penetrating inquisitiveness into the internal realm,’10 speaking thus of Berkeley and Hobbes: Berk’ley, inspired by the horror of crude senses examines hostilely their proofs and evidences: control of body, empty phantom, soul usurps. God only, nothing else, that human mind disturbs. Adjudged by Hobbes is human knowledge of all matter as cause of Being but without sensitive patter. God, Spirit, could they be? Mere words, no entities! – They are the whole! tis Berk’ley’s answer: Matter lies!11

218 Wolfgang Breidert

Prudhomme’s penning is remarkable in various respects: • Why does he write that Berkeley is inspired by the ‘horror of crude senses’ (horreur de sens grossier), while Berkeley is usually considered a sensualist, advocating great respect for the senses? (See, for example, NB 539.) Obviously the poem relies on a contrast between the refined eye of the mind and rough material senses. • Hostile examination attesting truth to the testimony of the senses perhaps refers to Berkeley’s careful investigation into the New Theory of Vision. But accurate exploration is not necessarily hostile. • Although Berkeley lets the soul rule the body (as Plato does), it seems wrong to interpret the origin of that relation as usurpation. Berkeley merely calls the relation a ‘connection of sensations’ or ‘correspondence’ of the same (DHP 241). The notion that the soul takes command by force is unfounded for Berkeley’s writings and is based, rather, on an interpretation of the history of ideas. • That the body is a mere phantom or an illusion in Berkeley’s philosophy would contradict any claim for reality in his work (cf. NB 80, 305, 391, 474a, 517a, 518, 535). • Prudhomme’s last sentence about Berkeley is paradoxical; it suggests that matter does not tell the truth, whereas Berkeley denies the existence of matter at all. Can a non-entity deceive? The paradox can be avoided if we view this last sentence as being metaphorical in two ways: (a) matter is personified, and (b) matter represents a nonentity. Of course, this is not the preferred terminology of serious philosophy – and should not become so. But it is a common manner of expression in poetry. Prudhomme marks Berkeley as a philosopher looking only inwards, and that is a very common misinterpretation. Berkeley looks outwards; he views the world, but all he sees depends on his mind. ‘Outside’ exists within the mind. Berkeley discovers the limits of knowledge, at which we are confronted with a problem similar to Wittgenstein’s problem of language: demarcation requires that we know about the ‘inside’ and the ‘outside,’ yet the mind knows of no ‘outside’ beyond itself, because every notion the mind may have about an ‘outside’ is itself within the mind; similarly, it is impossible to speak of anything outside of language without its being a paradox. While Prudhomme ponders Berkeley and Hobbes in one stanza, the author of another history of philosophy in verse, written in 1995, Ger-

Berkeley Poetized

219

man philosopher Lutz Geldsetzer, adopts the Locke–Berkeley–Hume triad, referring to them as English philosophers of the Enlightenment with regard to the theory of knowledge.12 Geldsetzer’s book, including the title and table of contents, is written in doggerels. While the form is playful, the contents are serious. After discussing John Locke he turns to George Berkeley: Now Bishop Berkeley by contrast, is moved by the same questions fast. But yet his doctrine, as it ends, to Platonism all honour lends.13

He labels Berkeley’s philosophy ‘Platonism’ and ‘idealism.’ But compared with Plato, Berkeley makes no use of an observer behind the eye, as Descartes had done. According to Berkeley only the spirit or metaphorical eye of the mind can see anything at all, and Geldsetzer correctly emphasizes the identity of complexes of ideas and things. The poem states explicitly, ‘to be is to be perceived,’ and where there is no perception, the world disappears (Sein, das heißt Wahrgenommenwerden). Berkeley denies Locke’s distinction of primary and secondary qualitites: ‘Again, the world is colourful’ (die Welt wird wieder bunt). In the next lines Geldsetzer deals with the problem of thinking in notions and the doctrine that there is no such thing as abstract thinking. It is that we perceive only concrete entities; there are no general entities. He then mentions the notion of a general triangle and that there is no object that could be identical to the percipere. For pious Berkeley, percipere could only be an action of the spirit, and God guarantees the stability of the world. ‘So far it was roughly speaking the doctrine also taught by Malebranche and Leibniz,’ Geldsetzer writes. But these philosophers mistook a spirit for an object. On Berkeley’s view, the only abstract thing around is spirit; thus it should be called a ‘notion.’ Geldsetzer not only sketches Berkeley’s doctrine, he also comments on it. Two kinds of readers, he says, will not understand Berkeley: first, the naturalists, who are convinced that they ‘see perception’ when they observe synaptic activity in the brain; and second, rationalists, who believe that we can think without using intuition or imagination (unanschaulich). Geldsetzer does not take up the problem of connecting sight and touch, nor the problem of distinguishing dreams and objectivity. But he admits that he addresses only a small part of the entire history of philosophy.14

220 Wolfgang Breidert

Of course, as hymns and gloomy poems testify, poetry is not always amusing. There are many serious poems in philosophy and about philosophy,15 and it would be a mistake to think that my discussion of ‘philosophy in poems’ is mere play. But now I turn to a humorous example, indeed. The idea of writing a history of philosophy in limericks, as Richard E. Aquila undertook it,16 suggests a playful nature in itself, since normally this poetic form makes use of single limericks for witty points, puns, or gags.17 Aquila uses limericks in an unusual way, connecting a chain of serious thoughts, namely a history of philosophy. But unsurprisingly his text brims with irony and humour. In this Limerick History of Philosophy, Locke and Berkeley turn up in a chapter containing twenty-two limericks about their theories of knowledge. After three stanzas about Locke’s empirical sensualism, Berkeley gets pondered this way: Or, like Berkeley, I might want to know: ‘What’s a tree when perceivers all go?’ Is it more than a label For the hope that we’re able To come back and see more of the show? (Though strictly, as Berkeley would state it, It’s a show never quite unspectated. For, glued to His screen There’s our God! – who loves seeing Each and all of the things He’s created.)

The mood is ironical, and common sense is attributed to ‘average voters.’ God is very busy as the creator of ideas (‘busier than even a bee is’). Berkeley, asserting ‘the world’s spiritual facet,’ is an impostor: How could a real bishop (As he is) ever dish up Such a view ...

The chapter closes with ironical consent to Berkeley: But if matter’s passé, Then Spirit’s Class A. And we’ve come to the end of our riddle.

Berkeley Poetized

221

Look up, then, and shout Alleluja! To that Spirit with power to rule ya.

So much for a satire on Berkeley! We find a very strange poem about Berkeley in an extraordinary work of poetry, Madoc (1990), by Paul Muldoon.18 The subtitle, ‘A Mystery’ serves as a first warning. At first glance the book is a collection of poems; on a second look it seems to be a reader in the history of philosophy, since except for some introductory poems and the first one called ‘The Key,’ all others are named after philosophers. (The names used as titles are placed in brackets.) The mystery lingers, even after one reads that this collection of poetry deals with the imaginary enterprise of poet-philosophers Samuel Taylor Coleridge and Robert Southey, who, in the nineteenth century had planned to found a utopian community (or ‘Pantisocracy’) on the banks of the Susquehanna. Muldoon wonders what would have happened had these two poets crossed the Atlantic, and in passing he offers ‘a mischievous guide to the history of Western philosophy.’ The book operates at various levels. It works through a romantic tour undertaken by Coleridge and Southey, a history of philosophy, a critical review of intercultural encounter, a criticism of the exploitation of nature, a prismatic reflection on drug problems, and so on. It is quite witty, abounding with allusions and jokes, many-layered in a postmodern way, but without being a patchwork. Here is the short poem entitled ‘[Berkeley]’: That tread-mill might have existed only in the mind Of a child, were the child not now a non-entity Under his flower-strewn, almost-imperceptible mound. BERKELEY COLERIDGE 1798–1799

The last of these four almost rhymed lines is to be understood as the tombstone inscription for Coleridge’s second son, christened Berkeley out of his father’s reverence for the philosopher. Confronted with that grave, Muldoon questions Berkeley’s view that the only existing substances are spirits. The whole world – in Muldoon’s words, ‘that treadmill’ – is nothing other than a complex of ideas in our mind or the mind of God. Thus the poem begins: The entire world might have existed only in the mind of a child, but – as the implicit criticism goes – the child is dead, he is a ‘non-entity.’ But if the child does not exist, then his spirit also does not exist. It would seem that this is what Muldoon

222 Wolfgang Breidert

infers, although it is not obvious that a dead person does not exist at all in light of our history of ideas and the immense discussion on immortality. Common sense says that the buried child is no longer perceivable and thus should not exist for Berkeley. Muldoon emphasizes this by making the mound almost imperceptible, too. Even the grave is in danger of non-existence. The paradox here is this: In this verse the author speaks of a child who is said to be a non-entity,19 but if the poet or philosopher is talking about a non-entity, he is thinking about it, and therefore it is perceived in some way; that is, it has to exist in a Berkeleian sense. It would seem as if Berkeley’s doctrine cannot be expressed negatively. Saying ‘the unperceived does not exist’ implies thinking about the unperceived; thus it is perceived and must exist. In philosophy we know that some statements (above all, metaphysical statements) cannot be held without contradiction.20 So we should – as Wittgenstein recommended in his Tractatus – either remain silent or use contradictions like these to create paradoxical and periphrastic poetry. Poetry may attempt to transcend ordinary language, the language of daily concerns and science. It may even attempt to transcend the realm of fiction and wild imagination. Whether that attempt makes sense or can be successful is another matter. Irving Layton, born in Romania in 1912, emigrated to Canada in 1913. He became a high school teacher in Montreal and later a professor of English literature in Toronto. His poetry makes use of romantic and ironic ingredients. Several of his poems are about famous persons, like the following one entitled ‘Bishop Berkeley Goes to Bed,’ which begins: God helping me Esse is percipi: So for an hour or so From this bay window I’ve looke At a metaphysical tower of snow;21

Layton once described his family’s poor lodgings during his childhood: ‘In the frosty winter nights – we couldn’t afford to heat the bedroom and I dubbed it the North Pole after my first geography lesson – I was very loath to leave the hard-earned warmth of my bed, run unslippered (slippers? who ever heard of them?) into the now cooled-off kitchen and make the same comfortless journey back.’22 The poem continues:

Berkeley Poetized

223

Now on returning to bed, To warm my miserable Self I slide my cold feet Beneath this rational sheet And cover myself With the patch of white in my head.

His poem on Berkeley begins with God and ends with Berkeley’s head. The well-known second line is quoted correctly: ‘is’ instead of ‘est’! God may be helping, when the bishop goes to bed, because God’s perception of things guarantees their existence during Berkeley’s sleep. ‘Helping me’ gives the text also a touch of everyday life, where we use expressions like ‘for god’s sake’ without any religious connotation. The bay window, through which Berkeley has a transcendent view, could be at Cloyne.23 The ‘metaphysical tower of snow,’ a metaphor for the rational metaphysics, is not warming and possibly melting – that is, it is useless and existentially jeopardized.24 After an intensive endeavour of metaphysical consideration (‘So for an hour or so / ... / I’ve looked’), the bishop returns to his miserable self. It is miserable because it needs God for security of existence. Berkeley covers himself with a patch of white, like a white patch on a map. The unconscious is used to remedy cold philosophical rationality, but God helps by the preservation of existence throughout unconsciousness. Charles H. Sisson (1914–2003) was an Anglican poet who believed rather in the existence of God than in the existence of human personality. Nevertheless he was convinced that our existence is based on the body, not on the mind.25 In the poem ‘Loquitur Senex’ (from Numbers, 1965), an old man, reflecting on his life, does not recognize the intellectual or spiritual character, metaphorically represented by a fish, of his former life as a businessman. But in his last days the ‘senex’ hopes with regard to his future that a mind, that is, God’s mind, will make sense of and ‘hold this world’ (including the old man himself). In this way the guarantee of existence will finally be founded on a spirit. The last stanza reads: Gracious God, when the tension gives And I am swept below the weir Do as Berkeley says Hold this world in your mind.26

224 Wolfgang Breidert

Few poems are concerned with a special passage in Berkeley’s writings. ‘The Fountain’ by Donald Davie (1922–95) is a splendid example of such an expounding poem. It refers to the end of Three Dialogues between Hylas and Philonous, where Berkeley uses the flow of the water in a fountain as a metaphor for the movement of men who are brought ‘up’ to scepticism and ‘back’ to common sense by following the same principles. The poem begins: Feathers up fast, and steeples; then in clods Thuds into its first basin; thence as surf Smokes up and hangs; irregularly slops Into its second tattered like a shawl; There, chill as rain stipples a danker green, Where urgent tritons lob their heavy jets.

In the first stanza a baroque painting is developed in a picturesque manner. It seems to be conventional poetry in a descriptive way, but the second stanza transfers to the metaphorical interpretation taken from the Dialogues. It finishes with the same words as the Dialogues do: ‘common sense’: For Berkeley this was human thought, that mounts From bland assumptions to inquiring skies, There glints with wit, fumes into fancies, plays With its negations, and at last descends, As by a law of nature, to its bowl Of thus enlightened but still common sense.

If the poem were supposed to be a mere translation of Berkeley’s text, it could close with these words, but it continues with a further stanza and comes to its end in the last, isolated, line: We who have no such confidence must gaze With all the more affection on these forms, These spires, these plumes, these calm reflections, these Similitudes of surf and turf and shawl, Graceful returns upon acceptances. We ask of fountains only that they play, Though that was not what Berkeley meant at all.27

Berkeley Poetized

225

In the third stanza the speaking ‘we’ – that is, poets and average human beings – remain on the level of common sense, looking from a distance on the play of the fountain.28 Unlike Berkeley, ‘we have no such confidence’ in the philosophical play of thought. In contrast to his intention, the post-metaphysical reader looks on these sophisticated forms of thinking as mere amusements without any profound meaning. The last line returns, so to speak, to the view as described in the first stanza. So the poem follows its course in the same way as the water in the fountain. In a passage of the long poem The Homestead Called Damascus (1920– 5), Kenneth Rexroth (1905–82) labelled Berkeley an illusionist. The lines refer to PHK 38, where Berkeley concedes that ‘it sounds very harsh to say we eat and drink ideas,’ and to his comments on the difference between dream and reality (PHK 41). In an erotic dream of a Picasso nude, dreaming and wakening are confused: Dreaming, Sebastian said, ‘That was a dream.’ And slowly he awoke and said, half Dreaming, ‘Bishop Berkeley’s cherries are Still weighted with me. I shall sit for Many a day eating illusions, And dream at night and in the morning Wake to the same worn-out cruelty.’29

The poems of Christian Morgenstern (1871–1914), a well-known German poet, have been compared to English nonsense literature, and some English renditions of his poetry do exist.30 He was philosophically well educated, but he loved queer fun and grotesque imagination. During his last years he was inclined to improve the world, assuming a rather mystical or anthroposophic attitude. Although the following poem does not refer to Berkeley specifically, it is founded on a Berkeleian argument: The Milestone Deep within the depths of wood, written black as ink in files, meant to do the wand’rer good, it reads: only thirteen miles! Funny – seriousness aside – that this text does not exist,

226 Wolfgang Breidert if there’s no one to provide it be the same text, that it is. The idea may be unfurled: What is it, when it’s unseen? What a strange thing that might mean! Just the eye creates the world.31

In the original version of this poem Morgenstern plays on some words: The milestone stands, and the inscription ‘stands’ on it. In German we also use the same word for both expressions, but once it has been used one way, it looks odd in the other. So the first lines, which seem to merely introduce the situation, actually anticipate the fun (literally, laughter) brought into the second stanza. The original second stanza presents the problem of absence in a reflexive manner, literally translated: ‘This text would not exist, if no glance wished to make a text out of it by itself’ (or ‘by himself’ – the German text allows both interpretations). 3. Conclusion Although we find sonnets on Descartes (by Durs Gruenbein), Spinoza (by Sully Prudhomme and Jorge Luis Borges), Kant (by Sully Prudhomme), Hegel (by Karl Krolow), and even the French moralists (by Jules Lemaître), as far as I can tell no sonnet has been devoted to Berkeley. Neglecting any very detailed interpretation of the poems in question, my brief survey of poetry that deals with Berkeley and/or his philosophy allows me to conclude that many of these poems were written more satirically than out of admiration. In many cases Berkeley has been misunderstood (and misinterpreted) by poets, and even philosopher-poets are occasionally guilty of distorting his ideas. It is nevertheless interesting to notice how elements of his philosophy (especially his immaterialism) and his biography make their way into poetry. Most of these poems mention God as creator, guarantor, and observer of the world, whereas it is astonishing that poets do not refer to either Berkeley’s general theory of language or his special conception of visual language. Notes 1 Johann G. Herder, ‘Tithon und Aurora,’ in Werke, ed. Heinrich Düntzer, 24 vols. in 14 (Berlin: Hempel, 1879), 15:214 (my translation).

Berkeley Poetized

227

2 For example, see Hans August Julius Liedtke, ‘Shelley durch Berkeley und Drummund beeinflusst?’ diss. Greifswald, 1933; E.D. Leyburn, ‘Berkeleian Elements in Wordsworth’s Thought,’ Journal of English and Germanic Philology 47 (1948): 14–28; D.J. Greene, ‘Smart, Berkeley, the Scientist and the Poets,’ Journal for the History of Ideas 14 (1953), 327–52; the reply by Karina Williamson, ‘Smart’s Principia: Science and Anti-Science in Jubilate Agno,’ Review of English Studies, n.s., 30 (1979): 409–22; Padraic Colum, ‘Berkeley and the Modern Artist,’ Saturday Review of Literature, 15 June 1935, 3–4, 14– 15; and Hans Joachim Oertel, George Berkeley und die englische Literatur (1934; repr. Walluf bei Wiesbaden: Sändig, 1973). 3 Alexander Pope, Epilogue to the Satires, Dialogue II, 70–3, in Poems, ed. John Butt (New Haven: Yale University Press, 1963), 697. 4 Friedrich Schiller, Sämtliche Werke, 4th ed., ed. Gerhard Fricke and Herbert G. Göpfert, with Herbert Stubenrauch, 5 vols. (Munich: Carl Hanser, 1965– 7), 1:298 (my translation). 5 Surprisingly, Goethe never mentions Berkeley’s theory of vision in his writings on optics. 6 George Gordon Byron, Baron Byron, Don Juan, ed. Leslie A. Marchand (Boston: Houghton Miffin, 1958). 7 W.B. Yeats, The Collected Poems (New York: Macmillan, 1952), 233. 8 Ibid., 236–7. 9 He is not even mentioned in the grand fourteen-volume German dictionary of literature, Kindlers Literatur Lexikon. 10 ‘Faustus consulte ceux dont l’oeil baissé promène / Dans le domaine intime un pénétrant regard.’ 11 René Françoise Armand Sully Prudhomme, Le Bonheur, part II, sec. 6, La philosophie moderne, in Oeuvres, 5 vols. (Paris: A. Lemerre, 1900–1), vol. 5, Poésies (1879–1888), 247: Berkeley, que l’horreur des sens grossiers inspire, Fait de leur témoignage un hostile examen: Du corps, fantôme creux, l’âme usurpe l’empire. Il ne reste que Dieu devant l’esprit humain! Hobbes n’avait à l’homme octroyé de connaître Que la ferme matière, unique fonds de L’Être: Dieu, l’esprit, que sont-ils? Rien! Des mots seulement. – Tout! répond Berkeley, car la matière ment! 12 Lutz Geldsetzer, Die Philosophenwelt /In Versen vorgestellt (Stuttgart: Reclam, 1995), 112. 13 Ibid., 114: ‘Den Bischof Berkeley dagegen / dieselben Fragen zwar bewe-

228 Wolfgang Breidert

14 15 16 17

gen, / doch macht die Lösung seiner Lehre / dem Platonismus aller Ehre’ (my translation). Ibid., 232. For example, by Lucretius, Boethius, Dante, Campanella, Leibniz, Goethe, Hölderlin, Rilke, Hesse, T.S. Eliot. Richard E. Aquila, Rhyme or Reason: A Limerick History of Philosophy (Washington, DC: University Press of America, 1981). Cf. the limericks by Ronald Knox in Bertrand Russell, A History of Western Philosophy (New York: Simon and Schuster, 1945), 648: There once was a man who said: ‘God Must find it exceedingly odd, If he finds that this tree Continues to be When there is no one about in the quad.’ ‘Dear Sir, your astonishment’s odd. I am always in the quad. And that’s why the tree Will continue to be Since observed by Yours faithfully, God.’

18 See Paul Muldoon, Madoc: A Mystery (London: Faber, 1990). 19 Cf. John Donne, ‘The Paradox,’ in Poems of John Donne, ed. E.K. Chambers, 2 vols. (London: Lawrence and Bullen, 1896), 1:74–5. 20 On how this point relates to how we must distinguish between different contexts of expression, see Roomet Jakapi’s essay in this collection. 21 Irving Layton, The Collected Poems (Toronto: McClelland and Stewart, 1971), 62. 22 Ibid., foreword. 23 Berkeley was not bishop during his residence at Newport. 24 Berkeley scholars may be reminded of the famous sight of the tower in New Theory of Vision. 25 Stefanie Christmann, ‘The Order of the Physical World: On the Poetry and Poetics of Charles H. Sisson,’ AAA: Arbeiten aus Anglistik und Amerikanistik 24, H. 2 (1999): 229–45. 26 Charles H. Sisson, Collected Poems 1943–1983 (Manchester: Carcanet, 1984), 75–6. 27 Donald Davie, Collected Poems, ed. Neil Powell (Manchester: Carcanet, 2002), 61–2. Davie was educated at Cambridge (St Catharine’s College). During the Second World War he served in the Royal Navy. After teaching

Berkeley Poetized

28 29

30 31

229

at Dublin and Cambridge he became Professor of English at the University of Essex, then at Stanford and Vanderbilt. In 1988 he returned to England. Davie was a poeta doctus. His poems are, intentionally, close to prose, and in virtue of many erudite allusions they seem to be enigmatic. In NB 392–4 Berkeley himself looked from a distance on his opponents’ speculations (‘We Irish men ...’). Kenneth Rexroth, Complete Poems (Port Townsend, WA: Copper Canyon Press, 2004), 30ff. Rexroth, the son of a pharmaceuticals salesman, was born in South Bend, Indiana. His parents died when he was young, and he lived for three years with an aunt in Chicago. He educated himself, worked in several jobs, and travelled through America and Europe. After the death of his first wife, he married three more times. He is reputed to be father to the Beats, and he attended to erotic mysticism, communism, pacifism, and Buddhism. E.g., Christian Morgenstern, The Gallows Songs: A Selection, trans. Max Knight (Berkeley: University of California Press, 1963). Christian Morgenstern, Galgenlieder usw. (Stuttgart: Philipp Reclam, 1978), 93 (my translation).

This page intentionally left blank

Index

accidents. See modes act–object distinction, 142–6, 151, 159 affection, 85, 203–5, 211–13, 224 afterlife, 191–3, 201, 222 analogy, 193–4 anti-abstractionism, 159–60, 163, 219 Aquila, Richard E., 220 archetypes, 32, 81 Aristotle, 11, 25, 98 Augustine, Saint, 200 Ayers, Michael, 5, 11, 75 Belfrage, Bertil, 8, 167–8, 196–7 benevolence, 200, 204 Bentham, Jeremy, 208 Berman, David, 196, 208 Bible, 190–7, 209 Boethius, Anicius Manlius Severinus, 228 Bolton, Martha, 115 Borges, Jorge Luis, 226 Bossuet, Jacques-Bénigne, 217 Breidert, Wolfgang, 9–10 Broad, C.D., 127 Brooke, Lord. See Greville, Robert Browne, Peter, 9, 193–4, 197 Brykman, Genèvieve, 182

Buddhism, 25 Burke, Edmund, 216–17 Burnet, Thomas, 212 Burthogge, Richard, 5, 22–8 Butler, Joseph, 202 Byron, George Gordon, Lord, 9, 215 Campanella, Tommaso, 228 causation, 5–6, 8, 15, 24, 69–70, 133–4, 140–6, 149, 151, 156–7 certainty, 181 Chisholm, Roderick, 43–5 Christianity, 188–98, 200–2, 210–11 Clark, Stephen, 208 Coleridge, Samuel Taylor, 221 complacency, 176 concept, 14–15 conscience, 9, 201–2, 205, 209, 211–12 Creation, 5, 20; Moses’ account, 64, 79 Cummins, Phillip, 7–8, 121, 170 Daniel, Stephen, 27–8, 164–5, 170, 182–3 Dante Alighieri, 228 Davidson, Donald, 12, 13, 21 Davie, Donald, 10, 223–5, 228

232 Index dependence, 124–9, 132–4, 149 Descartes, René, 215, 217, 219, 226; idealism, 5, 12–15, 22, 26; mind– idea relation, 7, 17–19, 24, 60, 148, 164; on perception 30, 39, 78 Desgabets, Robert, 11 disposition, 9, 157, 180–1, 188, 204, 207–8 Dobrée, Bonamy, 208 Donne, John, 228 Downing, Lisa, 12 dreams, 225 Eleatics, 12 Eliot, T.S., 228 emotivism, 188, 195–6 Empedocles, 214 empiricism, 11 existence, 127, 139, 149, 161–2, 164, 174; formal/objective, 5, 13–18 faculty, 8, 58, 68, 98–100, 102, 162–3, 176–7, 181, 185, 204, 207 Fénelon, François, 217 figure. See shape free-thinking, 190–2, 200 French moralists, 217, 226 Gadamer, Hans-Georg, 21 Geldsetzer, Lutz, 219 geometry, 19, 78, 108, 110, 114, 118 Geulincx, Arnold, 28 Glauser, Richard, 6–7, 27, 47, 106, 152, 167 God, 7, 9, 19, 22, 183, 185, 215–17, 219, 223, 226; benevolence of, 19, 24, 190–5, 200–1; cause of ideas, 4, 6–7, 10, 54–6, 62–72, 75–7, 173–7, 184, 219–23; ideas in, 3–4, 12, 24, 51, 158, 185; moral legislator, 209–11

Goethe, J.W. von, 215, 227, 228 Goldsmith, Oliver, 216–17 goodness, 202–4, 206 Gospel. See Bible grace, 9, 189–90, 193, 195 Grave, S.A., 160 Greville, Robert Lord Brooke, 22, 25 Gruenbein, Durs, 226 harmony. See ideas: order Hegel, G.W.F., 226 Herder, J.G., 214 Hesse, Hermann, 228 Hight, Marc, 6–7, 27, 73, 76, 77, 82, 119, 152, 167 Hobbes, Thomas, 25, 39, 176, 182, 217, 218, 227 Hölderlin, Friedrich, 228 Hume, David, 29, 39, 127, 149, 151, 174, 212 Hutcheson, Francis, 202, 206, 212 idealism, 3, 5–6, 8, 10–27, 156, 159, 169, 185, 186, 196, 213, 216, 219 ideas, 6–7, 13–15, 40, 46, 88–90; abstract, 29, 80, 109; collections of, 19–20, 50–6, 59–75, 77, 82–90, 93, 100, 105, 187; conformity to objects, 29–38, 46–8; God’s, 3, 51, 54–5, 81; order, 4–6, 10, 16, 24, 50, 54, 57, 67–71, 79, 111–12, 156, 185, 200–1; relation to mind, 4, 17, 47, 51–69, 75–6, 78, 81–2, 86, 122–6, 129, 132, 135– 66, 183; sensible, 57– 9, 66, 70, 72–5, 88–9, 94, 122, 142, 146; simple/complex, 4, 18, 33, 62, 65–7, 71, 111, 115; tangible/visible, 50, 59, 66, 68–70, 108–9, 112, 115, 118, 185 images, 58, 78

Index imagination, 6–7, 11, 56–9, 66–8, 70– 5, 77–8, 88–90, 102–4, 182 immaterialism, 10, 12, 35, 169, 226 individuation, 50–1, 76 inference, 29–30, 38–43, 45, 57, 77, 95, 102, 113, 156 inherence, 128, 134–7, 149 innatism, 19, 22 instrumentalism, 188 intentionality, 14–15, 17–18, 25, 111, 114 Jaffro, Laurent, 9, 198 Jakapi, Roomet, 9, 150, 228 Johnson, Samuel, 154–5 judgment, 102, 107, 202; perceptual, 42–5, 52, 167 Kames, Henry Home, Lord, 202 Kant, Immanuel, 5, 12, 13, 15, 16, 18, 19, 21, 23, 24, 26–8, 118, 226 knowledge, 33–42, 45–9; inferential, 38–45, 48; intuitive, 40, 170; sensitive, 29–30, 38–41, 45 Knox, Ronald, 228 Krolow, Karl, 226 language, 9, 20, 81, 188, 226; of nature, 54–5, 69, 90, 111, 226; noncognitivist, 188–98 laws of nature. See ideas: order Layton, Irving, 10, 222–3 Leclerc, Jean, 59, 74 Leibniz, G.W., 24, 217, 219, 228 Lemaître, Jules, 226 Lennon, Thomas, 11 likeness principle. See resemblance Locke, John, 5–6, 24, 212, 217, 219, 220; epistemology, 29–49, 110–11, 114–17, 119–20; on representation,

233

4, 18, 29–30; on substance, 164; on uneasiness, 175–6; on unity and number, 59, 60, 63, 68, 80 logical basis for reality, 25, 164 Lovejoy, Arthur O., 12, 22 Luce, A.A., 160, 214 Lucretius, 228 Malebranche, Nicolas, 30, 39, 78, 217, 219 Mandeville, Bernard, 9, 200 ‘master argument,’ 21–2 materialism, 11–12, 39, 42, 93 matter, 53, 129, 215, 217–18 McCracken, Charles, 154 McDowell, John, 12, 13, 23 memory, 6, 56–8, 66–7, 78 Migely, Genevieve, 8, 81, 107, 185 mind, 4–5, 7–8, 13, 17, 51–7, 121–5, 139–51, 172–87, 219, 221; active/ passive, 8, 24, 81, 130, 133, 139–46, 151, 153–71, 180–7; bundle theory, 172, 174–82, 185; immortality, 185– 6, 191–2; substratum, 123–4, 129, 132, 149 miracles, 24 modes, 13–15, 20, 21, 60, 131, 135–8, 149, 161. See also substance: predication Molyneux, Samuel, 58 Molyneux (William) question, 109, 114, 118 morality, 9, 199–213; consent, 203; divine command theory, 201, 210– 11; motivation, 202–3; rules, 204–5, 209–11, 213 moral sense/taste, 9, 201–13 Morgenstern, Christian, 225–6 Muehlmann, Robert, 76, 160, 163, 169 Muldoon, Paul, 10, 221–2

234 Index music, 216 mysteries, 9, 150, 179, 188–98 names, 6, 20, 65, 73, 84–6, 88, 106 Neoplatonism, 22, 24–5, 27–8 nominal essence, 68 notions, 23, 69–70, 219 Nuchelmans, Gabriel, 25 number, 59–62, 64–5, 72, 78–9, 106 objects, 18–19, 30–3, 42–6, 75–7, 82– 3, 105, 142, 218; individuation of, 50–1, 76; tangible/visible, 7, 20, 50, 108–10, 118; unity of, 50–107, 167 Olscamp, Paul, 209 one and many, 24 Pappas, George, 5–6, 29, 83, 96–102, 119 Parmenides, 16 Pascal, Blaise, 217 Paul, Saint, 191–2 perception, 5–8, 12, 17, 39–45, 100–1, 104, 108, 117–19, 142–3, 155; adverbial theory, 43–5; constitution theory, 43–5, 48–9; error, 71, 105, 107, 110–11, 116–17, 168; immediate/ mediate, 7–8, 32, 46, 47, 68–71, 77– 8, 80–1, 91, 94–103, 108, 110, 113– 14, 119, 145–6, 152, 155–8, 167–9; spatial, 120 person, 8, 177–8 Picasso, Pablo, 225 Pitcher, George, 97, 99–100, 154, 210 Plato, 217–19 Platonism, 11, 12, 16, 23–5, 219 pleasure/pain, 19, 48, 158, 165, 176, 181, 185–6 Plotinus, 217

poetry, 9–10, 214–26 Pope, Alexander, 215 powers, 19–20, 169, 206 Prudhomme, René François Armand Sully, 10, 217–18, 226 Putnam, Hilary, 21 qualities, 20, 47, 135, 206; primary/ secondary, 59–61, 109–10 Ramus, Peter, 164 rationalism, 11 Raynor, David, 63 realism, 3–6, 8, 26–27, 54–5; direct/ indirect, 30, 42–6; representative, 29–30, 37–8, 42–5, 48–9 Reid, Thomas, 46 relations, 55–6, 60, 62–5, 69–71, 116, 164 representation, 5–7, 14, 18, 30–1, 56– 8, 67, 71–2, 75, 77, 111, 113–14, 140, 186, 193; Stoic theory, 203, 207 resemblance, 31–3, 35, 46–8, 57–8, 64, 70, 78 resurrection, 191, 195 Rexroth, Kenneth, 225, 229 Rilke, Rainer Maria, 228 sameness, 92 scepticism, 12, 25, 31, 33–5, 46, 49, 74, 93, 151, 224 Schiller, Friedrich, 215 Schumacher, Ralph, 7, 79, 152 Scriptures. See Bible self-consciousness, 130, 150 Shaftesbury, Anthony Ashley Cooper, third earl of, 9, 199–213 shape: tangible/visible, 109–18 sight, 7, 81, 98–9, 227. See also ideas: tangible/visible

Index signs, 6, 7, 54–5, 60, 65, 69–70, 81, 85, 88, 180, 194, 196 Sisson, Charles, 10, 223 solipsism, 3, 10, 185 sorts, 59, 63–8, 71–2, 80–1 soul. See mind sounds, 70, 75, 98, 144–5 Southey, Robert, 221 space, 19, 22 Spinoza, Baruch, 11, 12, 25, 60, 63, 79, 214, 215, 217, 226 Stack, George, 154, 167–8 Stoicism, 9, 164, 203, 207–8 subject–mode distinction. See modes subsistence, 53, 164 substance, 5, 7–8, 13, 17, 38, 51–3, 96, 121–50, 153, 157, 161–5, 182, 186; autonomous, 131–2; corporeal, 52– 3; material, 128–30, 140–1, 179; mental, 121–52, 157–8, 161–2, 165, 182–4, 221; predication, 134–40; support of sensible objects, 127–31 suggestion, 6, 68–69, 71, 81, 95, 98, 102, 104, 107, 168 Swift, Jonathan, 217 tar-water, 216 things-in-themselves, 16, 18, 23, 26, 61, 72

235

time, 19, 22, 178–9, 185 Tipton, Ian, 170–1 Trinity, 9, 189–91, 194–5 truth, 37, 42, 189–90 understanding, 8, 153, 163, 169, 174, 177, 181 unity: idea of, 59–62, 67, 72, 78–81, 177; of ideas, 5, 7–8, 10, 59–62, 65, 72, 177; of mental substances, 126– 7 Urmson, J.O., 212–13 utilitarianism. See morality: rules virtue, 201, 203, 207–8, 212 vision. See sight volition. See will vulgar–learned distinction, 83, 91–3, 97, 105 Whiggery, 217 will, 4–5, 7–8, 140–5, 151–7, 163–4, 175–8, 180–1 Winkler, Kenneth, 97–8, 119, 167 Wittgenstein, Ludwig, 12, 13, 218, 222 Yeats, W.B., 9, 216