Philosophy, its History and Historiography (Royal Institute of Philosophy Conferences)

PHILOSOPHY, ITS HISTORY AND HISTORIOGRAPHY

Edited by A. J . H O LLAND Department of Philosophy. University of Lancaster

D. REIDEL PUBLISHING COMPANY A MEMBER OF THE KLUWER

ACADEMIC PUBLISHERS GROUP

DORDRECHT / BOSTON / LANCASTER / TOKYO

Library of Congress Cataloging in Publication

Data

Main entry under title: Philosophy, its history and historiography. (Royal Institute of Philosophy Conference; v. 1 983) Includes index. I. Philosophy, Modern-Congresses. 2. Philosophy, Modern-Historiography-Congresses. I. Holland, A.J. . II. Series. (Alan John), 1939190 B791.P44 1 985 85-11898 ISBN 90-277-1945-4

All Rights Reserved

© 1985 by D. Reidel Publishing Company, Dordrecht, Holland

TAB LE O F C O NTENTS

vii

SERIES PREFACE

ix

PREFACE P A RT I : CONCEPT IO N S O F P H I LO S O P H Y ' S H I S T O RY JONATHAN REEjThe End of Metaphysics: Philosophy's Supreme Fiction?

3

M. R. AYERS j 'The End of Metaphysics' and the Historiography of Philosophy

27

ANTHONY MANSER/The End of Metaphysics: A Comment

41

JONATHAN REE/Reply to Ayers and Manser

47

, MARY HESSE/Epistemology without Foundations

49

PHILIP PETTIT/ Philosophy after Rorty

69

ADAM MORTON / Comment on Rorty

85

MA.RY HEssE/'Heterodox', 'Xenodox', and Hermeneutic Dialogue

87

PHILIP PETTIT/Reply to Mary Hesse

91

PART I I : P H I L O S O P H Y I N THE S EVENTEENTH CENTURY G. MacDONALD ROSS/ Occultism and Philosophy in the Seventeenth Century

95

SIMON SCHAFFER/Occultism and Reason

1 17

G. MacDONALD Ross / Reply to Simon Schaffer

145

GARY HATFIELD/First Philosophy and Natural Philosophy in Descartes

· 149

DESMOND M. CLARKE/Cartesian Science in France, 1 660-1 700

1 65

vi

TABLE OF CONTENTS

FRANCKS / Caricatures in the History of Philosophy: The Case of Spinoza

RICHARD

STUART BROWN/Leibniz's EDWIN 1. R.

Break with Cartesian 'Rationalism'

McCANN / Lockean Mechanism Appendix: Was Boyle an Occasionalist?

MILTON / Lockean Mechanism : A Comment

1 79 1 95 209 229 233

PART I I I : P H I L O S O P H Y IN THE E I GHTEENTH CENTURY M. A . STEWART/ Hume and the "Metaphysical Argument A Priori" NICHOLAS

CAPALDI/ The Historical and Philosophical Signifi-

cance of Hume's Theory of the Self ECKART FORSTER/Kant's

P.

Refutation of Idealism

B. wooD / The Hagiography of Common Sense: Dugald Stewart's Account of the Life and Writings of Thomas Reid

INDEX

243 271 287 305 323

S ERIES PREFACE

The Royal Institute of Philosophy has been sponsoring conferences in alternate years since 1 969. These have from the start been intended to be of interest to persons who are not philosophers by profession. They have mainly focused on interdisciplinary areas such as the philosophies of psychology, education and the social sciences. The volumes arising from these conferences have included discussions between philosophers and distinguished practitioners of other disciplines relevant to the chosen topic. Beginning with the 1 979 conference on 'Law, Morality and Rights' and the 1 98 1 conference on 'Space, Time and Causality' these volumes are now ' constituted as a series. It is hoped that this series will contribute to advancing philosophical understanding at the frontiers of philosophy and areas of interest to non-philosophers. It is hoped that it will do so by writing which reduces technicalities as much as the subject-matter permits. In this way the series is intended to demonstrate that philosophy can be clear and worthwhile in itself and at the same time relevant to the interests of lay people.

General Editor, Royal Institute of Philosophy Conference Series

vii

STUART BROWN

PREFACE

This volume contains papers presented at the conference on Philosophy and its History organised by the Royal Institute of Philosophy and held at the University of Lancaster in September 1 983. On this occasion, philosophers, historians of philosophy and historians of science from Britain and overseas met together to consider not simply the history of philosophy but more especially its historiography. Received views of the history of philosophy were discussed and challenged both in general terms and through detailed studies focussing on the seventeenth and eighteenth centuries, when many of the categories and questions of so-called 'modern' philosophy were being formulated. Both the prepared contributions and the remarks from the floor yielded evidence in plenty that the rewriting of the history of philosophy is, once again, under way, informed by fresh findings and perspectives from the history of science in particular, but also from the histories of literature, theology, politics and other disciplines. Nor was there any mistaking the widely held view that only through the abandoning of ahistorical perspectives can the history of philosophy be the source of stimulus and inspiration for contemporary philosophy that it has the capacity to be. It is perhaps appropriate to record at this place that the conference was partly instrumental in provok­ ing the formation of a new British Society for the History of Philosophy. The conference format was a blend of commissioned symposia, all of which are included here, and submitted papers, of which a selection are included. Most of the papers in this volume were prepared specially for the Lancaster conference, one or two being written up later or extensively revised. Professor Hatfield's paper was accepted for the conference but is not the one he actually delivered there. The rejoinders by Professor Hesse, Professor Pettit and Mr Ree, and the chairman's comments of D r Milton and Professor Morton were written specially for the volume, and for these we are most grateful. None of the papers has appeared elsewhere. The volume is in three parts. In the first part past and present con­ ceptions of the history of philosophy are compared and one particular conception, that presented by Richard Rorty in Philosophy and the Mirror ofNature, is extensively explored. In the second and third parts individual ix

x

PREFACE

studies relating to the seventeenth and eighteenth centuries respectively provide varied points of reference for the preceding general discussions. I wish to thank first and foremost Dr Stuart Brown, who played a major role both in the conception of the conference and in its organisation; also Dr M. R. Ayers, Professor P. Jones, Professor S. Korner, Professor G. H. R. Parkinson and Dr M. A. Stewart for their guidance and help in planning the programme. I am indebted to Dr Stewart also for his ungrudging and patient advice throughout the preparation of this volume. ALAN HOLLAND

PART I

CONCEPTIONS OF PHILOSOPHY'S HISTORY

JONATHAN REE

T H E E N D OF M ETAP H Y S I C S : P H I LO S O P H Y ' S S U P REM E FICTIO N?*

We reason of these things with later reason And we make of what we see, what we see clearly And have seen, a place dependent on ourselves Wallace Stevens I

Most intellectual work, if not all, involves some sense of the past - with founding fathers , prodigal errors, fearful ordeals, internecine disputes, marvellous discoveries, and decisive battles, all leading up to the present state of the question : built-in historical maps, so to speak, with little flags stuck in, saying 'we are here'. These 'integral histories', as I shall call them, are sometimes explicitly elaborated, either as a quaint and colourful folklore or as an exactly controlled scholarly enterprise ; but often they are assimilated implicitly and unawares, by unconscious imitation. In literate cultures, in particular, the normal activities of the intellectuals - lawyers, teachers, priests, poets, scientists, or preachers - constantly confront them with texts: that is, with writings, which come, obviously enough, from the past and, in some cases, from time out of mind.2 And the very act of referring to these texts - the form and manner of the glosses, citations, and allusions, or indeed of misprision or neglect - will inevitably if un intention­ ally express some attitude to the relation between the past time of compo­ sition and the present moment of interpretation, and will tend to breed stories to fill, imaginatively, the interval. The stories will try to persuade you that whereas people once held this opinion, and then that, the view for today and the future is the one you are now being offered; they will present an argument whose logic ends where you are now being asked to stand. Of course, integral histories are not usually supposed to be taken very seriously: they are material for rhetorical padding in introductions and conclusions, or for remarks by the way. Allowing one's interest to be detained by them is, it is generally agreed, a mark of a minor thinker. Even granting their marginality, however, I think it is possible that integral histories exert a fateful influence nonetheless (or allthemore) on 3 A . J. Holland (ed.), Philosophy, Its History and Historiography, 3-26.

4

JONATHAN REE

intellectual life. By furnishing adoptive lineages for intellectuals, integral histories otTer them a sense of direction, an aspiration, and an identity around which to organise their work. It would, I think, be worth exploring how far intellectual history might be written as a history of integral histories, of the stories intellectuals have told about their place in history. In this essay, I shall attempt a small part of the task, confining myself to self-styled 'philosophical' writers within Western Christendom. I shall try to make a case for seeing transformations in the field of philosophical inquiry in terms of a succession of integral histories of philosophy. For memory's sake, I shall schematise the development into four phases, medieval, renaissance, enlightenment, and modern, and I shall try to explain how it has come about that philosophy, in its latest phase, has been characterised by its protracted arias about its own recent, overdue and richly deserved demise: the modern philosopher's curious theme, in other

words, of 'the end of metaphysics'.

I. MIDDLE AGES: PHILOSOPHY AS ANTIQUE PAGANISM

The magic of stories is that they make meanings blossom from the dry twigs of chronology, by marking events as belonging, intrinsically, to the beginning of a plot, or the middle, or - not least - the end, rather than simply occurring at a particular date. The elementary structure of a well­ made story - as Hegel (great if facile raconteur) knew all too well - has undivided possession at the beginning; wandering, loss and nostalgia in the middle; and return, retrieval and a sigh of relief at the end. And a most appealing kind of integral history, it seems , is a three-part invention which arranges its canon of texts and authors into a beginning and a middle whose significance is revealed, or perhaps conferred, by occupying a particular vantage point - this one - in the present. It will tell of an ancient age (perhaps golden); then a middle age (certainly dark); and now, at the beginning of the end, it must be the modern age (surely brave).3 Most of the weight of these tripartite histories, clearly, is borne by the concept of renascence, of anamnesis or revival of some original radiance and splendour. The literate intellectuals' daily business of textual inter­ pretation naturally intensifies their sense of the present as a time of reclamation of buried sense. But the idea of renascence had an additional pertinence for Christian thinkers, since, having translated their religion from the language of its founder into those of the Roman Empire, they found that the institutions which supported these languages - especially

THE END OF METAPHYSICS: PHILOSOPHY'S SUPREME FICTION? 5 the Imperial Schools - were crumbling.4 The Christian Grammar Schools, along with the Universities which eventually developed from them, were thus allotted the incongruous task of preserving a dead language whose classics were Pagan, and of piously performing the awkward duty of breathing new life into godless texts. And it was in this context that the word 'modern' began its interesting career s as a term which seems to designate a simple date, whilst actually implying an obligation too - the challenge to stop dozing amongst the time-wasting obscurities of the past, and wake up to what really matters. The most embarrassing point for a Christian intellectual, trying to be modern in a culture whose classics were Pagan, was philosophy: for anti­ quity, it seemed, had given to a theoretical discipline called philosophy the pre-eminence which Christians should be givmg to the spiritual disciplines of worship instead. The early Fathers of the Church had attempted to unite philosophy with Christianity, but in vain;6 and so, for the first thousand years of organised Christian culture, philosophy remained a stubbornly Pagan affair: the supreme philosopher was Aristotle, not Christ, and the study of his texts - Ethics, Physics, and Metaphysics, or 'the three philo­ sophies' - crowned the work of the Arts Faculties of the early Univer­ sities. ? Christian scriptures were segregated from the Arts, and belonged to a separate and senior teaching body, the Faculty of Theology. (The calamities of Peter Abelard warned of the risks involved in annexing Christianity to philosophy.) For medieval Christendom, therefore, the story of philosophy's past was mostly implicit in citations or explanations of Aristotelian texts, genuine or spurious. But explicit histories were also available, some of them staging stylised debates between representatives of established schools , on the pattern of Cicero's Academica, and others taking the form of prefatory surveys of past opinions on a given point, modelled on Book One of Aristotle's Metaphysics. In addition, there had been several ancient works completely given over to the history of philosophy, and one of these Diogenes Laertius' third-century Lives �f Eminent Philosophers - survived into the Christian era, and is perhaps the clearest representation of medieval Christianity's view of philosophy's past. Diogenes begins by mentioning 'barbarian' philosophers - Magi, Chal­ deans, Gymnosophists and Druids - but after a few lines these make way for eighty or so civilised Greeks and Italians. To the extent that Diogenes marshalls these characters into groups, his organising concepts refer to lines of succession personally transmitted from masters to pupils, rather

6

JONATHAN REE

than to points of theoretical agreement or difference. The principal dynasties of diadochai were the Ionic, terminating in Theophrastus, who could trace his pedagogical ancestry back, through teachers and teachers' teachers, to Thales' pupil Anaximander, and the Italic, stretching in a continuous tradition from Pythagoras down to Epicurus. Informative and entertaining as it often is, Lives of Eminent Philosophers is short on theory : as its modern editor observes, it is not so much a history of philosophy as a "contribution to the biography of men of letters who happened to be philosophers". 8 However, its conception of philosophy as having a history that begins with Thales and ends with Epicurus was adopted in its entirety by medieval Christians. Diogenes was eventually translated into Latin in the fourteenth century, and achieved enormous renown and popularity in printed versions in the sixteenth. If renaissance philosophers wished to distance themselves from Aristotle, it would most likely be by identifying with another character from Diogenes, such as Pyrrho, Carneades, Epicurus, or of course Plato.9 Up to the middle of the seventeenth century, philosophical erudition was devoted to establishing the texts of the authorities mentioned in Diogenes , and clarifying the issues between them, whilst so-called 'Histories of Philosophy' were no more than embellished, interpolated and annotated translations of his Lives. I O 2. RENAISSANCE: PHILOSOPHY AS CRYPTIC CHRISTIANITY Given Christianity's awkward relation to its Pagan classics, it is not sur­ prising that historians can argue, for almost any period of Christian intellectual history, that it was really a renaissance too, much like the one which, as was once taken for granted, was sparked off by the fall of Constantinople in 1 453. The epidemic renaissance of the fifteenth century can be distinguished from the chronic renascence of Christianity by its sense of the irretrievable distance of really venerable antiquity - a period which, as the renaissance progressed, was relegated to an ever earlier date, displaced backwards by an expanding middle age whose gloom, it seemed, was only darkened by the feeble flickerings of its scattered attempts to rekindle the old brilliance. But in the same movement, an unprecedented accommodation was arranged between Christian piety and ancient Paganism. At the beginning of the sixteenth century, Erasmus began to promote the programmatic concept of a 'philosophia ChristI", I I which was to strain and eventually snap the philosophers' attachment to the kind of history of

THE END OF METAPHYSICS: PHILOSOPHY'S SUPREME FICTION? 7 philosophy told bY'Diogenes Laertius. For Erasmian Christian humanism, the story had to be altered so that Christianity could emerge as its epitome, with Christ as a higher Socrates, and his followers as his philosophical pupils. And this project required that a process of deep interpretation be applied to the traditional philosophical texts, aimed at the discovery, behind their surface disagreements, of a concealed 'perennial philosophy' which could be taken to underlie Christian scripture too. 12 In the middle decades of the sixteenth century, Peter Ramus became an influential head of this movement. He made a career of denouncing Aristotle and the 'scholastic' dialectic which had dominated the curriculum of medieval schools, and he contrasted it with a virtuous 'true dialectic', essentially Christian, which, though known to Plato, Aristotle and their followers, had unfortunately not been entrusted by them to the written record. Galen, according to Ramus, had finally "closed the door of dialectic, which was never after opened until our day". 13 Ramus's ambition, therefore, was to restore what he called the 'single method' in logic : this, he argued, would enable all fields of knowledge to be reduced to simple principles, easily taught and readily recalled; and .this in turn - as he insisted with relentless practicality - would increase the productivity of teachers and reduce the time a boy would need to spend getting his education. 14 Ramist writings are full of the idea of a present strife between those who are moving with youth toward the future and those who cling to the habits of the scholastic past. It was the spirit, not the letter, of classical philo­ sophy that Ramus revered; in fact, to him, antiquity was identical with living spirit, as against the dead letter of scholasticism. So references to classical philosophers were made less as specific citations, more as open­ ended lists, designed to evoke the single, secret philosophy common to all authorities - "men like Pythagoras, Euclid, Plato and Aristotle," as Everard Digby put it, "by whose living voice we might be guided to knowledge step by step, as by a magic wand". IS Descartes too considered, or at least stated, that his own 'method' was only a revival of a knowledge kept secret by the ancients. 16 The idea that the true philosophy had been disguised and enigmatised by the classical authors enabled a decisive step to be taken in freeing the history of philosophy from the limits within which Diogenes Laertius had confined it. This was the publication in 1655 of the compendious Philo­ sophical Histories by Georg Horn, Professor of History in the Arts or Philosophical Faculty of the Protestant University at Leyden, "A man of vast reading, rather than great parts," according to an early account, which

8

JONATHAN REE

was more interested in telling its readers that the unfortunate Horn was "a little maniacal towards the end of his life ; which disorder was supposed to be occasioned by the loss of 6000 florins he had entrusted with an alchemist at the Hague" . 1 7 The condescension is characteristic, for Horn has seldom received due credit for his creation of a new model past for philosophy. IS Horn's originality is, first, that his history ranges, as its subtitle says, "from the creation of the world to the present day" ; and secondly, that it unhesitat­ ingly puts Holy Scripture on a par with the Pagan Classics. The story begins with Adam, blessed with complete innate philosophical wisdom 'naturalis sapientia ' or "what we now call Dialectic". With the fall, however, the human mind was "darkened by heavy shadows of ignorance"; philosophy could now be acquired only by arduous exertion, and even then, in this life, never better than partially. So began 'the old philosophy', and the school of Adam divided into the Cainites, who prospered, and the Abelites, who did not. In the second of his seven compact books, Horn tells his readers how Cainism was washed away by the flood, and philosophy could start again thanks to the teachings and writings of Noah. In Book Three, he presents the heroic and theological periods of Greek thought, before calmly condensing the 'philosophical period' - all the philo­ sophers recognised by Diogenes Laertius, the 'classics' of the medieval canon - into about forty pages. The B arbarians, Romans, and Jews came next, making way for the arrival of Christ in Book Five - as the S aviour of the Human Race, and also the Instaurator of True Wisdom, author of the knowledge which had been lost and polluted in Adam. Christ trans­ mitted his philosophy to the Apostles, who did their best to drive out all remnants of Paganism, until eventually overwhelmed by B arbarian invaders. Book Six deals with the three phases of 'scholasticism', from the ninth to the fifteenth century; and with that, in Horn's story, the 'old philosophy' came to an end. The 'new philosophy', for Horn, had its origins in a 'philological period' initiated by Petrarch and Erasmus ; but 'our age', or 'the period of philoso­ phy', had been opened by Vives and Ramus, and now flourished in the works of Campanella, Hobbes, Descartes and many others. Their achieve­ ment was to have set aside the 'sectarianism' of the 'old philosophy'; and Horn allowed himself to speculate that a new 'syncretism', unifying Plato, Aristotle and everyone else, might be under way, thanks to the work of Goclenius and Pico. But rather than risk an analysis of the rise of syncretism in the brave new

THE END OF METAPHYSICS: PHILO SOPHY'S SUPREME FICTION ? 9 world, Horn concluded with a seventh book, which dealt with the life-styles of philosophers through the ages, and then described the Degrees - Bachelors, Masters and Doctors - into which a modern Philosophy Faculty is organised. 19 Thu s Horn's textbook supplied Arts students with an integral history explicitly linking them, through their academic study of philosophy, not only to the classics of Ancient Greece, but to Eden and the philosophia Christi too. The antagonism of philosophy to Christianity seemed to have been overcome at last. 3 . ENLIGHTENMENT: PHILOSOPHY AS LOSS AND RECALL OF COMMON SENSE

The Renaissance division between the benign spirit of natural wisdom and the cryptic texts of the philosophical authorities could, however, cause discomfort as well as complacency. The story of how philosophers, from Cain to Campanella, had tried in vain to formulate what Adam had known by nature might be a comedy rather than an epic. Its protagonists, so it might seem, were simply puffed up with deluded enthusiasms arising from a ridiculous conflation of reality with romance ; the annals of philosophy, mere chronicles of wasted time, the adventures of some scholastic Don Quixote. Cicero had once written that "there is nothing so absurd as not to have been said by one of the philosophers", and, through a citation by M ontaigne, the formula became a commonplace of s eventeenth-century philosophy. 2o Descartes adopted it, for instance,2 1 and claimed that of all the "distinguished minds" which had ever been applied to the problem of method, "none have had the patience to find their way out of their difficulties". He compared the development of philosophy to a journey but one where the voyagers kept getting side-tracked, and never reached their destination - if indeed they ever had a destination outside their own deranged fantasies. He lamented that "nearly all" the characters in the history of philosophy "have followed in the footsteps of these travellers who, abandoning the main route in favour of a cross-road, find themselves lost amongst briars and p recipices". 22 Elsewhere, Descartes broke off from an enumeration of the four ways of knowledge (clear notions, sensation, conversation and reading) to reflect on those "who have tried to find a fifth road by which to arrive at wisdom, incomparably more elevated and assured than these other four. . . . I do not know that up to the present day there have been any in whose case this plan has succeeded", he continued,

10

JONATHAN R�E

adding that "those who have made this their special work have been called philosophers". 23 On the face of it, this denunciation of philosophers makes a paradoxical and self-confounding introduction to a book called The Principles ofPhiloso­ phy: the author explicitly claiming intellectual descent from a line of theoretical vagabonds and outlaws. But of course he was also putting himself in a good light by associating his own philosophy with a perennial common sense, whose steadiness he took to be a standing reproach to the "extravagant errors" of those who had, hitherto , "aspired to be philoso­ phers".24 Much like Ramus, Descartes claimed to be an innovator only in having renounced the obsession with novelty which had marred the achievements of the vain philosophers of the past. The principles of his own method, he felt, were so obvious that they must be "accepted by all men"; in fact, they must also have been approved, secretly, "by Aristotle and all the philosophers of every time; so that this (my) philosophy, instead of being new, is the most ancient and common of all".2 5 From Descartes onwards, the habit of decrying the wayward voyages of the old philosophers, who, as represented in Paradise Lost, "found no end, in wandering mazes lost" ,2 6 became a characteristic mannerism of philo­ sophical writings; and it enabled philosophers to make an intriguing invitation to their readers, of joining a democratic, commonsensical alliance against the baseless, self-deceiving conspiracy by which philo­ sophers had imposed on public opinion in the past. Bishop Berkeley - to take one striking example - proclaimed that "I side in all things with the mob", and pledged himself"to be eternally banishing Metaphisics, &c, and recalling Men to Common Sense". 27 The trouble with studying the history of philosophy - as he confided to his readers in the opening paragraph of the Principles of Human Knowledge - is that, however diligently we pursue it, "at length, having wander'd through many intricate mazes, we find ourselves just where we were, or, which is worse, sit down in a forlorn scepticism". But, he continued, we must not despair: "I am inclined to think that the far greater part, if not all, of those difficulties which have hitherto amused philosophers, and blocked up the way to knowledge, are entirely owing to ourselves. That we have first raised a dust, and then complain, we cannot see". Ifwe trust our author, though, we will soon have the comfort of rejoining "the illiterate bulk of mankind that walk the high-road of plain, common sense".2 8 In the Dialogues, the offer is repeated: Berkeley undertakes to dispel "the prejudices of philosophers, which have so far prevailed against the

THE END OF METAPHYSICS: PHILOSOPHY'S SUPREME FICTION ? 1 1 coinmon sense and natural notions of mankind". Some readers, he warned, would be disappointed that, "when they have taken a circuit through so many refined and unvulgar notions, they should at last come to think like other men : yet, methinks, this return to the simple dictates of nature, after having wandered through the wild mazes of philosophy, is not unpleasant. It is like coming home after a long voyage". 29 Ironically, though, the self-styled 'Philosophy of Common Sense' was created in deliberate reaction against the would-be common-sense philo­ sophies of Descartes and Berkeley - both of whom, according to Thomas Reid, had wandered blindly into scepticism. "A traveller of good j udgment may mistake his way, and be unawares led into a wrong track", he wrote in 1 764, " . . . but when it ends in a coal-pit, it requires no great j udgment to know that he hath gone wrong, nor perhaps to find out what has misled him". The truth was, that philosophy "has no other root but the principles of common sense"; and the trouble was that "men of genius in former ages" had made philosophy ridiculous by building castles in the air and making "their own dreams to pass for her oracles". In the time of Descartes, they had failed to deal with common sense "so favourably as the honour and interest of philosophy required"; in the time of Berkeley, they had gone even further and "waged open war with common sense". But now things were changing at last, and �'happily for the present age, the castle builders employ themselves more in romance than in philosophy". 3o So philosophy, for Reid, as for Berkeley and Descartes, was, or ought essentially to be, the dull sameness of perennial common sense; but the history of philo­ sophy was the record of repeated disobediences to this unvarying ideal. 4. MODERN: PHILOSOPHY AS CRITICISM OF METAPHYSICAL SYSTEMS

One of the things which helped philosophy to separate itselffrom empirical and technical fields of inquiry during the eighteenth century was its distinctive attitude to its past. Other subjects had integral histories which boasted a canon of exemplary classics. Philosophy, however, flaunted what one might call an anti-canon - a set of death's-head texts to wart). the unwary of the dangers of philosophical pride. For philosophers, a text accused of error need not be threatened with deletion from the canon: on the contrary, exemplary erroneousness was what made the history of their discipline special: it contained anti-classics, so to speak, rather than classics. And their chosen history furnished them, as they slowly began to Fi'�'"OS0P�V f":ibr
12

JONATHAN REE

perceive, with an intellectual programme : modern philosophers, it was implied, needed to be one degree more sophisticated than their errant predecessors - not in order to perpetuate sophistry, but in order to outwit it; and then, like intellectual Robin Hoods, they would be able to return philosophical treasures from the old philosophers who had appropriated them, to their rightful owners, the plain folk of common sense. This kind of integral history for philosophy was made explicit and cast into scholarly form by Johann Jakob B rucker, whose five-volume History of Philosophy was published in 1742-1744. Brucker's theme was that his work, as a "history of the human intellect", must also be "the index of its errors", and that philosophers had until very recently indulged in sectarian vanity, to the detriment of truth. He built this idea into the structure of his story, by dividing it into three ages: ancient philosophy, starting with the Romans, continuing with the philosophia Christi, and ending with the 'thorns and briars' of the three ages of scholasticism; and then a modern age, beginning uncertainly with the rediscovery of Plato and 'genuine Aristotle' in the fourteenth and fifteenth centuries, but gathering strength in the seventeenth and eighteenth so as to rise "above the unwholesome atmosphere of tyranny, superstition and bigotry, into the pure regions of freedom and truth". This consummation was described by Brucker as "the Universal Eclectic Philosophy", and its leader was Francis Bacon. But Bacon could not be compared with any of the anti-heroes of earlier philosophy, since the whole point of his work was that it showed that it was folly to be a sectary like the philosophers of the past. Thanks to Bacon, philosophical tributes were no longer to be paid to individual authors : they were due to universal human reason itself. Accordingly, the historian of the modern age must dispense with the terminology of sects, it being "incon­ sistent with the nature of Eclectic philosophy to admit of Sectarian subdivisions".3 1 Brucker had no ambitions as a philosophical thinker in his own right, but he confirmed contemporary philosophers in their sense of their for­ tunate modernity, and elaborated it with exceptional diligence, elegance, 'and philological skill. 32 His work was altogether more gratifying and more convincing than Horn's effort of a century before ; and Brucker, unlike Horn, is remembered for having met a death worthy of a scholar, tumbling from his stepladder as he reached for a volume from a high shelf of his library. 33 In the textbook tradition presided over by the incomparable Brucker, the development ofphilosophy until recent times could be charted in a neat

THE END OF METAPHYSICS : PHILOSOPHY'S SUPREME FICTION ? 1 3 and easily-learnt way, with 'schools' laid out along symmetrical avenues of vain disagreement - the sceptics against the dogmatists, the materialists against the immaterialists, the nominalists against the realists, and so forth 34 - all their lop-sided sectarian exaggerations, in the end, perfectly balancing each other out. The term 'metaphysics' gradually began to be used as a name for the ruined ancient city of sectarian philosophy, with each philosopher characterised by a special metaphysical idiosyncrasy called 'his philosophy' or 'his system' : the discipline's classics became "the dry systems of the Old Philosophers". 3 5 To the enlightenment thinkers, nostalgic for the age of common sense, the idea of philosophy as a tradition of pointless 'metaphysical systems' could, as we have seen, be excitingly democratic, urging the modern philo­ sopher to take a lead in overturning millenia of tyranny, so as to set plain good sense free again; or the predicament might equally, as Reid observed, be 'discouraging'. For some eighteenth century philosophers , beginning, I think, with Hume, it became a poignant but invigorating paradox: meta­ physical systems were certainly errors, but they were characteristic not of a passing stage of the progress of the race, but of human nature as such - weeds which we must ceaselessly uproot, though without any hope of eradicating them for ever. Philosophical wisdom, in other words, was only a response to philosophical error, and the story of philosophy did not belong to ordinary historical chronology ; its beginning, middle and end happened simultaneously, and, together, they would last for ever.

Burne: Philosophy and the Criticism of Fictions In the 'Introduction' to the Treatise, Hume describt!d the custom whereby phil�sophers "insinuate the praise of their own systems, by decrying all those, which have been advanced before them"; then he directly set about doing it himself, denouncing the "weak foundation even of those systems, which have obtained the greatest credit", and the incoherences which are "every where to be met with in the systems of the most eminent philo­ sophers and seem to have drawn disgrace upon philosophy itself'. But finally he coolly proposed, on his own account, "a compleat system of the sciences". 36 The explanation of the paradox was that the 'system' proposed by Hume was supposed to take its place not alongside those of its rivals , but on higher ground. Hume's system would account for human nature in general, and in particular for its weakness for the 'metaphysical reasonings' of

14

JONATHAN REE

philosophical system-builders. Hume affirmed, with the conventional allusion, that these contained "nothing which is not the subject of debate, and in which men of learning are not of contrary opinions". 3 7 As he put it in the Enquiry: Here indeed lies the justest and most plausible objection against a considerable part of metaphysics, that they are not properly a science; but arise either from the fruitless efforts of human vanity, which would penetrate into subjects utterly inaccessible to the under­ standing, or from the craft of popular superstitions, which, being unable to defend themselves on fair ground, raise these intangling brambles to cover and protect their weakness. Chased from the open country, the robbers fly into the forest, and lie in wait to break in upon every unguarded avenue of the mind . . . . The stoutest antagonist, if he remit his watch a moment is oppressed?8

In passages like this, by a slight adjustment of the traditional metaphor, Hume was ushering his readers into a new forum of philosophical dis­ cussion, in which the vanity of metaphysics served as the principal theme and the sole justification for philosophy, rather than as a reproach against it. The task of modern philosophy would be to explain how philosophy was composed of what Hume called 'fictions' - fabrications by means of which philosophers, seeking "a consolation amid all their disappointments and amictions", decorated the world with sympathetic human qualities which it did not really possess : substance, form, accident, personal identity, necessary connection, and the state of nature, for instance, would all turn out to be, according to Hume, "mere philosophical fictions". 3 9 So, though lamenting the vanity of metaphysics, Hume did not simply call for an attempt to recover plain common sense. For one thing, this would mean a loss 'in point of pleasure'. But also, the attempt would be bound to fail, for the propensity to dwell amongst consoling 'illusions' would prove to be an ineradicable instinct of human nature, which neither reason nor will could correct. We must, however, try to cope with it, and this counter-philosophical task was the duty of philosophy. There was, Hume said, "a gradation of three opinions, that rise above each other, according as the persons, who form them, acquire new degrees of reason and knowledge. These opinions are that of the vulgar, that of a false philosophy, and that of the true". And although we would find that "the true philosophy approaches nearer to the sentiments of the vulgar, than to those of a mistaken knowledge", there could be no short cut - the true philosophy could only be a commentary on the false.4o The rigorous examination of our dreams, according to Hume, was "an

THE END OF METAPHYSICS: PHILOSOPHY'S SUPREME FICTION?

15

excellent method o f becoming acquainted with our own hearts". In the same way, true philosophers could profitably pursue "a criticism of the fictions of the ancient philosophy", since these, "however unreasonable and capricious", nevertheless enjoyed "a very intimate connexion with the principles of human nature".4 1 The practice of criticism was the only defence against mistaken philosophy - which, though in itself "only ridiculous", could become dangerous by mingling with superstitious pre­ judices so as to give them "the air of science and wisdom". Strategies of deterrence, or even pre-emptive warfare, must be devised therefore: and the only weapons that could be effective against philosophy, were those of philosophy itself. The task involved "carrying the war into the most secret recesses of the enemy". Philosophy would then discover its true vocation : to apprehend false reason and denounce it, "to subvert that abstruse philosophy and metaphysical jargon, which, being mixed with popular superstition, renders it in a manner impenetrable to careless reasoners" the endless, Sisyphean task, in other words, of putting an end to meta­ physics.42

Kant: Philosophy and the Critique of Illusion One of the most important of Kant's agreements with Hume was his idea of philosophy as an antidote to the 'endless controversies' of metaphysics, in which "human reason precipitates itself into darkness and contra­ dictions". History showed, according to the Critique of Pure Reason, that there was progress in the empirical sciences, and even in the pure sciences of mathematics and physics; but in the case of metaphysics, there was only a record of unfulfilled longing, of a 'natural disposition' to engage in quarrels. Metaphysical quarrels, moreover, did not even possess the dignity of true polemic, since it always turned out that "both parties beat the air, and wrestle with their own shadows . . . . Fight as they may, the shadows they cleave asunder grow together again forthwith, like the heroes in Valhalla, to disport themselves anew in bloodless contests". It was the urgent duty of the Schools to criticise these vanities - and the criticism of metaphysics deserved public support because it would vindicate the opinions which recommended themselves to 'the great mass of humanity' anyway. Criticism itself would always be a scholarly rather than a popular pursuit, but its function was to dispose of the 'arrogant pretensions' with which the schools imposed upon public opinion.43 A governing image of the Critique of Pure Reason is of an island:

16

JONATHAN REE

an island, enclosed by nature itself within unalterable limits. It is the land of truth - enchanting name! - surrounded by a wide and stormy ocean, the native home of illusion, where many a fog bank and many a swiftly melting iceberg give the deceptive appearance offarther shores, deluding the adventurous seafarer ever anew with empty hopes, and engaging him in enter­ prises which he can never abandon and yet is unable to carry to completion.

In terms of this parable, the positive task of criticism is to make a survey of the island and to show "by what title we possess it"; and its negative task is to restrain crazy adventurers who take it into their heads to try and flee across the water - like the metaphysicians, poor lost souls, of philo­ sophy's pointless past.44 The largest part of the Critique ofPure Reason is devoted to this negative task - to exposing the "false, illusory character" of the "groundless pre­ tensions" of the metaphysicians, under the general heading of "the critique of dialectical illusion". The special domain of dialectical illusion, according to Kant, is constituted by 'Ideas', which refer principally to the great topics of God, Freedom and Immortality. All Ideas, however, lead into contra­ diction. They represent objects beyond experience; but this, according to Kant, is fraudulent, since it amounts to postulating objects which do not satisfy the conditions of obj ectivity, and can therefore lead only to dis­ appointment. The task of criticism, Kant wrote, was like that of a court before which such claims would be tried; the pages of the Critique itself were just "the records of this lawsuit", which he proposed to "deposit . . . in the archives of human reason".45 The Critique was published in 1781, and it concludes with an appeal to readers for assistance in opening up "the critical path" and making it "into a high-road", in the hope of achieving, "before the end of the present century what many centuries have not been able to accomplish; namely to secure for human reason complete satisfaction in regard to that with which it has all along occupied itself, though hitherto in vain". Kant's confidence, however, was limited in scope. Criticism might shortly be perfected, but that would not put paid to bad metaphysics, which would never be eliminated, since it arose "from the nature of universal human reason". The delusions of metaphysics were not accidental :

They are not fictitious (erdichtet) and have not arisen fortuitiously . . . . They are sophistications not of men but of pure reason itself. Even the wisest of men cannot free himself from them.

THE END OF METAPHY S I C S : P H I L O SO PHY'S S UP R EM E FICTION ?

17

After long effort he perhaps succeeds in guarding himself against actual error; but he will

never be able to free himself from the illusion, which unceasingly mocks and torments him.

Even "the severest criticism" could do no more than "neutralise . . . the harmful influence" of the delusions produced by our Ideas.46 Ideas did have their excellence, however, in that they were able to direct the under­ standing toward certain goals which, though imaginary, were worth striving for.47 And, for Kant, philosophy (as distinct from criticism) was itself such an Idea - "a mere Idea of a possible science . . . to which, by many different paths, we endeavour to approximate". We could never learn philosophy as such, but only how to philosophise ; and this we could do only by studying "actually existing attempts at philosophy" - even though, on Kant's definition, they must all have been doomed to fai1.4 8 Kant advocated respectful treatment of the philosophical classics, recommending, for example, that we "set aside the exaggerations in Plato's methods of expression" so as to be able to appreciate the "peculiar merits" of his doctrine, instead of joining with Brucker in uncomprehending ridicule.49 But the classic authors were admitted to the world of the Critique only to the extent that they exemplified tendencies - and erroneous ones at that - whose specification is read off not from the philosopher's own words, but from Kant's transcendental doctrine of elements. Kant even transformed their names, grammatically speaking, into general terms, with 'the illustrious Locke', 'the good Berkeley' and 'the celebrated Leibniz' each used to impersonate some predictable error. 50 The knots of their individual artistry and intelligence are smoothed away by the transcenden­ tal plane, and the shavings swept up as mere "exaggerated methods of expression". K ant's use of the classics was perhaps less respectful than he thought; and it left him troubled by the question of the progress of metaphysics, and of his own place (or rather that of criticism) in it. Hume might have been content with the idea that the chronological development of philosophy was generally devoid of theoretical significance, and he even took some pleasure in the thought that the effect of his critical labours was that "every thing remains precisely as before". 5 1 And Kant himself had conceded neatly inverting Berkeley's hope for a "return to the simple dictates of nature . . . like coming home after a long voyage" - that we would never be willing to settle on the island of truth: "we shall", he wrote, "always return to metaphysics as to a beloved one with whom we have had a quarrel". 5 2 Still - however inconsistently - Kant did cling to the hope that his

18

JONATHAN REE

'critical philosophy' would secure "an eternal peace",5 3 and he concluded the Critique of Pure Reason by trying to reconstitute the idea of philo­ sophical progress. Whilst excusing himself from the "attempt to distinguish the periods of history in which this or that change in metaphysics came about", he ventured to name the three issues - about the object, the origin, and the method of knowledge - which "gave rise to the chief revolutions in metaphysical theory", and which moved it through its three main phases : from the "infancy of philosophy" (where "men began where we should incline to end, namely, with the knowledge of God"), through periods of dogmatism and scepticism, to the present day, when, as regards method, "the critical path alone is still open". 54 Kant laboured to provide foundations for this view of the chronological development of philosophy not only in the Second and Third Critiques, but also in several political works, which replicated his view of the history of philosophy by presenting social history, too, as the drama of reason's self-emancipation from "self-incurred tutelage". 55 In the 'Idea for a Universal History', Kant bluntly asked how anyone could take the notion of progress seriously, or pretend to discover any "definite plan" at all in "this idiotic course of things human". It would, surely, be the idlest meta­ physical j uggling, for "it is evidently silly to wish to write a history in accordance with an Idea . . . . It seems that with such an Idea only a romance (ein Roman) could be written". 5 6 Still, the hope of "perpetual peace", which had hovered over the discussion of metaphysics in the Critique of Pure Reason, was made literal in the essay on international relations which Kant composed in his declining years, in 1 795. 57 And he came back to the theme still later, in the treatise on progress which appeared in his final publication, The Strife of the Faculties. There he posed the crucial question, 'How is a priori history possible?', and he meant it seriously, not rhetorically. The problem of progress, he affirmed, is not really historical, since the relevant sources belong to the future, not the past. An a priori history is possible insofar as the historian can divert the present towards the future which his reason divines as an Ideal. Moreover, Kant added, popular support for the French Revolution was an indication that ordinary citizens, equipped with mere unsophisticated natural reason, have a spontaneous 'disposition' toward improvement in conformity with reason - which, therefore, they can confidently be expected to will, freely, into existence. 5 8 This gallant and - it must be admitted - unconvincing attempt to give reason the initiative in history is based on a theory of the history of

THE END OF METAPHYSICS: PHILOSOPHY'S SUPREME FICTION? 19

metaphysics which Kant had worked out several years before, though it was not published till after his death (and has not been translated into English). Much of the essay, which is entitled 'Progress in Metaphysics', is a summary of the Critique ofPure Reason, and shows Kant still convinced that metaphysics is an illusion, an attempt to get obj ective theoretical knowledge concerning God, Freedom and Immortality. In metaphysics, we are all at sea - and it is "a sea with no coastline, on which one's progress leaves no trace": "any supposed knowledge in this domain is transcen­ dental, and therefore totally inane". 59 Metaphysics is, indeed, merely "the Idea of a science as system"; yet no one could stop taking it seriously. Like logic, metaphysics was a complete whole, but unlike logic, it was a delusive one, a fiction even ("eine blosse Erdichtung"). Still, it must be constantly lived in, and kept in good repair - for otherwise spiders and moths, which are never far away, are bound to take up residence there, and make it unfit for habitation by reason.60

But what could then be made of the history of metaphysics - seemingly the story, Kant says, of reason's transgressions, not its progression? Kant returned to this question in some notes accompanying the essay, where he argued that a history of philosophy, as distinct from a history of scholar­ ship, could not be written "unless one knows in advance what ought to happen . . . . For it is not the history of opinions cropping up here and there, but the history of reason's self-development from concepts". Might it really be possible, Kant inquires, to draft an a priori schema for the history of philosophy, with which the periods of philosophy, and the opinions of the philosophers, established from independent sources, will be bound to coincide - as if the philosophers themselves had actually had this schema before their eyes, and had developed their knowledge by reference to it?

Kant's reply was a bold 'yes!': an a priori history of philosophy was indeed possible, because the Idea of metaphysics was instinctive for human reason, which necessarily feels a need to make an attempt at the impos­ sible science which always lies, embryonically, within its own soul. The historical progress from dogmatism, through scepticism, to criticism was

20

JONATHAN REE

a . transcendental necessity, not a historical contingency. "This temporal sequence", Kant concludes triumphantly, "is grounded in the nature of the human cognitive faculty". 6 1 I t i s a somewhat delirious piece o f argumentation, n o doubt; and its conclusion, one would think, could be curtly refuted by a simple inspection of the philosophical record before Kant, not to mention since. Such a confrontation with the historical facts , however, cannot easily be arranged. For texts are admitted to the Kantian philosophical canon only as illustra­ tions to a modern compendium of spontaneous but self-confuting illusions. Such a compilation is of course 'dependent on ourselves'; and in that sense it may be granted that the history of philosophy can, as Kant affirms, be known a priori. When I first interested myself in theoretical problems about the history of philosophy, I was propelled by indignation at the insolence with which an author's works would be vandalised once they got counted as part of the fairy-tale called 'the history of philosophy', and conscripted into the contending ranks of mutually annihilating 'schools'. It seemed to me that if we were going to derive appropriate pleasure and benefit from these works, rather than specious grounds for congratulating ourselves on our modernity, then we would have to start paying attention to real history to the particulars both of the 'internal' growth of problems and solutions, and of their changing 'external' situation. I still think so now; but the points I have drawn attention to in this essay force me to complicate the judgment. I have been presenting the history of philosophy in terms of the history of its integral histories, as articulated by writers like Diogenes Laertius, Horn and Brucker, as well as by their more illustrious colleagues. And the theme of philosophy's involvement with error has dominated them all: for the middle ages, philosophy was a set of lineages going from Thales to Epicurus, educationally indispensable but indelibly Pagan and wrong; for . the renaissance, there was a true philosophy, the same from Adam to the present, but it masqueraded through the classics in a disguise of falsehood; for the enlightenment, philosophy was the desertion, by hubristic 'system­ builders', of the truths of common sense; and in the 'critical' attitude formulated by Hume and Kant, philosophy dealt with the antics of the instinctive metaphysician inside every one of us, incorrigibly tempted into error by the alluring mirage of idle theory. Our integral histories of philosophy have not changed much since. They

THE END OF METAPHYSICS : PHILOSOPHY'S SUPREME FICTION ? 2 1 all begin with the 'ancients', chiefly Plato and Aristotle; continue, some­ times with faltering conviction, through a 'middle age' dominated by scholasticism; and conclude with a modern age inaugurated by Descartes.

And they draw on the same re?ertory of common?laces too·. the

Ciceronian idea that nothing is too absurd to gull a philosopher; a set of journey-metaphors, with philosophically pretentious minds setting off across country in search of a royal road to truth, and needing modern philosophers to lead them back to the public highway; a collection of images of battles and peace-makers, a democratic rhetoric, and so on. Philosophy since Kant has thus mostly been predicated upon an integral history of philosophy as error, as the incontinent effusions of fine minds gone astray. Hegel, Marx or Comte (and the whole tradition of 'posi­ tivism'), Russell, Wittgenstein or Husser! (and the schools of analytical philosophy and phenomenology), have seen themselves, even obsessively, as inheritors of an empty and disreputable patrimony. 62 Naturally enough, many have then devoted themselves to a revolutionary programme aimed at the elimination of metaphysics - some even presenting the tournament as the apocalypse of Western culture. 63 The question of the accuracy of these integral histories is hard either to shirk or to answer. Is it true that philosophy began with Thales , or that the modern period was inaugurated by Descartes? Is it true that philo­ sophy deals exclusively in vain questions, confected by idle philosophers remote both from ordinary life and from worthy science? Is it true that ?hiloso?hy is mostly error? Clearly not, one might say·. at least it is clear

that attentive scholarship and sympathetic textual interpretation have established, again and again, that histories which present philosophy in this way distort or trivialise the work of the authvrs whom they treat. Such demonstrations, however, do not necessarily invalidate the old hisfories ; they may just show that the texts in question are not truly philosophical after all - that they are literary, psychological, mathematical, or scientific instead. For the secret of integral histories is that they function as definitions. They stipulate what the proper procedures of a discipline are, and draw its boundaries ; and so, contrary to appearances, they are not chronicles of an activity with an antecendently agreed identity: they are decisions about the future, rather than descriptions of the past. It is in this way - and with no pejorative intention - that the theme of 'the end of metaphysics' might be described as the 'supreme fiction' of modern philosophy - a tale philosophers tell to explain their peculiar fascination with error, and to poise themselves for an activity which wist-

22

JONATHAN

REE

fully defines itself as a disease which is its own only possible cure. In this light, the idea of putting an end to metaphysics has a significance rather different from what at first appears. It is not that philosophy is a subject which, because of the hardness of its problems or the softness of its practitioners , has till recently had no theoretical successes to show for hundreds of years of theoretical effort; nor is it that philosophy is an artistic rather than a theoretical activity, from which ordinary judgments of truth and error should be withheld. It is, rather, that philosophy specialises in seeing error as requiring diagnosis and understanding rather than summary dismissal: hence the seeming self-disdain of its integral history; and the perhaps uniquely negative canon of its classics (an 'anti-canon' of 'anti-classics', as I have termed it). It follows that those who deplore, lament or deride philosophy for having a past which is a catalogue of errors, are getting the cart before the horse. It is not that philosophers have failed to produce or deliver the goods which they promised. It is rather that they regard the erroneousness of a text, or the unreality and evanescence of the problems with which it deals, as engrossing rather than repellent. To be interested in putting an end to metaphysics, therefore, is simply to pay one's dues to philosophy as a distinctive discipline. But to mistake this convention for an original discovery, and announce an 'end of metaphysics' as if it were something new, is simply to be taken in by the oldest fiction of 'modern philosophy'.

Middlesex Polytechnic

NOTES *

I a m grateful t o Michael Ayers, Dick Leith, Anthony Manser, Gyorgy Markus, John Mepham and Claud Pehrson for comments on an earlier draft of this essay. 1 Wallace Stevens, 'Notes Toward a Supreme Fiction', Selected Poems (London: Faber, 1953): 99-129, p. 122. 2 On the effects of literacy on senses of history, see Walter J. Ong, Orality and Literacy: The Technologising of the Word (London: Methuen, 1 982), Chapter I . 3 On plot, see Aristotle, Poetics, 1 450b; on Hegel, Jonathan Ree, 'Descartes' Comedy', Philosophy and Literature, forthcoming; on endings, Frank Kermode, The Sense of an Ending (Oxford: Oxford University Press, 1966), and recent work by Paul Ricoeur, summarised in his essay 'On Interpretation' in Philosophy in France Today, edited by Alan Montefiore "(Cambridge: Cambridge University Press, 1983). 4 See H. 1. Marrou, A History ofEducation in Antiquity, translated by G. Lamb (London: Sheed and Ward, 1956); F. W. Farrar, History of Interpretation (London: Macmillan, 1 886). 5 Frank Kerrnode, The Classic (London: Faber, 1 975), p. 1 5.

T H E END OF M ETAPHYS I C S : P H I LO SOPHY'S S UPREME FICTION ? 23

6 See Werner Jaeger, Early Christianity and Greek Paideia (Cambridge: Harvard University Press, 1962), pp. 29, 65-66; Peter Brown, Augustine of Hippo (London: Faber, 1 967), pp. 40-42. 7 See Jonathan Ree, 'Philosophy as an Academic Discipline: The Changing Place of Philosophy in an Arts Education', Studies in Higher Education 3. No. I ( 1978): 5-23. 8 Lives of Eminent Philosophers, edited by R. D. Hicks, Loeb Classical Library (London: Heinemann, 1938), Volume I , p. xvii. 9 See for instance Gassendi's treatment of Epicureanism: Animadversiones in Librum X Diogenes Laertii (Leyden, 1649); see also Richard H. Popkin, The History of Scepticism from Erasmus to Spinoza (Los Angeles: University of California Press, 1979). 10 Thomas Stanley's very successful History of Philosophy, Containing those upon whom the A ttribute of WISE was confe"ed, 3 Volumes (London, 1655-1660) is a translation of Diogenes, embellished with engravings and expanded with translations of Sextus Empiricus, Gassendi and Pico della Mirandola. II See Marjorie O'Rourke Boyle, Christening Pagan Mysteries: Erasmus in Pursuit of Wisdom (Toronto: Toronto University Press, 1982), Chapter I . 12 The phrase 'philosophia perennis' was devised by Augustinus Steuchus in 1 540. On its implications for textual interpretation, see P. O. Kristeller, Renaissance Thought II (New York: Harper and Row, 1 955), pp. 98-99. 13 Dialectique ( 1 555), Preface, quoted in Walter J. Ong, Ramus, Method and the Decay of Dialogue (Cambridge: Harvard University Press, 1958), pp. 47-48. 14 See Neal W. Gilbert, Renaissance Concepts of Method (New York: Columbia University Press , 1960), Chapter 5. 1 5 De Duplici Modo, 53, quoted in Gilbert, Renaissance Concepts of Method, pp. 205-206. 6 1 'Reply to Objections II', in E. S. Haldane and G. R. T. Ross, The Philosophical Works of Descartes (Cambridge: Cambridge University Press, 1 9 1 1 ), Volume 2, p. 49; see also below, n. 25. 17 Encyclopaedia Britannica, Third Edition ( 1 792), s.v. 'Hornius'. 18 C. B. Schmitt, for instance, ranks Horn's History well below Stanley's homely version of Diogenes: s(;e The Totality of Thoughts', Times Literary Supplement, 6 May 1983, p. 463. 19 Georgius Hornius, Historiae Philosophicae Libri Septem, quibus de origine, successione, sectis et vita Philosophorum ab orbe condito ad nostram aetatem agitur (Leyden: Elsevier, 1 655). My quotations are taken from pp. 8, 14, 60-6 1 , 261 , 7, 323 and 386-387. 20 Cicero, De Divinatione II, 58, 1 19; Montaigne, Apologie de Raymond Sebond; see Etienne Gilson, Rene Descartes: Discours de la Methode, Texte et Commentaire (Paris: Vrin, 1930), p. 1 78. See also Hobbes, Leviathan, I, 5. 21 Discourse on Method, II, Philosophical Works, Volume I , p. 90. 22 The Search After Truth by the Light of Nature', Philosophical Works, Volume I , p. 306. 23 'Letter to the Translator' [of the Principles] , Philosophical Writings, Volume 1, pp. 205-206. The chief philosophers, according to this passage, were Socrates, Plato, Aristotle and Epicurus. 24 Moreover, "those who have learnt least about all that which has hitherto been named philosophy, are the most capable of apprehending the truth", 'Letter to the Translator', p. 208. 25 Principles of Philosophy IV 100, Philosophical Writings, Volume I , p. 296. See also the Letter to Father Dinet, Volume 2, pp. 359-360: "that there is nothing in all this philosophy in so far as it is termed Peripatetic and different from others, that is not new; and that on the other

24

JONATHAN R � E

hand there is nothing in mine that is not old." 26 John Milton, Paradise Lost II, lines 557-561 . 27 'Philosophical Commentaries', 405, 7 5 1 , in Berkeley: Philosophical Works, edited by M. R. Ayers (London: Dent, 1975), pp. 288 and 324. 28 Principles of Human Knowledge, Introduction, Philosophical Works, pp. 65-66. 29 'Three Dialogues between Hylas and Philonous', Preface, Philosophical Works, p. 1 32. 30 Thomas Reid, An Inquiry into the Human Mind ( 1 764), edited by Thomas Duggan (Chicago: University of Chicago Press, 1 970), pp. 9-20. 3 1 J.J. Brucker, Historia Critica Philosophiae a Mundi incunabulis ad nostram usque aetatem deducta, 5 Volumes (Leipzig, 1 742-44); second edition, with supplementary volume, 1 766-67. My quotations are taken from Volume I, p. 2 1 , Volume 3, p. 24 1 and Volume 5, pp. 4-9, and also from the abridged translation by William Enfield, The History of Philosophy, 2 Volumes (London, 1 79 1 ), Volume I , p. 1 3 and Volume 2, pp. 3 1 9 and 5 1 2. 32 Earlier accounts of this modernity had been given in works specialising in the history of morals, notably those by Pufendorf ( 1 678) and Barbeyrac ( 1 706). See Richard Tuck, Natural Rights Theories: Their Origin and Development (Cambridge: Cambridge University Press, 1 979),

pp. 1 74-1 75. Notable subsequent confirmations include D'AJembcrt's 'Discours Pre­

liminaire' ( 1 7 5 1 ) to Diderot's Encyclopedie (which itself borrowed copiously from Brucker) and Dugald Stewart's 'Dissertation First . . . on the Progress of Metaphysical, Ethical and Political Philosophy since the Revival of Letters in Europe', Encyclopaedia Britannica, Supple­ ment (Edinburgh, 1 824), Volume I : 1-166; Volume 5 : 1-257. 33 See Storia delle Storie Generali della Filosofia, Volume 2, Dall' Eta Cartesiana a Brucker, edited by Giovanni S antinello (Brescia: La Scuola, 1979), p. 5 3 1 . When complete, this five­ volume project should at last put discussions of the historiography of philosophy on a reliable scholarly ground. 34 The habit of marshalling philosophical texts into opposing 'schools' was of course ancient. Diogenes' mention of the division between sceptics and dogmatists (Lives, Volume I , p. 1 7 ) had been taken u p i n the fifteenth century (see Popkin, History of Scepticism, pp. 1 8-1 9); Proclus' distinction between the partisans of Body and the partisans of Mind (In Platonis Theologiam I , 3) was adopted in Berkeley's Siris (Berkeley: Philosophical Works, p. xvi); and in the fourteenth century, rival dialecticians had defined themselves as 'nominalists' and 'realists', to be retitled, on the defeat of the latter, the via moderna and the via antiqua respectively (see Anders Piltz, The World of Medieval Learning (Oxford: Blackwell; 198 1 ), pp. 25 1-253. 3S The phrase comes from Thomas Baker, Reflections on Learning ( 1 699); see Oxford English Dictionary, s.v. 'system'. For Galileo, 'system' had meant a set of objects, rather than a theory; it was in order to mark a different usage that Cudworth spoke of his "system of pre­ established harmony" ( 1 7 1 4) and Vico (First New Science II, Introduction) hoped to "establish this Science . . . as a System" ( 1 725). Condillac, however, added a tincture of sarcasm to the idea of 'systems' (Traite des Systemes, 1 749), as did Adam Smith (History ofAstronomy, 1 795). See also Walter J. Ong's suggestive essay, 'System, Space and Intellect in Renaissance Symbolism', in The Barbarian Within (New York: Macmillan, 1962). 36 David Hume, A Treatise of Human Nature ( 1 739-40), edited by P. H. Nidditch (Oxford: Oxford University Press; 1978), pp. xiii, xvi. In the Enquiry Concerning the Principles ofMorals ( 1 75 1 ), Hume remarked that "Men are now cured of their passion for hypotheses and systems in natural philosophy" and asserted that "it is full time they should attempt a like reformation

THE END OF M ETAPH Y S I C S : P H I LO S OPHY'S S UP R E M E F I CTI O N ?

25

in all moral disquisitions". See Enquiries, edited by L. A. Selby-Bigge and P. H. Nidditch (Oxford: Oxford University Press, 1975), pp. 1 74-175. 37 Treatise, p. xiv. 38 Enquiry concerning Human Understanding ( 1 748), Enquiries, p. I I . 39 Treatise, pp. 2 19-224, 254-259, 493 etc .. 40 Treatise, pp. 222-224. 4 1 Treatise, p. 21 9. 42 Treatise, p. 272; Enquiry concerning Human Understanding, Enquiries, pp. 12-13. 43 Immanuel Kant, Critique of Pure Reason, translated by Norman Kemp Smith (London: Macmillan, 1933), Preface to First Edition, p. 7; Introduction to Second Edition, pp. 56, 604, 30-32. 44 Critique of Pure Reason, p. 257. 45 Critique of Pure Reason, pp. 99- 1 0 1 , 325n, 570, cf. 9, 430, 549, 576, 593-594, 601 . 46 Critique of Pure Reason, pp. 669, 5 6 , 327-328, 532. For further comments o n the inexora­ bility of metaphysical illusion, see pp. 299-300, 329, 386, 485, 532 and 569, and on the idea that metaphysics is more than 'mere fiction' [blosse Erdichtung] , pp. 487 and 46; see also the discussion of 'heuristische Fiktionen', p. 6 14. 47 This "excellent . . . regulative employment" ofIdeas is described by Kant as "that of directing the understanding towards a certain goal upon which the routes marked out by all its rules converge, as upon their point of intersection. This point is indeed a mere idea, a focus imaginarius." (Critique of Pure Reason, p. 533). 48 Critique of Pure Reason, p. 657. 49 Critique of Pure Reason, pp. 3 12-3 1 3 . Perhaps this is the earliest statement of the policy of 'reconstruction' described in Michael Ayers, 'Analytical Philosophy and the History of Philosophy', in Jonathan Ree, Michael Ayers and Adam Westoby, Philosophy and its Past (Hassocks: Harvester, 1978). 50 Critique of Pure Reason, pp. 1 27, 89, 282. They are intended to represent failures to appreciate the transcendental deduction, the transcendental aesthetic, and the distinction between empirical and transcendental uses of the understanding, respectively. 51 Treatise, p. 25 1 . 52 "zu . . , einer Geliebten", Critique of Pure Reason, p . 664. 53 "einen ewigen Frieden", Critique of Pure Reason, p. 601. 54 Critique of Pure Reason, pp. 666-669. 55 'What is Enlightenment?' ( 1784), in Kant on History, translated and edited by L. W. Beck, (Indianapolis: Bobbs Merrill, 1963), p. 3. 56 'Idea for a Universal History from a Cosmopolitan Point of View' ( 1 784), Kant on History, pp. 12, 24. 51 'Perpetual Peace' (Zum ewigen Frieden) ( 1 795), translated in Kant on History. 58 'An Old Question Raised Again: Is the Human Race Constantly Progressing?' Kant on History, p. 1 37. This essay was published as part of Der Streit der Fakultiiten ( 1 798), with a view to demonstrating the superiority of the Philosophy Faculty over the Faculty of Law. 59 'Preisschriji aber die Fortschritte der MetaphysilC ( 1 793), Kants Gesammelte Schriften, Acade­ my Edition (Berlin: Walter de Gruyter, 1 942), Volume 20, pp. 255-35 1 , 296-309, 259, 301. 6O 'Preisschrijt', pp. 3 10, 259, 3 1 9, 3 10. 61 'Preisschrift', pp. 3 1 7, 342-343, 264. 62 There is an interesting presentation of this theme in Anthony Manser's inaugural lecture,

26

JONA THAN R E E

'The End of Philosophy: Marx and Wittgenstein' (University of Southampton, 1973). 63 Consider, for instance, Jacques Derrida's discussion of "the question of the privilege of voice and of phonetic writing in relation to the entire history of the West as represented in the history of metaphysics", Positions (Paris: Editions de Minuit, 1 972), p. 1 3. Such remarks may, however, be no more than a rhetorical decoration of Derrida's work.

M. R. AYERS

'THE E N D OF M ETAPH Y S I C S ' A N D T H E H I S T O R I O G RAPHY O F P H I L O S O P HY

No doubt most philosophers who spend time on the history of philosophy are familiar with that question asked to embarrass (and liable to be asked by scientists in particular) why the history of the subject should be thought a significant part of the subject itself. Either there is progress in philosophy, it is said, or there is not. If there is progress, why the laborious backward glances ? How can the past be so important? Why aren't philosophers like psychologists, given perhaps a short historical orientation before being brought up to the nitty-gritty of the present? If, on the other hand, there is no progress, if we might as well be discussing Locke as Quine, doesn't that imply that philosophy consists in a set of questions for which there is no way of establishing even that some answers are better than others? Wouldn't it be more profitable to pursue questions to which at least provisional answers can be established, approximating to the truth? Wouldn't it be better to be a scientist? To some philosophers such considerations represent a challenge to philosophy to turn itself into a science, and in so doing to turn its back on the past. But another kind of response is available, on the following lines. There is a kind of progress in philosophy: some old doctrines become unbelievable, and new thoughts do sometimes occur. On the other hand, being a priori, philosophy lacks the condition by which natural science (given a certain level of insight, common-sense and objectivity on the part of its practitioners) is prodded or channelled towards the truth, or at least away from falsehood. Philosophy makes no predictions, and there are no philosophical experiments. At the same time it is unlike mathematics or logic. Formal proof is of limited value, if only because premises and principles (not to speak of the interpretation of formal systems) are likely to be as much at issue as conclusions. That is not to say that there are no good answers in epistemology and metaphysics, but that philosophical beliefs are exceptionally liable to be swayed by such non-rational motives as fashion, wishful thinking, rhetoric, personal influence, and academic politics (conservative or revolutionary) as well as by insight. It is only too easy for progress in one respect to be matched by regression in another, for a sound point or promising approach to be pushed too far, for a

27 A. J. Holland (ed.), Philosophy, Its History and Historiography, 27-40. (c) l QR" hu n Rpinp! Puhli�h;no rnmnnnlJ

28

M . R. A Y E R S

paradox to stimulate the opposite paradox. And as it is no surprise that most Italians are Roman Catholics, so it is no surprise that most British philosophers are in the analytic tradition, while most French philosophers are not. All these things happen to some degree in science, no doubt, but are there to an extent regulated by the demand for repeatable observation and experiment. If the scandal of unresolved philosophical dispute is in the nature of the case unsurprising, it is also unsurprising that philosophy should be so deeply involved with its own history. As in other branches of historio­ graphy, much historiography of philosophy has consisted in the elabora­ tion and confirmation of myths which serve to bolster current doctrine, often by bestowing on it the authority of a suitably heroic figure or, on the other hand, by presenting it as the modern way of avoiding the appalling confusions and absurdities of the past. A mythical Hume was the hero of Logical Positivism, as a mythical doctrine of substance was the object of its ridicule. More reputably, perhaps the only discipline or method by which we can expect to improve our capacity to stand above the non­ rational determinants of our philosophical beliefs and concerns is the exploration, with as much objectivity as we can muster, of the origins and antecedents of those beliefs and concerns. We need to know how we got where we are. The deeper the presupposition, the more valuable to our philosophical health is the activity of discovering how it was acquired, and of considering as sympathetically and open-mindedly as we can times when presuppositions were different. Like other historians, we should seek to base our view of the rights and wrongs of old disputes on primary sources , rather than rely for our knowledge of them on the reports and traditions of the victors. If by so doing we can appreciate the reasonableness to its proponents of a view which has traditionally been mis-represented as prejudice or muddle, then our conception of the philo­ sophical possibilities, as well as our understanding of our own historical position, will have been broadened. Not that we can expect simply to resurrect past doctrine, integrated as even attractive doctrine tends almost universally to be with what is flatly unacceptable. But we may be prompted to see analogous lines of thought which point away from current dogma. Perhaps a rather strained and inadequate comparison may be drawn with some long-standing engineering or other practical problem to which no entirely satisfactory solution has yet been found. As more approaches are explored, each with its advantages and disadvantages , there is certainly progress of a kind, if only in our appreciation of the difficulties,

THE HISTORIOGRAPHY OF PHILOSOPHY

29

our perception of the problem. Moreover it is not out of place to bear in mind earlier, even (as it seems to us) primitive or palpably inefficient approaches to the same sort of problem. For by examining carefully, without contempt or illusion, the technology of other times - rope bridges, steam ploughs or airships - we may be led to some analogous possibility, as yet undeveloped, through the discipline of seeing things from a different point of view. If that happens, we do not have to pretend that the new idea is more than analogous with the old one. The idea of the end of philosophy as hitherto practised is inimical to such an optimistic account of the relationship between philosophy and its historiography. It is nevertheless an idea which, whatever form it takes, demands and promotes a certain sort of view of the history of philosophy. It is therefore an idea in which the historian of philosophy can feel a double interest. First, what is its provenance? Second, is any of the views of the history of philosophy which it promotes a credible one? Jonathan Ree begins his sketch of a part of an answer to the first of these questions in the mists of antiquity, but I should like to pick it up briefly at a l ater and, as it seems to me, a particularly significant point, before moving on to more recent developments. Seventeenth-century doctrines of method promoted a certain broad approach to the theories of opponents, epitomised by the Cartesians' treatment of error. If the adoption of proper rules allows the rational mind to make its own way to the truth, then rival doctrines must owe their existence to the distraction of men's j udgments by other forces than reason, i.e. by the senses, passions and imagination, by authority or by the web of words. Berkeley turns this general approacp to error against the New Philosophy itself. His recommended method is to strip away words and attend to naked ideas. By this means he promises to cut through the confusions, disputes and scepticism which arise from taking verbal distinctions for real ones - the sin of abstractionism. Not only philosophers are guilty of erroneous metaphysics, however, for (despite an occasional exaggerated expression of agreement with the opinions of the mob) Berkeley recognised a sort of 'prejudice of the senses' in our ordinary assumption that the things we immediately perceive are independent of the mind. Specifically philosophical error arises, he suggests, as an ill­ conceived response to the tension between vulgar belief 'lnd the perception by philosophers that vulgar belief will not do. This tension gives rise, by a dialectical process, to central doctrines of the New Philosophy, i.e. the distinction between qualities and ideas, the distinction between primary

30

M. R. AYERS

and secondary qualities and the notion of material substance. The true philosophy combines the elements of truth in the vulgar and philosophical beliefs, eliminating the error in each. The dialectical structure of B erkeley's treatment of the relationship between error and truth in vulgar and philosophical views seems to be a fresh departure. Even Malebranche's elaborate diagnosis of the causes of vulgar belief and of Aristotelian doctrine is, by comparison, a tissue of ad hoc insult. Hume in turn, although doubtless influenced by Malebranche's employment of the imagination in the explanation of philosophical error, is closer to Berkeley both in his targets and in the way he relates the three opinions� 'of the vulgar, . . . ofa false philosophy, and . . . of the true'. As with Berkeley, false philosophy arises from a sort of recognition of the inade­ quacy of vulgar belief, or rather from the tension caused by that partial recognition. The argument differs from Berkeley's in important respects, generally in a motivated, even carefully explicit way. But Berkeley's targets are all there : the belief that the immediate objects of perception are independent, the distinction between dependent ideas and independent qualities which resemble them, the notion of an underlying material substance, and the distinction between primary and s econdary qualities. 1 An important difference between Hume and Berkeley, however, is perhaps what makes Hume the first 'end-of-philosophy' philosopher in something like the modern mode. For Berkeley the true philosophy is a dogmatic metaphysics on the same level as the doctrine he rejects. But for Hume the true philosophy consists, on the one hand in scepticism (which is admittedly ancient enough), but on the other in a psychological theory of the imagination which is on a different level from the doctrines under criticism. On the one hand metaphysical questions are incapable of resolu­ tion, on the other hand an epistemology couched in the form of a natural history of belief and the imagination first accounts for the vulgar view and then goes on to explain the development of dogmatic metaphysics, ancient and modern. The explanation of error has in effect become the substantive theory on offer. A framework remarkably like Hume's reappears in the Critique of Pure Reason. There is, it is true, a great difference of principle in that Kant's theory of the imagination is a priori and is also a theory of human reason. Transcendental psychology is anything but natural history, and Kant could never have accepted Hume's suggestion that metaphysical theories are on a level with the everyday imaginative oonstructs of the morbidly fearful. There is, it is true, an interesting analogy with Hume in that, when

THE HISTORIOGRAPHY OF PHILOSOPHY

31

'Kantian' doctrines have survived i n the analytic tradition, they have generally done so in a strictly de-psychologised form. But de­ psychologising Kant is a rather different, perhaps less painfully radical business than de-psychologising Hume. One might say that the very point of Hume's psychological explanations is that they are strung together in a way which is more than a little accidental. They tell a story, but not one with the kind of coherence and inevitability of Kant's, for whom the generation of philosophical error is deeply characteristic of human rationality. No doubt that is why Kant's conception of the central opposi­ tions of philosophy remains interesting. Where Hume is condemned to parody, since for him rationally argued positions have to be presented as mere constructions of the imagination, Kant is interested in argument, and has arrived at his position by serious reflection on argument which really took place. Yet Kant's map of the central issues of philosophy is too neat by half. An a priori element is being fed into the history of philosophy, even if it derives from a philosophical theory which is itself the product of informed reflection on the history of philosophy. Individuals are squeezed into clothes which do not fit. Locke, for example, becomes an Empiricist, but a deplorably inconsistent one (a characterisation he has only recently begun to live down.) The Kantian conception of a fixed set of philosophical possibilities (among which, dialectically related to the others, but different in kind, lies the true critical path) is open to more objections than one. First, the trite thought that there is no progress in traditional philosophy can be met with an opposite intuition. There had, after all, been irreversible change during the seventeenth and eighteenth centuries, and more has occurred since then. Moreover the very a priori comprehensiveness of Kant has given his doctrine a vulnerability to the passage of time matching that of any dogmatic system. Non-Euclidean geometry, relativity and quantum mechanics each seems to disprove a central Kantian contention as to how we must think of an objective world. Consequently it is not surprising that neo-idealist theory has taken mutability very much to its heart. Within the analytic tradition, neo-idealism tends to see itself as concerned, not with the transcendental structure of the mind, but with 'our conceptual scheme' (or some equivalent) primarily conceived of as a mutable linguistic structure, with the pressure for change chiefly coming from science. On such a view nothing about the way we think is sacred, or truly necessary. The appearance of necessity is sometimes explained by the simile of sailors having to keep the conceptual ship afloat while engaged in rebuilding it.

32

M. R. AYERS

Because not too much can be changed at once, some things seem to be immutable by a kind of illusion. This gradualist model has come under strain as an alternative model of conceptual revolution has found favour, in particular among historians of science. But what is common to most present-day neo-idealist or conceptualist theories is the rejection of immutability, and so of any Kantian conception of synthetic a priori knowledge. With that, of course, goes the rejection of any fixed map of philosophical error. If nothing is fixed a priori, what is there left for philosophy to do? For philosophical argument is nothing if not a priori. Quine has given one popular answer which also draws on a different tradition. For of course not all 'end of philosophy' doctrines are what I have called 'neo-idealist' or 'conceptualist'. The doctrine of Language, Truth and Logic, one of the most famous of end-of-philosophy books, owes more to Russell and Hume (or to Russell's de-psychologised Hume) than it does to Kant, despite moments of flirtation with pragmatism. Ayer's distinction between 'meta­ physics' and the analysis of meanings is a distinction between bad philo­ sophy and good. The former, on his view, in endeavouring to speak of a reality beyond experience fails even to be meaningful, but the latter falls within the limits of possible knowledge. Language may change but mean­ ings are timeless, and in any case all meanings must be bottomed on 'sense-contents', the given foundations of empirical knowledge. Quine, on the other hand, rejects foundations and even the notion of analytic truth. What he retains is the notion of analytic semantics in another guise. The role of the good philosopher is complementary to that of the natural scientist. Both are concerned to make improvements to our 'web of belief' or total 'theory' in respect of its usefulness for prediction and explanation, but the good philosopher (as it seems to be Quine's view) does so with eyes focussed on the economy and determinateness of its logical structure. For this his best available tool is canonical Russellian logic. Bad philosophy is bad because it is not conducive to this purpose (not to speak of its commonly wrong understanding of ontology). Other neo-idealists, however, most notably Wittgenstein, are opposed to any conception of philosophy which assimilates it to science. Wittgenstein seems to have believed that philosophical doctrine and argument can embody a sort of understanding, as well as misunderstanding, of language­ involving practice. Nevertheless he held that philosophy always involves illusion, not simply in not perceiving itself as a function of socially­ conditioned 'grammar' (for that, it might seem, could be rectified), but in

THE H I S TO RIOGRAPHY OF P H I LO S O PH Y

33

its assumption that by a priori argument it might arrive at settled con­ clusions concerning some determinate abstract subject-matter. Language, one gathers from his discussions, is inherently and indefinitely flexible. Nothing is so fixed as to be available for capture by philosophy - or indeed for improvement, since the ideal of precision and economy is as much an illusion as the ideal of faithful correspondence to reality. The true critical path, one is left to suppose, is somehow to show whatever philosophy has been able to show, but in such a way as to avoid and reveal philosophy's characteristic illusions. It will avoid all doctrines and even proposals. In this way it will be radically unlike traditional philosophy, being only 'one of its heirs'. Wittgenstein's conception of philosophical thought suggests a view of the history of philosophy which is in its way as static as Kant's. It is made to seem accidental that philosophers have been impelled towards theory now by this, now by that aspect of linguistic practice. If practice changes, if new language-games develop, it is not philosophers who bring about the change. If they try, then language idles. Another view, however, offers to account for the appearance of progress in the history of philosophy - and the seeming relevance of some philosophical argument to undoubted advances in knowledge - by relating it to a more general conceptual history. For some have thought that philosophical doctrines and j ustifica­ tions constitute, or can constitute, a dialectic which accompanies, rationa­ lises, presages or even initiates significant changes in our ways of thought (although it is also capable of holding them up). Metaphysics and epistemology have lent authority to new ideas and practices. It is simply in so far as participants in the philosophical debate have imagined that they are establishing true, j ustified principles in place of false, unjustified ones that (on this account) they have suffered from the characteristic illusion of philosophy. Something like this view appears in Richard Rorty's recent monograph,2 which is a kind of eclectic compendium of neo-idealist thought. Rorty's eclecticism sometimes disguises the fact that the scandal of unresolved disagreement infects neo-idealist theory itself. But he is at any rate not one of those who sees the traditional illusion as perennially attractive, to be perennially dissipated. He rather looks forward to a future in which philo­ sophy is replaced by something which helps and cannot hinder our openness to conceptual variety and change, something purely 'edifying'. The present task is to destroy for ever the illusion definitive of traditional philosophy. It is no accident that his book, like the Critique ofPure Reason,

34

M. R. AYERS

ends with a programme for the historian, "a sort of prolegomenon to a history of epistemology-centred philosophy as an episode in the history of European culture" (op. cit. , p. 390). It will be a history of the time when people believed that there are absolute foundations to knowledge, that unconditioned justification is possible and describable, and when people saw "social practices ofj ustification as more than just such practices". Not just the direction of philosophy at any one time is accidental, but its existence as a whole. Now all this is a very different programme from Kant's, but it may raise rather similar suspicions in our minds. Not least we may ask how such an a priori prolegomenon to any history is possible at all? How can it be thought proper to admonish the historian to set out with the presumption of "the brute factuality of historical origins", and with the predetermined purpose of introducing "into the very roots of thought, the notions of change, discontinuity and materiality" (op. cit. , p. 39 1 )? The answer is, because Rorty's version of neo-idealism requires that we do so. His programme will only appeal to those who have allowed themselves to be led through certain philosophical arguments. It should not be thought that the stress on brute fact and accident gets us one step away from a priori history of philosophy. An anti-foundationist, conceptualist epistemology is as much the basis of Rorty's treatment of 'bad' philosophy as Trans­ cendental Idealism is the basis of Kant's. In illustration of this rather obvious point, consider Rorty's central thesis that Descartes introduced a representative theory of thought which made a sharp break with scholastic conceptions. This chance Cartesian brainstorm was supposed to have given rise to the possibility of founda­ tionist epistemology and so to a new sort of scepticism. There exist certain objections to such a simple view. One of them, voiced in a recent argument of John Yolton's (to the effect, to put it briefly, that Arnauld was not such an heroically original and solitary figure as Thomas Reid and Rorty would have it) is brushed aside rather than answered. If Yolton's argument is correct, we are told, then we shall simply have "to look further along in history for the emergence of what is now thought of as the epistemological problematic created by Descartes" (op. cit. , p. 50). Yet if one thing is not as simple as it was thought, perhaps nothing is as simple as it was thought. Perhaps the whole story of a new problematic and a sharply different conception of epistemology definitive of modern philosophy will collapse on further investigation. The trouble is that, because Rorty already knows in what the illusion of philosophy lies, he already knows what kind of

THE HISTORIOGRAPHY OF PHILOSOPHY

35

history of that illusion must be written, and what will be found if we look hard enough. It is not a ground for complaint that there is something philosophical in Rorty's treatment of past philosophy. That no doubt is inevitable if we are to get beyond the mere recording of sentences. Nor can it be said that his characterisation of Cartesian and post-Cartesian epistemology is without all plausibility. On the contrary, his analysis relies heavily on the authority of widely held, even standard interpretations, for he makes no claims to original scholarship. What causes unease is perhaps that there is so little openness in Rorty's attitude to history, that is to say, to the possibility of learning something significant from it, whether about the past or about philosophical truth. We are offered a programme both in philosophy and in the history of philosophy, not in order that his joint historico­ philosophical thesis about the origin and sources of foundationism and its role in the philosophical tradition should be subjected to serious examina­ tion or test, but just because these programmes are inherent in his 'end-of­ philosophy' thesis itself. It is simply ironical that what an interested public may perceive as a challenging, innovative call for a fresh kind of history of thought is based on such a conservative view of seventeenth-century theories of ideas and knowledge, and on such an orthodox notion as that of the end of philosophy. It may seem that in accusing Rorty of a. lack of openness to the possibility of being taught by past doctrine I am simply assuming what he denies : namely that there is a perennial philosophy or philosophical truth concerned with such abstract topics as knowledge and belief, substance, personal identity and so forth. Yet it is pertinent to reflect how deeply end-of-philosophy doctrines are themselves rooted in the western tradition of philosophical argument, to which indeed they make the currently dominant contribution. Philosophers are generally influenced by nothing so much as other recognisably philosophical writings, and neo-idealism is no exception. The very idea of 'our conceptual scheme' is itself indebted to the theory of ideas. Moreover, whatever brave new world is envisaged, free from all philosophical doctrine and argument in the old style, the new view itself can be established only by arguments which engage with its despised antecedents. There is a recurring weakness in end-of-philosophy theories epitomised in an early criticism of Logical Positivism. The verifica­ tion principle, it was obj ected, is either analytic or synthetic a priori. If it is the latter, then on its own authority it is metaphysical and meaningless. It' it is analytic, on the other hand, then as a definition of meaning it cannot

36

M. R. AYERS

consistently be allowed any more interest or authority than any other arbitrary definition. Quine too can be asked whether, for example, the principle of the indeterminacy of translation is open to revision. Wittgenstein deals somewhat unsatisfactorily with the same general problem when he tells readers of the Tractatus to employ his propositions as a ladder to be kicked away after use. In his later writings the problem seems to be one motive for the style which has gained him the reputation for insinuating his doctrines rather than arguing for them - for presenting doctrine and argument in disguise. For how could someone who claimed the right to conclusive argument for himself possibly refuse it in principle to others when they choose to deny what he asserts? Yet, as so often with end-of-philosophy theories, once the initial charm has gone from the new, non-argumentative way of philosophising we find that the arguments at its foundation not only look entirely in place in the context of the tradition, but rather obviously owe the possibility of their existence to that context. In Rorty's case the attempt to mark a discontinuity takes the form of a particularly narrow conception of 'epistemology', as if epistemology were something which could only appear to have a subject-matter if we believed that there are infallible foundations to knowledge, and as if the dispute between foundationists and anti-foundationists were not itself an issue in epistemology. It is a rather engaging feature of his book, however, that there is no more than perfunctory disguise. There is little serious attempt to present his own argument in the form of the promised new 'edifying' discourse. Again and again we are allowed to see how much of the thesis rests on this or that familiar, and familiarly doubtful philosophical position. A prime example is supplied by the claim, at the heart of Rorty's most general contentions, that Descartes created an incoherent, bastard category of the mental by running together the intentional with the pheno­ menal. This claim depends on the possibility of making sense of the notion of subjective phenomena which are not as such intentional (in the j argon, the notion of 'raw feels' or 'qualia'). Rorty accordingly defends it by a brisk endorsement of the 'obvious' view that physical pains are, on the ordinary conception of them, not intentional : that they are pure, blank 'phenomena'. But such a view, although common enough in the past, is far from obvious. It is far more plausible to suppose that pains can be veridical or falsidical. A veridical pain is a certain sort of awareness of a part of the body, integrated with our awareness of the body's disposition in space. Hence pains necessarily have location, or seeming location, rather' as colours do, although in the case of colours the location is characteristically perceived

THE HISTORIOGRAPHY OF PHILOSOPHY

37

as external. A veridical pain in a foot must be standardly due to some state of that foot (not necessarily, of course, to its being damaged). A pain in a non-existent foot (as Descartes reminds us) is falsidical. Let us suppose, however, that Rorty is not just wrong on this particular point - even, for the sake of argument, that he is right. Can we accept that the view of pain opposed to his is nothing but a monumental and accidental muddle? We might as well be accepting Malebranche's or Berkeley's or Hume's insult­ ing explanations of philosophical error as this extraordinary proposal. Calling it 'deconstruction' doesn't make it any better. Let us take another, related example. In opposition to foundationism Rorty rejects the allegedly pernicious notion of the 'naturally given', dissociating it from the acceptable notion of 'direct knowledge'. The latter, he says, is simply knowledge which is had without its possessor having gone through any conscious inference. But there is no suggestion that some entities are especially well suited to be known in this way. What we know non-inferentially is a matter of what we happen to be familiar with. Some people (those who sit in front of cloud-chambers) are familiar with, and make non-inferential reports of, elementary particles. Others are familiar with diseases of trees, and can report 'another case of Dutch elm disease' without performing any inference. (op. cit. , p. 1 06)

Rorty is gunning for the category of 'raw feels' as candidate foundations of knowledge, but what he says has implications for the boundaries of a rather less problematic category, the category of perceptual knowledge, or of what we know to be so because we literally see, feel, hear, smell or taste it to be so. Intuitively there is a distinction between our capacity to see that a table is oval and brown, and the capacity of some to 'see' that it was made about 1800 (neither capacity, of course, being infallible). The question as to whether there is conscious inference is quite inadequate to capture this distinction. On the one hand, someone who sees that an object is red might still for some reason go through the inference, 'It looks red and the light is normal, so it is red,' before he believes his eyes. The occurrence of his bit of reasoning does not prevent his knowledge from being immediate perceptual knowledge, since the reasoning serves simply to set aside a certain doubt as to the veridicality of the sensations which present the object as red. On the other hand, that an antique-dealer has no need to go through a bit of conscious reasoning does not mean that his sight is so much better than ours that he can literally see when something is a period piece. The relational attribute of being made before 1830 is in principle not the kind of attribute which could be visually presented, and much the same

38

M. R. AYERS

goes for Rorty's own examples. It may sometimes be true that familiarity determines how we perceive things : e.g. familiarity with cubes may be a condition of our seeing pictures of cubes three-dimensionally. Yet that is a question of empirical psy­ chology. It is moreover a question which presupposes the kind of subjec­ tive, phenomenal differences that exist between seeing such a picture now three-dimensionally and now as flat, now as j utting out of the paper and now as sinking into it - differences which have nothing to do with inference and everything to do with the content of sensation, or what is 'given' in perception. Otherwise it is simply false that it is our theories or 'concepts' which determine what we perceive. Someone who, with ignorant surprise, sees the northern lights for the first time is likely to see just what his companion sees, who is an expert on the phenomenon. Perhaps he sees more, since familiarity, as they say, can breed contempt. The general point relates as much to functionalist theory of mind as to holistic neo-idealism, since (as may be significant for their conjunction in Rorty) both face the same difficulty in tracing the boundary of sensory knowledge. For func­ tionalist theory, the intentional content of sensation has to be determined derivatively from the intentional content of the beliefs (and, ultimately, patterns of performance) to which it gives rise. The truth is rather the reverse. Roughly, perceptual beliefs are distinct from other beliefs in being those beliefs which derive their content from the intentional content of the sensations which cause them. They constitute perceptual knowledge when those sensations are non-accidentally veridical. It is therefore quite inade­ quate to define them merely negatively, as beliefs not based on conscious inference. If this excursus into the philosophy of perception is on the right lines (or even reasonable), we can draw conclusions for Rorty's historiographical programme. The philosophical part of his package possesses, it must be admitted, a considerable coherence: an anti-consciousness, functionalist theory of mind, an anti-foundationist, holistic epistemology, a relativist, pragmatist ontology. It is a part of the package that the whole enterprise should be carried through with that sIngular lack of feeling for the force of opposing argument which is characteristic of end-of-philosophy doctrine (and is reminiscent, ironically enough, of seventeenth-century explanations of error). Yet, as we have seen, the enterprise is less than invulnerable philosophically. How can such a package be an adequate basis for a new kind of historiography of philosophy, the historian's role being simply to draw out the implications of Rorty's philosophical doctrine, to "encapsu-

THE HI STORIOGRAPHY OF PHILOSOPHY

39

late and distance" such unfortunate episodes as the seventeenth century's "awkward and inconsistent use of the Aristotelian notion of 'substance'." (op. cit. , p. 1 25). We should do far better to try to understand the plausi­ bility and force of past theory as it appeared to its proponents themselves, in which case much of the awkwardness and inconsistency is liable to drop away. The call for a new sort of history, like the call for a new sort of 'edifying' philosophy, can safely be treated as a rhetorical concomitant of Rorty's complex philosophical package. It is a part of what he is claiming rather than a consequence of what he has established. Just as the only way to make any kind of critical response to his philosophical claims is to engage in the kind of philosophical thinking he wants us to eschew, so the only way to submit his historical claims to test is to approach past thought as objectively as possible, with methodological sympathy and an open­ minded reluctance to j ump to critical conclusions. Perhaps one of the sources of the attraction of Rorty's view of past philosophy is simply the extreme difference between the philosophical ingredients of his package and seventeenth-century doctrine. How could such different approaches to the same things or questions each have seemed plausible to reasonable people? How could one group of philo­ sophers have laid such emphasis on conscious representation and con­ scious 'grasp' in its treatment of knowledge, while another group defines or accounts for knowledge, even direct perceptual knowledge, without bringing the content of consciousness into it at all? One answer is that the two groups have simply not been concerned with the same thing: there are no perennial objects such as 'knowledge', 'perceptual knowledge', 'con­ sciousness', 'thought' or 'substance' for philosophers to be concerned with. There is a conceptual distance between us and them which it is the chief function of historians of philosophy to help us to perceive. But another answer is that, in considering abstractly what is involved in knowledge, in memory, in sense-perception, in what it is to be a substantial obj ect and so forth, philosophers have been considering real and, in at least some, rather fundamental cases, unchanging issues (although their perception of these issues may indeed change). If there are extreme differences in their theoretical re�ponses at different times, then that may be because each of the responses in question is one-sided or exaggerated. It may, for example, be wrong to give as much work to consciousness as was usual in seventeenth-century epistemology; but equally wrong to deny the re­ levance of consciousness to knowledge altogether. It may be right that knowledge has 'foundations' in that, for example, perceptual belief has an

40

M . R. AYERS

underived authority; but wrong that the foundations of knowledge are indefeasible. It may be wrong to read the natural, inescapable metaphor of 'seeing' a necessary truth as nearly literally as does Locke ; but no less limiting and perverse to reject it as dismissively as Rorty. Again, tradi­ tional doctrines of substance, essence and accident may be beyond re­ demption as they stand, but may yet stimulate us to achieve a more subtle and discriminating understanding of ontology than is embodied in the principle that to be is to be the value of a variable. To attempt to defend or develop these suggestions would be to go beyond the limitations imposed by the present occasion, but it can at least be said that they are consonant with the optimistic view of the relationship between philosophy and its historiography with which this paper began.

Wadham College, Oxford NOTES I Particularly relevant passages are Malebranche, D e fa Recherche de fa Verite l. xiv; Berkeley, Principles of Human Knowledge I. 56, 73; Three Dialogues between Hylas and Philonous III ( Works, edited by A. A. Luce and T. E. Jessop, Volume II, pp. 261-263); Hume, Treatise of Human Nature I. iv. 2-4. See my 'Berkeley and Hume: a question of influence', in Philosophy in History: Essays on the historiography ofphilosophy, edited by R. Rorty, J.B. Schneewind and Q. Skinner (Cambridge: Cambridge University Press, 1984): 303-327. 2 R. Rorty, Philosophy and the Mirror of Nature (Oxford: Blackwell, 1980).

ANTHONY MANSER

T H E END OF M ETAPHYS I C S : A C O M M ENT

Both contributors to this symposium talk of the fact that attacks on 'metaphysics' have been frequent in the history of philosophy, particularly in the modern period. The word has often been used to describe a philosophy of which the writer disapproved. However, this tendency is by no means universal ; that there are those of a different persuasion needs also to be remembered. In a past generation, Bradley and McTaggart could be mentioned; among contemporaries, J. N. Findlay is not unique in starting as a radical and ending by defending metaphysics as the crowning part of philosophy. And this involves him in according respect to his great metaphysical predecessors. They may not have always been right, but at least their work was on the right lines. It is worth noting that such a j udgment is more like that of practising scientists when they look back on their subject's past. The great figures may have been proved wrong, but nevertheless they were great, formed a vital part of that progress of 'science' to which the maker of the judgment hopes to contribute. One view of the nature of philosophy does liken it to science. Although it may be true, as Jonathan Ree asserts, that "much eighteenth century philosophical writing was predicated upon an integral history of philosophy as error" (and it is clear that the twentieth century outdoes the eighteenth in this respect), both he and Ayers argue that the history of philosophy cannot be properly understood on such a basis. It is significant that each closes his paper with an attack on the latest manifestations of the 'end of metaphysics', Ree referring to Derrida, and Ayers demolishing in some detail Richard Rorty's Philosophy and the Mirror of Nature. For both are concerned to rectify a certain view of the history of philosophy, one that sees it as a history of errors, and to replace it with one that is rather more Hegelian ; I use this term purely descriptively here. For the history of philosophy, as a serious academic exercise, was largely an invention of Hegel's. In the introduction to his Lectures on the History ofPhilosophy, he talked of the common view of the subject as the account of a battlefield covered with the bones of the dead; it is a kingdom not merely formed of dead and

41 A . J. Holland (ed. j. Philosophy. Its History and Historiography. 41-46.

42

ANTHONY MANSER

lifeless individuals, but of refuted and spiritually dead systems, since each has killed and buried the other. 1

This is an error, for all works normally included in the canon are philosophy: As to this reflection, the next thing to be said of it is that however different philosophies have been they had a common bond in that they were Philosophy. Thus, whoever may have studied or become acquainted with philosophy, of whatever kind, provided only that it is such, has thereby become acquainted with philosophy. That delusive mode of reasoning which regards diversity alone, and from doubt or aversion to the particular form in which the Universal finds its actuality, will not grasp or even allow this universal nature, I have elsewhere likened to an invalid recommended by the doctor to eat fruit, and who has cherries, plums or grapes before him, but who pedantically refuses to take anything because no part of what is offered is fruit, some of it being cherries, and the rest plums or grapes.2

It is possible to find two views of philosophy exemplified in its history. The first, which could be called the 'progressive' or 'Hegelian' view, is perhaps the commoner. It regards philosophy as a collection of arguments. Any philosopher thinks of his own as superior to those of his predecessors or opponents, otherwise he would not hold them. In a sense he hopes that what he has produced does bring to an end one area of philosophy, that he has said the last word on the topic. However, this hope is tempered by the knowledge that further arguments are possible, that he himself may be brought to admit that his views were erroneous in some respects. The other, or 'apocalyptic', view sees the enterprise of philosophy as being wholly mistaken, and the object of a 'philosophical' work produced in this tradition is to substitute something else for it, to bring about the 'end of metaphysics' (or of philosophy) in a genuine sense, so that people would cease to pursue it. There would still be room for a history of philosophy, but it would be like the history of any other discarded set of ideas, of alchemy or sorcery, for example. If one were to set out to write a history of alchemy it would be important to get the story straight, as Ree and Ayers both stress we need to do for philosophy. For it is clear that much of what is normally believed about the topic is a 'fairy-tale' (cf. Ree, p. 20) or a 'myth'. These may be the result of a commitment on the part of the author to an alternative view; here I am thinking of those once common histories of science which contrasted

THE END OF METAPHYSICS : A COMMENT

43

the foolishness of pre-Galilean work with the rationality of the approach of later scientists. It might be possible to show that the alchemists were not as silly as they had been thought to be, that there were structures of argument or of experimental results which, in the context of their time, j ustified what they then said. All this could be claimed without denying that alchemy was on a wrong track. It would be hard, given what we now know about the natural world, to write a 'progressive' history of alchemy, one which recorded the progress of the study towards the successful synthesis of the philosophers' stone, the universal alkahest or the elixir of life. In a sense it could be argued that the person setting out to write a history of alchemy in the belief that it was, in the last resort, an error would be writing an 'a priort history in the sense that Ayers gives to the phrase on p. 34 of his paper. B ut it is hard to see what else he could do. His belief in the falsity of alchemical doctrines will be founded on arguments, and these will have to engage to some extent with what the alchemists claimed. Hence I do not think that the historian need be unduly worried by this fact. Here I am thinking of a passage in Ayers' paper which runs: ,

Moreover, whatever brave new world is envisaged, the new view can itself be established only by arguments which engage with its despised antecedents. (p. 35)

It is not surprising that an 'end of philosophy' view should be in some respects recognisably like the philosophy it aims to abolish or 'decon­ struct'. To attempt to bring an intellectual pursuit to an end is to take it seriously. The only alternative would be the ignorant dismissal of the enterprise as a waste of time, a move which has been common among non-philosophers in all periods. It would be a parody to re-write a passage from Ayers' paper as follows : It may seem that in accusing the historian of alchemy of a lack of openness to the possibility of being taught by past doctrine I am simply assuming what he denies: namely that there is a perennial alchemy or alchemical truth concerned with such topics as the philosophers' stone, the alkahest, the elixir and so forth. (cf. p. 35)

The reason why it would be parody emerges in the next sentence, which I will quote as written : Yet it is pertinent to reflect how deeply end-of-philosophy doctrines are themselves rooted in the western tradition of philosophical argument, to which indeed they make the currently dominant contribution. (p. 35)

44

ANTHONY MANSER

If alchemy had attained its goals, it would have come to an end; there would have been no need to go on studying it, though it would still have been necessary to apply the techniques that it had discovered in a practical way. The reason why end-of-philosophy doctrines are rooted in the tradi­ tion is that philosophy necessarily questions its right to exist in a manner that is somewhat different from other disciplines. This questioning has been manifested throughout its history. Hence I cannot accept the 'cool' account that Ayers gives of the relation between philosophy and its historiography. I characterise it thus in view of a passage in his conclusion where he suggests that the issues which philosophers have traditionally confronted are real ones, though : If there are extreme differences in their theoretical responses at different times, then that may be because each of the responses in question is one-sided or exaggerated. (p. 39)

For it seems to me that the difference between those who argue for an 'end of philosophy' view and those who, like Ree and Ayers, think it is a matter of getting the history of the subject straight is a radical one. The question of the right of philosophy to exist can only be solved, if indeed it can be solved at all, by contemporary arguments. These may draw on what past philosophers have said, but that they believed there was a legitimate study on which they were engaged cannot constitute evidence of such legitimacy, because it is precisely that which is at issue. Consequently I cannot get very concerned about the question of whether a practitioner of the 'end of philosophy' school has got his history com­ pletely correct. If it is a conception of the nature of the subject which is at issue it may be useful to cite a particular historical example, e.g. Descartes's representative theory of thought, but what is under attack is not Descartes's views, but an account of thinking. Indeed it seems to me that philosophers, as distinct from historians, use names of past figures as devices to refer to theories. Given that there is often contemporary agree­ ment about what constitutes 'Cartesianism' this is a legitimate procedure, for philosophers are engaged in solving their own problems. These may have been first formulated by past philosophers, but the puzzles and worries to which they give rise are, for practising philosophers, objects of · current concern which require a present solution. Should it turn out that Descartes did not hold, or was not the originator of, Cartesianism, we shall have learnt an interesting historical fact. I am not sure it would make much difference to philosophy as I conceive it, or even whether it would involve

THE END OF METAPHYSICS : A COMMENT

45

using a different name to refer to the complex of 'Cartesian' views. The difference between a historical subject and one like philosophy is that the latter re-defines its technical terms constantly, and consequently they may come to mean something different from what they meant at their original introduction. To take an example from a different field, 'atom' now bears a sense for the physicist rather different from what it had for Newton. But there is no reason to suggest that modern physics is making a mistake in still using the word. How could we choose between a conventional and an 'end of philo­ sophy' view? It could be on the basis of arguments and these might well be of the kind that Rorty has adopted. Such arguments might echo those of Kant in certain respects, for he was also concerned with the legitimacy of the subject. There is also another method, that based on an examination of the motives or reasons for philosophising, and this seems more popular with the French at present, though I suppose Nietzsche and Marx were its main originators. Clearly there is not time, in a chairman's remarks, to deal with their arguments. However, in a symposium on this topic the question of the right of philosophy to exist as an academic discipline must be borne in mind. Indeed, this is the question that those who talk of the 'end of philosophy' raise in an acute form. The fact that the majority of those present, including myself, earn their living from teaching philosophy may make such a consideration more difficult to undertake, but surely philosophers ought to examine their speciality with greater care than should practitioners of other disciplines. It is interesting to note in this connexion that the great expansion of philosophy teaching in England and Wales, which took place after 1945, was largely predicated on a view of the subject which assumed that meta­ physics had been brought to an end or decisively refuted. The task of the philosopher was that of an under-labourer, clearing away confusions in other disciplines, not establishing views of his own. It was a belief in this therapeutic role which led to the establishment of large philosophy depart­ ments in the new universities, and their role was generally to give some teaching to the majority of students, possibly without any course in single honours philosophy. When the history of post-war British philosophy comes to be written, the collapse of this experiment will have to be documented. It would be perhaps over-cynical to suggest that it was the need to find another role which led to changes in the view of the nature of philosophy, turning it from an ancillary subject into a substantive one. In the context of teaching, the tendency to privilege single honours

46

ANTHONY M A NS ER

philosophy over cc;>mbinations is an example of this. The suggestion sounds cynical, but it represents a challenge which needs to be met, and a symposium with this title is perhaps one place where it should be considered.

University of Southampton NOTES I

Hegel, Lectures o n the History ofPhilosophy, translated b y E . S . Haldane (London: Routledge, 1892), Volume 1 , p. 1 7. 2 0p. cit. , p. 18.

JONATHAN R�E

REPLY TO AYERS A N D M A N S E R

I have just received a letter from a French colleague, expressing surprise about this conference. "II y a done, meme au pays de Russell and Moore, un tel interet pour 'l'histoire de la philosophie' qu'une conference puisse se tenir pendant trois jours . . . ?" he writes. Interest in the history of philoso­ phy is indeed unusual amongst analytical philosophers ; and it is worth reflecting on its growth in recent years. For one thing, there has been an accelerating movement in favour of discarding stereotyped caricatures of the old philosophers encased in the school-uniforms of 'rationalism', 'empiricism' and so on. Recent scholar­ ship has demonstrated that, when approached with a little sympathy and respect, the dead philosophers can turn out to have written subtle, imaginative, and still-interesting works about deep and lasting problems. This kind of scholarship is an act of reparation both in the sense of repairing the damaged classics, and in the sense of making mournful amends for the shocking abuse they have suffered in the past. It is to this task that Michael Ayers's argument about Rorty's cavalier use of the history of philosophy is principally addressed. The point which I particularly wanted to draw to the attention of this symposium, however, is rather different. It is that when we think of ourselves as doing specifically philosophical work, we are inevitably draw­ ing on a conception of the discipline which is determined, at least in part, by inherited versions of 'the history of philosophy'. We should hesitate, therefore, before dismissing philosophy's integral histories as marginal, for it may be that, as Wallace Stevens wrote, "what is central to philosophy is its least valuable part". ! My suggestion has been that philosophy has adopted a succession of 'integral histories' each, in a specific way, defining philosophy's past as error. It has consequently acquired a self-image as the search for a wisdom which is peculiarly intimate with folly: as a cure for error, but also a source of it. Hence, I suggested, the modern philosopher's frequent recourse to the idea of having to put an end to metaphysics. And I think it has been a mark of great philosophical writers from Hume and Kant onwards, to sustain the theme of the end of metaphysics with a precarious reflexive irony, 47 A. J. HoI/and (ed.). Philosophy. Its History and Historiography. 47-48.

48

JONATHAN R�E

without tumbling into crass and self-serving denigrations of the philosophy of every period except the present. These integral histories of philosophy, I argued, function as definitions rather than chronicles, and so are not open to immediate confirmation or refutation by comparison with the historical record. And even where they do demonstrably distort the past, they may as well be retained so long as - as Kant remarks about Ideas - we can avoid being deceived by the illusion. 2 This is why I have made use of Wallace Stevens's concept of a 'supreme fiction' in order to describe the formative role of integral histories in philosophy. It is important to recall that those who denounce past philosophy as 'metaphysical' may nevertheless designate their own intellectual goals as 'metaphysical' in another, better, sense. Kant, for example, attacked meta­ physics with a view to establishing the metaphysics of the future. And so did Hegel - who, I must admit, was the repressed topic of my essay. Hegel, on my account, appears as another 'ender of metaphysics', and not, as Anthony Manser reads him, as an opponent of the doctrine. It is true that Hegel held that every phase in the history of philosophy had to be traversed if philosophy was to become, at last, a science. But he did not conclude that the development of philosophy displayed a cumulative progress like that of the positive sciences, and neither, I think, did the British Hegelians, such as Bradley, McTaggart and Findlay. They all agreed that philosophy's past was error, but they held that each of its stages had to be grasped in all its plausible erroneousness : philosophic science - absolute knowledge - was simply the systematic understanding of the mistakes of the history of philosophy. So I would include Hegelianism in my reading of modern philosophy as centred on the theme of an endless 'end of metaphysics'; in fact I would give it pride of place. Success in such a readjustment of the canon - and there are plenty of signs that it is under way - would indeed be a kind of revolution, 'au pays de Russell and Moore'.

Middlesex Polytechnic NOTES I Wallace Stevens, 'The Figure ofthe Youth as Virile Poet' ( 1943) in The Necessary Angel (New York: Knopf, 195 1 ). 2 Critique of Pure Reason, pp. 533-534.

MARY HESSE

E P I S T E M O LO GY W I T H O UT F O UN D ATI O N S

1 . H I STO R I C I S ING E P I S TEMOLOGY

A quick history of philosophy from the 1 7th to the 20th century might start with the 'two books' metaphor: the book of nature and the book of God's Word. God's Word is received by authoritative revelation ; nature by man's 'glassy essence' or what Rorty calls the mind's mirror. Just as the doctrines of revelation and inspiration explain how we know God, so the philosophy of knowledge explains how we know nature. The doctrine of revelation has long succumbed to historical analysis, hermeneutics and cultural relativity, and philosophical grounds for ethics have similarly dissolved into the study of social norms. But there is still in Western consciousness a nostalgia for grounds for the only type of non-analytic knowledge now recognised as such, namely scientific knowledge and the natural phenomenological models upon which it is built. This is the context within which Rorty has launched his powerful attack upon the notion of foundations of natural knowledge in Philosophy and the Mirror ofNature. I He makes there a sharp distinction between philosophy as a search for universal and perennial foundations, and philosophy as merely local and edificationary - the 'conversation of the West'. I shall argue that such a sharp distinction, and such a global dismissal of foundations, depends on a certain style of philosophy which Rorty maintains despite his iconoclasm in other respects. This is roughly the analytic style for which logic and 'argument' are the backbone of philo­ sophy, rather than the more metaphorical and rhetorical style of the hermeneutic philosophers. Against this I argue that, even if Rorty's rejection of foundations is broadly correct, there remain local tasks of philosophy, and particularly successor subjects to 'epistemology', which can still be addressed in a non-foundational way. Moreover, the revival of such local epistemological issues leads to the recognition that real occasions of theory change, such as those in science, already imply a loosening of the criteria of 'argument'. Read in this light, hermeneutic philosophy escapes much of the critique that Rorty brings against it in his last chapter. 49 A. J. Holland (ed.), Philosophy, Its History and Historiography, 49-68. © 1985 b ' D. Reidel Publishin Company.

50

MARY H E S S E

In addition to his general thesis that there are no perennial foundations ofknowledge , Rorty has a second thesis regarding reduction. He holds that there are no grounds for fear of reduction of the 'human' to the 'mechanical', nor any need to defend the uniqueness of 'persons', 'mind', or 'spirit' by means of any sort of dualism of mind/matter or Geistes/Natur­ wissenschaften. According to Rorty, science may, and indeed is likely to, show eventually that causal mechanisms pervade all of human behaviour (p. 354), but this potential reducibility has nothing to do with such moral questions as whether to treat persons, or for that matter dogs, cats, fleas and foxes, as persons, which are questions for social decision and social norms. The connection between these two theses is that if there are no philo­ sophical foundations of knowledge, then there are no philosophical defences of an 'essence' of mankind which tell us that human beings are other than natural science tells us that they are. Apart from that tenuous connection, however, the two theses seem to be independent of each other. It would be possible to deny the general thesis, to hold, for example, that there are a priori underpinnings of natural science, but to accept reduci­ bility; and it would be possible to accept the general thesis but to hold that theological or ideological reasons of a non-philosophical sort require persons to be irreducible to causal mechanisms. The connection between the theses in Rorty's argument seems to depend on historical contin­ gencies : it has as a matter of fact been the case that the search for foundations and for human uniqueness have gone hand in hand since the demise of any comprehensive theological framework in the run-up to the Enlightenment. I do not have any fundamental disagreement with either thesis if it is philosophical dualisms that are in question. I accept the general thrust of Rorty's argument that philosophy has, at least since its professionalisation in the nineteenth century, had an overblown image of itself as the arbiter of knowledge in all domains of human concern, and as the guardian of a 'rationality' which has been supposed independent of time, place and circumstance. (This overblown image is not unconnected with one of Rorty's less attractive, but apparently approving, descriptions of the profession of philosophy: "relatively leisured intellectuals, inhabiting a stable and prosperous part of the world" (p. 359, cf. p. 389)). However, the course of the argument seems to me to have led Rorty into a number of dilemmas, largely because he has not fully faced the radical consequences of his historicism.

E P I STEMOLOGY WITHOUT FOUNDATIONS

51

Rorty professes at the outset his allegiance to the philosophical styles of Dewey, Heidegger and Wittgenstein, and there are several respects in which he is a faithful disciple of not only Wittgenstein but also Kuhn. First, he presents 'normal discourse' as the subject matter and framework of 'professional philosophy', just as Wittgenstein presents his forms of life, and Kuhn his scientific paradigms. Secondly, questions about the true, the real, the rational and the good arise only within social schemes and cannot be adjudicated between them on any perennial or autonomous basis. Third, normal discourses are presented as though they are spontaneously generated in history, and are homogeneous and free of change, conflict and scepticism. In other words, forms of life are not socially situated, nor accounted for in terms of either rational or social constraints. These three points are precisely those that have been found most inadequate in both Wittgenstein's account (specially as applied to alien belief systems by Winch) and in Kuhn's revolutionary account of the history of science. It has been objected that there is no society, particularly no modern society, in which social norms are so stable as not to be subject to critique, to internal uncertainty and debate (however forcibly suppressed), and ulti­ mately to the eruption of change in which some intellectual continuity can be found with what went before. In particular, and in the light of more detailed study of scientific change, Kuhn's paradigms have been found to be more like an indefinite sequence of Russian dolls : removal of one relatively superficial theory and its replacement by another takes place on the foundation of another deeper paradigm temporarily agreed between the protagonists, and so on. There is never in the history of science total discontinuity, or absence of debate between paradigms in terms of some canons of relative rationality temporarily agreed between them. Ifwe take this relation between scientific paradigms as a model, then the history of philosophy as Rorty recounts it can be seen in different perspec­ tive. Locke becomes, not so much the pursuer of an illusory perennial basis for science in terms of ideas in the mind, but rather the analyst of corpuscularian science who attempts to extrapolate this theory to a science of mind. Kant's philosophy becomes, not so much the discovery of transcendental conditions of knowledge, as the search for necessary pre­ suppositions of a particular acceptable form of knowledge initiated by Newton. Dilthey becomes, not so much the justifier of the Geistes­ wissenschaften, as the analyst of what was already acceptable as a form of knowledge distinct from physics, namely the discipline of history. There is no one 'right' way to interpret the nature and aims of philosophy, but it is

52

MARY H E S S E

at least a s plausible t o read its history i n terms o f relative and local reflections on the knowledge-industries of its time, as to equate it, as Rorty does, with the search for perennial foundations. Even if there is a case for holding that the actors, or some of them, saw it this way, this should not be regarded as determinative of a historical approach, and does indeed in itself constitute a problem for historical explanation. "Relative and local reflection on the knowledge-industry of its time" may be j ust what Rorty means by the "hermeneutic" task of philosophy as the continuation of the "conversation of mankind" (pp. 264, 389). But if this is so, then there is no j ustification for his brusque dismissal of the practice of epistemology as if none of it has had this hermeneutic element in the past, nor can he neglect the fact that it is the problems and techniques traditionally belonging to philosophy that are required in the mediating hermeneutic exercise. There is one place where Rorty does accept that philosophers have a useful function "made possible by their familiarity with the historical background of arguments on similar topics, and, most importantly, by the fact that arguments on such topics are punctuated by stale philosophical cliches which the other participants have stumbled across in their reading, but aboilt which professional philosophers know the pros and cons by heart" (p. 393). If this be so, however, it is likely that interesting local versions of even epistemological problems will persist and get some intellectual grip in the current 'conversation', even sometimes proving to be crucial elements in cultural change, dispute and conflict. My main task in this paper will be to try to relativise and historicise Rorty further than perhaps he wants to go, by replaying some of his themes in a context which is purely local, thus hoping to restore a function to some recognisable critical discipline, whether or not it be called philosophy or epistemology. 2. A R E T H E R E ALTERNATIV E CONCEPTUAL S C H EM E S ?

In Chapter VI Rorty takes the realism/idealism issue in modern philo­ sophy of language to be a latter-day attempt to restore the epistemology of foundations. For the realist the crucial question is "how language hooks onto the world". Once that is established in a theory of reference we have in principle the foundation of knowledge. On the other hand, in this re-run of the realism/idealism debate, idealists maintain that different languages embody radically different and undecidable conceptual schemes. Rorty's­ strategy is first to take Putnam's arguments for realism and refute them.

EPISTEMOLOGY WITHOUT FOUNDATIONS

53

Then instead of regarding this refutation as an argument for idealism, he moves to the Davidsonian argument that the very idea of alternative conceptual schemes is an empty one, and hence that there is no philo­ sophical issue between realism and idealism and no hope of restoring the foundations of knowledge. Let us take these two moves in turn. 1. Rorty does not deny that there are scientific issues about what are ,"our best notions of what there is in the world" (p. 276) - these are the notions that make us "better able to cope" (p. 269) in whatever ways our society assumes we need to cope. Rorty is a naturalist because our culture is generally speaking naturalist: '''our conceptual scheme' is . . . simply a reference to what we believe now - the collection of views which make up our present-day culture", and "our present views about nature are our only guide in talking about the relation between nature and our words" (p. 276). The notion of " better able to cope" is not developed in detail, except in Rorty's discussion of the relative claims to 'rationality' of Galileo and Bellarmine in their celebrated dispute about the Copernican system. Here Rorty points out that there is a value choice underlying the issue between "getting the heavens right" and "preserving the Church, and the general cultural structure of Europe" (p. 329), in other words different kinds of 'rationality' were operative, between which no 'foundational' rationality can decide. Natural science is the choice to seek "accurate, consistent, simple, and perhaps even fruitful" theories about nature for practical purposes. That seems to be the pragmatic content of "better able to cope" for Rorty, and it is the characterisation of science on which he relies in his replies to Putnam's arguments for scientific realism. Before considering these replies, however, we may ask whether this pragmatic definition of science is not itself a species of realism, since it certainly depends on some 'correspondence' of theories with reality, guaranteed by the test-and-correction method of experimental science. Rorty does indeed suggest, against Putnam, that the apparent success of science in converging on the real is nothing more than an "artefact of historiograehy" (p. 282) which writes up history as the story of progressive approximation to what we now believe. But this is not an adequate representation of what the history of science is about. For, at least in comparatively recent times, its pragmatic success has been an element in scientists' own self-understanding, and has led them to structure their theories in accordance with a principle of "correspondence" (in Einstein's and Bohr's sense, not that of the philosophers of truth), namely the prin­ ciple that new theories should correspond to old where the older theories

54

MARY H E S S E

were pragmatically successful. The value j udgment described by Rorty as the social choice to do science a certain way is a judgment found in history, not imposed on it by historians of science. The point against Putnam remains, however, because the progressive success of science can be ascribed to the local correspondences imposed by scientists, and does not require a strong realism of theories in the sense of the philosophers' correspondence theory of truth. In this modest and pragmatic sense, then, science is intrinsically realistic. Why is this modest sense not sufficient for a realist like the earlier Putnam, and why does Rorty think that in refuting Putnam he has disposed of realism as a serious philosophical issue? The answer is that for philoso­ phers like Putnam "realism" carries heavier baggage. There are at least two features of a realistic description of the natural world which go beyond the modest realism of "being able to cope". First it is assumed that the con­ comitant of 'real', namely 'truth' as applied to descriptions, satisfies propo­ sitional logic (p. 28 1 ). B ut, as Rorty rightly points out, the fact that we have the concept of propositional truth does not imply that it is applicable to the world. In a modest realism of "coping" it is perhaps not applicable to the world and does not need to be.2 Second, the full-blooded notion of realistic descriptions entails a notion of natural kinds which would fall with its demise. This seems to be the burden of Rorty's reply to Putnam's argument that we need to block the meta-induction that "no theoretical term ever refers" (p. 284). Here Rorty makes a distinction between 'reference' which hooks language onto the world, and which is currently defended in terms of some sort of causal theory, and, on the other hand, an intentional 'talking about' which depends on systems of beliefs in the head, and which can range over fictions and archaic entities like 'phlogiston' as well as the realities of our science. In the first sense, it is possible that no theoretical terms in our science refer ; they are nevertheless workable because we 'talk about' them in the second sense. The consequences of this distinction for a concept of natural kinds will be clearer if we make a further distinction between the potential reference of universal terms in language, and the loc�l reference of terms in particular instances. It would be absurd to maintain that 'cat' does not refer to an obj ect, or 'the cat is on the mat' to an event in the colloquial sense of 'refer' which presupposes that the object or the event exist. But this sense is neither Rorty's nor Putnam's 'realistic reference', nor is it just Rorty's 'talking about'. It is not just talking about, because it does have some relation to objects and to truth, again in colloquial senses.

EPISTEMOLOGY WITHOUT FOUNDATIONS

55

Nevertheless, it is not 'realistic reference' either. This is because use of the term 'cat' does not necessarily refer to a kind having a definite extension (or even a fuzzy extension) in nature. It is perfectly workable in language if it merely refers to some particular vaguely similar instances which may or may not all continue to reside in the extension of 'cat' in the changing idiom of English, much less to 'define' any such natural extension. In other words, general terms in working languages are family resemblance terms, not kind-terms. Moreover, 'the cat is on the mat' does not refer to the truth-value true (or false), not because it does not have a perfectly definite truth-value in the colloquial sense, under appropriate circumstances, but because there is no guarantee that truth-value connectives, entailments , etc. are applicable t o i t i n ordinary, o r even i n standard, useful and con­ venient English. If I am right that Rorty's arguments against realism entail the rejection of an applicable propositional logic, and hence of any theory of truth, argument, natural kinds and natural laws now on offer, then the conse­ quences for philosophy are likely to be much more far-reaching than he admits. In the first place, there are surely successor questions to the traditional concerns of philosophy of science to be considered. For example, how is it that any concept of 'truth', 'warranted assertability', 'correctness', 'consequence', 'argument' . . . get a grip in science, or indeed in ordinary language, as they undoubtedly do? It matters little whether such questions are called 'philosophical', or whether unreflective scientists feel any need or desire to ask them. They are, nevertheless, in some sense epistemological questions (i.e. to do with the theory of scientific knowl­ edge), though not those of an epistemology which seeks foundations. Secondly, and more important, there are consequences for an extension of the argument about 'truth' to what Rorty regards as the parallel case of 'goodness' and in general to the discussion of whether there might be conceptual schemes other than our naturalistic one. For the standard arguments regarding the applicability of 'truth', 'reason', 'realism', 'cor­ respondence', 'evidence', and the rest to morals or to alien belief systems, have ali presupposed a 'home' language in which these concepts are understood to satisfy truth-functional semantics, and to presuppose literal reference of general terms to 'kinds' in the world. If there are no such truth-values and no such kinds, the standard arguments are at least going to look very different, and it cannot be assumed, as Rorty seems to do, that the debunking of these concepts in regard to natural reality is going to carry away with it all possibility of rational scepticism and conflict in other

56

MARY H E S S E

areas o f social belief, o r remove all tasks o f philosophy with regard to them. 3 2. Rorty claims to have refuted Putnam's erstwhile realism, and Putnam himself has now recanted as a result of his own similar objections. But why does Rorty not take this refutation to be an argument for idealism, that is for the view that there are alternative but forever undecidable conceptual schemes? Here Rorty turns to Davidsonian arguments, to the effect that there is no special epistemological problem about truth, since it must be presupposed that any working language contains the concept of truth, and that most of its factual descriptive sentences must be true. Both Davidson and Rorty have argued that it follows that there are no radically untranslat­ able conceptual schemes, since differences of scheme are 'factored out' by translatability at factual level. Hence an idealism which asserts the existence of untranslatable conceptual schemes is false. Rorty considers the charge that such a view is verificationist (p. 306), and replies by distinguishing a sense of 'true' (the common sense), which is required internally for any factually descriptive language, from a philosophical sense, which is what is sought in a Putnam-type realism ("All our common­ sense 'truths' might turn out not to be True"). He goes on to argue that philosophical doubts about truth in this second sense may indeed arise, but they are not to be settled by verificationism or any other form of founda­ tionalism: they are indeed philosophically empty. They arise in abnormal conditions of scepticism, and in such conditions philosophy has nothing to say that is uniquely rational or foundational : If you ask . . . Davidson why he thinks that we ever talk about what really exists or say anything true about it, [he is] likely to ask you what makes you have doubts on the subject . . . [He] need not invoke verificationist arguments; [he] need simply ask why [he] should worry about the skeptical alternative until [he is] given some concrete ground of doubt (p. 3 1 1).

Rorty also considers the charge of relativism, and replies by simply reiterating that since there are no radically alternative schemes, there is nothing for a scheme to be relative to. Our language is not just relative to our culture, but is intertranslatable with all working cultures. There are, as I see it, two serious difficulties about this line of argument. The first concerns the parallel with moral concepts that Rorty invokes. 'Truth', he holds, derives at the commonsense level from natural percep­ tions, and likewise 'good' derives from moral intuitions. There is, to be sure, another philosophical or transcendental use of 'good' as there is of

EPISTEMOLOGY WITHOUT FOUNDATIONS

57

'true', but this is equally a mirage if thought to be a proper and specific goal of philosophy. But this parallel cannot work unless it is shown that intuitions of the 'good' are as cross-cultural as those of factual truth may be. It is at least plausible to suggest that natural evolution requires the latter but does not require the former, for although human societies can probably not survive without some adequate appreciation of their immediate natural environment, it is not obvious that they are unable to survive under radically different moral regimes. Some relatively advanced species of animals do so; the biological unity of mankind does not seem a sufficient basis to assert that different human societies cannot do so too. It is not this argument I want to pursue, however, since it is only a more extreme version of a second objection to Rorty's dismissal of alternative conceptual schemes. This dismissal does depend on a kind of verifica­ tionism, namely the assumption that it is 'factual truth' that is primary, because it is this that provides the basis of translatability. But the real challenge of the relativity of conceptual schemes is a concern with theoretical truth with the edifices of scientific or symbolic cosmology that all societies build upon the (perhaps contingent) universal factual basis (or, to put it less inductively, the viewpoints through which they interpret the mundane concerns of everyday life where they find we are all brothers under the biological skin). Contrary to much received opinion it is not radical indeterminacy of translation that is the sting of relativism, but the underdetermination of conceptual schemes, both scientific and symbolic. To suppose otherwise is to give the common natural world of biological survival an unwarranted priority over theories and symbolisms, and this form of naturalism is surely a relative and ungrounded presupposition of our culture. Rorty quotes Davidson : -

In giving up the dualism of scheme and world, we do not give up the world, but re-establish unmediated touch with the familiar objects whose antics make our sentences and opinions true or false (p. 3 10).

But we may ask 'What familiar objects?'. Sticks and stones , tables and chairs, stars and galaxies, seedtime and harvest, twins and birds, gods and spirits . . . ? And whose objects? Those of the Nuer, the Lele, the Pytha­ goreans, the alchemists, Locke, Berkeley, or late 20th-century natural man? Verificationism, naturalism and relativism are built into the very way in which "objects and their antics" are made to override such questions in the Davidson-Rorty conceptual scheme.

58

MARY H E S S E

Rorty has , however, shifted Davidson's argument onto a more universal plane, and by doing so has both made it more interesting and exposed its parochial character. Rorty has in effect abandoned the claim that relativism is an empty doctrine, and replaced it by an admitted social relativism. In his account this doctrine is implicitly distinguished from two other sorts of relativism. First it is not individualistic : there is no license for "everyone to construct his own little whole - his own little paradigm, his own little practice, his own little language-game - and then crawl into it" (p. 3 1 7). On the other hand it is not an ultimate philosophical relativism, because where there is no possibility of philosophical foundations there is no philosophical problem of absence of foundations either. What remains is what Rorty calls the 'hermeneutic' function of philosophy, to contribute to transitions from culture to culture, ethos to ethos, scheme to scheme transitions which in themselves owe little to reason and much to the flux of history and practice. 3. H ERMENEUTICS AND THE 'CONVERSATION OF MANKIND'

The discussion so far has identified two sorts of task for a 'successor discipline' to the epistemology of foundations. First, there are internal questions remaining for the philosophy of science, and doubtless for moral philosophy also, which result from the radical consequences of Rorty's critique. I mention here only the rejection of truth-functional semantics and the presupposition of natural kinds that I have argued to be entail­ ments of Rorty's position. These would seem at least to call in question the sort of view expressed by Van Fraassen when he denies that issues about meaning in philosophy of science have implications for received philo­ sophy of language.4 This is a vast subject, however, into which I do not want to enter here. I want rather to pursue a second set of problems concerning the prospects for non-foundational, but nevertheless rational and philosophical, discussion of conceptual change, conflict and critique. These problems presuppose the existence of alternative 'conceptual schemes' in the sense of scientific or symbolic cosmologies which are underdetermined but not necessarily untranslatable at some 'factual' level. Rorty considers these problems as subjects for philosophy-as­ hermeneutics. Let us then examine the claim of hermeneutics to be the framework for philosophy's new task. Rorty characterizes the hermeneutic programme of Habermas and Apel as that of using the concept of 'reflection' to replace traditional foundatio-

EPISTEMOLOGY WITHOUT FOUNDATIONS

59

nalism with a new transcendental epistemology. He interprets Habermas's 'reformulated transcendental philosophy' as a claim that there is a 'general synoptic way' of restoring universal categories for knowledge, although this will no longer be knowledge that 'objectivizes' persons in the manner of traditional realism. Rorty's reply is a restatement of his position that there is no such general analysis, and that in any case it has been demonstrated that such analysis is not necessary for the purpose of refuting traditional realism, or of rescuing the concept of 'person' from empirical obj ectiviza­ tion. I believe that Rorty has not come to grips here with the import of the hermeneutic concept of'reflection'. In Habermas and Apel we may identify three important functions of this concept which I shall now examine. They may be classified as the transcendental, the reintroduction of the subject, and the critical dissolution of relativism. 3. 1 .

Transcendental Reflection

It would require a much longer and deeper analysis to do justice to the subtle relationships between the 'transcendental' and the 'empirical' in Habermas. But there are indications in both Habermas and Apel that neither would want to claim the kind of universality and finality that is associated with a Kantian account of the categories of knowledge. For example, in 'Postscript' Habermas makes it quite clear that it is a "new and transformed transcendental philosophy" that is required, and asks whether a 'transcendental' foundation can "be meaningful at all in view of the fact that despite the attempts to rationally reconstruct competences we cannot assume a transcendental subject . . . apart from the 'empirical', naturally generated and socially formed subjects ?". 5 Habermas goes on to make a distinction between two senses of 'reflection', namely, "reflexion upon the conditions of potential abilities of a knowing, speaking and acting subject as such", and "reflexion upon unconsciously produced constraints to which a determinate subject . . . succumbs in its process of self-formation".6 This second sense is developed in the critical theories of Hegel, Marx and Freud, but in his 'transformed transcendental philosophy' Habermas warts to give an equally important place to the first sense, which he believes can now be detached from its Kantian dependence on the inevitable subjective conditions pertaining to a transcendental subject. The first sense of reflection is illustrated in recent reconstructive studies of cognitive schemes, of which Chomsky's generative grammar and Piaget's reconstruction of early learning are prime examples. Such studies are partly empirical and

60

MARY HESSE

partly normative, and as such Habermas argues that they are to be distinguished from the descriptive natural sciences. It is this, non-Kantian, sense of 'transcendental reflexion' that comes nearest to what Habermas claims to be doing in his 'universal pragmatics'. It is not that all contingent conditions are transcended as such, as for Kant, but that empirically given discourses in given social and cultural circumstances are reconstructed to reveal the contingent normative conditions of their possibility. In similar vein Apel describes the 'transcendental language-game' as a dialectical mediation between transcendental idealism and historical materialism. ? The language-game is an evaluative commitment to the unlimited communication community (cf. Habermas's 'ideal speech com­ munity'), which is constitutive of all argumentative discourse. The activity of self-reflection, which replaces epistemological foundations, does not reinstate transcendental categories, but elicits the presuppositions of the search for truth itself. That men search for truth in argumentative discourse is an empirical and perhaps contingent fact; given that they do so, they thereby constitute certain foundations for the search. Whether one agrees with the account that Habermas and Apel give of these presupposi­ tions in terms of ideal speech, it is clear that their programme is not that of reinstating the Kantian conditions for all possible experience. Perhaps men can live without argument, or even if they cannot, this seems to be a contingent fact of evolution, not a necessary fact. Apel reinforces this conclusion in his discussion of Peirce.8 Like Habermas, he adopts Peirce's model of the ideal limit of the scientific self-corrective methodology as the definition of consensus truth, and extra­ polates it to the 'infinite community of investigators' which is held to constitute truth in relation to norms and values as well as natural facts. But there is no 'realism' or 'correspondence truth' presupposed in this account, and therefore no attempt to reinstate a realist foundation of knowledge. Where Peirce sometimes implies that the 'ideal limit' is a reality in future history which would reconstitute correspondence truth, neither Apel nor Habermas need to speak in this way of an ideal realizable in history. The commitment to ideal consensus which would constitute truth, is a subjec­ tive conditional claim, whose bite consists of the present regulative prin­ ciple which is necessarily entered into if what we mean by rational argu­ ment is adopted. How historically local and contingent the details of that commitment are may be judged from Habermas's references to the European Enlightenment, and Apel's to the survival of the human species qua communicative community. Not to adopt the (relatively) trans-

EPISTEMOLOGY WITHOUT FOUNDATIONS

61

cendental commitment i s t o abandon philosophy, and from the Apel­ Habermas point of view it may be said that Rorty is guilty of precisely that ultimate treason. 9 3.2.

Reintroduction of the Subject

The intention of the Apel-Habermas programme then begins to look more like an analysis of the normal discourse of Enlightenment rationalism - a discourse within which Rorty also situates himself. Further agreement with Rorty's approach can be found in the second use of 'reflection' in the Apel-Habermas programme, namely their reintroduction of the s ubject. It is not sufficient, Apel argues, for positive science to extend its explanations to human intentionality and to thereby suppose it has replaced the human subject within its total picture of the world. It is worth pursuing for a moment the arguments by which Apel reaches this conclusion, because it is important to notice that they are not undermined by Rorty's anti­ mind/body dualism. Apel distinguishes between what he calls the descriptive and the participa­ tive methods of understanding language-games such as that of human intentionality. lO The descriptive method is currently adopted in cultural anthropology, linguistics, and sociology, and, he might have added, post­ Quinean philosophy of l anguage. This approach takes intentionality to be an autonomous phenomenon, not necessarily reducible to neuro­ physiology, and it supposes that intentions, beliefs, desires, etc. can be elicited from behaviour. Apel points out that the process presupposes the existence of intentionality, and does so even in the extended case of ethological studies of animal behaviour such as those of Lorenz. The 'introspective' model of human intention is inescapable. So far, descriptive theorists would not dissent, and many of them even make a virtue of the alleged inescapability from 'commonsense' linguistic usage about human intentionality, arguing that this usage is not to be compared to a falsifiable theory of human behaviour. But Apel goes on to object that the language­ game in which hermeneutic analysis of this kind is carried out (i.e. the descriptivists' own language-game) is not a neutral meta-language in which other language-games are simply recognized and objectively described. Hermeneutic analysis has to make contact with what is to be understood: 'description' of another life-form is also its constitution or reconstruction. Hermeneutic understanding has to mediate between the analyst's game and that of the subject-matter - understanding therefore involves partici­ pation.

62

MARY H E S S E

The argument may perhaps be illustrated more perspicuously by considering the currently popular analysis of 'rationality' in terms of a 'rational agent' model. The ideal rational agent is supposed to be one whose actions are explained as the most efficient way of realizing his desires given his factual beliefs. Given this model, it is possible to observe actions and derive testable hypotheses about unobservable beliefs and desires, thus giving a descriptive theory of intentionality. But the rational agent model itself is part of the philosopher's language-game. It is itself subject to criticism as being inapplicable in various ways to both what is generally meant by 'rationality', and to the way people actually behave. In other words, it is already a reconstruction of behaviour - it is an analytic ideal that needs to make more intimate contact with actual forms of human rationality, that is, it needs to be seen as just one move in the mediating game of hermeneutic participation. Apel has a second argument to the effect that hermeneutic dialogue about human intentions cannot itself be reduced to scientific analysis. I I This argument does not at all depend on any dogmatic dualism or rejection of the possibility of mind/brain reduction. Apel says nothing that would contradict Rorty's own assumption that there are no ghosts - all mind states are brain states. Apel's argument depends rather on the creative, self-generating and non-reproducible character of human communication. But with regard to the second argument there is a difference of emphasis between Apel and Rorty that all but nullifies their agreement. Rorty thinks the self-generating and self-reflective character of human communication is a trivial fact: The sense in which human beings alter themselves by redescribing themselves is no more metaphysically exciting or mysterious than the sense in which they alter themselves by changing their diet, their sexual partners, or their habitation. It is just the same sense: viz., new and more interesting sentences become true of them (p. 3S 1 ).

Rorty agrees that humans exhibit their freedom in such open-ended com­ munication, but his conception of human freedom is severely restricted by his assumption, several times stated, tha� there is complete naturalistic predictability of every physical and mental event : Every speech, thought, �heory, poem, composition and philosophy will turn out to be com­ pletely predictable in purely naturalistic terms. Some atoms-and-the-void account of micro­ processes within individual human beings will permit the prediction of every sound or inscription which will ever be uttered (p. 387).

EPISTEMOLOGY WITHOUT FOUNDATIONS

63

In these passages, and especially those following the second quotation above, Rorty seems to be making three claims which deserve more careful consideration. These are : (I)

All physical manifestations are in principle predictable by purely scientific means.

(2)

This poses no danger to human freedom because there is no possibility in practice of predicting, for example, the content of decisions or inventions now being made. There are two types of argument for this : first, it will never be possible to assemble all the necessary initial conditions for prediction (p. 354), and second, we can only use the terms 'decide', 'invent', etc. when the outcome is now uncertain for us (p. 387).

(3)

Even if we had the complete set of predictive laws and com­ plete initial conditions, this "would not yet be the whole 'objec­ tive truth' about human beings, nor the whole set of true predictions about them" (p. 388). This is because there are many different normal discourses between which there are no rules for settling disputes (this is what Rorty means by 'incom­ mensurable' discourses (p. 3 1 5 ). These discourses are mutually irreducible but not necessarily incompatible with the complete basic 'atoms-and-the-void' description.

Claims ( 1) and (2) are no more than the standard analytic philosopher's position on freedom and determinism, according to which complete physi­ cal determinism is compatible with everything we need to mean by freewill. I do not find this position acceptable, but this is not the place to pursue that issue. 12 In any case I think Rorty unnecessarily overstates his case here, for all he needs for the 'no ghosts' thesis is the assumption that there is complete description in principle of all physical and mental states in naturalistic terms, 1 3 whether the descriptions be determined absolutely or only statistically or not determined at all. Reducibility may be of particu­ lars without being lawlike and therefore predictive. What is more important and interesting is claim (3). If Rorty means here to refer to irreducible but compatible scientific discourses (different but compatible scientific paradigms), then (3) is consistent with ( 1 ) only on the assumption that these discourses are about different and mutually irre-

64

MARY HESSE

ducible aspects of material reality. But in that case, putting all possible such discourses together (all being compatible with the atoms-and-the­ void description) would give the "whole 'objective truth' about human beings". This would include not only the scientific description, but also all possible alternative language games ('ways of cutting up the world') that are compatible with it. Moreover, if the scientific description were true, complete and determinist, all the language games that actually appear in history would in principle be determined, though presumably not predict­ able in practice. If we knew such a scientific description, we should have objective foundations for all possible knowledge. Since Rorty would certainly reject such objective foundations , we can only make sense of (3) by taking seriously something that is almost an aside in his account, namely a reference to the cognitive status which goes "as much for the aesthetic worth of poems as for the beliefs of persons, as much for virtues as for volitions" (p. 3 88). He goes on to suggest that the assemblage of all objective natural truths "might be a picture of the world without a sense, without a moral". Only individuals ascribe values to the world, and "it would not be objectively true or false that it 'really did', or did not, have a sense or a moral". Comparison of this sentence with the quotation in (3) suggests that when Rorty puts 'objective truth' in quotes on page 388, he is referring to the value-laden accounts that human beings give of their situation when these are consistent with but irreducible to the 'atoms-and-the-void' accounts. When he omits the quotes he means objec­ !ive truth which is the province of science alone. Although no privilege is overtly accorded to naturalism in Rorty's account (because to do so would be to reintroduce essences into knowledge), nevertheless there is a kind of Weberian decisionism - a separation of fact and value and the relegation of the latter to individual whim, or, in more guarded passages than those on page 388, to socially acceptable but ungrounded ethical discourse. There is no room for systematic mutual critique of discourses. 1 4 In this sense the Geisteswissenschaften are irreducible to naturalism, but at the cost, for Rorty, of making them value-laden, and hence not subject to mutual epistemological constraints (p. 354). The contrast with Apel is clear: Apel holds that naturalism is itself subject to reflective critique, and rejects the ethical relativism entailed by a naturalistic separation of fact and value. Apel claims to do philosophy from a theoretical standpoint above the 'normal discourses' of unreflective speech (cf. Habermas on levels of discourse), whereas Rorty both denies the possibility of sllch a standpoint, and in fact lives relatively happily

EPI STEMOLOGY WITHOUT FOUNDATIONS

65

within naturalism. Rorty's two positions mutually reinforce each other. 3.3.

The Critical Dissolution of Relativism

This brings us to the question of whether it is possible to dissolve relativism by taking a meta-standpoint above the level of normal discourses. To answer this in the affirmative is the main thrust of the philosophies of both Habermas and Ape!. According to their view, argu­ mentative or theoretical discourse is common to all discourses that seek truth and values, and the uncovering of the presuppositions of argumenta­ tive discourse is a normative exercise involving evaluative critique. As is well known, Habermas carries out this normative task by describing the ideal speech situation that he claims is presupposed in every commitment to truth-seeking discourse, namely the situation in which no participant is constrained either by external factors of social domination, nor by internal factors of psychopathology. There has been much discussion of objections to this idea, not only in regard to the apparent impossibility of realizing ideal speech, but also because the very conception itself can be presented as ideologically loaded. Rather than add to the discussion of these objec­ tions, however, it is perhaps illuminating to look at Apel's different, but parallel, discussion of some of the grounds for critical theory. Apel describes philosophical critique as operating at a level above that of unreflective discourse whenever it attempts to mediate between lan­ guage-games . 1 5 In particular, critique is required of objective-science discourse for the reasons already discussed, namely the impossibility of 'objectivizing' open-ended human communication, and the inescapability of evaluations in the understanding of human intentionality. Critique is also required of historical or hermeneutic discourse which seeks to under­ stand actors on the basis of their behaviour and their speech. Apel argues that if motives and meanings were always transparent to actors (that is, if actors were already in ideal speech situations), then hermeneutics could take their own self-understandings for granted, that is, would not need to be critical. But motives and meanings are not transparent; actions always embody contradictions ; and hence hermeneutic 'understanding' must be better than actors' self-understanding, and must be based on external grounds. Apel goes on to develop the significance of the psychoanalytic model for critical theory. In the psychoanalytic situation there is a constant dialectic between the patient's expression of his own self-understanding and the therapist's demystification of this in the light of his external theory. Sometimes the external theory that grounds the critique of false conscious-

66

MARY H E S S E

ness may be technical science itself, for example economic science has been called in to correct actors' self-understanding of the causes of the First World War. In such cases technical science itself becomes an ideological weapon, and reveals its own moral dimension as a source of persuasion and power. In discussing the foundations of ethics, Apel gives a second example of the search for grounds of critique. He argues that neither the ethical commitment to ideal speech, nor the value presuppositions of technical science, 16 are sufficient to ground ethical principles in practical conditions of power and conflict of interest. Here as in his earlier recognition of the self-delusion endemic in all self-understanding, Apel shows a somewhat more realistic appreciation of the human situation than does H abermas. His attempt to supplement the concept of ideal communication as a foundation for ethics involves, however, an equally ungrounded ultimate commitment: namely commitment to the survival of the human species qua single example of the communication community. Such a principle is, of course, itself ethically contestable, but Apel should perhaps be interpreted not as seeking ultimate grounds, but rather procedures for rational com­ munication at every level of commitment. Given a commitment to the survival of the human species as a matter of de facto consensus, we could get on with a rational ethics. I give these two examples of the operation of critical philosophy from Apel to illustrate a general point. That is, in brief, that Rorty has given up too soon. Critical philosophy suggests a hierarchical picture of the possi­ bilities of philosophical discourse that he has overlooked. Ground-level discourses such as those offered by, as he puts it, "poets, novelists, depth psychologists, sculptors, anthropologists and mystics" (p. 362) as well as scientists, may be mediated at a higher level by various philosophical devices such as those used by Habermas and Apel. These include commit­ ment to ideal speech, critique of ideology, science as corrective of illusions of self-understanding, and so on. These critiques themselves rest on grounds that are contestable at the same or higher levels. There is the same potentially infinite process as occurs in the unending dialectical movement between thesis, antithesis and synthesis, or in the practice of sociology of knowledge, where one asks 'And what are the social conditions that explain the prevalence of the sociology of knowledge in the late 20th century?' Does this unending or even apparently self-refuting character of critical discourse render it invalid or unnecessary as Rorty would claim? Surely not, because at each level new types of argument and new types of

E P I STEMOLOGY WITHOUT FOUNDATIONS

67

cntIque can be brought to bear, without at any time supposing that "foundations" or "essences" have been reached. And because some of these new types of critiques uncover social power-relations from which philosophy cannot insulate itself, philosophical discourse is not just more disciplined than mere "free and leisured conversation" (p. 389), but also vastly more socially significant. l 7 University of Cambridge NOTES I Oxford: Blackwell, 1980. Page references to this book will be given in brackets in the text. 2 This does not, of course, imply that 'logic' is irrelevant to science (certainly mathematics is directly relevant), but the relevance oflogic will be as an ideal and an approximation - notions that themselves require philosophical elucidation. See, for example, Kuhn's discussion of translation forthcoming in the last section of his "Commensurability, comparability, com­ municability", PSA 1982, Volume 2, edited by P. D. Asquith and T. Nickles (East Lansing, Michigan: Philosophy of Science Association, 1983), and my reply, ibid, Section 2. 3 In various places (Consequences of Pragmatism, Harvester, Sussex, 1982, p. xvi, and private correspondence) Rorty suggests that there are no philosophical questions about 'How does morality work?' or 'How does art work?'. But this is surely to narrow the scope of philosophical ethics and aesthetics unnecessarily. Just as there are internal philosophical questions about the 'logic' of scientific methodologies, so there are questions about, for example, the relative desirability of various types of distributive justice, or the effectiveness and function of metaphor in language. 4 B. van Fraassen, The Scientific Image (Oxford: Clarendon, 1 980), p. IO. 5 1. HabeTmas, 'A Postscript to K1lDwledge and Human Inleresls' , Philosophy if Ihe Social Sciences 3 (1973), p. 165. 6 0p. cit. , p. 1 82. 7 K.-O. Ape!, Towards a Transformation of Philosophy, translated by G. Adey and D. Frisby (London: Routledge, 1980), p. 140. 8 Op. cit. , p. 83 ff. 9 Rorty reveals his awareness of a moral issue at this leve!: Suppose that Socrates was wrong, that we have not once seen the Truth, and so will not, intuitively, recognize it when we see it again. This means that when the secret police come, when the torturers violate the innocent, there is nothing to be said to them of the form 'There is something within you which you are betraying. Though you embody the practices of a totalitarian society which will endure forever, there is something beyond those practices which condemns you.' This thought is hard to live with. (Consequences ofPragmatism, p. xlii). 10 Op. cit. , p. 3 1 . II Ibid. , p. 1 16. 12 Of all the recent writers on this tangled issue, perhaps D. Dennett comes closest to doing justice to the intuition of the libertarians that freewill requires undetermined forks in physical world-lines (Brainstorms, Harvester, Sussex, 1 978, Ch. I S). 13 'Naturalistic terms' must mean here 'within the framework of the currently accepted best science'.

68

MARY H E S S E

14 There i s no place for "the notion of a presiding discipline called 'epistemology' or 'transcen­ dental philosophy' - reducible [either1 to Naturwissenschaji (psychophysiology) [or 1 to Geistes­ wissenschaji (the sociology of knowledge)" (p. 354). Rather than accusing Rorty of covertly reintroducing a fact-value distinction, it would perhaps be better to say that he adopts the distinctions (fact/value, Natur/Geisteswissenschaji, scientific description/description in terms of 'meanings') as he finds them implied in the Western Enlightenment tradition. But Ape! is also a Western Enlightenment man, and he would disagree that they are implied by the tradition. The point illustrates the impossibility of appealing to a 'social consensus' in cases where disputes precisely disrupt the consensus. 1 5 Op. cit. , p. 66 fT. 1 6 Ibid., pp. 278, 226. 1 7 Rorty's Introduction to Consequences of Pragmatism brings out a feature of his work which is implicit in Philosophy and the Mirror ofNature, namely the absence of a critique of the social power relations that make his type of pragmatism possible. To the dissolution of a legion of 'philosophical problems' in Pragmatism one wants to ask 'What is the motivation for telling people not to engage in a certain kind of activity (the 'search for foundations')?'. It cannot be the conclusive character of the arguments against foundations, for such a notion would be self-defeating, and Rorty would himself agree that there are none, and that his enterprise is persuasive towards paradigm change rather than knock-down refutation. It cannot be just that the search has hitherto been unsuccessful and looks likely to remain so: a search for 'moral perfection' is similar, but that does not entail that we should cease to engage in it. It cannot be the Western consensus, because there is none - some people even persist in searching for theological foundations. Rorty's motives are, certainly, philosophical - he is in the tradition of those who recommend a 'new turn', and who in the past have been happily assimilated into the tradition by their-successors. But there is another, latent, function of this sort of pragmatism and naturalism. It serves to legitimate what Habermas calls the techno­ cratic systems analysts - the technological and social programmers. And it can only appear as an ideology to live by where the power issues are not seriously joined, that is where philosophy can be "free and leisured conversation" in a "stable and prosperous part of the world". I am indebted to private correspondence with Richard Rorty for clarification of several points, and to the helpful comments of Nicholas Lash, Michael Power and John B. Thompson, specially in regard to my interpretation of Habermas and Ape/. I am grateful to John Thompson in particular for his insight into the absence of the dimension of social power in Rorty's book. Needless to say remaining misunderstandings are all my own.

PHILIP PETTIT

P H I L O S O P H Y AFTER R ORTY

1. In her paper 'Epistemology without Foundations' Mary Hesse tries to rescue philosophy from the elusive, even evanescent, future to which Richard Rorty has condemned it. I Rorty's claim is that since Descartes and Locke philosophy has been the search for foundations of knowledge; that such an enterprise is misconceived, being based on an unnecessary and historically unfortunate picture of the mind as an inner arena of self-intimating certainties ; and that all there remains for philosophy to become is the hermeneutic auditor and advocate of discourses which, lacking routines of conflict resolution, are apt to be ignored in our scientistic culture. Hesse has little complaint about Rorty's tirade against the epistemological aspirations and psychological assumptions of tradi­ tional philosophy. Her criticism is directed only at the characterisation of philosophy as having henceforth a novel and purely hermeneutic task. She makes a number of protests : that even if philosophy is assigned a new hermeneutic aim, we shall probably be able to reinterpret the history of the subject in the light of this goal;2 that Rorty's anti-foundationalist arguments raise questions which actually outrun the hermeneutic brief that he envisages;3 and that there is more in any case to the hermeneutic task than he ever allows. The last criticism is the fundamental one and will be the focus of our concern. On Rorty's conception, philosophy as hermeneutics tries to arbitrate, or at least practise shuffle diplomacy, in theoretical conflicts which suffer from evidential and methodological underdetermination : from under­ determination by data and by agreed criteria of theory selection. Preferring to speak of reflection - sometimes hermeneutic reflection - Hesse argues that in three ways, high-lighted in the Apel-Habermas view of philosophy, reflection transcends the bounds of Rorty's image of it. Where he is taken to envisage a mundane, scientistic and ultimately relativistic enterprise, she argues for something transcendental, non-scientistic and non-relati­ vistic: transcendental, because it displays conditions for the possibility of the discourse, or whatever, in question; non-scientistic, because it offers 69 A. J. Holland (ed.), Philosophy, Its History and Historiography, 69-83. trI\ 1 00 , L�.

n

D_:..1_1 D•. LI.:_L':. __

r _____ ___ . _

70

PHILIP PETTIT

more than a second-rate status to the Geisteswissenschaften ; and non-rela­ tivistic, because it seeks to uncover common procedures of rational com­ munication and conviction in all the discourses we countenance. There are two topics to which I address myself in this paper. First, in the section following, I examine Rorty's image of philosophy; then, in the final section, I ask whether the criticism of that image which is implicit in Hesse's last protest is fair. My comments are made on the assumption, shared with Hesse, that Rorty is certainly right to despair of an epistimo­ logically foundational role for philosophical reflection. The question is, what other role can philosophy have.

2. Rorty ascribes two distinguishable and complementary functions to philo­ sophy in the final part of his book. First it is held to do for abnormal or incommensurable discourses something that corresponds to what episte­ mology - in a downgraded, nonfoundational sense - does for discourses which are normal and commensurable. Where such epistemology tracks down criteria and methods of conflict resolution within normal realms, philosophy strives to ensure that the absence of such procedures in abnormal areas does not mean that those discourses are silenced or shut otT. Philosophy is the hermeneutic auditor of abnormal kinds of talk and serves to keep the conversation going between the speakers in question and more pedestrian interlocutors.4 Secondly philosophy is said to play the role not so much of hermeneutic auditor as of hermeneutic advocate. It sets out to construct a novel abnormal discourse and then to press this discourse into use, reinter­ preting our ordinary experience in its terms. In this service philosophy ceases to be an underlabourer to other intellectual enterprises ; it becomes an entrepreneur in its own right, advancing a distinctive 'poetic' product. For Rorty the twin roles of philosophy involve it in a process of Bildung or edification, rather than in a systematic programme of research. "The attempt to edify (ourselves or others). may consist in the hermeneutic activity of making connections between our own culture and some exotic culture or historical period, or between our own discipline and another discipline which �eems to pursue incommensurable aims in an incommen­ surable vocabulary. But it may instead consist in the 'poetic' activity of thinking up such new aims, new words, or new disciplines, followed by, so to speak, the inverse of hermeneutics: the attempt to reinterpret our

PHILOSOPHY AFTER RORTY

71

,, familiar surroundings i n the unfamiliar terms o f our new inventions. 5 It is difficult to comment on the entrepreneurial brief which Rorty gives philosophy. If there are new illuminating things to be said, even in alien and abnormal discourse, then certainly let them be said; and if it pleases, let those who manage to say them be called philosophers. In what follows I shall ignore the possibility of such poetic philosophy and concentrate on the more properly hermeneutic form of the discipline, which Rorty envisages in his first underlabourer brief. Is the job of interpretation which is mentioned in this role-assignment a coherent and challenging task to assign to philosophers ; or indeed to anybody? Challenging, it may not at first seem to be. After all, interpretation across cultures, epochs and disciplines, such as Rorty mentions in the quotation above, is a relatively smooth practice and has long been a non-philosophical province. Why should it rouse colonial pretensions among philosophers, even philosophers deprived of their epistemological homeland? For Rorty I think that there is no reason why it should, except so far as the interpretation called for is of forms of discourse which are, as he says, abnormal or incommensurable. The hermeneutic task to which he wishes to recruit otherwise un-employed philosophers is not the regular job of interpretation but a rather more problematic one. As he says, "hermeneutics is only needed in the case of incommensurable discourses" ;6 "Hermeneutics is the study of an abnormal discourse from the point of view of some normal discourse - the attempt to make sense of what is going on at a stage where we are still too unsure about how to describe it, and thereby to begin an epistemological account of it". ? A discourse is abnormal or incommensurable - for Rorty the terms are synonyms - if it occasions disputes for which there appear to be no standard procedures of resolution. "By 'commensurable' I mean able to be brought under a set of rules which will tell us how rational agreement can be reached on what would settle the issue on every point where statements seem to conflict".8 Is incommensurability an intrinsic failure in a discourse, a failure on the part of participants to develop standards of conflict resolution? Often Rorty takes it as such but sometimes he suggests that the fact of its being alien is enough to make a discourse incommensurable and here he must be thinking of a relational, not an intrinsic, feature. This is suggested when he speaks above of hermeneutics being required at a stage where we are uncertain about the discourse under description and, more generally, when he says that epistemology caters for the familiar but hermeneutics is required to deal with the unfamiliar.9 In view of these

72

P H I L I P PETTIT

indications we may take 'incommensurable' to mean 'at least incommen­ surable-by-us and perhaps incommensurable-by-anyone' ; 'abnormal' to signify 'at least with norms different from ours - i.e. alien - and perhaps without any norms of its own.' We were concerned with why interpretation should be thought to be a challenging job for philosophers. The fact that the matter for interpretation is always an abnormal discourse may be taken to introduce an appropriate challenge, for two reasons. The first is that given abnormality we cannot expect in Rorty's view to be able to translate discourse smoothly, or even Whiggishly, into the normal discourse from which we start. "The fact that hermeneutics inevitably takes some norm for granted makes it, so far forth, 'Whiggish'. B ut in so far as it proceeds non-reductively and in the hope of picking up a new angle on things, it can transcend its own Whiggishness." 1 0 A first challenge then is to find out whether we are confronted with a way of speaking and seeing which transcends our present perceptions, our current image of the world. "Our wonder, stripped of mirror-imagery, is simply about whether somebody or something may not be dealing with the world in terms for which our language contains no ready equivalents. More simply still, it is just wonder about whether we do not need to change our vocabulary, and not just our assertions." l l Rorty also suggests a second reason for thinking that the interpretation of abnormal discourse should be challenging. This is that in his view such interpretation may lead, not just to self-expansion, so to speak, but also to self-correction. The consideration is hinted at in the remark just quoted, where Rorty takes for granted that understanding an unfamiliar discourse may lead us to change our assertions. It is explicit in the following description of the hermeneutic philosopher at work. "In his salon, so to speak, hermetic thinkers are charmed out of their self-enclosed practices. Disagreements between disciplines and discourses are compromised or ,, transcended in the course of the conversation. 1 2 The important point is that between the normal, interpreting discourse and the abnormal, inter­ preted one there are always assumed to be disagreements. This means that the interpreter is faced with the challenge, not just of seeing whether his horizons can be expanded, but of discovering whether his present views are even adequate. The hermeneutic philosopher is no mere interpreter concerned only with comprehension. He is also a critic, both of his own discourse and that of his interpretees, for he has an equal interest in determining which of the disagreeing sides is in the right. 1 3 Let us grant that the task assigned to the philosopher is a challenging

P H I L O S O P H Y AFTER RORTY

73

one. The next question is whether it is coherent. I wish to argue that it is not, on the following grounds. If a discourse is so different that its under­ standing involves self-expansion, then it cannot conflict with one's present beliefs in such a way as to encourage self-correction. Conflict of the kind which means that at most one of two rival beliefs or belief sets can be true requires that the beliefs can be expressed - or that they entail beliefs which can be expressed - as the assertion and denial, respectively, of the same proposition or propositions. Difference of the kind which means that in passing from one discourse to another one expands one's horizons requires that the propositions to which the discourses are addressed are not the same. Thus it is incoherent to envisage that an inquiry should at one and the same time force self-expansion and prompt self-correction. The point can be put as follows. The normal discourse from which the hermeneutic inquirer starts represents his orthodoxy. There are two ways in which the discourse he investigates can fail to be orthodox : one, it may be heterodox, contradicting his original beliefs ; or two, it may be 'xenodox', involving beliefs of a different and unfamiliar kind, beliefs addressed to novel propositions. No single discourse can be at once xenodox and heterodox and so no single inquiry can simultaneously yield the fruits of xenodox exploration - self-expansion - and heterodox contemplation self-correction. This point can survive the Quinean claim that there is no fact of the matter as to whether a discourse under interpretation is one or the other; the point is that it cannot be represented simultaneously as both. That Rorty thinks of hermeneutics as the investigation of the xenodox is abundantly clear. He consistently contrasts hermeneutics with epistemo­ logy - that is, nonfoundational epistemology - on the grounds that the latter holds that "all possible descriptions can be rendered commensurable with the aid of a single descriptive vocabulary". 1 4 Again he often makes explicit that in doing hermeneutics one gets into contact with propositional contents - he would not call them that - which are radically alien. The following two passages are examples. There is no special reason to think that any given one-word expression in one culture can be matched with a one-word expression in a very different culture. Indeed, we may feel that even lengthy paraphrases will be oflittie help, and that we must just get into the swing of the exotic language-game. 1 5 Producing commensurability by finding material equivalences between sentences drawn from different language-games is only one technique among others for coping with our fellow humans. When it does not work, we fall back on whatever does work - for example, getting the hang of a new language-game, and possibly forgetting our old one. 16

74

PHILIP PETTIT

But if Rorty sees hermeneutic inquiry as investigation of the xenodox, he is equally happy to cast it as an attempt to come to terms with the heterodox. "Hermeneutics sees the relations between various discourses as those of strands in a possible conversation, . . . where the hope of agreement is never lost so long as the conversation lasts." I 7 Here what is envisaged as important is not getting in tune with something novel and expansive, but finding the adjustments required to ensure a harmony of opinions across rival and conflicting accounts. The image of the explorer has given way to the image of the diplomat. But Rorty does not hold two alternating pictures of the hermeneutic philosopher. He consciously merges the pictures, arguing that the philo­ sopher-as-interpreter will seek the agreement we want with the heterodox, while respecting the gulf that separates us from the xenodox. He writes as follows of the hope for agreement. This hope is not a hope for the discovery of antecedently existing common ground, but simply hope for agreement, or, at least, exciting and fruitful disagreement. Epistemology sees the hope of agreement as a token of the existence of common ground which, perhaps unbeknown to the speakers, unites them in a common rationality. For hermeneutics, to be rational is to be willing to refrain from epistemology - from thinking that there is a special set of terms in which all contributions to the conversation should be put - and to be willing to pick up the jargon of the interlocutor rather than translating it into one's own. For epistemology, to be rational is to find the proper set of terms into which all the contributions should be translated if agreement is to become possible. I S

How does Rorty come to maintain inconsistent things? My own belief is that he is ill served by the term 'incommensurable', for it misleads him in two ways. He stipulates that it means 'in currently non-resoluble conflict', whether non-resoluble by us or by anyone. Applying it then to another discourse he is misled into assuming, without argument, not just that the discourse is in internal conflict, but also that it is in conflict with his own. This unargued assumption has 'incommensurable' come to mean 'in currently non-resoluble conflict with orthodox views'; it takes on the sense, more or less, of 'heterodox'. But secondly, and more significantly Rorty is misled so far as he never shakes off the ordinary usage of 'incommensurable', though he explicitly disowns it. 1 9 On this usage the word means 'untranslatable, at least in a straightforward way' ; it has the sense, roughly, of 'xenodox'. That he never rids himself of the common usage appears in his constant application of the epithet, not to bodies of

PHILOSOPHY AFTER RORTY

75

doctrine, but to the terminologies in which such doctrine is spelled out ; it is unfortunate that 'discourses' may refer to either.20 Because hermeneutics is said to deal with incommensurable discourses, and because this may mean either of two things, it is cast at once as exploration of the xenodox and examination of the heterodox. Such at least is my diagnosis of what must be seen as a surprising slip on Rorty's part. The slip comes to the surface on one occasion. In a context where the term has its standard rather than its stipulated sense, he says : "incommensura­ bility entails irreducibility but not incompatibility".2 1 If he had added that it positively entails compatibility, then he would have said that xenodox cannot be heterodox; he would have admitted that no single inquiry could explore the xenodox and examine the heterodox. A further comment in concluding this line of criticism. Mary Hesse has suggested that from the point of view of one discourse 0 1 , another 02 may be at once xenodox and heterodox.22 This happens, she thinks, when 0 1 is current medical theory and 02 faith-healing lore. She would therefore defend the coherence of Rorty's view of hermeneutic philosophy, even on a strict reading of his remarks. But her suggestion cannot be pressed into service on Rorty's behalf. The reason is that two discourses like medical and faith-healing wisdom are not incommensurable in his sense. From the point of view of each, the other is simply false, since it maintains things on which the standards of resolu­ tion in the home discourse ambiguously require a negative j udgment. If the discourses can be said to be incommensurable, it is only in the weaker and irrelevant sense that from an outside point of view, there may be no standards available for j udging between their competing claims. What are we to make then of Rorty's image of philosophy? We can interpret his remarks charitably along either of two lines. The one is my own invention, the other a suggestion made by Rorty in a private com­ munication : it is unclear whether he regards this as a proposal that is inspired by charity however and not one that is perfectly straightforward. At one point in his book Rorty speaks of "incompatible points of view, in the sense that we cannot be at both viewpoints simultaneously".23 My own suggestion is that we could take the agreement of which he speaks in connection with hermeneutics to consist just in the recognition that though the point of view of another discourse may not be simultaneously tenable with that of one's own, it may be a view point to which one can at any time emigrate. On this picture discourses are pragmatically in conflict, so that one cannot participate in them at the same time, just as one cannot

76

P H I L I P PETTIT

simultaneously see the duck and the rabbit in Wittgenstein's switching gestalt. The expansion involved in understanding a discourse alien to that from which one starts is sufficient to enable one to see the way beyond such pragmatic conflict. It holds out the hope of agreement in the sense, not of mutual accommodation, but at least of peaceful co-existence. Rorty's own proposal for what I take to be a charitable reading starts from the admission that what is heterodox cannot be xenodox, and vice versa. It represents hermeneutics as the investigation, not of a discourse which is assumed to be both, but of one which may turn out to be either. "No single discourse can be at once heterodox and xenodox, but it's not true that we know right off the bat which it is. It may take a long time to figure out which, and what happens in the interval is the sort of shuffie ,, diplomacy I want to recommend under the name of 'hermeneutics . ,24 On the proposed reading of the text, as on the previous one, the contradiction which we isolated is removed but the spirit of the hermeneutic vision of philosophy is still preserved. We may be happy to find that such an interpretation is available without feeling obliged to decide which we ought to endorse. 3.

W e return now t o t h e m atters raised in Mary Hesse's main protest. She complains that even if the pretensions of a foundational epistemology are rejected, we may still hope for a form of philosophical reflection which is transcendental, non-scientistic and non-relativistic. Is this just? Does Rorty's image of philosophy-as-hermeneutics entail that no transcendental form of argument is available? And does it mean that he commits himself to a scientistic and relativistic view? I would lik� to consider first scientism, second relativism, and finally the charge about transcendental argument. Rorty is explicitly naturalistic in believing that the world can be com­ pletely - in the sense of exhaustively2 5 - described and even predicted in the terms of basic, physical science. "Every speech, thought, theory, poem, composition, and philosophy will turn out to be completely predictable in ,, purely naturalistic terms. 26 By 'naturalistic terms' he refers to the terms of an 'atoms-and-the-void account of micro-processes'. Hesse claims that he is not just naturalistic in this sense however, but scientistic: specifically, that he is committed to the view that the only really objective truth about the world is the truth revealed in basic science.

PHILOSOPHY AFTER RORTY

77

She admits that he himself claims that objective truth is accessible in non-basic forms of discourse, forms of discourse which may be incommen­ surable in terms of basic theory, and therefore not reducible to such theory. However she argues that this claim is inconsistent with the joint assertion of naturalism and anti-foundationalism. The assemblage of basic and non­ basic discourses would exhaust objective truth. "We should have objective foundations for all possible knowledge.'027 This argument does not seem to carry force however. The assemblage in question is an unattainable ideal for Rorty, and not even an ideal regulative of anything. At no point could we be in a position to say that we had reached the end of all discourses, and that henceforth any knowledge claims would have to be shown to be derivable from what we had in hand. At no point could we in that sense have established a foundation. "Given leisure and libraries, the conversation which Plato began will not end in self-objectivation - not because aspects of the world, or of human beings, escape being objects of scientific inquiry, but simply because free and leisured conversation generates abnormal discourse as , the sparks fly upward.' 2 8 The view to which Hesse says that Rorty is committed is that whereas basic science reveals objective truth, non-basic discourse - at least if it is irreducible to basic - is warped by values whose acceptance is ultimately a matter of decision. This is unfair. Rorty thinks that basic science is just as value-laden as non-basic.29 This need not worry him, since he believes that values themselves may be objective. in the only available sense of that term. "The application of such honorifics as 'objective' and 'cognitive' is never anything more than an expression of the presence of, or the hope for, , agreement aruong inquirers.' 3o In connection with the charge of scientism it is worth considering Rorty's view of the relation between basic and non-basic forms of discourse. He denies that the latter are reducible to the former. Con­ sistently with doing so, he might wish to maintain the supervenience of the non-basic: that is, that no non-basic change can take place without a basic one; that the basic level fixes the non-basic. Supervenience does not entail reducibility, since it allows that non-basic truths may not be expressible at the basic level: this, because basic terms do not allow us to formulate translational or even nomic equivalents for non-basic ones. 3 ! However it is unclear whether Rorty wishes to assert even such supervenience. He says, for example : "alternative biochemical (e.g.) as well as alternative psychological theories will be compatible with all and only the same move-

78

PHILIP PETTIT

ments of the same particles". 32 This suggests that he envisages non-basic change without any basic counterpart. It may be that Rorty is guilty here of careless phrasing and that he does accept supervenience. 33 Yet it has to be said that he often writes as if non-basic discourses ride freely or non-superveniently on the basic atoms­ and-the-void account. His attitude is one of laissezjaire. Equal rights for all discourses and a rejection of "invidious comparisons between these modes of description".34 Such comparisons are said almost always to be ill conceived, as in the case of factual and evaluative discourse. "The usual excuse for invidious treatment is that we are shoved around by physical reality but not by values. Yet what does being shoved around have to do ,, with objectivity, accurate representation, or correspondence? 3 5 Rorty looks for tolerance, asking us to judge each discourse and vocabulary in its own right "on pragmatic or aesthetic grounds alone". 36 The rejection of supervenience may be important to Rorty's argument for a purely hermeneutic philosophy. If we allow supervenience, then the following task will call to be performed for any non-basic form of discourse: that of showing how the discourse can get going compatibly with basic theory, without being reducible to such theory. Arguably, a good deal of contemporary work in philosophy - most conspicuously, in philosophy of mind - is of this kind. What we may need, if Rorty's image of philosophy is to be protected, is a defence of non-supervenience. And yet it is hard to see where that may come from. All our intuitions point to supervenience. More importantly, it is obscure how non-basic forms of discourse can provide the causal explanation if they are not supervenient on basic theory: supervenient in such a way that the causal chains invoked may run through links charted at the basic level. We shall return later to this theme. In the meantime, we have to ask whether Mary Hesse's charge of relativism sticks any better than that of scientism. She suggests that in insulating discourses from one another, arguing that they need not be mutually commensurable, Rorty deprives us of any good reason why we should assert our own discourse-bound com­ mitments as against those of alternativ� discourses. The idea is that since we cannot j udge between the commitments, and since they still compete with one another, we can prefer our own only on the grounds of embracing the devil we know. There is no room for us to enter even the presumption that ours are correct, the others false. In the absence of procedures of adj udication, there is nothing that we could mean by this other than that only ours are true-for-us, or satisfy some other relativistic ersatz. 37 Hesse

PHILOSOPHY AFTER RORTY

79

sees the exit from relativism in the possibility of unearthing devices whereby the claims of different discourses can be settled. "Ground-level discourses . . . may be mediated at a higher level by philosophical devices such as those used by Habermas and Apel. These include commitment to ideal speech, critique of ideology, science as corrective of illusions of ,, self-understanding, and so on. 3 8 Rorty invites the charge of relativism, so far as he represents alternative discourses as mutually incommensurable, not just in the sense of xenodox but also in that of heterodox. If I have to see discourses other than my operative one as making claims incompatible with mine, but claims so foreign that I cannot judge between them and mine, then certainly relativism, with all its self-defeating paradoxes, is in prospect. However, we have already seen that there are two different ways in which Rorty's remarks may be charitably interpreted. On neither does he posit the existence of discourses which are at once mutually xenodox and mutually heterodox. If either is endorsed, then we no longer need to see him as inviting the charge of relativism. As a matter of fact Rorty supports Davidson's well-known argument about conceptual schemes and so positively disavows relativism. That argument is that the only available ground for thinking of another lan­ guage, and therefore another discourse, as a language or discourse is to think of it, if not as translatable into mine, at least as interpretable - say, by paraphrase, neologism and even linguistic importation - in such a way that I can still hold onto most of my own; I can still retain my present sense of reality. We may not actually be able to translate or otherwise interpret but "we cannot make sense of the claim that there are more than temporary impediments to our know-how".39 Thus I can think of other discourses as discourses only if I think of them as saying things which, however novel, are mostly true by my present lights. This is a far cry from relativism.4o And yet the claim that Rorty is not a relativist may be resisted. H e rejects the idea that when confronted with challenging claims from a n alien discourse one can rise above the confrontation with one's own tenets and appeal to a higher court for resolution. He would have no sympathy with Hesse's invocation of philosophical devices which can mediate from on high between ground-level discourses. Does it not follow that in facing alien claims all he can do is to be ethnocentric and affirm his parochial vision? Rorty's views do entail a certain ethnocentricity but not one involving

f!'''r: llosophv

i

. i "f��

[.'�;'. M.jrtor': � �i b"

.

80

PHILIP PETTIT

relativism. Of the following ethnocentric principles, they support only the second.

1. The principle of bullish relativism There are many incompatible views of the world, all equally good, but piety and practicality dictate that you stick with your own. 2. The principle of sheepish absolutism In making sense of other incompatible views of the world you have to start from your own but there is no saying, given the effect of challenge and self-criticism, where you may end. Even if we rescue Rorty from the charges of scientism and relativism, there remains the question of whether he leaves room for a transcendental mode of inquiry, such as Hesse envisages. This investigation would reveal, not the conditions for the possibility of supposedly inevitable features, as in the Kantian model, but only conditions for the possibility of particular, contingent discourses. Thus the Apel-Habermas analysis of the supposi­ tions of argumentative discourse is held to be transcendental but non­ Kantian : 'Their programme is not that of reinstating the Kantian condi­ tions for all possible experience. Perhaps men can live without argument, or even if they cannot, this seems to be a contingent fact of evolution, not a necessary fact".4 1 Although he explicitly despairs of anything other than philosophy-as­ hermeneutics, I agree with Hesse that there is nothing in Rorty's anti­ foundationalism to undermine the sort of analysis which she envisages. Earlier I adverted to the common philosophical pursuit of trying to display the compatibility of non-basic forms of discourse vis-a-vis basic theory. That may be redescribed as showing that the conditions for the possibility of those forms of discourse are realisable in a world described by the atoms-and-the-void account; it can be cast as involving a sort of transcendental analysis. I see nothing in Rorty's arguments against foundationalism to undermine such an enterprise. Like Hesse I think that Rorty discards a measure of wheat with the chaff of foundationalism.42 As a matter of fact Rorty does object to the very analysis of the condi­ tions for the possibility of argumentative discourse which Hesse hails. It may be however that this objection is not so much that the analysis is transcendental in form as that it is designed like Kant's analysis to serve a critical purpose reminiscent of foundationalist epistemology; here his

P HI LOSOPHY AFTER RORTY

81

view of it would differ from Hesse's : "The notion that we can get around overconfident philosophical realism and positivistic reductions only by adopting something like Kant's transcendental standpoint seems to me the basic mistake in programmes like that of Habermas".4 3 The reason for suggesting that this may be the real source of Rorty's objection to Apel and Habermas is that he does himself sometimes seem to endorse arguments of a loosely transcendental character. For example, he endorses Davidson's argument that to interpret is necessarily to find a good deal of truth and agreement and that the ordinary notion of truth does not allow us to imagine that what we find ourselves commonly maintaining is actually, in the main, false: "Only in the context of general agreement does doubt about either truth or goodness have sense".44 In conclusion to this piece I would like to suggest that there is no anti-foundationalist reason why Rorty should not extend his sympathy for such forms of argument and that if he does so he will encompass much more of contemporary philosophical practice than is reflected in his strictly hermeneutic brief. Consider Wittgenstein's argument that rule-following only makes sense in the context of a set of social practices ; or Wiggins's, that interpretation presupposes that certain utterances can be isolated as possessing a property hardly distinguishable from regular truth-value ;45 or the commonly mooted argument that to explain human actions in the ordinary intentional way is to assume that agents satisfy certain norms of rationality.46 These are distinctive and substantive claims which have appeared in recent philosophy: they are distinctive because one can hardly imagine them coming up with other disciplines of inquiry; and substantive, because they have important ramifications for various practices. I see no reason why Rorty should not countenance such paradigms of philo­ sophical research. Were he to do so, he would have to paint a more nuanced picture of philosophical activity than is depicted on his her­ meneutic canvas. He might even satisfy Mary Hesse's desire to see philoso­ phy represented in transcendental, or quasi-transcendental, colours.47 Australian National University, Canberra

NOTES 1 Mary Hesse's paper is in this volume; henceforth it is referred to as EF. Richard Rorty's condemnation is in the third part of his book Philosophy and the Mi"or of Nature (Oxford:

82

PHILIP PETTIT

Blackwell, 1980); henceforth it is referred to as PMN. EF, pp. 5 1-52. 3 EF, p. 55. This protest survives what I see as a misreading of Rorty's views on truth. Mary Hesse takes Rorty to commit himself, presumably not without an awareness of what he is doing, to "the rejection of an applicable propositional logic, and hence of any theory of truth, argument, natural kinds and natural laws now on offer" (p. 55). This does not fit with his endorsement of Davidson's theory of truth (PMN, p. 295 ff.). 4 This is a summary of PMN, Chapter 7. 5 PMN, p. 360. The second role for philosophy is discussed in Chapter 8. 6 PMN, p. 347. 7 PMN, pp. 320-321 . The epistemological account mentioned would be such only in a non­ foundational sense. 8 PMN, p. 3 16. 9 For example PMN, pp. 352-353. Elsewhere Rorty speaks of mutually incommensurable but intrinsically normal discourses (p. 388). 10 PMN, p. 321 . II PMN, pp. 352-353. 12 PMN, p. 3 17. 13 It may be for this reason that sometimes Rorty casts the hermeneutic philosopher not as an interpreter but as someone working at a higher order: "hermeneutics is, roughly, a description of our study of the unfamiliar and epistemology is, roughly, a description of our study of the familiar" (p. 353). 14 PMN, p. 378. 1 5 PMN, p. 267. 16 PMN, pp. 355-356. See also p. 3 19: "We play back and forth between guesses about how to characterise particular statements or other events, and guesses about the point of the whole situation, until gradually we feel at ease with what was hitherto strange". 17 PMN, p. 3 1 8. 18 PMN, p. 3 1 8. 19 PMN, p. 302, n. 35; p. 3 16, n. I . 20 See for example PMN, pp. 350, 360, 362, 376-378, 386-387. Rorty's comment, p. 3 16, n. 1 , does not serve to legitimate this usage. 21 PMN, p. 388. 22 In discussion at the Lancaster conference. 23 PMN, p. 385. 24 Private communication June 1983. 25 See PMN, p. 205. 26 PMN, p. 387. 27 Hesse, EF, p. 64. 28 PMN, p. 389. 29 See for example PMN, p. 329. 30 PMN, p. 335. I think that Hesse mistakenly imputes a decisionist view of value to Rorty, on the basis of his remark that there is no objective truth as to whether the world as presented in a certain way has a sense or moral for an individual (PMN, p. 388). All that Rorty wants to stress in saying this however is that the individual may always generate a novel discourse in the light of which the given ones will lose their hold on him. 31 For a good characterisation of supervenience see J. Kim 'Supervenience and Nomological 2

P H I LO SO PHY AFTER RORTY

83

Incommensurables', American Philosophical Quarterly 15 ( 1 978): 149-164. 32 PMN, p. 203. See also p. 247. 33 In the private communication mentioned, Rorty says that he accepts supervenience in the sense given. 34 PMN, pp. 204-205. 35 PMN, p. 375. 36 PMN, p. 208. 37 This is a loose and perhaps imaginative summary of EF, Section 2.2. 38 EF, p. 64. 39 PMN, p. 355, n. 25. Notice one assumption in the claim: that the foreign speakers are sensorily and otherwise equipped more or less as humans are. 40 Rorty might avow relativism in something like the sense distinguished by Bernard Williams in The Truth in Relativism', Proceedings of the Aristotelian Society 74, ( 1 974-75): 2 1 5-228. 4 1 EF, p. 60. Contrast Rorty's softening of transcendental argument, mentioned below, note 47. 42 For a more systematic account of one view of such wheat see my 'Philosophy and the Human Sciences', University of Bradford, Inaugural Lecture, 1979. 43 PMN, p. 382. 44 PMN, p. 309. 45 'What Would be a Substantial Theory of Truth?' in Philosophical Subjects, edited by Zak Van Straaten (Oxford: Oxford University Press, 1 980): 189-221 . 46 For one version of this see Graham Macdonald and Philip Pettit, Semantics and Social Science (London: Routledge and Kegan Paul, 1 98 1 ). Some ramifications of the argument are also traced in this book. 47 In an article from 197 1 - endorsed PMN, p. 305 - Rorty supports the sort of transcendental argument which tries to establish that specific proposals to revise certain conventional ideas are actually parasitic on those ideas and self-destruct: this, as distinct from attempting to show that no revisionary proposals whatsoever could hope to succeed (,Verificationism and Transcendental Arguments', Nous 5: 3-14). The paradigms of philosophy mentioned in this paragraph can surely be cast to fit this model of argument. What is not so clear however is that they can be squared with the view that philosophy is just hermeneutics.

ADAM MORTON

C O M M ENT ON R O RTY

Hesse and Pettit present somewhat different reconstructions of Rorty's suggestions about the discipline that might survive the collapse of foundationalistic epistemology. They both treat Rorty's argument very respectfully, as opening the way to an interesting new possibility. I think that they are both too charitable to him; I think that there are a lot of bad arguments in Philosophy and the Mirror ofNature, and a quantity of simple silliness. This is not to say that the openings up of the subject that Hesse and Pettit derive from Rorty are not genuinely attractive. What I would challenge is the suggestion that they are forced upon us by the collapse of the rest of the subject. I see no reason to believe this. According to Rorty, something central disappears from philosophy if (a) we cease to see human knowledge as based on indubitable evidence, and (b) we accept that intellectual life inevitably divides into different more-or-less autonomous currents, traditions, or discourses, each with its own presuppositions and procedures, and between which translation is beset with difficulties of principle. It is pretty clear why if both of these are true then a certain 'Cartesian' enterprise of grounding all we know on foundations no-one would doubt is hopeless. The truth of either of them would be fatal enough. But this is a pretty small target: Descartes on a traditional but historically by no means uncontroversial interpretation, 1 Russell for a few moments during the 1 920's, perhaps, a few incautious people in Vienna between the wars, . . . Most of the enterprise of trying to sort out how people do acquire their knowledge of the world, and how they can best do so, is not directly incriminated. Several of the papers of this conference have shown how wide and open the live questions about scientific method in seventeenth and early eighteenth century Europe were. An orthodoxy defined by the Descartes of first-year philosophy certainly did not rule then. And the issues that were broached then have obviously developed into questions which continue to have a place in the scientific enterprise. Real questions of method are an intrinsic part of our situation ; we attack them from within ; this is worth doing; it is called epistemology. How does Rorty get from his attack on foundationalism to his rejection of all of epistemology? I'm afraid I have trouble avoiding libel here. As far 85 A . J. Holland (ed.), Philosophy, Its History and Historiography, 85-86. "

86

ADAM MORTON

as I can make out, he does it with three manoeuvres : (i) identifying foundationalism with Descartes, describing Descartes' position in terms of a Rylean optical metaphor, and then refocus sing the optics so that the metaphor encompasses a much more comprehensive project; (ii) s addling non-foundationalistic epistemology with a 'realist' programme in the analysis of truth and reference, and then launching a free-wheeling attack on this programme ; (iii) appealing to the authority of a number of modern masters, particularly Dewey, Heidegger, and Wittgenstein, all of whom are supposed to have expressed the same exhortation to abandon philosophy as we know it and move over to some sort of hermeneutics. B riefly: (i) needs no comment. (ii) is potentially interesting. I don't think that Rorty says enough to show that the epistemologist must adopt any particular theory of truth or reference. And I think that his attacks on Field, Putnam, and others are confused. But there is an intelligible strategy here, and if it succeeded something significant would have been shown. (iii) seems to me to turn the fact that these thinkers had enemies in common into a claim that they were making the same attacks on these enemies. It would be easier to settle this if Rorty's taste in heroes did not focus so consistently on such elusive writers. So you can see why I am not impressed. That is not to say that Rorty's alternatives to philosophy are illusions. What seems to me an obvious illusion is j ust the suggestion that they are alternatives we must follow.

University of B ristol NOTE I See for example Margaret D. Wilson, Descartes (London: Routledge, 1978), especially Chapter 1, Sec. 8, Chapter 4, Sec. 1 , and the Conclusion. As I read Wilson, Descartes, as she reads him, does not possess a working notion of evidence, and sees the grounds for assent in more archaic terms. Descartes, if this is right, did not really take a stand with respect to (a). He does not deny (b) either, as I have stated it, though he certainly thinks that all systems except one are (demonstrably) wrong.

MARY HESSE

' H ETE R O D OX ' , 'XENODOX', AND H E R M EN E U T I C D I A L O G U E

In his discussion ofthe 'heterodox' and the 'xenodox' Philip Pettit mentions a comment of mine in the discussion at Lancaster, but without developing either the point of the comment or the example in terms of which I explained it. The matter is of some importance, since it lies at the centre of the concept of 'hermeneutic' philosophy, and according to the private response Pettit quotes from Rorty, it seems that it is misunderstood by Rorty himself. Let me therefore briefly recapitulate my argument. Pettit's point is this. If 'heterodox' discourses are those that entail mutual contradictions, and 'xenodox' discourses are those that are con­ ceptually quite different but not incompatible, then no pair of discourses can be both heterodox and xenodox. (By a 'discourse' I mean a set of internally consistent doctrines that can be expressed in a particular con­ ceptual vocabulary). In other words, if discourse A entails statement S, and discourse B entails statement not-S, then A cannot be 'expanded' to take in B. On the other hand, if there are no such contradictions between A and B, then A cannot be subject to 'self-correction' from B. And yet her­ meneutic philosophy is often said to imply both an 'expansion of horizons', and 'self-correction' in the light of other discourses. How can this be so? It is true that Pettit does not give the above definitions of 'heterodox' and 'xenodox' in so many words, but they seem to be implied in what he says about entailment, assertion and denial on p. 73, and if this is not what he means I am unable to see how his argument even gets off the ground. I shall argue against him that although a pair of discourses cannot, except trivially, be both heterodox and xenodox according to his definitions, nevertheless in real hermeneutic situations discourses can be both expand­ ed and self-corrective. There is one trivial way to see how discourses may be both heterodox and xenodox given Pettit's definitions. This is to regard at least one of the discourses, say 'A', as separable into two logically independent parts A l and A 2 , such that A l is inconsistent with B and A2 is conceptually different from but consistent with B. A consistent extension plus self-correction of A may then be effected by conjoining A 2 and B. B ut this is far from being a good model of what Gadamer, for example, means by the 'fusion of horizons', and the way in which it falls short is instructive 87 A . J. Holland (ed.), Philosophy, Its History and Historiography, 87-90.

88

MARY HESSE

for our purpose. The trouble is that discourses are bodies of doctrine having a certain unity, both of conceptual vocabulary and of mutual coherence of doctrines. Thus the mere splitting of a discourse and conjunction of part of it with another is an artificial and ad hoc procedure. Let us look at the example of orthodox western medicine (A) and the lore of faith-healing (B). Sup­ pose that initially B is taken to include the doctrine 'laying-on of hands in particular ritual circumstances has physically therapeutic effects' (S). A includes the doctrine 'Such laying-on of hands does not have physically therapeutic effects'. It looks as though the discourses are heterodox and not xenodox, and so on Pettit's understanding they are. But wait. The situation has been misdescribed, because it is not correct to say that the relation between B and S, and A and not-S respectively, are relations of 'entailment'. Both A and B are fairly loose and flexible sets of doctrines, held together not (or not only) by relations of premise and entailment, but by mutual plausibility, mutual support by analogy, historical precedent, experience and natural inference, in short by the family resemblances among various doctrines expressing a particular unified view of the world. If we are faced with beliefs in, say, divine intervention in the natural world, the saving effects of worship, penitence and prayer, a sacred text containing accounts of divine healings, etc. etc., together with a belief in the laying-on of hands, we should have no difficulty in guessing that all these beliefs and practices go together in a relatively coherent and unified way of life and thought. Similarly if we came upon beliefs that the cosmos is a closed network of cause and effect, that nothing can be known except what is directly or indirectly confirmed by experiment, that the influence of the mind on bodily ills is minimal, etc. etc.. In neither case do these beliefs make up necessary and sufficient sets of premises for the discourse, which can then be defined as the closure of premises and entailments. And yet the beliefs are not arbitrarily stuck together - their coherence is exhibited by developing the discourse by natural analogies, tradition, negotiation and argument from experience. A discourse is a dynamic system - some of its doctrines may be dropped or added without it ceasing to be recognisable as the same or a closely related discourse. In terms of this picture of discourses, can orthodox medicine and faith­ healing exhibit both extension into each others' domains (,fusion of horizons') and mutual correction? Surely it happens all the time. Consider the following scenario. Proponents of orthodox medicine come to recognise the 'power of mind over body'. They notice how social and ritual

'HETERODOX'. 'XENODOX·. AND HERMENEUTIC DIALOGUE

89

circumstances affect people's mental attitudes. This leads them to drop some of their previous doctrines (which they can consistently do, since the doctrines are not entailed by any doctrines that are retained), specifically 'the influence of mind on bodily ills is minimal', and 'laying-on of hands does not have physically therapeutic effects'. They have thus extended their horizon to assimilate psycho-somatic effects that were previously foreign to their thinking, and have corrected some of their previous beliefs in the light of a foreign discourse. A similar story might be told about how faith-healers take in some aspects of orthodox medicine and drop some of their own initial doctrines. No doubt in practice there will remain irre­ ducible differences of concept and contradictions between the two discourses, but since both are dynamic sets of beliefs, there are no necessary limits to the process of fusion. There is no clearly defined core of orthodox medicine or of faith-healing which cannot be breached by negotiation and argument. Moreover, any interesting analysis of the pro­ cess will remain completely unmoved by a philosophical request to 'give the criteria of identity of a discourse'. There are no such criteria, and why should we have any need of them? The practical upshot may be that A comes to incorporate more of B than B of A , or vice versa, or a new discourse C may emerge from the dialectic of A and B , or important areas of conflict may remain. It is almost embarrassing to have to rehearse such elementary matters. They are of a sort that is exemplified every day not only in social, legal, and political argument, but also in philosophy when philosophy is actually being done, as opposed to being talked about in inappropriate logical straight-j ackets. And yet neither of Pettit's suggested resolutions of the heterodox/xenodox dilemma takes account of real scenarios of this kind. His first suggestion is that A and B are merely pragmatically distinct in not being easy to live in together, but not inconsistent, as in duck/rabbit perceptions. The pragmatic distinction is true of the medical example as far as it goes, but it fails to take account of the elements of fusion and self-correction crucial to hermeneutic dialogue. The second suggestion, quoted from Rorty, is that we have to engage in hermeneutics (,shuffle diplomacy') to discover whether a pair of discourses is heterodox or xenodox, presupposing that they are one or the other, and that this is what it is important to discover. But this is completely to neglect the fact that discourses are not -static and well-defined systems of logical entailment, but loosely structured and dynamic bodies of doctrine. There need not be any answer to the question 'Are they heterodox or xenodoxT unless they

90

M A RY H E S S E

are artifically frozen into a form never realistically held b y anybody a t any time. And to do this is to trivialize the process of negotiation and argu­ ment, and to degrade the concept of human reason.

University of Cambridge

PHILIP PETTIT

REPLY TO MARY H E S S E

The debate which Mary Hesse opens up deserves an early death. I t i s born of confusion - some of it spawned, no doubt, by me. Here are the relevant claims made or implied in my paper: 1.

2.

3. 4.

If a discourse D 2 i s incommensurable i n Rorty's sense with D I - i.e., its claims cannot be judged there - then D2 is not heterodox from the standpoint of Db i.e. it does not contradict DI·

Rorty's discussion of incommensurable discourses suggests that in such a case D2 will be xenodox - strictly speaking, completely xenodox - from the standpoint of D I : it will address different propositions. Given 1 , the hermeneutic investigation of D2 by someone in D I cannot lead to self-correction. The same result follows from 2, since (complete) xenodoxy entails non-heterodoxy.

Assuming the truth ofthese claims, I offered a conjecture about a reason why the hermeneutic investigation of D 2 by someone in D I could lead to something like self-correction: that D I and D2 are in pragmatic conflict. Conceding the truth of the claims, Rorty made the point (in his private correspondence) that the hermeneutic investigation of D2 is typically pursued in ignorance of whether D2 will turn out to be (completely) I xenodox or heterodox. Mary Hesse ignores the claims and discusses a different issue. Or rather issues. She asks whether the fact that D2 (faith-healing) is heterodox from the point of view of D I (orthodox medicine) means that it is not xenodox in any degree. She answers, correctly, that it does not : D2 may be incompletely xenodox. Secondly, she asks whether in such a case the hermeneutic investigation of D2 by someone in D I may lead to the sort of self-correction described by Gadamer as fusing horizons. She answers, and again I agree, that it may 91 A . J. Holland (ed.), Philosophy, lts History and Historiography, 91-92.

92

P HILIP PETTIT

do so, illustrating the claim by reference to the example of medicine and faith-healing. There is no debate here, but there may be a lesson. This is that hermeneutic inquiry presupposes a sense of 'incommensurability' - if it is to be conceptualised by the use of that term - which allows heterodoxy; that it is motivated by a concern, not j ust with the completely xenodox, but also with what is incompletely so. The lesson can be illustrated by a matrix. It is that hermeneutics may be pursued under any of the three assumptions represented by boxes 2, 3 and 4 but not under that associated with box 1 : this, unsurprisingly, since the assumption is incoherent. Heterodox

Non-heterodox

Completely xenodox

1.

2.

Incompletely xenodox

3.

4.

Australian National University, Canberra NOTE I The problem addressed by Rorty and me is raised by claim 1 (and 3) on its own. In my paper I failed to make this clear, relating the problem to the stronger claim 2 (and 4).

PART II

PHILOSOPHY IN THE SEVENTEENTH CENTURY

G. MacDONALD ROSS

O C C U LT I S M A N D P H I L O S O P H Y I N T H E S EVENTEENTH CENT U RY

I . THE NEGLECT OF OCCULT I NFLUENCES ON P H I L O S O PHY

It is well known that the heroes of the seventeenth-century scientific revolution were themselves by no means free of the occultist modes of thought from which they were supposed to be rescuing the human mind. Far less attention has been paid to occult tendencies in the philosophy of the time. I Since there was no sharp distinction between philosopher and scientist in the seventeenth century, it would be most surprising if the savant wearing his philosophical thinking-cap were somehow immune from occult influences to which he was prone as a scientist. The main purpose of this paper will be to suggest a few such influences. A secondary purpose will be to draw some more general conclusions about the defina­ bility of occultism, and its demarcation from philosophy and science. One of the reasons why historians of philosophy have tended to over­ look occult influences is that there is much greater indeterminacy of inter­ pretation in philosophy than in science.2 In the history of science, there used to be considerable ideological prej udice against the very idea that occultism could co-exist with rational science in one and the same mind, let alone be inextricably bound up with it. The prejudice eventually receded in the face of overwhelming, unambiguous empirical evidence, such as the Portsmouth collection of Newtonian manuscripts on alchemy and on the prisca sapientia. 3 Philosophy, on the other hand, is too abstract for it to be generally possible to pin down a philosopher's meaning as unambiguously occultist. And the area of potential contamination with occultism is precisely the area of greatest indeterminacy in interpretation. A further complication is that many non-philosophical beliefs intruding into a philosopher's writings can be interpreted as religious. For certain positivists, anthropologists, and others ,4 this makes no difference, granted that metaphysics, religion, and occultism are all equally meaningless superstitions anyway. For them, the only demarcation which matters is that of all three from genuine science. Of course, such an attitude cannot be shared by a religious historian of philosophy. S urprisingly, perhaps, even atheist philosophers tend to treat religious beliefs with special 95 A. J. HoI/and (ed.). Philosophy. Its History and Historiography. 95-1 15.

G. MacDONALD ROSS

96

respect. They take off their hats when entering churches, and they allow religion to be intellectually respectable when occult superstition is not. By piously labelling extraneous beliefs as religious, historians of philosophy side-step the awkward issue of demarcating metaphysics from super­ stition.

2.

OCCULT I S M AND RELIGION

I shall therefore start by considering certain problems over the demarca­ tion between philosophy and occultism which are subject to contamination by a religious dimension. I shall discuss three topics in particular : disembodied spirits, embodied spirits, and causality. First, that of dis­ embodied spirits. It is difficult enough to draw a line between religious and non-religious spirit beliefs in twentieth-century England. Presumably angels belong to religion, fairies not. But what of ghosts ? And even among religious people, an active belief in angels might seem odder than a belief in the possibility of communicating with the souls of the departed. In the seventeenth century, the categorisation of beliefs was very different. All philosophers believed in angels, if only on Biblical authority.5 Conversely, on the same authority, any attempt to communicate with the dead constituted the dreadful sin of necromancy. It was in any case generally considered impos­ sible for the spirits of the dead to return, or to communicate with us in any way. The single exception was the miracle of Christ's return after the resurrection.6 The Biblical report of the success of the Pythoness of Endor in summoning the soul of S amuel for King S aul was frequently dismissed as a delusion.? Similarly, ghosts were variously explained away as melan­ cholic delusions,8 as evil demons taking on the shapes of the dead,9 or as detached 'astral bodies' (the semi-material vehicles of the now departed immortal souls). 10 As for seances, the modern fashion is for summoning the spirits of the dead. In the seventeenth century, mediums avoided such overt necro­ mancy, and communicated instead with angels, or with good fairies. I I The distinction between the two was far less significant than we might expect. The essential religious belief was that there were all sorts of spiritual beings in Heaven and on Earth. It hardly mattered if one added various ranks not explicitly mentioned in the Bible - the spirits and genii of Graeco-Roman mythology, the celestial hierarchies of Cabalism and of Neoplatonism, or the fairies and gnomes of folk-lore. Thus, Robert Boyle approved the

OCCULTISM AND PHILOSOPHY

97

project of the Reverend Robert Kirk of Aberfoyle to provide empirical confirmation of Christian spiritualism by assembling reported sightings of fairies, elves, and fauns in the Highlands of Scotland. 1 2 If we now turn to theories of embodied spirits, we find it no easier to put a finger on what makes a belief superstitious rather than religious or philosophical. Descartes' account of the soul is an excellent example of the conflicting ways in which a single theory can be classified and evaluated. Most of his critics have accused him of superstition, but on a variety of incompatible grounds. Ryle's catch-phrase 'the ghost in the machine' catches the spirit all rightY However, Ryle himself was only interested in the logical mistake which led Descartes to postulate mental events and entities underlying dispositional properties. He did not expand on the locus of the ghost in Cartesian ontology. Most twentieth-century dualists would approve of Descartes' insistence on the absolute immateriality of the soul, precisely because it raised dualism above the primitive superstition of the soul as a quasi-material ghost trapped within the spatial confines of the human body. If Descartes was superstitious, it was because he was not wholly successful in escaping from the old modes of thought. In particular, he retained the traditional belief in a tenuous spirit mediating between mind and body.1 4 And though he insisted on its strict materiality, this did not distinguish him from the majority of occultists, who also treated spirits as consisting of a highly rarefied, and ghostly form of matter. As for his mechanical explanations of mental events, they owed as much to analogy and sheer invention as rival spiritualist accounts , such as those of the Paracelsians. Materialists, on the other hand, commend Descartes for his attempt, however half-baked, to explain mental functions in terms of the motions of small particles of matter obeying the same mechanical laws as gross bodies. For them, his superstitiousness consists in the pious retention of an occult, immaterial soul , . which could have no intelligible function in explaining the behaviour of the whole person. Although Descartes himself probably saw the unity of thought and extension in Man as a religious mystery,15 he is often accused of inconsistently treating the soul as if it were a little ghost sitting in the pineal gland, watching what was going on in the head, and magically moving the gland from time to time. 1 6 Commen­ tators like Kenny and Ree emphasise the naIvety of this position by labelling it the 'homunculus concept'P Somewhere between the two extremes of dualist and materialist critics, Henry More ( 1 6 1 4-1 687) complained, not about his belief in a spatially

98

G. MacDONALD ROSS

extended ghost inhabiting the body, but precisely about the lack of it. For More, Descartes' immaterial substance existing nowhere was the ultimate occult entity. He insisted that spirits must occupy space, and ridiculed Descartes' version of immaterialism as nul!ibism, or 'nowhereism'. 1 8 The third topic i n which religion complicates the demarcation between occultism and philosophy is that of causation. As long as attention is confined to extremes, there is a clear enough contrast between the new, mechanistic philosophy, and occultism. Mechanists believed that ulti­ mately the only medium of causal interaction was pushing, or impact between material bodies. Occultists believed in all sorts of interactions : mutual attractions and repulsions (sympathies and antipathies); influences operating immediately over a distance (those of the stars, or of magic spells); the effects of incantations; purposeful, vital principles, and so on. Scholasticism fell s omewhere between the two extremes, with its final causes, substantial forms, and hidden 'virtues'. Mechanists got into considerable difficulties trying to explain attractive forces, such as the cohesion of solids, gravitation, and magnetism, in terms of corpuscles pushing against their neighbours. One can sympathise with Newton's refusal to 'invent hypotheses' about how gravitational forces were mediated.1 9 On the other hand, one can equally sympathise with Leibniz's accusation that Newton was admitting occult forces, by explain­ ing gravitation as due to 'gravitational forces', operating immediately at a distance, and knowable only by their effects.2o We may now prefer the public Newton who confined himself to finding mathematical formulae to describe the phenomena - yet the private Newton was as convinced as Leibniz that there had to be some underlying mechanism involving aetherial particles.21 On the other side, many occultists went out of their way to translate magic powers into strictly mechanist terms. The favourite medium for occult interaction was the World Spirit transmitting the influences of individual spirits. By conceiving spirits as rarefied matter, they obliterated any difference in kind between their World Spirit and the aether of the mechanists.22 Magic powers were thus brought into the same class as attractive forces - given (or so it was believed) in experience, and explic­ able by a supposed hidden mechanism. Religion adds an extra dimension of complication. It had long been believed that miraculous effects could be brought about in religious con­ texts : a priest's words could transform bread and wine into the flesh and blood of Christ, and holy water or saints' relics could cure the sick. More

OCCULTISM AND PHILOSOPHY

99

peripherally, the kings of England and of France were believed to have the power of curing scrofula (the 'King's Evil'). In England, at least, the normal view was that this was a divine sacrament performed by God's annointed, rather than magic. The ceremony took place in the Royal Chapel, with a priest officiating, and in accordance with a special form of prayer included in some editions of the Book of Common Prayer. Further removed from the aegis of the Church were other healers, such as the famous 'stroker' Valentine Greatrakes ( 1 629-1 683). He himself believed that God was working through him ; but others attributed his powers to diabolical agency, or to natural forces mediated by rarefied material effiuences akin to magnetism.24 Witches were in much the same category. The orthodox view was that their powers were due to devils, and their existence was often cited as evidence for the spiritualist world-view of Christianity (for example by Robert Boyle, Henry More and Joseph Glanvill).25 Conversely, others explained their powers in purely mechanistic terms. In the seventeenth century, only a small minority believed either that their powers were irreducibly magical, or that they were wholly delusory.26 Even archetypally occultist practices such as alchemy were given a strong religious aura. No doubt, at one end of the spectrum, some alchemists carried out their experiments without any thought of God except, perhaps, for a prayer at the beginning of The Work. But equally, many a village wise-woman would charm away a wart without worrying whether she was doing it with demonic assistance, with spiritual effiuences, or just doing it. More reflective alchemists were very concerned with the theory of what they were doing. They generally believed that they had a special understanding of the natures of things through divine illumination. It was this that enabled them to see behind the appearances of things to their occult virtues, which they could then manipulate.27 At the other end of the spectrum, for example in the so-called 'Rosicrucian Manifestoes', the actual practice of alchemy was even forbidden: the extraction of gold from base metals was not to be understood as a physical process, but as an allegory of the purification of the sou1.28 In effect, the way of alchemy was little other than a fringe religion. Again, in any particular case it is far from clear whether we have to do with practical technology, superstitious magic, mystical religion, or some blend of all three. There was a comparable ambiguity within religion itself. Most Pro­ testants denied all miracles: not only was the 'Age of Miracles' past, but

G. MacDONALD ROSS

100

no magical changes were brought about in the Eucharist or other sacra­ ments.29 Even Catholics tended to go along with the denial of religious magic, and to attribute any non-natural event directly to God himself. However, this did not necessarily constitute any real limitation on the powers of the priesthood. There is no practical difference between God's granting a priest the power to change wine into blood on uttering certain words, and himself bringing about the change on the occasion of the utterance of those words. Indeed, an occasionalist would have to say that there was no difference at all between the sacramental event, and anything else that happened in nature. During the seventeenth century, a significant body of philosophers, including Descartes, Malebranche, Spinoza and Leibniz, came to believe that there was no such thing as a real causal connection at all. All talk of 'influence' was equally superstitious. The only question was the ultimately theological one of the principles by which God governed the evolution of the universe. So, Leibniz in particular often phrased questions about what was natural or possible in overtly theological terms : would God create a universe which needed re-winding? Would God create two identical atoms ?30 The conclusion to be drawn from the above examples is that it is virtually impossible to force any confrontation between occultism and the mechanical philosophy. When challe.nged, occultists can easily dress up their beliefs either as essentially religious, or as not different in kind from mechanist theories. At the same time, many religious believers wanted to off-load religious magic, and at least some scientists wanted to dismiss all supposedly natural influences as superstitious, and attribute everything directly to God. The resultant conceptual chaos makes it difficult either to assert or deny occult influences on the philosophy of the time. Yet there are other areas in which at least the religious dimension is relatively subordinate. 3.

OCCULTISM

AND

PHILOSOPHY

One way of showing the importance of occultist modes of thought in seventeenth-century philosophy would be to take a range of writers who were all equally regarded as philosophers at the time, and demonstrate that many of them were heavily influenced by occultism. Such a strategy would be almost too easy: for every unsuperstitious philosopher such as Hobbes, Pascal, Descartes , Malebranche, Locke, Spinoza, or B ayle, there is a Campanella, Fludd, Kircher, Comenius, Henry More, Cudworth, or

OCCULTISM AND PHILOSOPHY

101

F. M. van Helmont. And whereas the first list is nearly exhaustive, the second is only a small, random selection. However, the evidence would not persuade anyone who was not already convinced of the importance of occult influences on philosophy. It would immediately be countered by the denial that the writers in the second list were philosophers at all. A more effective strategy would be to take figures who are now still recognized as genuine philosophers, such as B acon and Leibniz, and show how their philosophies are permeated with occultism. In particular, Leibniz's system was deliberately intended as a sort of highest common factor of mechanism and occultist vitalism. 3 1 Yet this too would meet with the reply that, insofar as Leibniz's thought is capable of contamination by occultism, he is not really functioning as a philosopher: producing theories about the world is not a proper part of philosophy. I do not myself agree that philosophy should be defined so narrowly that there were only half-a-dozen philosophers throughout the seventeenth century - and then only part-timers. Nevertheless, in order to circumvent what one might call the Thrasymachus gambit',3 2 I shall leave the obvious examples on one side. I shall limit myself to more ambiguous cases in certain topics central to our philosophical tradition : the theory of per­ ception, primary qualities, the Cartesian cogito, what was later known as the synthetic a priori, and empiricism. Before the seventeenth century, the nature of visual perception was not a point at issue between the occultist and scholastic philosophers. It was generally agreed that objects emitted infinitesimal surfaces or effiuences, often called 'intentional species', which entered the body through the pupils of the eyes. On mingling with the spirit which animated the body, they ended up as the sensory images immediately present to consciousness. The main point at issue among scholastics was whether or not there was also a simultaneous emission of a 'visual stream' searching out its objects. This was an essential element of Plato's theory, intended to account both for the phenomenon of projection, and for the observer-relativity of the objects of perception. 33 As for the occultists, the visual stream theory was virtually essential for the conceptual grounding of various superstitions, such as fascination, the evil eye, the influence of a mother's imagination on an unborn child (even Kepler, Gassendi and Descartes believed in this ),34 and perhaps also the astrological influences of the intelligences governing the heavenly bodies. The early mechanist philosophers, such as Galileo and Descartes, took great pains to explain the radical differences between their theories, and

102

G. MacDONALD ROSS

the effluences and visual streams of their scholastic and occultist contem­ poraries. 3 5 Above all, they insisted on an absolute dualism of appearance in the brain or mind of the observer, and the qualitatively different reality of the object of perception; they explained all reality, including whatever carried sensory information from object to observer, as consisting of nothing but matter in motion ; and they took it for granted that the mind was fundamentally passive in perception. Paradoxically, the new philosophers following the 'way of ideas' were even more dependent than the occultists on hidden or occult entities. According to the traditional theories, perceptible qualities were located in external objects themselves. Occultists differed from the more sober scholastics in their stress on additional, occult qualities underlying the surface ones. However, as Berkeley saw, the new dualism deprived objects of all their surface qualities by locating them in the mind, so that reality itself consisted entirely of hidden qualities known only to the philosopher or scientist. 3 6 Therefore the mechanists of the seventeenth century had a considerable problem if they wanted to maintain that they were different in kind from the magicians of old, and were not simply the first generation of successful magicians. Some occultists exploited this weakness in the new philosophy. For example, John Webster's Displaying of Supposed Witchcraft attacked the view that witchcraft provided evidence for the existence of evil spirits. For Webster, the new philosophy of the Royal Society had shown how little could be known about the true causes of things; consequently, there were no grounds for denying that witches were genuinely able to tap hidden magical virtues. 3 8 Locke, on the other hand, seems not to have noticed how little separated him from the enthusiasts he so despised. He believed that we can only guess at the 'real essences' of things. 39 Perhaps he assumed that his scepticism distanced him enough from the occultists who thought they knew. B ut as it happens, occultists themselves prudently avoided rash claims about the details of the hidden natures of things. Furthermore, Locke made knowledge of essences only contingently unattainable, since angels could become miraculously acquainted with them.4o And what could happen miraculously in a religious context, could conceivably be achieved magically in a secular one. Locke did at least see that his empiricism gave him the problem of explaining how our guesses about hidden essences could be meaningful. His solution was that, although our ideas of secondary qualities give us no

OCCULTISM AND PHILOSOPHY

103

insight into the powers that cause them, our ideas of primary ones are qualitatively identical with their archetypes, thus affording us a conceptual bridgehead in reality itself.41 Now, in view of Descartes' sharp distinction between real and perceived geometrical properties ,42 the casualness of Locke's assumption might seem quite remarkable. It is less surprising, however, if we bear in mind the survival of the traditional theories of perception. Locke himself admits that he sometimes lapses into talking as if ideas and qualities were numerically as well as qualitatively identical.43 This tendency is reinforced by his frequently describing perception in terms of ideas travelling from objects into the mind via the sense organs.44 In theory, his language should be taken as metaphorical; but in practice, it is a sign that he was allowing himself to be seduced back into the magical imagery of the effiuence theory of perception, according to which our minds are fed with the very qualities of objects themselves. It might be claimed that the real differentia of mechanist accounts of perception (as indeed of the mechanical philosophy in general) is precisely that they were mechanist, and thereby closer to the truth than their rivals. However, although mechanics itself was a strikingly successful piece of science, we should not close our eyes to the complete arbitrariness and inadequacy of the various corpuscularian hypotheses thought up by the mechanists. It is a hangover from Victorian prejudice to see theories as especially virtuous simply because they happen to be couched in materialist, atomistic terms. If the claim is diluted, so that the differentia becomes that of understanding the world in mathematical terms, then it fails to differentiate the new philosophers from occultists and NeoPythagorean mystics such as John Dee and Kepler,45 who also saw reality as fundamentally mathematical. Ironically, in the seventeenth cen­ tury itself, mathematics was popularly regarded as one of the principal occult sciences. As Hobbes wrote in Leviathan, Chapter V: "Geometry [the most part of men] have thought Conjuring".46 Advances in the understanding of perception were due to developments in geometrical optics that owed nothing to the replacement of visual streams and effiuences by corpuscles. The laws of linear perspective had been developed long before by thinkers like Leonardo da Vinci, who was working within the framework of the older concepts.47 The theory that light consisted of cohorts of material corpuscles reflected from surfaces differed little from the effiuence theory, beyond making corpuscles colour­ less.48 Effiuences and corpuscles were both on the interface between the material and the immaterial, and they both gave rise to similar problems,

104

G. MacDONALD ROSS

such as why images were invisible sideways on, and why there was no mutual interference when they crossed paths. The alternative theory that light consisted of wave motions in a material aether had its own insuper­ able difficulties. The corpuscular hypotheses were invented in order to satisfy an ideological need to account for everything materialistically; but if anything they hampered progress in scientific optics. It is far from obvious that the new philosophers' wrong guesses about the nature of light should be regarded as of a different order of rationality than those of their predecessors. Corpuscularianism was not even fatal to the visual stream theory. Just as fascination could be explained as an emission of rarefied matter,49 so was there no reason in principle why visual streams should not be accounted for in a similar way. The same goes for other alleged phe­

nomena, such as extra-sensory perception. John Wilkins , for example,

explained short-range ESP as due to material, magnetic effiuences. Long­ range ESP, on the other hand, he attributed to the assistance of guardian angels, who were capable of instantaneous communication without any material medium. 50 This all goes to show that the availability of a corpuscular explanation is really irrelevant to whether a belief is to be classed as superstitious or not. Thus, Robert Fludd ( 1 5 74- 1 637) claimed that in a state of ecstasy, a 'geomancer' could emit mental rays which would penetrate the macrocosm and return with warnings of the future. 51 This belief would not deserve to be taken any more seriously ifhe had added that the rays consisted of very fine material particles. Similarly, in the following century, Franz Mesmer ( 1 734- 1 8 1 5) tried to make hypnotism intellectually respectable by explain­ 2 ing it as a form of magnetism. 5 Ironically, it is precisely his materialist explanation that has dated. There is an interesting and instructive twist to the changing fortunes of mechanist and visual-stream theories of perception. The mechanists can fairly be criticised for emptying the baby with the bath: they over-reacted against visual streams by refusing the mind any active role whatever in the process of perception. Kant's theory of constructive imagination was an important corrective. The irony is that some modern magicians have seen Kant, and his follower S amuel Taylor Coleridge, as having contributed towards the reinstatement of a magical world-view in which imagination rivals science as '1 source of knowledge and power. 5 3 I shall now move to what Descartes had to say about our knowledge of the hidden reality underlying the perceptible qualities of things. Super-

OCCULTISM AND PHILOSOPHY

105

ficially, his approach is very similar to Plato's in its dependence on the magical principle that like is known by like. For Descartes, the world is really only matter in motion. Particular motions are represented by sensory images, which are themselves motions of spiritual matter in the brain of the observer. General truths about extended substance are known by the immaterial, rational soul. Such knowledge includes that of 'first philoso­ phy' itself, together with the sciences of geometry, mechanics, and physics. In order to explain how the soul becomes acquainted with the essence of matter, Plato had postulated an ante-natal acquaintance by contact. Descartes' account owed much more to the Christian tradition of mystical meditation,54 and the related approaches of intellectualist magicians and alchemists. Will-Erich Peuckert, a leading historian of occultism, defines a magus as one who follows the 'light of nature', as contrasted with the , cabalist, who follows the 'light of grace .55 If he is right, then the core of Descartes' philosophy is both magical and cabalistic.56 H aving arrived at the indubitable proposition, 'I am', Descartes appealed to the light of nature in order to prove that the information carried in the idea of a perfect being could not have come from himself. Just as the magus used the light of nature to read the 'signatures' of the macrocosm in the microcosm, so Descartes argued back from the signa­ ture or idea of a perfect being which he discovered embedded in his own essence, to the macrocosmic existence of that being. It was then by the grace of the non-deceiving God that he was assured that his clear and distinct ideas (those unsullied by perceptual imagery) corresponded to the archetypes on which God had modelled reality. Descartes did not himself call the criterion of clear and distinct perception the 'light of grace', but this is undoubtedly what he meant; and he was heavily criticized for appealing to as subjective a criterion as that of any enthusiast. 5 7 Around the time of his famous dreams, Descartes first heard about the Rosicrucians, and tried to make contact with them in case they had forestalled him in the discovery of the new wisdom. He was rapidly disillusioned. 58 On the other hand, the main thrust of his epistemology is hardly distinguishable from that of the alchemist Oswald Croll, in the 'Admonitory Preface' to his Basilica Chymica. Croll also slates the sterility and authoritarianism of university teaching, and commends the reader to turn inwards to discover God, "the one Rector of that great university in the sky . . . Everyone learns by going back to the source, and hence to God, who created knowledge for man . . . The true route to true wisdom is the [v6)S. 0'f:cxut6v of the Delphic Oracle . . . By starting out from his own self,

106

G.

MacDONALD ROS S

and understanding himself, Man can see and understand everything in himself as in a sort of divine mirror . . . The notion of all things is born with us . . . God is known by the light of nature . . . Man can derive most of his knowledge of what he is from contemplation of himself'.59 Neither sceptics nor empiricists should be surprised at Descartes' using the conceptual apparatus of magic and occultism to justify his rationalist epistemology. Both sorts of philosopher deem it a superstition to suppose that human reason alone can attain non-trivial knowledge of external reality. Locke (or at least the Locke of the first two books of the Essay) saw himself as both sceptical, and what we now call empiricist. He correctly diagnosed the crucial rOle played by innate ideas in justifying rationalist claims to bridge the gulf between knower and known. Yet by Book IV, Locke had himself lapsed back into comparable claims. Unlike philosophers such as Hobbes and Leibniz, who reduced all logical truth to definitions ,6o Locke wanted to preserve a contrast between 'trifling' verbal propositions,61 and the real knowledge we had in mathematics, ethics, and theology. He based it on intuitions of the 'agreement', and 'disagreement' or 'repugnancy' of ideas.62 However, he could give no account of what the intuition of agreement or disagreement consisted in, beyond appealing to the usual metaphor of light.63 It is no accident that the terms he chose were, in his day, standard synonyms for the 'sympathy' and 'antipathy' of the occultists.64 In the context of the rest of his philosophy, any necessary connections between distinct ideas had to be either non-existent or magical; and his choice of terminology is a tacit admission of that fact. Strict empiricism was in an even worse position than rationalism for avoiding superstition, since it could provide no a priori criterion for judging between credible reports of experience. This aspect of empiricism would have been more evident in the seventeenth century than now, since it is only since the nineteenth century that the term 'empirical' has acquired its present, commendatory sense. Previously it meant only 'anti-scientific', or 'quack', as in the expression 'empirical remedy'.65 In antiquity, the empirical school of medicine rejected all forms of treatment based on theory, and confined itself to what had been found successful in practice. Consequently, the empirical doctor did not differ in kind from the village wise-woman, or witch. Broadened into a general, sceptical philosophy (as with Sextus the Empiric),66 empiricism had no basis for excluding generally accepted beliefs about occult phenomena. Certainly some empiri­ cists were less credulous than others ; but this was more a question of temperament than of theoretical stance. It is therefore hardly surprising

OCCULTISM AND PHILOSOPHY

1 07

that there should be as much superstition in the Sylva sylvarum of Francis Bacon, the so-called father of modern empiricism, as there is in the Magia naturalis of the occultist della Porta, and rather more evidence of practical experimentation in the latter.67

4.

THE RHETORIC OF THE OCCULT

The examples that I have given, and others that could be adduced, show that in the seventeenth century there was no clear line of demarcation between occultism, philosophy, religion, and science. Over large areas of belief, it simply was not the case that the modern philosophers had a monopoly of truth, meaningfulness, evidence, reasonableness, or even of a rational scheme of concepts. It would be possible to cobble together some sort of a definition of occultist as opposed to rational philosophy, by taking extreme examples, and setting up contrasting paradigms in terms of families of overlapping characteristics. However, such a descriptive procedure would have limited value. It would be tied to a particular historical period and culture, and it would be incapable of clarifying either the area of overlap between superstition and rationality, or the process of conceptual change. Rather than trying to understand the contrast merely through characte­ ristics internal to the belief-systems themselves, it is necessary to take into account an additional, intentional dimension, namely the perceived relation between the occult philosopher himself, and orthodoxy.68 The position of the occultist is closely analogous to that of the heretic, whose status depends as much on the fact of his being branded as a religious deviant, as on what he actually happens to believe. The occultist is essentially an intellectual deviant. So in order to understand the contrast between occul­ tism and orthodoxy, we have to take into account not merely its internal logic, but also the rhetoric of the labels. As a deviant, the occultist is faced with three possible strategies : he can go underground,69 he can protest his innocence, or he can embrace the label. 7o During the Middle Ages and the Renaissance, the occult sciences were practised in relative secrecy, if only out of fear of religious or secular persecution. By the seventeenth century, however, the accelerating col­ lapse in respect for the authority of the Church and the universities allowed advocates of occultism an unprecedented freedom of thought. For a while, many occultists saw themselves as a legitimate wing of the intellectual revolution against the scholastic establishment.72 As experimentalism and

108

G. MacDONALD ROSS

mechanism gradually took root, a significant body of occultists tried to make their beliefs intellectually respectable by formulating them in corpuscularian terms, and by trying to verify them experimentally.72 Then, as now, occultists were often at least as meticulous about their experi­ mental methodology as orthodox scientists. At the same time, others willingly described their own philosophy as 'occult', and positively gloried in anti-rationalism and anti-materialism. When the resurgence of occultism was at its height, at around the middle of the century, it was far from clear what the orthodoxy of the future was going to be : it could have been represented by vitalist physics, astrology, Hermetic medicine, and alchemy.74 Only towards the end of the century did it become evident that mechanism had replaced scholasticism as the new orthodoxy. An integral part of the process of establishing the new orthodoxy was a rigid defining of its borders with religion and superstition, and the ruth­ less purging of all occult tendencies. For the first time in nearly a century, it was clearly decidable what was to count as irrational belief. The immediate consequence was that the survivors of extreme occultism had once more either to go underground, or to identify themselves openly as committed opponents of philosophy and science rather than as fellow­ travellers. This is precisely what we find at the beginning of the eighteenth century, with the emergence of consciously anti-rational occult societies, such as Freemasonry and Rosicrucianism, and of anti-scientific romanticism.15 An excellent illustration of the changing climate is provided by Leibniz's attitude to alchemy.16 In his youth (in the 1 660s), Leibniz was heavily involved in alchemy: his first job was as secretary to a society of alchemists at Nuremberg, he participated in a number of gold-making enterprises (though, unlike Newton, he seems to have done no actual laboratory work), and he probably owed his posts at Mainz and Hanover to his reputation as an adept. He maintained at least a casual interest in alchemy for the rest of the century. It is quite possible that he deliberately destroyed documentary evidence of his activities at Nuremberg, and he certainly lied to his biographer about how he felt when he was young. In his early days, there was the single art known indiscriminately as 'alchemy', or 'chemistry'. By the time he was an old man, chemistry was an established science, and alchemy an outmoded and shameful superstition.?? Just as the term 'alchemist' had acquired a pejorative sense, so a whole range of beliefs became defined as beyond the pale for a serious philo-

OCCULTIS M AND PHILOSOPHY

109

sopher or scientist.1 8 All the same, the concept of deviance by itself no more provides an adequate line of demarcation between orthodox and occult sciences than a purely internalist criterion. There are many forms of intellectual deviance which have nothing to do with occultism or superstition. Yet it is possible to identify a group of 'occultist' attitudes and concepts, often clustered around particular ancient and mediaeval texts, which have historically tended to remain on the fringes of the intellectual world. Internalist historians would say that occultism has inevitably remained underground because it is objectively antipathetical to rational thought and successful science. Others would explain its ostracism as resulting from the anti-authoritarianism implicit in its stress on individual enlightenment. But such questions cannot sensibly be discussed without involving live philosophical controversies about the nature of rationality itself My historical conclusion is that the concept of occultism is of little use when there is no well-defined orthodoxy for it to deviate from. In a period of transition and intellectual ferment, the widespread mingling of different approaches robs the term of much of its meaning. As far as the seventeenth century is concerned, its most valuable use is the negative one of reminding the historian that it was above all a century in which philosophy and science cannot be sharply demarcated from the occult.

University of Leeds NOTES 1 I treat the occultist background to seventeenth-century philosophy much more fully in my chapter, 'Okkulte Stromungen im 1 7. lahrhundert', in the forthcoming new edition of Ueber­ wegs Grundriss der Geschichte der Philosophie, Series V, Part 1. 2 In the present paper, I have space only to interpret seventeenth-century philosophers, not to discuss the very difficult and important problems about the theory of interpretation - in particular the challenges raised by Donald Davidson in his 'On the Very Idea of a Conceptual Scheme', in Proceedings of the American Philosophical Association 47 ( 1 973-74): 5-20. I say something li�out them in my paper, 'Angels', delivered to the Northern Universities Philoso­ phical Society conference at York in 1982. Space also prevents me from discussing how we should transcend the limitations of the by now familiar distinction between 'internalist' and 'externalist' explanations. See, for example, 1. R. Ravetz, 'Francis Bacon and the Reform of Philosophy', in Science. Medicine and Society in the Renaissance, edited by A. G. Debus (London: Heinemann, 1972), Volume I I : 97-1 1 9, esp. pp. 97-98; 1 . R . R. Christie and J.V. Golinski, The Spreading of the Word: New Directions in the Historiography of Chemistry 1600-1 800', in History of Science 20 ( 1 982): 235-266, esp. pp. 235-237. Nor can I go into the fascinating question of why Anglo-Saxon historians of philosophy have tended to confine

1 10

G. MacDONALD ROSS

themselves to intemalist history ('why' both in the sense of understanding their rationale, and of placing their intemaiism in the context of philosophy as a social institution). See David Bloor, Knowledge and Social Imagery (London: Routledge, 1976), Chapter 1 . 3 Cf. Betty Jo Teeter Dobbs, The Foundations of Newton 's Alchemy (Cambridge: Cambridge University Press, 1975). 4 For example, A. J. Ayer, Language. Truth and Logic (London: Gollancz, 1936); Bronislaw Malinowski, 'Magic, Science and Religion' ( 1 925), reprinted in Magic, Science and Religion and Other Essays (New York: Doubleday, 1954): 1 7-92. 5 Most sceptical are Hobbes, Leyiathan, Chapter XXXIV, and Spinoza, Cogitata metaphysica, in Opera, edited by C. Gebhardt (Heidelberg, 1924-1926), Volume I, p. 275. 6 For example , Marin Mersenne, Quaestiones celeberrimae in Genesim (Paris, 1623), p. 583ff. And even if necromancy were possible, it was still a sin: Leyiticus XIX.3, XX.27; Deuteronomy, XVIII . l O-1 1. 7 For example, John Webster, The Displaying of Supposed Witchcraft (London, 1677), p. 1 65. Webster also criticises the Authorized Version for referring to her as the 'witch' of Endor in the headings to I. Samuel XVIII. The translators may well have been influenced by King James' having called her a witch in his Daemonologie (Edinburgh, 1 597), p. 29. 8 For example, Reginald Scot, Treatise upon the Nature and Substance of Spirits and Diyels, etc. (London, 1584), Chapter XXXII. 9 For example, King James, op. cit. (n. 7), p. 6 1 ff. ; Johann Heinrich Decker, Spectrologia (Hamburg, 1 690), p. 145ff. 1 0 For example, John Webster, op. cit. (n. 7), pp. 3 1 1-320. 1 1 For example, Meric Casaubon, A True and FaitJiful Relation of . . . Dr John Dee . . . and some Spirits (London, 1659), and William Lilly, History of his Life and Times, from the Year 1602-1684, Written by himself, edited by Elias Ashmole (London, 1 7 1 5), reprinted as The Last [sic!} of the Astrologers, edited by K. M. Briggs (London: Folklore Society, 1974). 12 Robert Kirk, The Secret Commonwealth, edited by S. F. Sanderson (Cambridge & Totowa: Brewer, for the Folklore Society, 1976). Cf. Robert Boyle, Works, edited by T. Birch (London, 1 744, Volume 1 (Life, & c.), pp. 1 19-1 30. 13 Gilbert Ryie, The Concept of Mind (London: Hutchinson, 1949), p. 15. 1 4 A good example of Descartes' fudging of the problem of interaction between thought and extension is to be found in his First Thoughts on the Generation of Animals (OeuYres, edited by C. Adam and P. Tannery, Volume XI, pp. 5 1 8-5 19), where he talks as if spirits were such that only an infinitesimal force is needed to deflect them - in other words, they are exactly on the interface between the material and the immaterial. 15 This is how I take his remarks to Princess Elizabeth: Adam & Tannery, Volume III, esp. pp. 665-666 and 691-693. 1 6 Not only materialists have mlide this point. It is expressed very sharply by Leibniz in the Theodicy, Part I, §§60-6 1 . Leibniz's diagnosis of Descartes' mistake as due to a defective formulation of the laws of motion is not only a very perceptive piece of historical reconstruc­ tion, but a shining example of sympathetic historiography. It is, however, just possible that Descartes was more explicit in lost M S S studied by Leibniz when in Paris. 1 7 A.J. P. Kenny, 'The Homunculus Fallacy', in Interpretations of Life and Mind, edited by M. Grene (London: Routledge, 1972): 65-74; Jonathan Ree, Descartes (London: Allen Lane, 1974), pp. 63-64 and 1 21-123. 18 Henry More, Enchyridion fnetaphysicum (London, 1 67 1 ), Volume I, p. 27. 19 Principia, 3rd Edition (London, 1 726), p. 530.

OCCU LT I S M AND PHILOSOPHY 20

111

This is a Leibnizian commonplace. See, for example, the Preface to the Nouveaux Essais, Academy edition (VI.6), edited by Robinet & Schepers, pp. 64-65. His comments on occult qualities on p. 68 are identical in spirit with Moliere's joke in the 3rd Interlude of Le Malade Imaginaire ( 1 673): 'BACHELlERUS : "Mihi it docto doctore I Domandatur causam et ration em quare I Opium facit dormire. / A quoi respondeo, / Quia est in eo / Virtus dormitiva, / Cujus est natura I Sensus assoupire.'" 2l Opticks, Bk. III, Pt. i. Qn. 3, ad fin. (Dover Edition, 1979), p. 40 1 ; letter to Bentley of 25.2. 1692/3, in Four Letters from Sir Isaac Newton to Doctor Bentley (London, 1756): 23-32, reprinted in Isaac Newton 's Papers and Letters on Natural Philosophy, edited by I. Bernard Cohen (Cambridge: Cambridge University Press, 1958): 300-309. 22 Marielene Put scher, Pneuma, Spiritus, Geist. Vorstellungen vom Lebensantreib in ihnen geschichtlichen Wandlungen (Wiesbaden: Steiner, 1973). 23 Helen Farquhar, 'Royal Charities. Touchpieces for the King's Evil', in The British Numismatics Journal 12 ( 1 9 16): 39-135; 13 ( 1 9 1 7): 95-163; 14 ( 1 9 1 8): 89-120; 15 ( 1 919) 141-184; Marc Bloch, Les Rois Thaumaturges (Strasbourg: Librairie Istra, 1924); Keith Thomas, Religion and the Decline ofMagic (Harmondsworth: Penguin, 1978), pp. 227-244; G. MacDonald Ross, The Royal Touch and the Book of Common Prayer', in Notes and Queries 30 ( 1983): 433-435. 24 Valentine Greatrakes, A Brief Account ofMr. Valentine Greatrak's, and divers of the strange Cures by him (London, 1666); Henry Stubbe, The Miraculous Conformist (Oxford, 1666); Joseph Glanvill, Saducismus Triumphatus (London, 1 6 8 1 ) Pt. I , p. 90fT.; Pt. II, p. 247; Robert Boyle, Works, edited by T. Birch (London, 1 744) Volume I (Life, & c.), pp. 45-53; Anne Conway, Conway Letters. The Correspondence of Anne, Viscountess Conway, Henry More and their Friends, 1642-48, edited by M. H. Nicolson (London: Milford; Oxford University Press, 1930), pp. 244-275. 25 For example, Henry More, An Antidote against Atheisme (London, 1653); Joseph G1anvill, op. cit. (n. 24). 26 For the seventeenth century, the important contrast was between 'natural' and 'demonic' magic. Boyle, More, and Glanvill were among those who insisted on the reality of demonic magic. Among those who explained witchcraft away as at most natural magic were Johann Wier, De praestigiis daemonum et incantationibus et veneficiis (Basel, 1 563); Reginald Scot, op. cit. (n. 8); Tobias Tandler, Dissertationes physicae-medicae (Wittenberg, 1 6 13); Thomas Ady, A Candle in the Dark (London, 1656); John WagstatTe, The Question of Witchcraft debated; or a discourse against their Opinion that affirm Witches (London, 1669); and John Webster, op. cit. (n. 7). It is one of the theses of this paper that what was intentionally dubbed 'natural magic' cannot be straightforwardly divided into the 'natural' and the 'magical'. Insofar as such a division can be made, Webster is probably on the magical side: assuming (with the Dictionary ofNatiol)al Biography) that he was identical with the author of Academiarum examen (London, 1654), he was described by Thomas Hall, in his Vindiciae literarum (London, 1654), p. 199, as belonging to the "Familiasticall-Levelling-Magicall temper". Cf. P. M. Rattansi, 'Paracelsus and the Puritan Revolution', in Ambix 1 1 ( 1 963): 24-32, esp. p. 29. 27 The variety of approaches to alchemy is reflected in modern interpretations. At one extreme alchemy is seen primarily as the matrix of modern technology (e.g. Robert Mu1thauf, The Origins of Chemistry (London: Oldbourne, 1966» , at the other primarily as mystical and symbolic (e.g. e. G. Jung, Psychologie und Alchemie (Zilrich: Rascher, 1 944» . Most other historians fall between these two extremes. An alternative emphasis on the rhetorical and didactic dimensions is given by Owen Hannaway's The Chemists and the Word (Baltimore &

1 12

G. MacDONALD ROSS

London: Johns Hopkins University Press, 1975). 28 Allgemeine und General Reformation. der gantzen weiten Welt. Beneben der Fama Fratemitatis. dess Loblichen Ordens des Rosenkreutzes (Cassel, 1 614). Cf. Will-Erich Peuckert, Die Rosen­ kreutzer (Jena: Diedericks, 1928); Paul Arnold, Histoire des Rose-Croix et les origines de la Franc-MQI;onnerie (Paris: Mercure de France, \(55); Frances A. Yates, The Rosicrucian Enlightenment (London: Routledge, 1972), and my review-article, 'Rosicrucianism and the English Connection', in Studia Leibnitiana 5 ( 1973): 239-245. 29 A typical Protestant attack on the magic of catholicism was that of Jean-Baptiste Thiers, Traite des superstitions (Paris, 1679); and there is much of the same throughout Hobbes' Leviathan. Cf. also Thomas, op. cit. (n. 23), Chapter 2. Note also Descartes' attempt to demystifY transubstantiation by explaining it in materialistic terms: Replies to Objections IV, sub fin. 30 This tendency is particularly evident throughout the correspondence with Clarke. 31 I expand on this theme in my Past Masters: Leibniz (Oxford: Oxford University Press, 1984). For Bacon, see Paolo Rossi, Francesco Bacone: Dalla Magia alla Scienzia (Bari: Laterza, 1957), translated by S. Rabinovitch as Francis Bacon. From Magic to Science (London: Routledge, 1968); and l. R. Ravetz, op. cit. (n. 2). 32 I refer, of course, to Thras}T'achus' ploy in reply to Socrates' point that rulers sometimes make mistakes (Plato, Republic 340C). 33 A singularly clear account of the contemporary state of play is to be found in Francesco Suarez, De anima III.ii, xvii, in Opera omnia, edited by Andre & Berton (Paris, 1 856-6 1 ), Volume III, pp. 525-526, 670-673. Plato's theory is expounded in Theaetetus 156A-E, Timaeus 45B-46C. 34 On fascination in general, see Daniel Sennert, Medicina practica VI.9, in Opera (Paris, 1641); Johann Christian Frommann, Tractatus de fascinatione novus et singularis in quo fascinatio vulgaris profligatur. naturalis confirmatur. et magica examinatur (Nuremberg, 1675); and the otherwise markedly anti-occultist Etienne Chauvin, Lexicon rationale. sive thesaurus philosophicus. s. v. 'fascinatio' Iff. FF to Ff2V]. Bacon, on the other hand, attributed any genuine cases to demonic agency; Sylva sylvarum, Cent. 10, §950. Belief in the influence of imagination on the foetus seems to derive from Avicenna and Algazel: Lynn Thorndike, A History ofMagic and Experimental Science (New York: Columbia University Press, 1958), Volume VII, p. 476. Cf. Johannes Kepler, Opera 01lJnia, edited by C. Frisch (Frankfurt & Erlangen, 1 858-7 1 ), Volume II p. 726, Volume V p. 263; Descartes, Traite de ['homme, Adam & Tannery, Volume XI, p. 1 76; Pierre Gassendi, De phantasia, in Opera, edited by Habert de Montmor & Henri (Lyons, 1658), Volume II, p. 424. 35 See in particular, Galileo, Il saggiatore (Rome, 1623), pp. 196-199 [The Assayer, translated by S. Drake (New York: Doubleday, 1957), p. 273ff.] ; Descartes, Traite de la lumiere, Chapter 1 : 'Of the difference between our sensations and the things that produce them', in Adam & Tannery, Volume XI, pp. 3-6. 36 Berkeley diagnosed Locke's failure to see the absurdity of his position as due to his belief in the 'wonderful [ = magical?] faculty' of abstracting ideas: Principles, Introduction, § 10. Leibniz avoided the problem in the first instance by claiming that ideas of secondary qualities were really confused perceptions of primary ones: Nouveaux essais, II.viii. ! 1-21 . But ultimately his solution was an almost Berkeleian phenomenalism: see my 'Leibniz's Pheno­ menalism and the Construction of Matter', in Studia Leibnitiana, Sonderheft 13 ( 1 984): 26-36. 37 I suspect hidden dimensions to the irony of J. M. Keynes' description of Newton as "the last of the magicians": cf. 'Newton, the Man', in The Royal Society: Newton Tercentenary

OCCULTI S M AND P H I L O S OPHY

1 13

Celebrations (Cambridge: Cambridge University Press, 1947): 27-37, esp. p. 29. 38 Webster, op. cit. (n. 26), pp. 17, 267-270. 39 E.g. Essay, IV.iii.24. Strictly, "we cannot so much as guess"! 40 E.g. Essay, lI.xxiii. 1 3 ; IV.iii.6, ad. fin. . 41 E.g. Essay, lI.viii.I S ; xxx.2. 42 Cf. the example of the wax in Meditation 2, and of the sun in Meditation 3. 43 Essay, II.viii.8: "Which ideas, if I speak of sometimes as in the things themselves, I would be understood to mean those qualities in the objects which produce them in us". I am aware that some commentators have interpreted this sentence differently. 44 E.g. Essay, l.i. I S ; lI.ii. l ; iii. l ; ix. I S ; xi. 1 7. 45 E.g. John Dee, Monas hieroglyphica (Antwerp, 1564), repro in Ambix 12 ( 1964): 1 12-221 ; Johannes Kepler, Mysterium cosmographicum ( 1 596). 46 Cf. Thomas, op. cit. (n. 23), pp. 426-427, 430-43 1 . 4 7 Leonardo da Vinci, Th e Notebooks, edited b y Ed. MacCurdy, 2 Volumes (London: Cape, 1938), Volume II, pp. 363-380. 48 Gassendi, for example, insbted that "species, i.e. images emanating from bodies" were material: Epistle II De apparente magnl�udine solis humilis et sublimis §5, in Opera (Lyons, 1658), Volume III, p. 425. 49 Cf. n. 34, above. 50 John Wilkins Mercury. or the Secret and Swifi Messenger. shewing how a Man may with privacy and speed communicate his Thoughts to a Friend at any distance (London, 1 64 1 ), pp. 1 1 8-122, 148. 5 1 Robert Fludd, Tractatus de geomantia, pp. 3-170 of Fasciculus geomanticus. in quo varia variorum opera geomantica continentur (Verona, 1687). 52 Franz Mesmer, Memoire sur la decouverte du magnetisme animal (Geneva, 1779). 53 Gareth Knight, A History of White Magic (London & Oxford: Weiser, 1978), Chapter 1 . 5 4 Surprisingly little attention has been paid to the important similarities between Descartes' Meditations, and Ignatius Loyola's Spiritual Exercises. In my view, they are still understated by Pierre Mesnard, 'L'arbre de la sagesse', Cahiers de Royaumont. Philosophie. No. II (Paris: Editions de Minuit, 1957). Cf. esp. Versio vu/gata, edited by Calveras & Dalmases, Monumenta Ignatiana Series 2, Volume I (Rome, 1 969), pp. 140, 160, 1 72. On the other hand, however close the analogies, it remains the case that the principal purpose of Descartes' Meditations is the discovery of natural science rather than wisdom. This puts him firmly in the camp of occultist enthusiasts as contrasted with religious mystics. 55 Pansophie: Part 2: Gabalia. Ein Versuch zur Geschichte der magia naturalis im 16. bis 18. Jahrhundert (Berlin: Schmidt, 1 967), p. 20. 56 There is another demarcation problem between mysticism, heresy, and cabalism. One way of taking Descartes' Meditations is to see him as discovering his being as a mode of the substance Thought. But how is unlimited, infinite Thought to be distinguished from God himself? Since his introspection discovers the concept/essence/ being of God embedded in his very thinking, how is he not himself identical with, or at least part of, God? That would be the Heresy of the Free Spirit, which developed in Holland as an off-shoot of the mysticism of Meister Eckhart. Cf. Robert E. Lerner, The Heresy ofthe Free Spirit in the Later Middle Ages (Berkeley: University of California Press, 1 972). No wonder most of the proof of God's existence in Meditation III is really a proof that Descartes is not God! As for Cabalism, it is perhaps not just a coincidence that Descartes repeats the phrase 'Ego sum' almost as a meditational mantra - and EGO SUM is one of the names of God (Exodus 111. 14). Perhaps ,

1 14

G. MacDONALD RO S S

instead of the slogan, 'cogito ergo sum', which does not occur in the Meditations anyway, we should substitute the slogan : 'Dubito, ergo non sum Deus'. 51 E.g. Leibniz, Philosophischen Schriften IV, edited by c. 1. Gerhard (Berlin, 1 880), p. 328. 58 Adrien Baillet, La vie de Monsieur Des-Cartes (Paris, 1 69 1 ), Volume I p. 87ff.. See also Henri Gouhier, Les premieres pensees de Descartes (Paris : Vrin, 1958), pp. 1 1 7-14 1 ; Arnold, op. cit. (n. 28), pp. 273-299. Modern historians have correctly nailed the many myths about Descartes, such as his membership of the Rosicrucian brotherhood, and his discovery of the secret of prolonging life, as in Gabriel Daniel, Voiage du monde de Descartes (Paris, 1 690), p. 45ff.: 'Que M. Descartes n'est pas mort'. On the other hand, it is an important historical fact about how Descartes was perceived by his contemporaries that such myths could have arisen. For other suggestions of Cartesian occultism, cf. Ravetz, op. cit. (n. 2), pp. 99-100, and Westfall, ibid. , pp. 1 86-187. It is also worth adding that the concept of an evil genius, while fanciful to us, would have seemed quite commonplace in the spiritualist climate of the seventeenth century. 59 Oswald Croll, Basilica chymica (Frankfurt, 1609), pp. 1 3 , 26-27: "Supremae universitatis unicus Rector [translated by H. Pinnell, Philosophy Reformed and Improved (London, 1 657), p. 23, as "the One onely Governour ofthe supream Universe"] . . . Omnes per retrogressionem discunt a primo, et hic a Deo, qui ei scientiam creavit rVWel m;cw't6v . . . vera via (secundum Apollinis Delphici oraculum . . . ) ad veram Sapientiam . . . a se ipso primum exordiri: sic quod Homo intelligens se ips urn, in seipso, tanquam in quodam Deifico Speculo intueatur . . . concreata est nobis omnibus Rerum Notio . . . Lumine Naturae . . . cognoscitur Deus . . . et potest Homo bonam, magnamque partem cognitionis eius Qui est, ex sui ipsius cons ide­ ratione venari . . . ". 60 It could be argued that the belief that conceptual analysis can uncover new truths about reality is as superstitious as any other epistemology. From this point of view, Leibniz is very much in the tradition of Lullian mysticism. 6 1 Essay, IV.viii. 62 Essay, IV.i.2, & c. 63 Essay, IV.ii. l : "This part of knowledge is irresistible and, like bright sunshine, forces itself immediately to be perceived, as soon as ever the mind turns its view that way; and leaves no room for hesitation, doubt, or examination, but the mind is presently filled with the clear light of it". 64 Cf. the English translation of della Porta's Magia naturalis: Natural Magick (London, 1658) [repr., New York: Basic Books, 1 957] , Book I , Chapters 2, 7 : pp. 2, 8. 65 Cf. Samuel Johnson, A Dictionary of the English Language (London, 1 755), s. v. EMPIRIC: "A trier; an experimenter; such persons as have no true education in, or knowledge of physic aI practice, but venture upon hearsay and observation only . . . Known only by experience; practised only by rote, without rational grounds . . . Experimentally; according to experience . . . without rational ground; charlatanically; in the manner of quacks. EMPIRICISM: Dependence on experience without knowledge or art; quackery". 66 The Outlines of Pyrrhonism and Against the Professors. 61 In particular, Books XVII-XIX: De catoptricis imaginibus. De staticis experimentis, and De pneumaticis. 68 I do not see this dimension as essentially sociological, as would be implied by calling it 'externalist' (cf. n. 2, above). A merely externalist interpretation cannot take proper account of how different intellectual factions viewed each other - a question which belongs neither to sociology nor to the internal logic of theories and concepts considered in abstract. The ' "

OCCULT I S M A N D PHILO S OPHY

1 15

notion of intention serves as a useful bridge by incorporating a rhetorical dimension. I have not gone into the question of the secretiveness of some of the new philosophers. Descartes planned the appearance of his philosophical system in the manner of a political dissident organizing a coup, and he rebuked Regius for being too open about the ultimate objectives of his philosophical revolution. Cf. Hiram Caton, The Origin of Subjectivity (New Haven & London: Yale University Press, 1 973), Chapter I. 70 E. M. Schur, Labeling Deviant Behavior (New York: Harper & Row, 1971). For its applica­ tion to the closely related topic of witchcraft, see C. J. Lamer, Enemies of God (London: Chatto & Windus, 198 1 ), pp. 98-100. 71 An interesting example of the alliance is the way in which astrological almanacks helped spread the new ideas. Cf. Bernard Capp, Astrology and the Popular Press (London & Boston: Faber, 1979), Chapter 6. 72 I have already given many examples of the mechanisation of the occultist world-picture, and of disagreements as to what was to count as occult. A well-known example of an atteinpt at empirical verification is the Brief Lives of John Aubrey ( 1 626-1697), which was intended as a mass of observational data of' which to base astrological correlations. 73 In modern times, one might contrast the parade of statistical orthodoxy of J. B. Rhine and S. G. Soal in the field of ESP, and of Michel Gauquelin in the field of astrology, with the disdain of logic shown by high-status scientists such as nuclear physicists (even some logicians concede that logic is the handmaiden of physics). 74 For example, in 1665 the Hermeticists made a nearly successful bid to found a 'Noble Society for the advancement of Hermetick Physick' in opposition to the medical establish­ ment: P. M. Rattansi, 'The Helmontian-Galenist Controversy in Restoration England', in Ambix 12 ( 1 964): 1-23, esp. p. 13. For a while there was a flourishing 'Astrologers' Club', which was not unlike the future Royal Society: Elias Ashmole (1617-1692), edited by C. H. Josten (Oxford: Clarendon, 1966), Index, s. v. 'Astrologers' Club'. 75 See n. 28, above, and D. Knoop & G. P. Jones, The Genesis of Freemasonry (Manchester: Manchester University Press, 1957), and Karl R. H. Frick, Die Erleuchteten (Graz: Akademische Druck-u. Verlaganstalt, 1973-78). 76 See my 'Leibniz and the Nuremberg Alchemical Society', in Studia Leibnitiana 6 ( 1 974): 222-248; 'Leibniz and Alchemy', in Magia naturalis und die Entstehung der modemen Natur­ wissenschajien, Studia Leibnitiana, Sonderheji 7 ( 1978): 166-180; 'Alchemy and the Develop­ ment of Leibniz's Metaphysics', in Studia Leibnitiana Supplementa 22 ( 1982): 40-45. 77 At the beginning of the century, any distinction between alchemy and chemistry was not a distinction between superstition and science. For example, Andreas Libavius's Alchemia is praised by Hannaway (op. cit. , n. 27, above) for its modernity in comparison with Croll's Basilica chymica. Libavius himself makes a distinction within 'alchemy' between 'chymia' and 'encheria' - but this has nothing to do with the modern chemistry/alchemy distinction. Indicative of the change in meaning is a casual remark by Gottfried Thomasius, in an unpublished letter to Leibniz of 3 1 .7. 1 696: "one of the Chemists, or, if you prefer, Alchemists" (Niedersachsische Landesbibliothek, Hannover: Leibniz-Briefe 925, Bl. 1 2V). 78 Including a number of beliefs and attitudes which have subsequently been reinstated. The eighteenth-century scientists of Paris were much harder on superstition, with their refusal to accept reports of heavenly bodies landing on earth, than their modern counterparts in the Groupe d'Etudes des Phenomenes Spatiaux (Sunday Times, 20.2.83). 69

SIMON SCHAFFER

O CC U LTI S M A N D R EA S O N

G . MacDonald Ross (,Occultism and philosophy i n the seventeenth century') tells us that "it is virtually impossible to force any confrontation between occultism and the mechanical philosophy" and that the seven­ teenth century "was above all a century in which philosophy and science cannot be sharply demarcated from the occult". These are very welcome conclusions for historians of philosophy and of science. They allow us to abandon the lengthy attempts to impose unhelpful criteria of demarcation on a period to which these criteria are irrelevant, and to forget the anguish of historians who have feared that they were abandoning reason in analysing the history of systems of thought. In this response to Ross's paper, I shall try to single out some areas of seventeenth-century philo­ sophy and natural philosophy which pose these issues of occultism and reason, and suggest how Ross's argument might be put to work in historical practice. I argue first that in this period the realm of spirit was crucial in making possible a philosophy of nature, and that in the process even reason itself could become an 'occult' principle. Next, I indicate how knowledge of spirit was made safe and how this knowledge could deal with attacks on its propriety or its possibility. Finally, I reaffirm Ross's claim that the mechanical philosophy did not function as the obvious and progressive replacement of occult philosophies. Historians must ask how 'occultism' worked as an effective label in philosophical dispute rather than how 'occultism' was overthrown. 1 1 . THE DANGER OF REASON

In histories of seventeenth century philosophies which set themselves the task of locating the birth-pangs of modern rationality, the category of reason played a central role. But as Ross shows, it is the case that the Cartesian statements on the light of the rational and the divine can easily be compared with those of a Paracelsian such as Croll , and that these strategies by no means demarcate putatively 'progressive' workers. Fur­ thermore, reason itself was made out as both tyrannic and subversive. The authority of individual judgment based on reason was seen as a source of 1 17 A. J. Holland (ed.), Philosophy, Its History and Historiography, 1 1 7-143. .

:.... - �----- ..

1 18

S IM O N S CHAFFER

civil strife, heresy and sectarian dissensions. We might assume that practices which appealed to individual reason and to private experience as the basis of knowledge, uncoupled from special grace or inspiration, would be closely linked with the orthodoxy of seventeenth century philosophy. I n fact, the reverse is the case. Mid-century radicals i n England, such a s Winstanley o r Overton, argued for Reason a s a "pure spirit i n man", a s "the great Creator" through which each individual citizen learnt o f his share in the commonwealth, and, above all, as the most powerful resource against establishments of authority in church and state. "God manifests himself in actual knowledge, not in imagination", wrote Winstanley, and used this claim to belabour clerics who had appropriated individual clear judgment to the obscure arcana of scholastic occultism. The witnesses of Christ's teaching, he wrote, were of humbler, more "experimental" character. "What wee know, wee receive it by degrees, now a little and then a little", wrote Overton in 1 645, and "is it not frequent amongst us, that the thing that we judged heresie, wee now beleeve is Orthodox". So the masters of knowledge "uphold the accursed Doctrine of Persecution, lest liberty of printing, writing, teaching, should discover their deceipts". So in turn, this 'Right Reason' would be celebrated by the radicals as the individualist basis of truth: "God is not a God of irrationality, and madnesse or tyranny : Therefore all his communications are reasonable and just, and what is so, is of God". These radicals could use such rationalism for their own purposes : in the work of Overton, it was deployed to emphasise the most ruthlessly individualistic politics, "every man by nature being a King, Priest and Prophet in his owne naturall circuite and compasse".2 In this context, Restoration philosophers in England could not safely join in the celebration of the light of pure and rational experience. A series of prohibitions surrounded what could be said about nature and about the soul, and amongst these that on appeal to unaided testimony of nature was strong. To understand the actual categories of the occult is thus to under­ stand how philosophers of nature could safely but effectively describe the world and the means of access to that world. Writers in the 1 660's who propagandised for experimental natural philosophy, mechanists or not, could not leave spirit out of nature without being branded as atheists ; they could not describe experimental work as an unaided and solitary practice for fear of being linked with enthusiasm; and they could not exclude authority and the power of the community for fear of being seen as radicals. One common response to such propaganda was precisely that which made

OCCULTISM AND REASON

1 19

out the new philosophies as overly authoritarian, seizing too much power over judgment for itself. The eccentric Margaret Cavendish complained of many who "will not admit of rational arguments, but the bare authority of an Experimental Philosopher is sufficient to them . . . to pronounce the Truth without any appeal to Reason". Apologists for the Royal Society responded by satirising the "deformed Monsters, or at best pretty but impossible Monsters" which pure Reason would breed. 3 The commonest targets for critics of philosophical tyranny and obscurantism were the most obviously rationalist: Hobbes and Descartes. Most critics of Hobbes pointed to his wish to have Hobbist principles preached from all pulpits and colleges; Huygens worried early in his career about Descartes's promulgation of a new scholastic dogmatism. Critics complained that Descartes was "governed by his Phantasy" and that he was "one whom excessive pride and self-conceit . . . hath absolutely deprived of his witts", a "Fanatick", in fact. So Meric Casaubon, for example, coupled Hobbes with Descartes and then glossed the method of scepticism as a means whereby disciples would "forgett and forgoe all former precognitions . . . then they must adheare to him tooth and nayle, or acknowledge themselfes to have been fooled". For these writers it was rationalist mechanical philo­ sophy which appeared occultist and enthusiast.4

2.

EXPERIM ENTS AND S PIRITS

Apologists for experimental philosophy were compelled to emphasise its humility before nature and revelation, and its capacity to generate suf­ ficiently stable knowledge of inherently unstable objects. Hutchison re­ cently argued that these writers did not abolish occult qualities, therefore, but admitted what had been occult qualities to the realm of natural philo­ sophy: the insensible and the immaterial were by no means excluded from knowledge.5 The complex struggles around this practice were dominated by the dangers which knowledge of the hitherto occult would pose. Glanvill's constructive scepticism was built as an alternative for those who otherwise "have been willing to accept Mechanism upon Hobbian prin­ ciples"; he assailed dogmatism as "the great disturber both of ourselves and the world without us" and argued that "hence grow Schisms, Heresies and Anomalies beyond Arithmetick". This propaganda carefully limited the claims of natural philosophy precisely in order to bolster the authority of experiment, to secure a place for experimental philosophy, and to explain why experimental knowledge of spirits, for which Glanvill also

1 20

S I MO N S CHAFFER

campaigned, was not threatening for civil or ecclesiastic authority. How could knowledge of spirit fail to be linked with enthusiasm or illegal occultism ?6 This question itself raised the issue of the state of the philosopher's soul. Writers explored the qualities of this soul to determine the source of that special access which such philosophy claimed, and the special authority and security with which this philosophy was endowed. When Henry More argued with William Petty in 1 648 - 1 649 about the status of Descartes's philosophy, they disputed the faculties of the philosophic mind: More compared the "blinde molewarpes" who claimed that "the first and most generall principles of Nature" could be "hermetically imprison'd in some narrow neck'd glasse . . . like a Jack in a Box to astartle the vulgar" with the "higher and nobler" philosophers. The following year Walter Charleton prefaced a Helmontian text with a similar attack on "purblind Moles whose imperfect Opticks could never endure to pry into the mysteries of the Intellectual and Spirituall World". 7 But when spirit was to become a matter of communal and experimental work, the claims shifted from a celebration of the more elevated philosophical faculty to an examination of the way in which such spirits could be fixed, replicated and reliably reported. Here scepticism was a powerful resource: Charleton now ( 1 654) confessed that "we are Men, i.e. Moles, whose weak and narrow Opticks are accomodated only to the inspection of the exterior and low parts of Nature". Spirits were to be known through "the refracted and reflected" line "of Effects".8 These changes between an emphasis on the elevated philosophical faculty and the reliance on the 'refracted' world of effects had important consequences for the treatment of what had hitherto been occult. The terrors of enthusiasm (assaulted by More in the 1 650's) and of mechanic atheism (denied by Charleton at the same time) compelled the production of spirit inside philosophy but made that production dan­ gerous. A repertoire of resources existed which were exploited to explain how such work was possible. While indebted to Descartes, for example, Henry Power's Experimental Philosophy ( 1 664) reaffirmed More's point that philo­ sophers "may be well placed in a rank specifically different from the rest of grovelling Humanity". The souls of philosophers, for Power, were structured just like the "Elastical" particles of which he wrote, "in a kind offermentation", endowed with a " Springy Intellect". The themes of public and communal knowledge, of the authority of the community of workers, and of the soteriological power possessed by that philosophy were all

O CCULTI S M AND REASON

121

displayed during the 1 660's. Hermetic authors were attacked for their suspicious secrecy or else treated with the caution appropriate to their claims and dangers. Writers such as Boyle or John Edwards wrote of the hopes that natural philosophy would restore men to a prelapsarian state by Baconian communal action, and that in the future state "a Vertuoso shall be no rarity" and men's faculties would be enlarged in an appropriate manner for philosophy. Henry Power insisted that with the relevant "Engines" the "faculties of the soul of our Primitive Father Adam" might be reproduced and surpassed.9 Once again, therefore, it would be wrong to read these claims as marking a definitive break with occultism : instead, the manner of production of knowledge of that realm was changed, and redefined by a set of powerful appeals to new areas of authority in experience. 3. S PI R IT TE STIMONIES

Natural philosophy came to provide a fruitful source for the work of theologians in its evidence of spirit: the sermons of many Restoration divines were filled with examples drawn from such work. Even the arch­ opponent of the Royal Society, 'Robert South, used images of the "aerial nitre" to illuminate a sermon on grace. John Moore, Bishop of Ely, followed many in citing Boyle, Thomas Browne, Helmont and the Oxford physiologists to offer proof of resurrection, palingenesis, "ye plastick power of ye soul"; "How infinite are ye storys we may recount of sym­ pathy?", he asked. Notoriously, too, the Cambridge Platonists persistently mobilised natural philosophy as the most fundamental evidence of the incapacities of matter, and of "the Vicarious Power of God upon this great Automaton, the World". The alliance forged at this period between one form of philosophY and one form of religion was a dominant feature of late seventeenth century thought, notably in the realm of natural theology. l O However, that alliance was by no means as simple or secure as historians have recently assumed. Many writers asked "why bodies only should engross and monopolize natural philosophy", viewing the exclusion of the soul as a "philosophical kind of tyranny". In contesting such tyranny, Henry More nevertheless sought to speak to the natural philosophers "in their own dialect", as he put it. In 1 664 he outlined "how a man is to behave himself in this Rational and Philosophical Age for the gaining of men to or retaining them in the Christian faith", promulgating a set of rules of argument which sought to match those of the experimenters, but to con-

1 22

S IMON

SCHAFFER

vince an audience of real divine power. Recent studies of More, for example, have shown how early he broke with Descartes on issues which drew their force from just this effort to produce spirit in real experience, and that he often contested the views of those such as F. M. van Helmont who adopted too immaterialist a philosophy. In this case, at least, the realm of the 'occult' was at the heart of a confessedly rational philosophy, yet that philosophy did not willingly seek to differentiate itself from experi­ mental philosophy. 1 1 It is a caricature, therefore, to portray the Platonism of this Cambridge school as inherently occultist in contrast with the sanity of experimentalism. Many philosophers argued that the practice of experiment must and could cope with the experience of spirits. In his edition of H elmont's Ternary of Paradoxes, Walter Charleton simultaneously affirmed that "in the infinite Magazine of Nature are to be found various Agents not obliged to the dull conditions of an immediate Corporeall Contact, but richly endowed with an Influentiall or Radiall Activity", and reported in detail a controlled trial of the effects of the weapon salve which allegedly converted one "Dr. of Theology" from "a praevaricate Adversary" to a "fervent Proselyte to our Doctrine of Magnetisme". Charleton stated that before this trial the divine formally disavowed all confederacy with the Devil, and that the trial had worked powerfully on his mind. 1 2 The capacity of the practice of experi­ ment to work on private belief was one crucial resource for those who affirmed the effects of spirits. Hence the importance of witch testimony and its manner of reporting during the century; as Ross observes, these reports are not clearly those of a practical technology, or of superstitious magic, or of mystical religion. Henry More mobilised such testimony, in one case "more than six hundred true testimonies", in his responses to Descartes in summer 1 655. S uch testimony was necessary first because without acceptance of spirit "it is impossible there should be any God, or Soule or Angel, Good or B ad, or any Immortality or Life to come . . . ". Glanvill attacked the only alternative, that all "our Conceptions are but the thrust­ ing of one part of Matter against another". Second, experimental testimony directly compelled assent. More told Glanvill that the existence of "such fresh examples of Apparitions and Witchcrafts" were "a special piece of Providence" because only thus could "benummed and lethargick Mindes" be won over. Glanvill agreed that "we know not anything of the world we live in but by experiment and the Phaenomena: and there is the same way of speculating immaterial Nature by extraordinary Events and Appar­ itions". Finally, testimony of spirit allied the project to that of the experi-

OCCULTI SM AND REASON

123

menters, and sorted out the true atheists and sceptics. More praised the "more perfect Philosophy" of the Royal Society, "which is so far from tending to Atheism, that I am confident it will utterly rout it and the Mechanical Philosophy at once". Once again, the realm of spirit was contrasted with that described by the unreasonably confident mechanists: these latter were "hardened against Conviction", possessed of a "mighty Confidence grounded upon nothing, that swaggers and huffs and swears there are no witches". Just because witches were part of the object of experiment, denial of them was irrational. 1 3 . As Ross observes, such work was not limited to More's allies and their ilk : Boyle too played an active role in the licensing and encouragement of testimony. In denying Webster's claim that Boyle was now a sceptic of witchcraft, he told Glanvill that "to grant (or deny) that there are intelligent beings that are not ordinarily visible does much conduce to the reclaiming (or confirming) of atheists", and, interestingly, wrote that Glanvill's reports of "the invisible marks of witches" would "suggest odd speculations to the naturalist", would "help to enlarge the somewhat too narrow conceptions men are wont to have of the amplitude of the works of God", and, finally, might "advantageously enlarge our knowledge, though not, perhaps in physics, strictly so called, yet in philosophy". Boyle collaborated in the publication of reports of the celebrated Demon of Miicon, and, as we shall see later, used much of this work in his own philosophical pUblications. 14 At the same time, however, it is important to begin dis aggregating the work on witchcraft reports. To deny such reports was by no means to deny the power of spirits , but, instead, more commonly represented an intervention in legal and political debate. The Act of 1 650 included important limits on statements on witchcraft which had some effect; eminent legal theorists such as John Selden and Thomas Hobbes denied witches' power but not the legitimacy oflaws against witchcraft, "for the false belief they have that they can do such mischief, joined with their purpose to do it if they can", as Hobbes put it. Alternatively, writers such as Webster and Wagstaffe did "firmly believe there are many thousands of spirits made of an incorporeal matter too fine to be perceived by the senses of man". Yet they denied that spirits could "play mad pranks amongst us" merely "upon the account of a contract made with any man or woman". In Britain this tradition culminated in the texts of Hutchinson and Defoe, whose histories of superstition attacked legal rather than ontological issues of the effects of spirits. Here we must be sensitive to the forms in which occult powers could legitimately be produced, tested and reported. I S

1 24

S IMON SCHAFFER

Finally, as Ross properly argues, to give effluvial or purely material accounts of the action of such spirit did not of itself exclude such power from nature or from knowledge. In 1 654 More wrote of one healer, Matthew Coker, and explained his successes in terms of "a power partly naturall and partly devotionall" , in which "the blood and spiritts of this party is become sanative and healing by long temperance and devotion". He also warned that Coker should not be told of this rational account of his powers, lest "it come to his knowledge to the disturbance of him in his way, and the weakening of him in his present faculty". We may begin to see how debates about body-soul interactions were understood in the context of precisely such 'occultist' obsessions. This theme has emerged clearly from recent controversies about the interpretations offered of the cures of the healer Valentine Greatrakes in the 1 660's, where a range of explicitly political interests represented the material and spiritual causes of this phenomenon in contrasting manners. 16 Such an understanding also draws attention to the persistence and, in fact, the ultimate domin ance of spirit and of active principles in natural philosophy itself. I shall briefly consider the importance of this work in matter theory later in this paper: around the problem of enthusiasm and of 'miraculous' healing, however, it is proper to emphasise here how important 'occult' resources were for accounts offered of these social problems. Further, it was in this context that the proper meaning of that term, abusive or approving, was established. Michael MacDonald has shown how philosophical accounts given of enthusiasm and of healing changed socially to accounts given of melancholy, madness and of the condition of the poor. Philosophers such as Thomas Browne, Meric Casaubon and John Locke all paid consider­ able attention to the accounts which might be offered for madness and melancholy, and their work on this issue was assessed in terms of its use of spirit and of mental powers. Browne offered a philosophical account of mass delusion and riot: because of domination of the collective "irrational and brutal part of the soul", Browne argued, "their single numbers once , huddled together, they will be error itself . 1 7 One important test case for the ability of philosophers to represent such mass delusion, evidence of spirit, and social subversion, occurred in London in 1 706- 1 7 1 0, when Camisard refugees appeared as part of an enthusiast movement. Camisard allies included Newton's associate Fatio de Duillier, and propagandists made out the power of enthusiasm in terms of 'a fl ash of fire' which could materially move the soul. Rival accounts of these phenomena depended on rival philosophical and political commitments, so while deists wrote of

OCCULTISM AND REASON

125

"poisonous and melancholy vapours streaming from an enthusiast", more orthodox churchmen and natural philosophers argued that "whether it be a good or bad spirit that these men do act, yet so long as it is by any spirit it does serve to convince the being of invisible things, and the reasonable­ ness of religion". It was a commonplace, therefore, that immaterial spirits had real effects : without this apparently 'occult' evidence, religion and philosophy might fail. Most importantly, such testimony was not evaluated in terms of some simple notion of what could allowably be attributed to nature alone, but persistently in terms of the persuasive effect of reliable experience of hidden powers and spirits, which by no means disappeared by the end of the seventeenth century. 1 8 4 . DIVINE A N D MATERIAL S O U L S

Ross has shown that over a wide range of philosophical topics, issues of religious mystery and spiritual action played a central role. Such issues centred on the problem of authority and control: the authority of the philosopher in a realm set apart for special knowledge, and the authority of God and of mind over man and over matter. In this section I consider briefly some of these issues to show how philosophical investigation of such control worked to remove the dangers inherent in such a project, and how this investigation produced a space for philosophical understanding. Notably, the problem arose in the interpretation of some of the more lapidary statements of Descartes, Ross views Descartes's exchanges with Elizabeth as testifying to the view of "the unity of thought and extension in man as a religious mystery". He also argues that Leibniz's analysis of the Cartesian doctrine of scalar motion was "not only a very perceptive piece of historical reconstruction, but a shining example of sympathetic historiography". The responses of Elizabeth, and, later, of Leibniz and of Newton, show how important the understanding of spirit was to the establishment of philosophy at this period. In his notebook of 1 664-1 665, Newton titled one entry 'Occult Qualityes', struck this out and replaced it with the title 'Philosophy', and then wrote down a paraphrase of a text of Hobbes in De Corpore, where the argument is based on an understanding of the proper realms of matter and spirit: "The nature of things is more securely and naturally deduced from their operations one upon another than upon our senses. And when by the former Experiments we have found the nature of bodies, by the latter we may more clearly find the nature of our senses. But so long as we are ignorant of the nature of both soul and

126

SIMON SCHAFFER

body we cannot clearly distinguish how far an act of sensation proceeds from the soul and how far from the body". This notebook also contained comments on More's pneumatology and on Glanvill's reports of 'fascina­ tion' and of the power of mind over body. By the time Newton composed his critique of Descartes in the late 1 660's, he could affirm that "the usefulness of the idea of body that I have described is brought out by the fact that it clearly involves the chief truths of metaphysics"; that Cartesian metaphysics did "manifestly offer a path to Atheism"; and importantly, he had developed the main points of his model of the divine sensorium, for which our ability to move bodies was the central analogy of divine power over the universe.1 9 It is important, therefore, to see how fruitful putatively mechanist philo­ sophical texts were for this programme. The exchanges between Descartes, Elizabeth and Arnauld in the 1 640's, together with his Passions

de l'iime, composed from 1 645, already raised those analogies which would plague the Newton-Leibniz debates later in the century. Here the analogy between the action of gravity and the action of mind was explored; and the modes by which philosophers might know the world were explicitly defined and compared with divine understanding and power. For Descartes, the philosophy of the garden, of isolation and contemplation, offered the sole path to knowledge of the union of mind and body: this was to be contrasted with the uses of understanding and of imagination by which soul and body were known separately. These differentiated forms of knowledge matched differentiated forms of interaction between the soul and the spirits : in the Passions de l'iime Descartes differentiated passions and representations through a differentiation of the nature of the conflict between these two motions, viewed alternatively as simple interaction or more complex con­ flict. He told Elizabeth in May 1 643 that "we have hitherto confounded the notion of the force with which the soul acts on the body, with that by which one body acts on another" ; and in July 1 648 he told Arnauld that "the action by which the mind moves the nerves, in so far as such action is in the mind . . . is no other than the inclination of the will to this or that movement". It was at this point that he exploited the analogy with gravity, and that the issue of scalar conservation played its part. He wrote to Arnauld that "it is no more difficult for us to apprehend how the mind moves the body" than it was for "the majority of philosophers" to "appre­ hend how such a quality carries the stone downwards" in free fall. So two problems emerged in these texts : the appeal to gravity as an example of occult quality and as useful analogy, and the appeal to the work of the soul

OCCULTISM AND REASON

127

in changing the direction rather than the quantity of bodily motion by inclination. Both problems, in turn, were closely linked to the ulimate analogy with divine will. It was at this point that Leibniz offered his reconstruction of Cartesian mechanics in terms of the effects of the soul on motion. Contests between "the natural appetite and the will" had been described by Descartes as a 'repugnance' between motions in the pineal gland excited by the spirits and the will; Descartes often wrote in terms which could, nevertheless, be interpreted as affirming that the will could actually change the quantity as well as the direction of motion. The relation between soul and body here was made out as the result of conflict, not of harmony, "when the cause exciting a certain passion in the soul also excites quite independently of the soul certain movements in the body, and when the soul, immediately on apprehending the movements arrests or strives to arrest them". Further­ more, this conflict itself relied on a problem in understanding divine will. Descartes told More that he had not developed an extreme form of voluntarism because that would "seem to be supporting the opinion of those who regard God as anima mundi, united to the matter of the world".2 1 Yet his claim that motion was conserved from moment to moment in the form of scalar motion did depend on divine origin. This set of claims stimulated the refutation of the laws of impact at the Royal Society in the 1 660's. Because of the relation between divinely validated conservation and the perfect rigidity of colliding bodies , Leibniz was then able to point to two ways in which the impact laws were connected with the Cartesian doctrines of God and of soul : fir st, he observed that Descartes postuiated scalar conservation because otherwise "the modifi­ cations of the soul would seem to violate this law" and that "thence occur voluntary motions". The mind-body problem led to the peculiar structure of Cartesian mechanics. Leibniz also pointed out that in this mechanics the universe would become more chaotic since the velocity differences between larger and smaller bodies would increase. Only with divinely supported scalar conservation could this be avoided, and even then "it would be necessary to suppose that the same state of the universe would always return precisely after some period". Cartesian cosmology, just because of this central and obscure relation, was cyclical, chaotic and inaccessible to human reason, because all possible states of matter would be occupied without divine intervention, since only scalar motion was conserved, and since there was a disjuncture between mind and body and because of the arbitrary sustenance of divine action. Newton was not,

128

SIMON SCHAFFER

therefore, the first target for Leibnizian accusations of occultism, and that label was still a powerful resource in analysing the deep structure of cosmology and of mechanics.22 Once again, it is possible to extend the claim that soul and spirit played a fundamental role in the interpretation of mechanics, and that therefore hitherto occult qualities were placed ambiguously at the heart of natural philosophy. Newton's early notes 'Reflections' and on 'The lawes of motion' from the 1 660's and early 1 670's raised just these problems with Cartesian accounts of spring and impact, when he established, against the vortex theory, that motion was lost in all but indubitably rare perfectly elastic collisions. The example he used in 1 672 (two equal globes in col­ lision at right angles) was identical in import to that which he used in the Queries to the Optice of 1 706, to demonstrate that motion might be "lost or got", where motion was specifically defined as scalar in character. The sensitive reconstruction offered by Leibniz for Descartes's conservation law was also a central concern for Newton, therefore : on the one hand, he concentrated on the implications of scalar conservation of motion for the interaction of deity and material world, and on the other hand, he placed this problem in the wider context of his model ofthe divine sensorium. This latter was the crucial political and philosophical resource in Newtonian philosophy, and was recognised as such by critics such as Leibniz who brandished the label of 'occultism'. Work on the "contagious communica­ tion of the strong imagination" was a fruitful source for this part of New­ ton's programme, and he read widely on reports of spirit testimony and of fascination, the "moral and spiritual eye".23 In the early text now called 'De gravitatione', as has been noted, Newton wrote in the context of laws of mechanics of the sensorium and the control of nature. Newton asserted against Cartesian impact rules and metaphysics that he had "deduced a description of this corporeal nature from our faculty of moving our bodies", and that God "may appear (to our innermost consciousness) to have created the world solely by the act of will, just as we move our bodies by an act of will alone". "Therefore", Newton concluded, "the world should not be called the creature of that soul [of the world] but of God alone, who creates it by constituting the soul of such a nature that the world emanates [from it]". Throughout his career, Newton stressed the central role this notion must play. In correspondence with Locke he examined the control 'phansy' exercised in man, and discussed "a disposition ofye sensorium to move ye imagination strongly". Drafts for Queries and for texts on space and motion all used ranges of accounts of spiritual influence, of fantasy and

OCCULTIS M AND REASON

1 29

of the control of the soul, to affirm that "we cannot say that all nature is not alive", and that if "all space be the sensorium of an immaterial, living, thinking being", then "the laws of motion arising from life or will may be of univers all extent". When he actually published these accounts, however, Newton was compelled to modify most of these claims : the insertion of the 'tanquam' in some print runs of the Optice is notorious, while in 1 7 1 7 a reference to "our spirit which is in us" became the simpler "we". Again and again in the battles of the early eighteenth century, as Ross points out, Leibniz's attack prompted modifications of Newtonian positions which explicitly drew on earlier testimony of the action of spirit, and which while crucial for his whole concept of the extent of the laws of motion, and of natural philosophy, turned out to be too dangerous to print. It was not only in the suppressed alchemical work, therefore, that charges of 'occultism' could stick. 24 Such charges worked in philosophical debate because of the resources which philosophers exploited, and, consequently, because of the status which they claimed. The examples Ross discusses show clearly how deeply such resources worked in the most apparently mechanist or rationalist areas of research ; it is also important to recall the function of philosophy in these debates. Writers such as John Wilkins or Edward Stillingfleet insisted that a philosophical 'faculty' existed which it was the duty and the 'happiness' of man to exercise, and whose function was to discover God in the world. Hence, for example, the central place of miracles and spirits in this philosophy. The former category came to dominate philosophical exchanges in the early decades of the eighteenth century: but throughout the previous century the definition and demarcation of the miraculous formed part of the project which made spiritual action appear visibly in nature. Again, the dangers of enthusiasm (which would be seen as a "wild amuzing men's minds with Prodigies and conceits of Providences") and of atheism (against which Boyle reminded his readers of such phenomena as magnetic effiuvia or invisible winds) were used as warnings to control philosophical endeavours.2 5 Most importantly, the exact description of allowable disruptions of the course of nature was determined by the writer's conception of the proper office of the philosopher. To emphasise the essentially divine will evident in the common concourse of nature was to celebrate the sacred work of the natural philosopher. Boyle sometimes argued that "the works of God are not like the tricks of jugglers or the pageants that entertain princes, where concealment is necessary to won­ der; but the knowledge of the works of God proportions our admiration

130

SIMON SCHAFFER

of them", thus affirming the importance of philosophical penetration of and understanding ofthe hidden realm of nature. Thomas Sprat agreed that "it is a dangerous mistake" to "discover in every turn of human actions many supernatural Providences and miraculous events". Conversely, to deny all suspension or intervention in the course of nature was equally 'dangerous': Boyle claimed that "true philosophers are sharp-sighted enough to discern" the "subtler characters and flourishes" of divine wisdom while never ignoring the palpable exercise of providence. Further, as historians have recently emphasised, these issues of clear marks of providential action played a part in political assessments of good government and of proper civil authority, as they did in England after 1 688 and again in 1 7 14. Here the role of the philosopher, the apologetic function of that philo­ sophy, and the exact discrimination of spiritual, supernatural and natural action became a vital source of conflicting interest. 26 5. L I MITS ON MECHANICAL P H I L O S OPHY

The use of mechanical accounts of nature by no means excluded a commit­ ment to the reality of occult powers ; further, no historical assessment of this philosophy can ignore the place of spirit in its cosmology. Ross refers, for example, to Locke's statements on angelic perception and the knowledge of spirit in the Essay. Here spirit played two complementary functions in a philosophy which emphasised that "the foundation of all our knowledge of corporeal things lies in our senses". On the one hand, the real existence of nevertheless imperceptible spirits was evidence of the limits on this 'foundation' ; on the other hand, the more penetrating knowledge which spirits might possess showed that these limits were contingent. Ross observes that Locke "was allowing himself to be seduced back into the magical imagery of the effiuence theory of perception", while in fact this seduction performed crucial roles in his philosophical strategy. Locke claimed that "angels do sometimes assume bodies", that nevertheless "all those intelligences" were "things whereof our natural faculties give us no certain knowledge at all", and, finally, that nevertheless "spirits who see and know the nature and inward constitution of things" would have "as clear ideas of the radical constitution of substances as we have of a triangle".27 Most of these claims were shared by Robert Boyle and his colleagues. Firstly, in texts and manuscripts connected with his Origin of Forms and Qualities ( 1 666), Boyle did affirm the scandalous state of knowledge which relied on unobservables. He made out this scandal of

OCCULTISM AND REASON

131

school philosophy a s a necessary limit o n the proper role o f the philo­ sopher: in responses to Henry More and to Henry Oldenburg in the 1 670's, for example, Boyle used the parable of the Jesuits who presented a watch to the Emperor of China, who then took the watch for a living animal, to show what could properly be demanded of natural philosophy: "I should have thought I had fairly accounted for it", he wrote, "if I had shewn that an engine of such structure would necessarily mark the hours, though I could not have brought an argument to convince the Chinese monarch that it was not endowed with life". Thus, secondly, Boyle's picture of philoso­ phers as 'priests of nature' worked to emphasise the precise limits and compulsions on their work: they could not exclude life from the cosmos, but they could only treat that life with care. 28 These are familiar aspects of seventeenth century experimental philo­ sophy; but they must not prevent historians realising how important the priestly model was in this period. The model was used to bring out the effect of philosophy on the soul of witnesses and also the need of philosophy to treat the supernatural. The boundaries of pure mechanism were constantly renegotiated, therefore. A long series of texts of the 1 670's showed how these boundaries could be acceptably extended and, ulti­ mately, abandoned, by saving spirit from alleged occultists. Boyle's and Newton's texts on alchemy which date from this period are well-known, as is the care with which these writers felt the subject should be treated when it entered the public domain. In 1 674 Boyle produced a tract on "some hidden qualities in the air" which insisted on the need to "discourse like a naturalist" about hitherto unacceptable phenomena "that appear not to have been considered, not to say thought of, either by the scholastick or even the mechanical philosophers". In his essay on final causes, sent to Oldenburg at the same period, Boyle discoursed on the chemical and spiritual powers in living bodies which were clearly insensible, as were magnetic and celestial spirits with which Boyle had been concerned since his earliest exchanges with Hartlib, some of which he reprinted in his General History of the Air: "you did not expect, I am sure", he wrote, "I should have adventured so particular an apology for astrology".29 In such texts, Boyle argued forcefully that it was crucial that natural philosophy begin work on such actions, while affirming the necessary and self-imposed limits on the command which those philosophers might exercise in such an ambiguous realm. He pointedly compared the Cartesian materia subtilis with "the universal spirit of some spagyrists, not to say the anima mundi of the Platonists", not because of its spiritual character but because of the

1 32

SIMON SCHAFFER

practical manner in which its effects were produced: the boundary between occult and acceptable qualities was drawn in terms of the work required to produce knowledge, not just in terms of the claims of mechanical philosophy, as the Cartesian case showed. 3o The most spectacular case of this careful analysis of the boundary between acceptable and unacceptable qualities was in the posthumous second part of Boyle's Christian Virtuoso. In this confused text, Boyle offered at least two maps of the philosophical world, in one of which he divided "the operations of God below" into "supernatural, natural in a stricter sense, that is, mechanical, and natural in a larger sense, that which I call supra-mechanical". Boyle's examples of supra-mechanical action were all significant : they included "divers motions of a man's limbs", discussed in terms redolent of Descartes's views of the 1 640's, for which Boyle argued here that "to put a check at pleasure to the motion of the body . . . is not a mechanical operation, as may appear by all the changes of determination, that one body makes in the motion of another". Boyle's analysis of such power and determination was analogi sed with the standard incommensurables of geometry, which were linked with "man, as he is a creature that results from the strict and vital union of an incorporeal and intelligent mind, with a corporeal and organical body". This soul, viewed here as "but an imprisoned angel", was then connected, via Boyle's second possible taxonomy of creation, with the spirits to which Locke also referred. "Three worlds or grand communities" of divine works could be seen as spiritual, visible and 'dioptrical', the first of which, populated by 'abstracted intelligences', played the same function as that which it per­ formed for Locke. Such intelligences were beyond sensible experience, but spirit testimony evinced their power, and their own knowledge showed how contingent were the boundaries which divided the visible and the dioptrical realms, which they could penetrate. 3 l This text demonstrates the way in which experimental philosophy could extend its proper space by absorbing most of the hitherto occult: it preserved the virtues of spirit and of the knowledge of spirit only by remaining sensitive, as 'priests of nature', to the proper import of the divine and animate realm. These constraints on natural philosophy prompted many of the most important struggles of the late seventeenth century. Henry Power, John Mayow and Robert Hooke may all be interpreted as representative mecha­ nical philosophers, for example, but each of them worked to extend and infringe the boundary around such practice. Power was a disciple of Thomas Browne, and close colleague of Boyle: as Charles Webster has

OCCULTISM AND REASON

133

pointed out "Power followed the Stoic philosophers in effacing this boundary. In the infinity of gradations of fineness of materia subtilis, the corporeal was allowed to merge imperceptibly into the incorporeal". Mayow, Boyle's assistant in air pump trials in Oxford and eminent theorist of the aerial nitre, also effaced the boundary. He was attentive to the dangers of an abandonment of such constraints : he attacked notions of a 'lucid soul', writing that "fires of this sort and new lights, no less in Anatomy than in Religion, have always seemed to me vain and fanatical". But firstly he drew on Paracelsian texts such as those of Michael Sendivogius to describe "the nitro-aerial Mercury which entered the territory of his enemy sulphur and robbed him of his saline consort" ; then, secondly, he controlled the dangers of iatrochemistry by experimental production of spirit, and used the terms spirit and particle interchangeably to confirm this change ; and, finally, wrote of the nitro-aerial particles/spirit as the "chief instrument" of "the sensitive soul", which he described as "some more divine aura endowed with sense from the first creation, and co-extensive with the whole world", and which he claimed was a "spiritual material existing out of the bodies of living things". Here the controlled practice of Oxford physiology, admirably documented by Robert Frank, captured and then celebrated the world of spiritual powers without losing either the authority of experimental philosophy or the virtues of real and extended spirits. 32 In texts by Robert Hooke we can also make out the flexible character of such boundaries round experimental and mechanical philosophy. Bennett has shown how important a resource was magnetical cosmology, drawn from the explicitly animist texts of William Gilbert. This resource was used with some care by John Wilkins, Christopher Wren and John Wallis in their own philosophies, and it played a central role in Hooke's cosmology. Hooke was sensitive to the context from which Gilbertian magnetism had been drawn, writing in his Cometa that magnetic virtue "(which may be called one emanation of the Anima Mundi)" was "more spiritual" and seemed "to act according to magical and mystical laws". In texts of the 1 670's, Hooke announced to his colleagues in London the way in which these "magical and mystical laws" could serve to explicate the action of gravity in heaven and earth. Hooke applied the model to cometary action and to his favourite aether, to combustion, fermentation, and to the property of 'congruity' in bodies. In turn, cohesion and spring were modelled as part of Hooke's 'harmonic' cosmology, in which musical harmony played the fundamental analogical and experimental role. In his

1 34

SIMON SCHAFFER

discourse on comets of 1 682, Hooke described m atter and motion as "the male and female of nature", and then bolstered this appeal to apparently occult powers by showing it as "very consonant to the sense I understand of the History of the Genesis of the World". Matter was "materia, material substance, or mater, while motion, the second principle, was "Paternus, Spiritus, or hylarchic spirit, as some call it". In Cometa, too, Hooke used the Hexaemeron as a source of authority for his account of matter, its powers and its motions. Ultimately, in lectures of the 1 6 80's on the soul and its structure, Hooke compared the properties of memory and under­ standing with astronomical events such as eclipses, appealed to the action of light, "the most spiritual action of all we are sensible of in the world", and then told the Royal Society that the soul "may have a much bigger sphere of influencing power, and possibly may extend even out ofthe body, and that to some considerable distance". Just the examples Newton used in his construction of the model of the divine sensorium were used here by Robert Hooke to illustrate this conception of animate power over matter.33 However, once again it is important not to ignore the obstacles confront­ ing any such attempt as that of Hooke to extend radically the boundary of natural philosophical knowledge. He was immediately accused of attempting "to prove the soul mechanical" (a charge promptly denied) and he was assailed by Henry More iIi 1 67 1 for failing to introduce enough of spirit in his own optical theories. His response to More, given in a lecture at the Royal Society in 1 676, illustrates the concerns which differentiated the philosophies which sought to treat spirits. For Hooke, as for Boyle at the same period, the criterion of acceptable philosophy lay in command over nature, and, consequently, in the security of that philosophy. For example, Hooke complained that More's hylarchic spirit, a term which Hooke himself used, nevertheless was not acceptable within a natural philosophy. "What were we the better or the wiser unless we also know how to rule and govern this spirit?", Hooke asked. Unless "we could like Conjurors, command this spirit", Hooke argued, "I am yet to learn by what charm or incantation I should be able to incite the spirit to be more or less active". So the boundary which divided these philosophies of spirit did not discriminate between pure mechanism and pure vitalism. It discriminated between various practical methods of displaying spirits at work. Natural philosophers demanded control over such spirits when they were displayed within their practice ; they excluded the 'hylarchic principle', for example, only because it gave these 'priests of nature' insufficient command over that nature.34

OCCULTISM AND REASON

1 35

6. OCCULT I S M AND PHILO S O PHICAL A B U S E

All the issues raised i n an examination o f occultism and philosophy i n the seventeenth century were disputed in the prolonged debate between English and continental philosophers in the early eighteenth century, in which the Leibniz-Clarke correspondence has come to acquire pride of place. Hutchison's recent paper on occult qualities terminates in this correspondence, and Ross argues that both the Leibnizian accusation of occultism and the public Newtonian disavowal of hypotheses about the cause of gravity are sympathetic positions. The analysis of these polemics must, however, involve more than a contrast between public declarations of the conflicting camps. Two points have emerged in recent study: first, the rival models of the work of the natural philosophy at stake in the controversy, and second, the manner in which such models were con­ nected with vital political interests in Augustan Britain. Natural philoso­ phers defined their work so as to show its utility and its power; Newton and Leibniz , rival court philosophers, competed for a secure place in the new regime. This insight helps us place the function of 'occultism' as a term of rhetorical abuse. To show Newtonian philosophy as occult was to destroy its basis in civil and religious support. For example, the appeal of mechanical philosophy was made out through its correspondence with concepts accessible to all subj ects, and a philosophy displayed as non­ mechanical would therefore be ineffective in moral and epistemological appeal. A philosophy which relied too heavily on the miraculous was labelled as 'scholastic' or 'Papist', serious charges in 1 7 1 5, when lacobitism was a real threat. Furthermore, as I have argued, resources existed both in English and in continental philosophy which could all too easily be made out as occult, just because the display of spirit was such a valuable and unavoidable aspect of philosophical practice. Finally, decisions to publish and to intervene in debate allow the historian to distinguish quite carefully between manuscript and public statements : while, for example, it is clear that Newton was 'convinced' of the need for a material account of gravity, the range of his drafts for responses to Leibniz show how public positions were painstakingly constructed to suit the polemical context. 35 In the 1 6 70's, for example, Leibniz had used his diplomatic missions to London and to Paris to absorb much of the work in the realm of spirit which English and French natural philosophers had developed: this in­ cluded the work of Anne Conway and F. M. van Helmont, as well as the more orthodox texts of Boyle, on which Leibniz commented in a critique

1 36

SIMON SCHAFFER

of Boyle's text on the resurrection to develop the ideas of his Elementa de Mente. Leibniz's remarks about "the soul" as "firmly implanted in this flower of substance", together with his aetherial cosmology presented in the Hypothesis Physico Nova, were later, during the controversy with Clarke, to be portrayed as sober mechanical philosophy, and linked with an equally sober English tradition, now corrupted by Newton's allies. "In the time of Mr. Boyle", Leibniz now wrote, "and other excellent men, who flourished in England under Charles II, no body would have ventured to publish such chimerical notions. I hope that happy time will return under so good a government as the present". Leibniz appealed to Boyle's notion that "every thing was done mechanically in natural philosophy". In debates with H uygens on the rules of impact in inelastic collisions, too, Leibniz used the language of 'perpetual miracles' while H uygens appealed to the universal comprehensibility of his mechanical principles of atoms and the void. Terms such as 'mechanical philosophy' and 'miracles', therefore, were highly evaluative labels of considerable use in philosophical debate.36 On the British side, philosophers such as Keill or Clarke also assessed rival texts in these terms. Keill attacked the sacred physics of Thomas Burnet because it relied too heavily on "Mechanical Principles and natural causes", while in his Introduction to the True Physics, he wrote that "in most of the writings of the Philosophers there is scarce anything Mechanical to be found besides the Name". Cartesians used nothing but "the Miracles of their Subtile Matter", Keill wrote; Keill, in turn, following his paper on the laws of attraction ( 1 708), was assailed for "a certain fantastical scholastic quality or even enthusiasm such as that of Fludd" and other "great scholastic monstrosities". In his anonymous Recensio ( 1 7 1 5), Newton wrote of his own fury at being coupled with occultists : the editors of Acta Eruditorum had suggested he was using "th.e Hylarchic principle of Dr. Henry Moor": "by this sort of Railery they are perswading the Germans that Mr. Newton wants Judgment".37 So the deployment of these labels in philosophical dispute linked those debates with assessments of the propriety and authority of the philoso­ phers themselves. Leibniz used his own terminology of the miraculous and the occult to define the boundary of proper philosophy: he used it against Newton from the moment he read the Principia in Rome in 1 689. Newton and his allies argued that miracles "happen seldom and by happening seldom create wonder"; for Leibniz a miracle was an event inexplicable by a law of nature. He affirmed this distinction in the correspondence with Hartsoeker which, in turn, prompted the response of Newton and Cotes

OCCULTISM AND REASON

1 37

in 1 7 1 2- 1 7 1 3. Here he outlined "the difference between reasonable natural miracles and miracles properly called, or supernatural, or rather . . . between a reasonable explanation and the fictions to which one has recourse to sustain badly founded opinions". The distinction between the two concepts of the miraculous, and thus of the occult, was fundamental, since it mapped two widely different models of proper natural philosophy: for Newton, in his public statements of 1 704- 1 7 1 7, natural philosophy would progressively diminish the apparently miraculous, and until that point it was possible to appeal to substances "of a different nature from mechanism, and yet acting regularly, [which] may well be called natural". For Leibniz, any event now inexplicable from mechanism was ipso Jacto excluded from philosophy. "It must be a perpetual miracle", wrote Leibniz, recalling the term he had used against Huygens, "and if it is not mira­ culous , it is false. 'Tis a chimerical thing, a scholastic occult quality" . In this

context, Newton's lengthy drafts for Conti, in which he prepared a public response to this assault, could legitimately point out the evaluative basis of Leibniz's own terms: "he colludes in the signification of words, calling those things miracles which create no wonder, and those things occult qualities whose causes are occult though the qualities themselves are manifest". 3 8 One final observation follows from the recognition of the evaluative and polemical meaning of terms such as 'occult'. It is in fact unhelpful to adopt an essentialist reading of the corpus of a philosopher's texts. That is, it is unlikely that all such texts will be reducible to a single, consistent view of problems such as those of occult qualities : when, in the contest with Leibniz for position and for authority, Newton discovered his publications attacked for occultism, he responded by working out with Roger Cotes a complex set of changes in the General Scholium he wished to insert in the new edition of the Principia. During spring 1 7 1 3, Cotes and Newton planned an ever longer list of 'hypotheses' which must be excluded from philosophy: to 'mechanical' and 'occult qualities' were successively added 'metaphysical' and then 'physical' hypotheses, until the completed text changed this prohibition from "Hypotheses . . . fugio" to "hypotheses . . . non sequor" to "hypotheses . . . locum non habent".39 But we also know how keen Newton was to find a hypothetical account of natural action such as electricity or gravity, that Newton told Leibniz in October 1 693 that "if someone explains gravity ' " by the action of some subtle matter" then he would be "far from objecting". By 1 697, referring to his correspondence with Richard Bentley and with David Gregory, Newton was preparing a

138

SIMON SCHAFFER

set of scholia which would present an explicitly 'occultist' account of gravity and of spirit and by 1 705-1 706 his drafts contained statements that "we find in ourselves a power of moving our bodies by our thoughts . . . and see the same power in other living creatures, but how this is done and by what laws we do not know. And by this instance and that of gravity it appears that there are other laws of motion (unknown to us)". Not all Newton's colleagues accepted these suggestions: Gregory reported that Christopher Wren, in 1 698, was "in possession of a method of explaining gravity mechanically. He smiles at Mr. Newton's belief that it does not occur by mechanical means, but was introduced originally by the crea­ tor".40 Historians have smiled in similar terms at the apparent switches and agonies discoverable in Newton's manuscripts and public declarations on causality and mechanism: commenting on the wild twists and turns of Newton's occult or mechanical aethers, M. B. Hall has speculated that "ultimately, perhaps, he did not require it", while elsewhere A. R. Hall and M. B. Hall have asked how "Newton, the great abstainer from hypotheses" could "not have seen how pressed with difficulties these hypotheses of his own are?". Yet these are false problems for historians. Instead, by being attentive to the context of use in which each statement was designed to appear, and the effect it was to have in polemic, such terms as 'occult quality' might lose their over-arching consistency yet recapture their full historical import.4 1 The context of use of the terms 'occult quality' and 'perpetual miracle' in the Newton-Leibniz exchanges was deeply political. It posed in an acute form the issue of the proper mode of work for philosophy. The Court Whig support for Newton and his allies determined their statements on the production of knowledge of God, nature and morality. The care with which these statements were disseminated testifies to this influence. Caroline told Leibniz that "Neither Dr. Clarke nor Mr. Newton wishes to be thought a follower of Mr. Locke, but I cannot and would not wish to be one of theirs", while Clarke himself discriminated between the 'true' Locke, and the unfortunate Lockean doctrine of the material soul, "but herein he has been followed only by some materialists, enemies to the mathematical principles of philosophy, and who approve little or nothing in Mr. Locke but his errors".42 Again and again in these exchanges, protagonists used the political basis of the struggle to make their point and to display their opponents as interested and corrupt. Bernoulli listed for Leibniz the heretical, visionary and corrupt character of the members of Newton's clique (including Keill, 'Newton's ape' and Whiston, 'who strives to revive

OCCULTISM AND REASON

1 39

Arianism'), while Leibniz wrote to Bernoulli that "certain persons have acted not as mathematicians and Fellows of the Royal Society against a Fellow, but as Tories against a Whig". Hence Leibniz's own claims that Newton had revived "a more than papistical theology" and "resuscitated the occult qualities of the schools by their attractions" were coupled with the remark, disingenuous enough, that "I had not imagined that the spirit of faction would extend itself to the mathematical sciences". Ultimately, the meaning of such terms can only be understood historically through their use, and in these debates their use changed and became powerful. Many writers could present the Leibniz-Newton battle as a return to occultism which had been abandoned for over half a century, or as a break with occultism which had dominated philosophy for j ust that period. "Everything has its fashions", wrote Reaumur on the contest, "nor is philosophy itself an exception to it: those occult qualities, those sympathies and antipathies which nobody would have dared to name in physicks fifty years ago, have, since that time, showed themselves again with splendour". 43

Department of History and Philosophy of Science, Cambridge University

NOTE S I Compare M. B. Hesse, 'Reason and evaluation in the history of science', and P. M. Rattansi, 'Some evaluations of reason in sixteenth and seventeenth-century natural philosophy', in Changing Perspectives in the History of Science, edited by M. Teich and R.M. Young (London: Heinemann, 1973): pp. 1 27-147 and 148-166, with e. Webster, From Paracelsus to Newton (Cambridge: Cambridge University Press, 1 982). 2 O. H. Sabine, The Writings of Gerrard Winstanley (Ithaca: Cornell University Press, 1 94 1 ), p. 61 1 ; O. Winstanley, The Law of Freedom and Other Writings, edited by e. Hill (Harmonds­ worth: Penguin, 1977), pp. 347-348; R. Overton, The Passionate Remonstrance made by His Holinesse in the Conclave at Rome (1 641 ), pp. 48-49; idem. , Araignment of Mr. Persecution ( 1 645), pp. 24-25; idem., An Arrow against all Tyrants ( 1 646), pp. 3-5; idem. , Appealefrom the . . . commons of England ( 1 647), p. 2. 3 M. Cavendish, Observations upon Experimental Philosophy ( 1 666), 'Further observations', pp. I-4; A. Cowley, Proposition for the Advancement of Experimental Philosophy ( 166 1 ), 'Preface'. See B.1. Shapiro, Probability and Certainty in Seventeenth Century England (Prince­ ton: Princeton University Press, 1983), Chapter 2; M. Hunter, Science and Society in Restora­ tion England (Cambridge: Cambridge University Press, 198 1 ), Chapter 7 ; M. e. Jacob and J. R. Jacob, The Anglican origins of modern science', Isis 71 ( 1 980): 25 1-267. 4 J. Wilkins and S. Ward, Vindiciae Academiarum ( 1654), p. 6 1 ; J. Wallis to e. Huygens, 1 January 1659, in Mathematical Work ofJohn Wallis edited by J. F. Scott (London: Taylor and Francis, 1938), pp. 1 70-1 7 1 ; J. Wallis, Hobbius heauton-timorumenos (Oxford, 1 662). For

140

SIMON SCHAFFER

Huygens's notes on Descartes, see Oeuvres Completes (The H ague: NijholT, 1 888-1950), 22 Volumes, Volume X, p. 403. For Descartes as fanatic, see Lord Conway to Anne Conway, 1 July 165 1 , in The Conway Letters, edited by M . H. Nicholson (New Haven: Yale University Press, 1930), pp. 29-30; John Sergeant, Non Ultra, or a Letter to a Learned Cartesian ( 1 698), pp. l08-109. Casaubon is discussed in M. R. G. Spiller, 'Concerning natural experimentall philosophie': Meric Casaubon and the Royal Society (The Hague: NijholT, 1980), p. 205 and M. Hunter, 'Ancients, moderns, philologists and scientists', Annals ofScience 39 ( 1 982): 1 87-192. 5 K. Hutchison, 'What happened to occult qualities in the scientific revolution T , Isis 73 ( 1 982): 233-253 and 'Supernaturalism and the mechanical philosophy', History of Science 21 ( 1 983): 297-33. 6 J. Glanvill, The Vanity of Dogmatizing ( 1 66 1 ), pp. 228-229; idem., Scepsis Scientifica ( 1 665), 'To the Royal Society'. 7 W. Charleton, A Ternary of Paradoxes ( 1650), sig. e3'; C. Webster, 'Henry More and Descartes: some new sources', British Journalfor the History of Science 4 ( 1 969): 359-377. 8 W. Charleton, Physiologia Epicuro-Gassendo-Charltoniana ( 1 654), pp. 50-5 1 . 9 1. Edwards, Compleat History of all the Dispensations and Methods of Religion ( 1 699), pp. 744-745; E. Stillingfleet, Origines Sacrae ( 1 662), pp. 103-104; H. Power, Experimental Philosophy ( 1 664), 'Preface' and pp. 1 9 1-192. 1 0 John Moore's sermons, Cambridge University Library M S S.Dd. 14.9 f. 49; Henry More, Antidote against Atheism (third edition, in Collection of Several Philosophical Writings, 1662), pp. 40-46; idem., Modest Inquiry into the Grand Mystery of Iniquity ( 1 664), pp. 483-489. II More, Antidote, ibid., N. Culverwel, Elegant and Learned Discourse of the Light of Nature ( 1 652), p. I S . For More, see C. Staudenbauer, 'Platonism, theosophy and immaterialism', Journal of the History of Ideas 35 ( 1 974): 1 57-169; A. Coudert, 'A Cambridge Platonist's Kabbalist Nightmare', Journal of the History of Ideas 36 ( 1 975): 633-652; A. Gabbey, 'Philosophia Cartesiana Triumphata: Henry More 1646-167 1', in Problems in Cartesianism, edited by T.M. Lennon et al. (Montreal: McGill-Queens University Press, 1 982), pp. 1 7 1-250. 12 Charleton, A Ternary of Paradoxes, sig. dlV, d4v-eF. 1 3 Henry More, Immortality of the Soul ( 1 659), p. 56; Joseph Glanvill, A Blow at Modem Sadducism ( 1 668), p. 1 16; idem., Praefaratory Answer to Mr. Henry Stubbe ( 1 6 7 1 ) pp. 1 55-157; idem. , Essays on Several Important Subjects ( 1675), pp. 3-4, 14, 59; idem. , Saducismus Triumphatus ( 1 68 1 ), p. 1 6. See also Gabbey, op. cit. , p. 2 1 3, on spirit testimonies. 14 Boyle to du Moulin, in Robert Boyle, Works, edited by T. Birch, second edition (t772), 6 Volumes, Volume I, pp. ccxxi-ii; Hartlib to Boyle, 14 September 1 658, ibid., Volume VI, p. 1 14; Boyle to Glanvill, 18 September 1677 and 1 0 February 1678, ibid., pp. 58-60. For the Demon of Macon, see C. Webster, From Paracelsus to Newton, pp. 92-94. 15 Thomas Hobbes, Leviathan ( 1 6 5 1 ), p. 7; J. WagstalTe, The Question of Witchcraft Debated (second edition, 1 67 1 ), pp. 1 12-1 1 3 ; F. Hutchinson, Historical Essay Concerning Witchcraft ( I 7 1 8); Daniel Defoe, A System ofMagick ( 1 727). See K. Thomas, Religion and the Decline of Magic (London: Weidenfeld, 197 1 ), Chapter 1 6 and T. H. Jobe, The Devil in Restoration science', Isis 72 ( 1981): 343-356. 16 For Coker, see More to Anne Conway, June 1654, in Conway Letters, pp. 101-102; for Greatrakes, see N. H. Steneck, 'Greatrakes the Stroker: the interpretations of historians', Isis 73 ( 1982): 1 61-177 and J. R. Jacob, Henry Stubbe. Radical Protestantism and the Early Enlightenment (Cambridge: Cambridge University Press, 1983), Chapter 3. 1 7 Thomas Browne, Pseudodoxia Epidemica (third edition, 1658), p. 7. On enthusiasm and social control; G. Rosen, 'Enthusiasm, a 'dark lanthorn of the spirit", Bulletin of the History

OCCULTISM AND REASON

141

of Medicine 42 ( 1968): 393-42 1 ; M. Heyd, 'The reaction t o enthusiasm in the seventeenth century', Journal of Modern History 53 ( 198 1): 258-280; M. MacDonald, 'Religion, social change and psychological healing in England 1600-1800', Studies in Church History 19 ( 1982): 10 1-126. 1 8 John Lacy, Letter to Dr. Josiah Woodward ( 1 708), p. 1 8 ; John Trenchard, Natural History of Superstition ( 1 709), pp. 1 3-20; John Humfrey, An Account of the French Prophets ( 1 708), p. 8. See Hillel Schwartz, Knaves, Fools, Madmen and that Subtile EfJluvium: A Study 0/ the Opposition to the French Prophets (Gainseville: Florida University Press, 1978); M. C. Jacob, 'Newton and the French prophets: new evidence', History 0/ Science 16 ( 1 978): 1 34-142. 1 9 J. E. McGuire and Martin Tamny, Certain Philosophical Questions: Newton's Trinity Notebook (Cambridge: Cambridge University Press, 1983), pp. 238 and 376; A. R. Hall and M. B. Hall, editors, Unpublished Scientific Papers of Isaac Newton (Cambridge: Cambridge University Press, 1962), pp. 142-143. 20 Descartes to Elizabeth, 31 May 1 643 and 28 June 1 643, in Descartes, Oeuvres, edited by C. Adam and P. Tannery (Paris: 1 897- 1 9 1 0), Volume III, pp. 663-668 and 695; Descartes to Arnauld, 29 July 1648, ibid. , Volume V, pp. 221-223. These texts are translated in N. Kemp Smith, Descartes: Philosophical Writings (New York: Modern Library, 1 958), pp. 249-263. 21 Descartes to Henry More, August 1649, Oeuvres, Volume V, pp. 403-404; Descartes, Passions de ['arne, art. 47 in Oeuvres, Volume XI, pp. 364-366; Kemp Smith, Descartes: Philo­ sophical Writings, pp. 283-285. 22 Leibniz, Philosophischen Schriften, edited by C. Gerhardt, (Berlin: 1 875-1 900), Volume IV, pp. 340-34 1, 497, 54 1 (to Bayle), (on the violation of the conservation law); ibid., Volume V, pp. 208-209 (Nouveaux Essais); ibid., Volume VI, p. 620 (Monadology). See Y. Belaval, 'Animadversions de Leibniz sur les Principes de Descartes', in Melanges Alexandre Koyre, edited by I. B. Cohen and R. Taton, 2 Volumes (Paris: Hermann, 1964), Volume II, pp. 29-56; A. R. Hall, 'Mechanics at the Royal Society 1 668-1670', British Journal for the History of Science 3 ( 1 966): 24-38 ; D. M. Clarke, 'Impact rules of Descartes's physics', Isis 68 ( 1 977): 55-66. 23 Newton, 'Reflections', Cambridge University Library, M S S.ADD.4004, If. 1 3-14; 'Lawes of motion', ibid. , M SS.ADD.3958.5, tf. 8 1-83. See Isaac Newton, Co"espondence (Camb­ ridge: Cambridge University Press, 1959-1977), 7 Volumes, Volume III, pp. 60-64; idem., Opticks (fourth edition, 1 730), (New York: Dover, 1952), p. 398; J. Herivel, Background to Newton's Principia (Oxford: Clarendon, 1965), pp. 1 4 1-179; R. S. Westfall, Force in Newton 's Physics (London: MacDonald, 197 1 ), pp. 343-355. 25 Cambridge University Library, M S S .ADD.396S . 1 3 , If. 541-2/545-6; ibid. , MSS.ADD.3970, 1f. 252v, 6 19-620r; Newton, Opticks, p. 403; Hall and Hall, Unpublished Scientific Papers, pp. 14 1-142; Newton to Locke, 30 June 1691, Co"espondence Volume III, pp. 152-1 54. See A Koyre and I. B. Cohen, 'The case of the missing tanquam: Leibniz, Newton and Clarke', Isis 52 ( 1961): 555-566; J. E. McGuire, 'Newton on place, time and God', British Journal /or the History 0/ Science 11 ( 1978): 1 14-129; M. Tamny, 'Newton, creation and perception', Isis 70 ( 1 979): 48-58. 25 John Wilkins, Of the Principles and Duties of Natural Religion (16'15), pp. 18-19 and 395; E. Stillingfleet, Origines Sacrae ( 1 662), pp. 357 and 422; Thomas Sprat, History of the Royal Society ( 1 667), p. 362; Boyle, Works, Volume VI, pp. 752-753 ('Christian Virtuoso'). 26 Boyle, 'Usefulness of natural philosophy', in Works, Volume II, p. 30; Sprat, History, p. 360; for providence in politics, see M. C. Jacob, The Newtonians and the Eng/ish Revolution (Hassocks: Harvester, 1976), Chapter 2.

142

S IMON S C H AFFER

27 Locke, Essay, II.x.9; II.xxiii. 1 3 ; III.xi.23; IV.iii.6; IV.iii.27. Boyle, 'Hydrostatical discourse', in Works, Volume III p. 627; 'Final causes of natural things', ibid. , Volume V, pp. 401 and 443. See H . Fisch, 'The scientist as priest: a note on Robert Boyle's natural theology', Isis 44 ( 1 953): 252-265; J. E. McGuire, 'Boyle's conception of nature', Journal of the History of Ideas 33 ( 1 972): 523-542; .T. G. Lennox, 'Boyle's defense of teleological inference in experimental science', Isis 74 ( 1983): 38-52. 29 Boyle, ' Suspicions about some hidden qualities in the air', in Works, Volume IV, pp. 96-97; 'Final causes of natural things', ibid., Volume V, p. 442; 'General History of the Air', ibid., Volume V, p. 641 . 30 Boyle, 'Excellency and grounds of the mechanical hypothesis', in Works, Volume IV, p. 73. 3 1 Boyle, 'Christian Virtuoso', in Works, Volume VI, pp. 754-756 and 770-773. 32 Henry Power, Experimental Philosophy ( 1 664), p. 6 1 ; C. Webster, 'Henry Power's Experi­ mental Philosophy', Ambix 14 ( 1 967); 1 50-1 78, p. 1 74; and T. Cowles, 'Dr. Henry Power, disciple of Thomas Browne', Isis 20 ( 1933-34): 345-366; John Mayow, Tractatus Quinque (Oxford, 1 674), translated as Medico-physical Works, edited by A. Crum Brown and 1. Dobbin (Edinburgh: Livingstone, 1957), pp. 35-36, 109 and 252-253; H. Guerlac, Essays and Papers in the History of Modern Science (Baltimore: lohns Hopkins, 1977), pp. 245-274 and R. G. Frank, Harvey and the Oxford Physiologists (Berkeley: California University Press, 1980), Chapter 1 0. 33 Robert Hooke, 'Comet a', in R. Gunther, Early Science in Oxford (Oxford: privately printed, 193 1 ), Volume VIII, pp. 227-23 1 , 249 and 263; R. Waller, editor, Posthumous Works ofRobert Hooke ( 1 705), pp. 145-147 and 172-176. See I.A. Bennett, 'Cosmology and the magnetic philosophy 1 640-1 680', Journalfor the History ofAstronomy 12 ( 1 98 1 ): 1 65-177 and P. Gouk, The role of acoustics and music theory in the scientific work of Robert Hooke', Annals of Science 37 ( 1 980): 573-605. 34 Hooke, 'Lampas', in Gunther, Early Science, Volume VIII, pp. 1 87-1 88. 35 See G. H . R. Parkinson, 'Science and metaphysics in the Leibniz-Newton controversy', Studia Leibnitiana (Supplementa) 2 ( 1 969): 79-1 12; C. litis, The Leibnizian-Newtonian debates : Natural philosophy and social psychology', British Journalfor the History of Science 6 ( 1 973): 343-377; A. R. Hall, Philosophers at War : The Quarrel between Newton and Leibniz (Cambridge; Cambridge University Press, 1980); S. Shapin, 'Of Gods and Kings: politics and the Leibniz-Clarke dispute', Isis 72 ( 1 98 1 ): 1 87-215. 36 Huygens, 'Sur Ie Conservation du mouvement', ( 1686), Oeuvres, Volume XIX, pp. 162-165; Leibniz to Huygens, 26 September 1 692, ibid., Volume X, p. 319 and Huygens to Leibniz, 12 January 1693, ibid., pp. 384-387 ; Leibniz, 'Hypothesis physica nova', in Philosophischen Schriften, Volume IV, pp. 1 8 1-219. On occultist sources for some of these claims, see C. Merchant, The vitalism of Anne Conway: its impact on Leibniz's concept of the monad', Journal ofthe History ofPhilosophy 17 ( 1 979): 255-269 and The vitalism ofF. M. van Helmont: its influence on Leibniz', Ambix 26 ( 1 979): 170-183; R . E. Butts, 'Leibniz's monads: a heritage of gnosticism and a source of rational science', Canadian Journal of Philosophy 10 ( 1980): 47-62; for notes on Boyle's 'Physico-theological considerations', see L. Loemker, 'Boyle and Leibniz', Journal of the History of Ideas 16 ( 1 955): 22-43. For Leibniz on Charles II and mechanical philosophy, see The Leibniz-Clarke Correspondence, edited by H . G. Alexander, V. 1 14, p. 92. 37 John Keill, Examination of Dr. Burnet's Theory of the Earth (Oxford, 1698), pp. 36-37; Introduction to the True Physics (edition of 1745), p. 4; review of Keill, 'The laws of attraction', in Acta Eruditorum 29 (September 1 710): 4 1 2-413; [Isaac Newton], 'Account of the book 28

OCCULTISM AND REASON

143

entitled Commercium Epistolicum', Philosophical Transactions 29 (February 1 7 1 5); 1 73-224. These texts are discussed in Hall, Philosophers at War, Chapter 7-9. 38 Samuel Clarke, in Leibniz-Clarke Correspondence, IV.45, p. 53; Leibniz, ibid. , V. 94 and V. 120-122, pp. 86 and 94; Leibniz to Hartsoeker, February 1 7 1 I , in Philosophischen Schriften, Volume III, p. 5 1 6; Newton to Conti, February 1 7 16, in Correspondence, Volume VI, p. 285. Compare the drafts printed in A. Koyre and I. B. Cohen, 'Newton and the Leibniz-Clarke correspondence', Archives Internationales d'Histoire des Sciences 15 ( 1962): 64-126, p. 73. 39 Cambridge University Library MSS.ADD.3965 fT. 357v and 363'; ibid., M SS.ADD.3968 f506 (on the drafting of Cotes's preface); Newton to Cotes, 6 January and 2 March 1 7 1 3, and Cotes to Newton, 1 8 March 1 7 1 3 , in Newton, Correspondence, Volume V, pp. 361 , 384, 391-393. The revisions are discussed in Hall and Hall, Unpublished Scientific Papers, pp. 350 and 353, 356 and 360 and in I. B. Cohen, Introduction to Newton 's Principia (Cambridge: Cambridge University Press, second edition, 1979), pp. 240-245. 40 Newton to Leibniz, 16 October 1693, in Newton, Correspondence, Volume III, pp. 285-286; Gregory memorandum, 20 February 1 698, in Newton, Correspondence, Volume IV, p. 267; Cambridge University Library MSS.ADD. 3970 f620', discussed in J. E. McGuire and P. M. Rattansi, 'Newton and the Pipes of Pan', Notes ana Records of the Royal Society 2 1 ( 1 966): 108-143 and J. E. McGuire, 'Force, active principles and Newton's invisible realm', Ambix 15 ( 1 968): 154-208. 4 1 A. R. Hall and M. B. Hall, 'Newton's matter theory' in Newton 's Annus Mirabilis, edited by R. Palter (Cambridge, Massachusetts: MIT Press, 1970), p. 64 and M. B. Hall, 'Newton and his matter theory in the eighteenth century', Vistas in Astronomy 22 ( 1978): 453-459. 42 Samuel Clarke, in The Leibniz-Clarke Correspondence, 1.2, p. 12; Caroline to Leibniz, 26 November 1 7 1 5, ibid. , p. 190. 43 Leibniz to Bernoulli, 8 August 1 7 1 3 and Bernoulli to Leibniz, 17 February 1 7 14, in Newton, Correspondence, Volume VI, pp. 21-22 and 67-68; R. A. F. de Reaumur, The Art of Hatching and Bringing Up Domestick Fowls ( 1 750), pp. 462-463 in E. Gasking, Investigations into Generation 1651-1828 (London: Hutchinson, 1967), pp. 83-84.

G. MacDONALD ROSS

R E P LY T O S I M O N S CH A F F E R

I am most indebted t o D r Schaffer for adding such a wealth o f scholarly detail to what was often only general and sketchy in my paper. If I have any criticism to make of his comments, it is that he is too tactful in avoiding explicit criticism of my claims. The nearest I find to straight rejection of any of my positions is in his section 'Occultism and Philosophical Abuse'. We both seem to hold that the contemporary disagreements as to what was to count as occult make it a waste of time trying to define occultism in terms of a particular kind of belief, specified by internal criteria. We certainly agree that at least an important component of the meaning of the word 'occult' is its rhetorical function as a term of abuse. We also see this fact as having socio-political significance; but here we diverge over the nature of this significance. Dr Schaffer brings the term into the political realm by claiming that occultism was associated with Popery and J acobitism. This is no doubt true to a certain extent - indeed I myself emphasized the difficulty of demarcating secular from religious, especially Catholic, magic. But as a matter of historical fact, seventeenth-century Catholics were no more prone to occultism than Protestants. If anything it was the other way round: one of the greatest flowerings of occultism was during the Commonwealth in Protestant England, and the acceptability of occultism began to fade away first in Catholic France. It may be that British polemicists misconstrued the situation in their own age, and that an association of occultism with Catholicism is therefore relevant to the understanding of contemporary British politics. But the peculiarities of British politics cannot begin to explain the ·astonishingly rapid demise of occultism over the whole of Western Europe almost simultaneously. My gesture towards the socio-political was slanted more towards the sociological end of the spectrum. In particuJar, I suggested that what is known as 'labelling theory' might shed some light on the function of terms like 'occult'. One reason why it might be useful is that it offers a relatively neutral way of describing people's attitudes to a particular label, and it becomes more readily intelligible how different groups or individuals could simultaneously use it in a pejorative and in a commendatory sense. Too many questions are begged by the use of expressions like 'philosophical

1 45 A. J. Holland (ed.), Philosophy, Its History and Historiography, 145-147. (c) 1 QR" }m 1) RbiAol PtlhJ;,,].iMn r....""" .... _�.

146

G. M acDONALD ROSS

abuse' or 'polemical meaning'. In my view, throughout the seventeenth century there was considerable ambivalence as to the extent to which occultist labels were in fact derogatory. People like Fludd and Vaughan were obviously proud of the esoteric wisdom to which they were party, and which was a closed book to adherents of the currently fashionable mecha­ nist and materialist philosophy. My difficulty is that the concept of a label is, in this context in particular, closely related to the concept of deviance. As I put it in my paper (Section 4): "The occultist is essentially an intellectual deviant". But my conclusion was : "that the concept of occultism is of little use when there is no well-defined orthodoxy for it to deviate from". In the eighteenth century there is no great problem, since there was an easily-definable orthodoxy relative to which underground occult societies were clearly deviant. But can the notion of deviance get a foot-hold at all in the seventeenth? If not, then the problem of how to make sense of the concept of the occult in that context seems almost intractable. It is not even possible to describe the term 'occult' as abusive, since it was a brickbat that had as often as not turned into a bouquet by the time it arrived. The best one can do is, if not to define an orthodoxy, then at least to indicate a rough area of consensus as to where a line was to be drawn between the limits of reasonable disagreement, and the outrageously outre. Even so, the notion of straying beyond this line into overt occultism or whatever, will have less historical work to do than in other centuries, when orthodoxy is more sharply defined. But it is surely a paradoxical conclusion that we should abandon the concept of the occult j ust for the very century in which the occult, whatever it may be, is of greatest interest. The solution to this paradox is, I believe, that while the general label has no useful work to do, there is much to be said using more specific concepts, such as those of astrology, alchemy, spiritualism, and so on. Another problem with the concept of deviance, which I merely touched on in my paper, is that it is essentially a sociological concept, leaving out of account what is peculiar to occultism in particular. To define specifically occult deviance, as contrasted with heresy or political dissidence, for example, would seem to require an internalist definition of occultism, which is precis.ely what an externalist approach was supposed to avoid. In my paper ( Section 4) I claim that "it is possible to identify a group of 'occultist' attitudes and concepts, often clustered around particular ancient and medieval texts, which have historically tended to remain on the fringes of the intellectual world". But I have not offered any detailed support for

REPLY TO SIMON SCHAFFER

147

this claim, and I have not even established that this is the most promising approach for defining a cluster of 'occultist' beliefs, which would be valid for as difficult a century as the seventeenth. In my paper, I drew attention to the need to relate beliefs to institutional factors, and if I may turn a mirror onto the present debate, I think it is relevant to note a certain institutional difference between Dr Schaffer and myself, namely that while he belongs more to the world of the history of science, I am employed in a more purely philosophical capacity. My central message is that it is high time the history of philosophy woke up to the sort of understanding that is now commonplace in the history of science. As primarily a historian of science, Dr Schaffer has reinforced, indeed embel­ lished, what I had to say in connection with the history of science; but he has not queried my extension of the same sort of approach to the traditional heartlands of philosophy - not that either of us will be very happy at going along with the purist philosopher's sharp dichotomy between philosophy and science. I have made a number of claims which I would expect to raise a number of eyebrows in certain circles. I have interpreted Descartes' cogito as an example of relatively conventional occultist mysticism ; I have interpreted Locke's way of ideas as a hang-over from belief in the effiuence theory of perception, and the existence of magically knowable occult virtues ; and I have related empiricism to the philosophy of the viIIage wise-woman plucking her herbs in the light of the moon. I personally do not think this is any more radical than drawing attention to the alchemical activities of a Newton or a Leibniz, or to the religious and even prophetic dimension of seventeenth-century science. On the other hand, I do not believe that such issues have had a sufficient airing in philosophical circles, and it was my purpose in writing my paper to draw attention to a perspective which is still out of fashion in such circles.

University of Leeds

GARY HATFIELD

FIRST PHILO SOPHY A N D NATURAL P H I L O S O P H Y I N D E S CARTE S ·

Descartes was both metaphysician and natural philosopher. He used his metaphysics (among other things) to ground portions of his physics (or natural philosophy, or science of nature). However, as should be a com­ monplace but is not, he did not think he could spin all of his physics out of his metaphysics a priori, and in fact he both emphasized the need for appeals to experience in his methodological remarks on philosophizing about nature and constantly appealed to experience in describing his own philosophy of nature. 1 It remains unclear exactly what he took to be amenable to empirical support, and how his appeal to experience was squared with his notorious demand for absolute certainty in matters philo­ sophical. It is illuminating to consider Descartes' exploits in physics and meta­ physics against the background of scholastic Aristotelianism. By: focusing on what was novel in Descartes from this perspective, there emerges a different than usual picture of his work and its significance. For, while it may be that to the present-day philosophical mind the most troublesome and perplexing side of Descartes' dualistic ontology is the purported existence of a special mind-stuff, when Descartes first proposed his division of creation into mind and matter, the most troublesome claim for his scholastic audience would have been the conception of matter as a substance whose sole essence is extension. Descartes not only promoted the existence of such a substance, but he contended that all of nature is nothing but passive, inert, extended substance, thus denying in one fell swoop the scholastic Aristotelian conception of nature as populated with active principles and substantial forms. This radical rejection of the scholastic ontology not only embodied a substantively new conception of nature; it carried with it a new conception of the relationship among the three traditional branches of theoretical philosophy - physics, mathe­ matics, and metaphysics - as well as a new ideal of 'scientific' reasoning about natural things. In the Aristotelianism of the high scholastic tradition, as manifested in the commentaries of Thomas Aquinas and conveyed in Descartes' day by the Coimbra commentaries and the textbook of Eustace of S aint Paul,2 it 1 49 A. J. Holland (ed.). Philosophy. Its History and Historiography. 149-164. (C) 1 985 hv D. Reid"l P"hli.hinu r()mnnnv.

I

1 50

GARY HATFIELD

was regarded as in principle impossible for a primary substance - a 'this­ such' - to possess as its form or essence merely the property of mathe­ matical extension. This is to say that in the scholastic ontology, mere extension could not play the role of form in the various form-matter combinations that constitute individual things in nature. Although the scholastics granted that all bodies do have extension, they denied that any given type of natural thing can be defined in terms of its geometrical properties alone. Rather, things are defined in terms of their nature or form, which is their principle of change or motion. As Aquinas said in his commentary on Aristotle's Physics, "Natural philosophy is about natural things , and natural things are those whose principle is a nature. But a nature is a principle of motion and rest in that in which it is. Therefore the science of Nature deals with those things which have in them a principle of motion". 3 This "principle of motion" is otherwise known as the substan­ tial form of a thing. It determines the thing as whatever kind of thing it is - specifically determining its development and its characteristic modes of activity - whether the thing be a man, a rabbit, an oak tree, or a magnet. The high scholastic tradition accepted a neat division among the branches of theoretical philosophy as regards their degree of abstraction from individual existing things. According to high scholastic doctrine, all human knowledge (apart from revealed knowledge) comes by way of abstraction from sensory particulars - even knowledge of God. Physics or natural philosophy is at the first level of abstraction; it abstracts from individual things, and considers bodies insofar as various of them exhibit the same form or nature. Hence natural philosophy studies the several kinds of things, as distinguished by their specific principles of change. Mathematics comes next, and abstracts from all sensible properties of bodies, and indeed from their principles of change, and considers bodies only insofar as they are extended. As it was commonly put, mathematics considers the 'intelligible matter' of objects - their pure extension. These geometrical properties were regarded as mere accidents in that they might be otherwise without changing the substantial form of the thing. Meta­ physics is the most abstract, for it abstracts from all matter, sensible or intelligible, and considers 'being as being', which is to say that it considers the fundamental principles and predicates pertaining to things that exist. The law of contradiction, the law of the excluded middle, the distinctions between substance and accident, necessary and contingent, act and potency, the general predicates 'one', 'true', and 'good', and the investiga­ tion of such forms as may exist independently of matter (God and angels),

FIRST PHILO SOPHY AND NATURAL PHILOSOPHY

151

all are i n the purview of metaphysics, or first philosophy.4 In this scheme of things, natural philosophy does not stand to derive much of its content from first philosophy. Although there was disagree­ ment among various commentators and authors on this matter, 5 a minimal common position can be delineated. First philosophy can certify general principles that are used in natural philosophy, such as the distinctions between substance and accident, and between potency and act. It does not, however, provide knowledge of the particular forms that constitute the natures of things. That is the province of natural science. This knowledge is obtained by the cognitive separation of the universal from the particular, presumably as described in Book II, Chapter 19 of the Posterior Analytics.6 Hence, one can discover the forms extant in nature only through experience with natural things. The number of such forms is indefinitely large, for it must be remembered that included here are not only the forms of the traditional four elements, but also, as Aquinas remarked, "every sort of animal, and their parts , such as flesh and blood, and also plants". 7 There are as many active principles to be investigated as there are organized bodies in nature, including the organized parts of bodies. Against this picture of nature teeming with many substances distin­ guished by their distinct natures, forms, or essences, Descartes promoted a program of ontological austerity. He reduced the number of forms or essences in the visible world to one. If one were to put his point in scholastic terminology, one would say that Descartes had drastically reduced the diversity of form-matter combinations in high scholasticism by decreeing that in all of nature matter has a single form or essence : extension. And in fact Descartes did cast his point in these terms. Thus in the Principles of Philosophy (Pt. I, art. 53) Descartes declared that "extension in length, breadth and depth constitutes the nature of corporeal substance", and that with regard to matter extension is the "one principal property which constitutes its nature and essence". 8 In direct contradiction with the scholastic doctrine that a body's extension is a mere accident, Descartes makes extension the universal essence of natural things. This equation of matter with extension - with the attendant stipulation that nature (including the entire universe, both celestial and terrestrial) is to be conceived as containing nothing but this matter - provided Desc'artes with the fundamental elements or 'framework principles' (so to speak) of his natural philosophy. lIe would allow into his explanations of natural phenomena reference to nothing but extended matter with its various 'modes' (modifications): size, shape, position, and motion. The task of the

1 52

GARY HATFIELD

natural philosopher would then be to discover or to posit various configu­ rations of material particles (of a given size and shape) through which, together with the motions attributed to them, the diverse phenomena of nature can be derived or explained. The proposed explanations specifically cannot attribute to matter as fundamental properties such qualities as color, sound, odor, taste, heat, and cold. Descartes explicitly denies that these sensory qualities are real properties of matter, and maintains that they must be explained in terms of the dispositions of objects, by virtue of their size, shape, and motion, to cause a pattern of motion in the nervous system that gives rise to sensations such as color in the mind. Moreover, the motion of the parts of bodies, which must be invoked in any explana­ tion of natural change, is to be explained by reference to a few simple laws or rules of motion.9 These hold for all matter, as opposed to scholastic principles of motion or change that were specific to each kind of natural thing. Within the framework of Descartes' physics, the characteristic properties of natural bodies must be accounted for by the size, shape, and motion of their parts ; not by a 'substantial form', i. e. , a scholastic nature or essence. These fundamental properties - size, shape, and motion - all seem to derive directly from the introduction of geometrical modifications into pure extension. At first the inclusion of motion may seem problematical it may not seem like a geometrical property. But it can be treated in a purely descriptive or kinematic way, without violating the dictum to attribute nothing to matter that cannot be conceived as a modification of pure extension. And indeed Descartes enjoins that this is the proper understanding of motion. In the Principles (Pt. II, art. 25) he defines motion as "the transference of one part of matter or of one body, from the vicinity of those bodies immediately contiguous to it and considered as at rest, into the vicinity of others", and goes on to draw attention to the fact that motion is defined as the "transference" of a body, rather than the "action" which transfers. Regarding the various particular modifications of extension actually found in the world, one may ask how they arose. How did extension, which in its bare concept contains no blueprint for being carved up into the physical universe, come to be divided into just those parts that it did come to have? For the explanation of these facts, Descartes must go beyond the essence of matter, for in his world the essence of matter does not contain its own principle of motion or change. To explain the diversity of moving particles in the universe, Descartes introduced the 'fictional' hypothesis (as

FIRST PHILOSOPHY AND NATURAL PHILOSOPHY

153

opposed to the 'credal' reality of instantaneous creation of the world much as it is now) that God had introduced a chaotic pattern of motion into the universe at its creation, and that merely by the jostling of these particles the world had sorted itself into what we now see.l O It might seem that once having created this matter and put it in motion, God could retire. Descartes was sometimes read in this way during the seventeenth century, for example by Pascal who accused him of limiting God's role in governing the world to an initial flip of the finger, which set in motion the cosmic mechanism, never to run down. But consider again pure geometrical extension. From the fact that an extended portion of matter is at one instant in motion, does it follow that it will remain in motion over time? What about collisions ? When two geometrically­ conceived bodies collide, what can be determined merely from the fact that they each have a certain size, shape, and motion, except perhaps that they will not interpenetrate? I have said that Descartes' matter lacked a guiding principle of change within itself, an active principle or entelechy. Motion regarded kinematically may allow for descriptive laws of motion, but the laws themselves are not thereby explained, nor is there provided an under­ standing of the causal agency behind the kinematically described motion. This being so, the principles of motion and impact had to be sought elsewhere, and were found in God, acting at each instant in accordance with the three laws of motion as stated in The World and extended in the Principles ofPhilosophy. I I Descartes thus reduced the scholastic menagerie of active principles in nature to one. Of course, Descartes is in accordance with scholastic doctrine in making God the ultimate cause of natural change. But his creative application of traditional doctrine makes for a conception of .nature quite different from that of the scholastics. By banishing active principles from matter, he makes motion considered purely as natural a geometrically (kinematically) conceived property, and restricts actual force or causal agency to a being outside of nature, whose causal efficacy presumably is beyond doubt and requires no humanly intelligible explanation. H aving catalogued Descartes' clean sweep of the fundamental explana­ tory principles and entities of high scholasticism, one may ask how Descartes justified such radical housecleaning. He employed two stra­ tegies, the better known of which depended on his metaphysics. His meta­ physical ground for rejecting substantial forms and real qualities was that they cannot be clearly and distinctly understood by the human intellect. As he wrote to Mersenne in 1 643 , "My principal reason for rejecting these real

1 54

GARY HATFIELD

qualities is that I do not see that the human mind has any notion, or particular idea, to conceive them by; so that when we talk about them and assert their existence we are asserting something we do not conceive, and doing something we do not understand". 1 2 The intellect can intuit the geometrical properties of extension, but it can form no direct conception of a real quality conceived as a principle of change. These forms can be known only by their manifest effects, by alterations or motions that are perceivable. Descartes' claim, then, is that kinematically conceived motion and geometrically describable shapes and sizes are intelligible, whereas specific patterns of growth and alteration - principles for altering the geometrical and other properties of natural objects - are not. Of course, a scholastic sympathizer might well ask for the criterion of intelligibility at work here. Descartes' response would draw upon the basic elements of his metaphysics and epistemology. For Descartes' meta­ physical justification of the assertion that extension is the essence of matter depended upon the epistemic criterion of clear and distinct per­ ception, where this is ultimately interpreted as clear and distinct per­ ception by the intellect alone, operating independently of the senses and therefore of sensory abstraction. The metaphysical justification of the fundamental principles of his physics depended upon a novel conception of metaphysical knowledge - one that bases such knowledge on pure rational intuition as opposed to abstraction from sensory particulars - as developed in his Meditations on First Phi/osophy. 1 3 The second type of justification that Descartes gave for the framework principles of his physics was, perhaps surprisingly, based on an 'empirical' criterion of explanatory adequacy. Thus, in the letter to Mersenne quoted above, Descartes continued the explanation of his rejection of 'real qualities': The second reason is that the philosophers invented these real qualities only because they did not think they could otherwise explain all the phenomena of nature; but I find on the contrary that these phenomena are better explained without them. 14

Although this letter does not elaborate further the claim of having provided a superior explanation, that theme is expounded in numerous other places. Specifically, in the Discourse, the Meteors, letters from 1 638, and in the Principles, Descartes argues that the first principles of his physics can in fact be given an a posteriori justification, based upon their explanatory success and their comparative unity and simplicity. I S

FIRST PHILOSOPHY AND NA TURA L PHILOSOPHY

155

This a posteriori mode of establishing his principles depended upon reasoning from 'effects to causes.' Descartes made bold claims for its efficacy. Thus, in a letter to Vatier from February of 1 638, he contended that he could 'demonstrate' the principles introduced in the Meteors basically, the use of only particles of various sizes and shapes colliding in a plenum in framing explanations - through their comparative success in explaining a range of known phenomena (,effects'): I cannot demonstrate a priori that which I have supposed at the beginning of the Meteors without expounding my whole physical theory; but the phenomena [experiences] which I have deduced necessarily from them, and which cannot be deduced in the same way from other principles, seem to me to demonstrate them sufficiently a posteriori. 16

One might object, as did another correspondent, Morin, that this mode of reasoning proves nothing, since "nothing is easier than to fit a cause to an effect". Descartes defended his practice in a letter to Morin of 1 6 3 8 : "It is true that there are many effects to which it is easy to fit many separate causes; but it is not always so easy to fit a single cause to many effects, unless it is the cause which truly produces them. There are often cases in which in order to prove what is the true cause of a number of effects, it is sufficient to give a single one from which they can all clearly be deduced".17 We shall see below that this sort of reasoning might be used to justify the postulation of a particular causal mechanism in the explana­ tion of, e.g. , magnetism or the properties of water. Of interest here is that Descartes used it to justify the most general principles of his account of nature, insofar as they are implicated in his unified account of natural phenomena. In fact, Descartes invited one to "compare all their [the scholastics'] real qualities, their substantial forms, their elements and their other countless suppositions with my single supposition that all bodies are composed of parts". 18 The unity and simplicity of his explanatory prin­ ciples and particular explanations he regarded as an a posteriori demon­ stration of their merit. One may remark that to the extent that the effects are not rigorously but only loosely 'derivable' from the supposed causes and principles - here, in the sense that these causes and principles provide the fundamental elements from which purportedly successful explanations have been constructed - this a posteriori 'demonstration' has the less force, a fact that was not lost on Descartes. Before following up this last remark, let us pause to consider what may seem to be the strangeness of Descartes' procedure. It may seem that there

1 56

GARY HATFIELD

is something odd - even contradictory - about providing an empirical j ustification for fundamental principles that one believes have been or could be derived a priori. The 'contradictoriness' of this practice would be more apparent to a present-day philosopher than to Descartes, for we are likely to link 'a priorI' with 'analytic' with 'necessarily true', and to sit pat with the notion that no j ustification that shows something to be 'neces­ sarily true' can be improved upon. However, Descartes' use of 'a priort is less fixed than ourS. 1 9 And in fact, while he believed that his metaphysical results provided the highest certainty possible to human reason - more certainty even than geometrical demonstration - for all that he claimed only that they were the best we could do, not that it would be impossible for some being (e.g., an angel) to do better.2o If one has the best single 'proof possible, but not the best proof possible tout court, it is not contradictory to reinforce it with evidence that is persuasive even if the reinforcement is not, by itself, as persuasive as the first bit of evidence. But Descartes' primary motivation in providing a posteriori arguments for his first principles was not simply to strengthen his position overall. He wanted to explore a different style of argumentation, one that he would use more fully in his reasoning about particular explanatory mechanisms. He explained this motivation in the same letter to Vatier as was quoted above: "I wanted to try whether the simple exposition of truth would be sufficient to carry conviction without any disputations or refutations of contrary opinions". 21 Descartes was consciously testing an alternative to 'a priort or 'metaphysical' modes of 'proving' his principles (to indicate his words). This discussion of the motivation for an a posteriori j ustification invites fuller specification of the motivation for the metaphysical justification. In the letters from 1 638 in which Descartes emphasizes the a posteriori proofs of his principles, he goes on quickly to add that nonetheless he could give an a priori demonstration of his first principles. What distinctive role could such a 'demonstration' play? The answer, I believe, was that he considered this a priori demonstration the only means of disposing of other explana­ tory principles, such as substantial forms, once and for all. The distinctive role of metaphysics, then, is to establish as certainly as possible what could be claimed as merely empirically plausible if left to its own, a posteriori, devices. This reading of the role of metaphysics is supported by the fact that in the Meteors and in a letter to Regius of 1 642, Descartes was careful to say that the a posteriori fruitfulness of his own mode of explaining nature does not necessitate the rejection of substantial forms, but that it simply shows that they are unnecessaryP

FIRST PHILOSOPHY AND NATURAL PHILOSOPHY

1 57

If empirical modes of reasoning merely supplement metaphysical (here, a priori rational) modes of reasoning in establishing the fundamental prin­ ciples of natural philosophy, they play a much stronger role in reasoning about the particular mechanisms that are invoked to explain specific natural phenomena (such as the properties of oil and water). Avowedly, Descartes proposed (or claimed he could provide) explanations for "all Natural Phenomena".23 Notoriously, the explanations he did propose were soon regarded as unfounded and doubtful. Of present interest is the fact that, as a number of commentators have remarked, in both the Discourse and the Principles Descartes insisted that a priori derivation of the whole of physics is impossible. Since many mechanisms - many combinations of shape, size, and motion - might account for the same phenomenon, just as there might be two clocks that appear the same on the outside and yet contain quite different works, one must seek to discern the actual mecha­ nisms via experimentation. 24 This use of experiment is a species of the reasoning from effects to causes discussed above, as applied to the postu­ lation of specific mechanisms to explain specific phenomena. The similari­ ty between this type of reasoning and hypothetical-deductive reasoning, complete with 'model-fitting', has been explored with care and penetration by several authors.2 5 Moreover, it has also been noted that in conformity with the hypothetical character of particular mechanistic explanations , Descartes was prepared to accept that the truth of his explanatory hypo­ theses can be ascertained only with less than metaphysical certainty. His technical term, adopted from scholastic terminology, was 'moral certainty' : enough certainty to proceed in using one's hypotheses as if they were true. 26 What has perhaps not been sufficiently appreciated is the extent to which Descartes was himself grappling to alter the scholastic conception of science as demonstrative in order to accommodate the hypothetical character he ascribed to the knowledge of his particular explanatory mechanisms. In order to clarify this point, it may be helpful to review and to extend my comparison of Descartes and high scholastic Aristotelianism. We have seen that the Aristotelians believed that metaphysics could provide only the basic framework within which scientific knowledge of nature must fit; explanations of specific natural things depend upon the physicist's abstrac­ tion of principles of change peculiar to the specific thing. These principles then serve as the premises in providing an account of the phenomena associated with the type of thing under investigation. Descartes was

1 58

GARY HATFIELD

seeking to derive an alternative framework by establishing that the one universal form of all natural bodies is extension. He claims to have gotten more from his metaphysics than just his framework principles, for beyond arguing that any explanation in natural philosophy must be framed in terms of moving particles , he purports to have derived specific laws of motion a priori from God's immutability, and to have established the plenum (denying the vacuum) by reflecting on the fact that extension is the essence of matter (there is no place in Descartes' ontology for space distinct from extension). Yet just as the Aristotelian natural philosopher must investigate individual principles of change empirically, so too for Descartes one must discover the particular mechanisms that yield specific phenomena through empirical investigation. These mechanisms may then be used to explain the phenomena in question. Looking over the changes and noting the parallels, it may seem that Descartes has not gained much. For, it might be argued, it is difficult to see why God's immutability must lead to the particular conservation principles espoused by Descartes, and this difficulty calls into question the supposed 'derivation' of the laws of motion from metaphysics; moreover, in his investigation of particular (kinds of) natural things, Descartes has simply substituted many empirically-investigated mechanisms for many empirically-investigated forms, with no gain. Such an assessment would miss the point in two ways. First, even if the 'derivation' of the laws of motion from metaphysics is questionable, and even if Descartes' actual laws needed to be revised, the metaphysical vision behind the idea that all of nature, celestial and terrestrial, might be con­ structed from a small set of properties of matter and a few laws of motion was historically efficacious, bearing fruit in Newton. Second, it misses the way in which Descartes altered the conception and ideal of scientific 'demonstration' or 'proof' maintained by scholastic Aristotelianism. According to the high scholastic tradition, the forms , natures, or essences discriminated as principles of change in natural things provide the basis for premises in demonstrative arguments. Physics is a science, and scientific knowledge is demonstrative knowledge of the reasoned fact - the facts being the phenomena of nature.2? In altering this conception, Descartes didn't just ruefully admit that 'moral certainty' is the best one can get; he went to work on the idea of 'demonstration' itself. Here the language that Descartes used is quite striking. Consider the following oft-quoted passage from the Discourse:

FIRST PHILOSOPHY AND NATURAL PHILOSOPHY

1 59

If some of the matters of which I spoke in the beginning of the Dioptrics and Meteors should at first sight give offence [to readers] because I call them hypotheses and do not appear to care about their proof, let them have the patience to read these in entirety, and I hope that they will find themselves satisfied. For it appears to me that the reasonings are so mutually interwoven, that as the later ones are demonstrated by the earlier, which are their causes, the earlier are reciprocally demonstrated by the later which are their effects. And it must not be imagined that in this I commit the fallacy which logicians name arguing in a circle, for, since experience renders the greater part of these effects very certain, the causes from which I deduce them do not so much serve to prove their existence as to explain them; on the other hand, the causes are proved by the effects. 28

Several things should be noted. First, Descartes wryly remarks on the form of Aristotelian demonstration of the reasoned fact, the conclusions of which are the phenomena to be explained. Certainly, he suggests, one doesn't want to prove that the phenomena exist, for they are the alleged starting point of all knowledge of nature (one recalls Aristotle's twinkling stars). Second, he uses the key Aristotelian term 'demonstration' with two implied senses : he glosses 'causes demonstrating effects' with causes explaining effects (as in effects being 'derived' from causes); he glosses 'effects demonstrating causes' with effects proving causes. But 'prove' (prouver) had at that time not only the sense of 'logical proof, but also the sense of 'confirm', or even 'test'. Support for this reading is provided by correspondence explicating this passage. In a letter of July 1 638, Descartes responds to Morin's charge that there is "a vicious circle in proving effects from a cause, and then proving the cause by the same effects": I agree: but I do not agree that it is circular to explain effects by a cause, and then prove the cause by the effects; because there is a big difference between proving and explaining. I should add that the word 'demonstrate' can be used to signifY either, if it is used according to common usage and not in the technical signification given it by the Philosophers.29

The word 'demonstrate' is here stretched in two ways. First, in moving from causes to effects, it is used as above in the sense of 'to explain', rather than 'to prove' (or 'to demonstrate'!). Second, it is used in the sen�e of hypothetical reasoning from effect to cause, in which a postulated cause is 'proven' by the fact that it can serve as a unified explanation for a (wide) range of effects. To this point, we have followed two shifts by Descartes away from scholastic Aristotelianism. The first pertains to Descartes' altered con­ ception of nature. For, although he retained the Aristotelian terminology

1 60

GARY HATFIELD

of substance and essence, he broke the rules of Aristotelian ontology by attributing a passive essence, pure mathematical extension, to natural, corporeal things. He thereby maintained that geometrical descriptions fully capture the properties of all corporeal things, as opposed to the Aristotelian doctrine that geometrical properties are accidental. This may be called a substantive shift away from Aristotelianism, for it changed one's conception of what constitutes nature : conglomerations of whirling particles as opposed to bodies organized by principles of growth and change. The second shift pertains to Descartes' conception of the method of investigating nature. Here, two movements away from Aristotelian doctrine occurred, in different directions , as it were. The first was the movement away from the doctrine that metaphysical knowledge is based upon abstraction to the notion that the essence of matter (as indeed of mind) can be known independently of abstraction from the senses, by means of direct intuition. The second was the movement from the idea that natural things are investigated by abstracting their natures from sensory particulars to the idea that natural things are investigated by positing (often sub-visible) mechanisms to explain a range of known phenomena (or effects). These together might be termed a methodological shift, for they change one's conception of how to go about investigating nature: via pure contemplation on the one hand; and via conjectural positing of hidden mechanisms on the other. These substantive and methodological shifts complete Descartes' revolution of the sciences, by altering the relationships among physics, metaphysics, and mathematics. Descartes' first revolutionary step (which may now have a regressive ring) was to claim that metaphysics may determine the first principles of physics. A second step pronounced that the properties of corporeal things are to be accounted for using mathe­ matical properties alone. But a third step decreed that although the properties of bodies are geometrical, and therefore that particular geometrical configurations may be understood with geometrical conse­ quence, one cannot expect geometrical certainty about which constel­ lations of properties actually do occur in nature. The intelligibility of specific kinds of natural things - such as the magnet or the action of light depended upon what they had in common, their constitution as various modes of extension. Apprehension of what is peculiar to magnets or light depended upon attributing to the magnets or light a constellation of geometrical properties such that the known phenomena could be ex-

FIRST P H I L O S OPHY AND NATURAL PHILOSOPHY

161

plained. 1t might be no easy job to make a satisfactory conj ecture (and just . as hard to decide what is 'satisfactory', although Descartes was opti­ mistic), and one could always go wrong (as further experience might show). These two shifts fit hand in glove. For in conceiving extension as the essence of corporeal substance, Descartes attributed to natural bodies a property that had been the age-long model of the object of rational compre­ hension; whence its apparent suitability for an a priorist conception of how it is known. Moreover, in replacing the many forms of the Aristotelian catalogue of nature with many mechanisms, he specified as that-which-is­ to-be-known in specific bodies something other than an essence, form, or nature. In shifting away from the scholastics' conception of demonstrative science, Descartes may have been guided by the consideration that their ideal of scientific knowledge requires apprehending a 'form', or 'essence', which is an eternal principle. The ideal of demonstrative physical science sets up a standard of certainty that Descartes himself demanded in the domain of metaphysics, but which he doubted could be attained with respect to the particular constitution of nature. Hence, the shift from 'essences' to 'postulated mechanisms'. Essences serve in demonstrations. Mechanisms serve in hypothetical explanations, and in Descartes' 'demon­ strations'. In consequence, the notorious 'Cartesian quest for certainty' does not extend to the particular mechanisms posited in Descartes' mechanistic physics.

Johns Hopkins University NOTES * An earlier version of this paper was presented before a joint session at the meetings of the History of Science Society and Philosophy of Science Association in Philadelphia, October, 1982. The current version has been revised as the result of further work, still in progress. I wish to thank Owen Hannaway for helpful discussion of the earlier version of this paper, and Wilda Anderson for detailed comments and criticism of recent drafts. 1 A long line of articles and books have provided forceful arguments for and characterizations of Descartes' explicit avowal of the need for experience and experiment in natural philosophy. Gerd Buchdahl, in his Metaphysics and the Philosophy of Science (Oxford: Blackwell, 1969), Chapter 3, provides a penetrating philosophical discussion of these issues. Desmond Clarke's Descartes' Philosophy of Science (University Park: The Pennsylvania State University Press, 1982) provides the most extensive discussion to date, and includes references to several of the earlier studies (on p. 16, n. 1 5 ). 2 I have chosen Eustace of Saint Paul's Summa Philosophiae (Coloniae, 1620) and the University of Coimbra's Commentariorum . . . in octo libris Physicorum Aristotelis Stagiritae

162

GARY HATFIELD

(Coloniae, 1 599) as representative of scholasticism in Descartes' day; Descartes described the first of these as the best book of its kind "ever made", and remembered the second from his school days (Oeuvres de Descartes, edited by C. Adam and P. Tannery [hereafter AT), new edition, Paris, 1964-1974, Volume III, pp. 232, 185; the passages cited are among those in A. Kenny's [hereafter K) translation of Descartes' Philosophical Letters, Oxford: Clarendon, 1970, pp. 82, 78). In characterizing a generalized high scholastic position, I will often use quotations from Thomas Aquinas, who was frequently cited (often with favor) by Coimbra, and who also was cited by Eustace (who, however, maintained in the preface to his work that his chief aim was to interpret Aristotle's philosophy, as opposed to Thomas' or Occam's, and who was not averse to disagreeing with Thomas). In addition, I will sometimes give references to passages in Aristotle's works that could be read as supporting the scholastic position. Of course, I do not intend to suggest that the scholastic reading is philosophically or historically the best interpretation of Aristotle. 3 Thomas Aquinas, Opera Omnia (Parma, 1 865; reprint: New York, 1949), p. 227a, translated by Richard J. Blackwell, Richard J. Spath, and W. Edmund Thirlkel as Commentary on Aristotle's "Physics" (New Haven: Yale University Press, 1963), p. 4. Eustace characterized the subject matter of physics as "Corpus naturale ut naturale", rather than "ens mobile" (Pt. 3, Physica, p. 4); on the definition of nature he quotes Aristotle from the Physics, 11. 1 : "Princi­ pium & causa ut id moveatur & quiescat, in quo primo, per se, & non ex accidenti inest" (ibid. , p. 26). Coimbra accorded with Aquinas on both counts, cols. 33 and 249-252. 4 A helpful source on the division of the sciences and the doctrine of abstraction is The Division and Methods of the Sciences, A. Maurer's translation of Questions V and VI of Aquinas' commentary on the De Trinitate of Boethius. The division of the sciences is made with reference to the doctrine of abstraction by Eustace, Pt. I , Dialecticae, pp. 1 55-157; Pt. 3, Physica, pp. 1-4; Pt. 4, Metaphysica, pp. 1-3; and by Coimbra, col. 6. 5 Edith D. Sylla, in The A Posteriori Foundations of Natural Science', Synthese 40 ( 1 979): 147-187, discusses a variety of medieval commentaries that address, among others, the question of whether metaphysics can prove the first principles of physics. 6 Aquinas' commentary on Book II, Chapter 19 of the Posterior Analytics occurs in his Opera, Volume 18, pp. 222-225. Eustace touches upon the process of abstraction in his discussion of universals, Pt. I , Dialectica, pp. 24-3 \ . 7 Aquinas, Physics, p. 69. 8 AT VII, 25; the translation is from V. R. Miller and R. P. Miller, Principles of Philosophy (Dordrecht: Reidel, 1983), which I generally will follow. Descartes' language in making a similar point in Le Monde is of interest since he specifically contrasts his proposed 'nature' of matter with scholastic active principles as well as with the scholastic notion of 'prime matter' (which was typically defined as 'pure potentiality'): it is worth quoting several sentences, which occur as Descartes is introducing matter into his 'imaginary' world, con­ structed at the center of an 'imaginary space': "Now, since we are taking the liberty of imagining this matter to our fancy, let us attribute to it, if you will, a nature in which there is absolutely nothing that anyone cannot know as perfectly as possible. To that end, let us expressly assume that it does not have the form of earth, nor of fire, nor of air, nor any more particular form (such as of wood, or stone, or metal); nor does it have the qualities of being hot or cold, dry or moist, light or heavy, or of having some taste, or smell, or sound, or color, or light, or suchlike, in the nature of which one could say that there is something that is not clearly known by everyone. Let us not think, on the other hand, that our matter is that prime matter of the Philosophers that has been so well stripped of all its forms and qualities that

FIRST PHILOSOPHY AND NATURAL PHILOSOPHY

1 63

nothing more remains that can be clearly understood" (AT X, 33; translated by M. Mahoney, Le Monde, New York: Abaris, 1 979, pp. 5 1-53). 9 On the project of explaining all natural phenomena in terms of only the modes of extension, see Principles, II, 64; III, 46-47; IV, 199-200. On the exclusion of sensory qualities from among the real properties of matter, see the passage quoted in n. 8, as well as Principles, I , 66-74; on the treatment o f these qualities as 'dispositions' o f the geometrical properties, see ibid. , IV, 196-199. On the laws of motion, see ibid. , II, 37-42; Le Monde, Chapter 7. 1 0 The 'hypothesis' of a chaotic beginning for the universe is introduced in the Principles, III, 45-46, and the text continues under its aegis until Pt. IV, 1 88, where Descartes cuts off his account ofthings on the earth with an apology for not including projected fifth and sixth parts, on animals and plants, and on man. The hypothetical chaos is introduced in Chapter 6 of Le Monde. II I have discussed Descartes' kinematic conception of motion, and his role for God in explaining the 'force' of moving bodies, in 'Force (God) in Descartes' Physics', Studies in History and Philosophy of Science 10 ( 1979): 1 13-140. Clarke, in Chapter 4 of Descartes' Philosophy of Science, presents an alternative point of view. 1 2 AT III, 649 and K 1 35. The above claim that the geometrical properties of pure extension are known clearly and distinctly whereas active principles are intellectually obscure consti­ tutes, I believe, the metaphysical foundations for his physics that Descartes claimed to Mersenne to have provided in his Meditations (AT III, 233, 297-298, and K 82, 94). More­ over, it is the type of justification provided for restricting his account of nature in Le Monde to modes of extension, thereby excluding real qualities and substantial forms. There he invokes repeatedly the idea that he is introducing into his world "nothing that anyone cannot know as perfectly as possible" (Chapter 6). Perhaps this conception, that extension can be clearly understood while active principles and real qualities cannot, constituted the meta­ physical insight into the foundations of physics that Descartes wrote to Mersenne he had achieved in 1629, when at work on what is thought to have been an early version of the Meditations ( 1 5 April 1 630, AT 1, 144, and K 1 1 ). n 1 have discussed Descartes' claims for clear and distinct perception by the intellect alone in ascertaining the framework principles of physics in 'The Senses and the Fleshless Eye: The Meditations as Cognitive Exercises', in Essays on Descartes' Meditations, edited by Amelie O. Rorty (Berkeley: University of California Press, forthcoming). 14 0p. cit. , n. 12. 1 5 Discourse, Pt. VI, AT VI, 76-77; Meteors, 1 st Discourse; letters to Vatier, 22 February 1638, and to Morin, 1 3 July 1638; Principles, IV, 205. 1 6 AT 1, 563, translation altered from K 48. 1 7 AT II, 199, and K 58. 18 AT II, 299, and K 59. 1 9 Descartes often uses 'a prion< in the sense of 'known through its causes', but since he believed that such knowledge might be derived from pure intellectual contemplation of God, the soul, or the . essence of matter, this a priori knowledge might nonetheless be known independently of the senses. Gerd Buchdahl, in Metaphysics, Chapter 3, brings out the philo­ sophical interest of the 'fluidity' of Descartes' use of such terms as 'a priod. 20 Clarke, Descartes' Philosophy of Science, p. 56, emphasizes the hypothetical character of Descartes' metaphysical remarks. 21 To Vatier, op. cit., n. 1 6. One cannot help but remark Descartes' swipe at scholastic 'disputations'.

164

GARY HATFIELD

22 Meteors, AT VI, 239; to Regius, AT III, 492, and K 127. Also, in Chapters 6 and 7 of Le Monde, Descartes leaves out substantial forms and real qualities for being unintelligible and of no use in framing explanations. 23 Principles, II, 64; IV, 199. 24 Ibid. , IV, 204; Discourse, AT VI, 64-65. 25 For two such discussions with additional references, see B uchdahl, Metaphysics, Chapter 3, esp. pp. 1 1 8-126, and Clarke, Descartes' Philosophy of Science, Chapter 6. 26 Principles, IV, 205-206; Descartes of course consistently maintained that 'the more general' aspects of his natural philosophy were more than morally certain. Clarke, op. cit. , Chapter 8, has emphasized the weakening of the demand for certainty on Descartes' part; cf. also Buchdahl, op. cit.. 27 The doctrine that physics is a demonstrative science - grasping eternal principles and natures of mutable things - was commonplace; see Eustace, Pt. 3, Physica, pp. 2-3, and Coimbra, col. 16-26. Clarke, op. cit. , pp. 65-66, discusses Descartes' weakening of the term 'demonstration' . 28 AT VI, 76; translation slightly modified from E. Haldane and G. Ross, The Philosophical Works of Descartes (Cambridge: Cambridge University Press, 1 93 1 ), Volume I, p. 1 29. 29 AT II, 197-198, and K 57 ; cf. to Mersenne, AT II, 1 4 1 , and K 55. "The Philosophers" is a reference to scholastic philosophy.

DESMOND M. CLARKE

CARTE S IA N S CI EN C E I N FRANCE, 1 660 - 1 700

In the years between the death of Descartes and the gradual introduction of Newtonian science into France, there was a distinctive French school of Cartesianism which propagated the central ideas of Descartes' philo­ sophy and science, including · his concept of scientific knowledge. Male­ branche seems to have emerged from this period as the most interesting figure for philosophers today, possibly because of his obvious influence on Hume. 1 However, he was surely not so dominant in French science in the second part of the seventeenth century; both Jacques Rohault ( 1 620-1 672) and Pierre-Sylvain Regis ( 1 632- 1 707) had a comparable influence on French science, and either Rohault or Antoine Ie Grand ( 1 620- 1 699) were much better known in England at that time as exponents of Cartesian science. Given the number of different philosopher-scientists who deserve to be included among the influential Cartesians of this period, it seems preferable not to concentrate attention on one, such as Malebranche, and to attempt instead to characterise some features which were common to the French Cartesians as a group. There are three features of the context in which this school developed which I shall mention at the outset: ( 1 ) The most obvious one - the proliferation of scientific societies in France, for example those of Montmor and Thevenot, and their replace­ ment, by Colbert, by the Royal Academy of Sciences ( 1 666).2 Huygens, Mariotte, Claude Perrault, Leibniz and many other scientists were invited to be members of the academy. None of the Cartesians was invited until, after the reorganisation of 1 699, both Malebranche and Regis were made members in a subsidiary category. Thus, during the period under dis­ cussion, Cartesian science was not endorsed by the scientific establish­ ment in France. (2) The official philosophical teaching of the universities was a scho­ lastic melange usually attributed to Aristotle. Any attempt to introduce alternative philosophies was stubbornly resisted; the preferred strategy of the Sorbonne was to rely on a royal decree or on Church-enforced censor­ ship. Cartesianism was one of the primary objects of this kind of censor­ ship.3 1 65 A .J. Holland (ed.), Philosophy, Its History and Historiography, 165-178. &:'\

1 ll0 .c L • . 1"'\

n .� � _ J n

J " .1.,_ _ _ " • . . __ -_. . .

1 66

D E S MOND M . CLARKE

(3) It is possibly anachronistic and misleading to distinguish this as a separate item from No. 2; however, theological disputes played a very important part in explaining the disfavour in which Cartesians were generally held. The most obvious example of theological objections centred on Desgabets' attempt to explain the Eucharist in Cartesian terms ;4 how­ ever, the reasons for suspecting Cartesian thought were even more funda­ mental. Opponents argued that if the soul of beasts were reducible, it was only a matter of time before the mind of man would be equally reduced. More generally, there was enough overlap between Cartesians and the main theological enemy of the Papacy and Louis XIV - namely Jansenism - that the Cartesians were suspect both for their theological views and even as a source of political instability. Given this very synoptic sketch of the context in which the Cartesian school struggled for acceptance, the rest ofthe paper is concerned with two points : (A) the influence of theological and metaphysical assumptions on Cartesian science, and on the Cartesian concept of science ; (B) the role of experimental evidence in Cartesian science during this period. A. THEOLOGY & M ETA P H Y S I C S

As everyone knows, Cartesians claimed to explain all physical phenomena as complex effects of collisions between small particles of matter in motion. The basic unit of scientific explanation, therefore, should provide a successful, mathematical analysis of simple impact situations. The Car­ tesians, in general, repeated or slightly modified Descartes' unsuccessful impact rules. By contrast, Huygens and Mariotte developed successful alternative accounts of collisions - and it is clear that Malebranche and Regis, at least, knew of these alternatives. The question arises : why did the Cartesians hold fast to what looked like a blind alley when alternatives were available? I want to argue that part of the reason for their apparent intransigence was their understanding of the role of theology and metaphysics in science. One might expect a clear separation of disciplines as a consequence of Cartesian dualism - that physics would explain material things and that metaphysics or theology would take care of the non-material. There is even a suggestion of this in the title of Cordemoy's Discernment ( 1 666): The

CARTESIAN SCIENCE IN FRANCE, 1 660-1 700

1 67

Distinction of the Body and Soul. in Six Discourses. to serve as a clarification ofphysics. 5 The appeal of such a clear distinction is overshadowed by the fusion of theology, metaphysics and physics which one finds in Cartesian texts. One of the worst offenders here is Malebranche; he begins Chapter 1 of the Recherche with a discussion of the results of Original Sin on the human mind. One might apply a principle of charity at this point and describe the Cartesians' penchant for conflating disciplines as an 'integra­ tive' approach. This integrative approach had definite repercussions on their physics ; specifically, it helped determine their solution to the problem of collisions between so-called 'hard' bodies. The Cartesians denied the efficacy of secondary causes and, by implica­ tion, of the reality of force as something which is distinct from the quantity of motion of a moved body. Their defence of this view was motivated, it seems, at two different levels. The first one was a theological view of the universality and total efficacy of God's actions. This seems like a simple transposition of Jansenist theories of grace onto philosophical problems about causality. In theo­ logical terms, God's grace is both necessary and sufficient for salvation; human effort cannot be understood as affecting God's completely free decision to grant salvation to those whom he elects. The opposite is the case ; what looks like meritorious activity in human terms is an effect, rather than a cause, of God's freely conferred grace. There is an obvious sense, then, in which human salvation is completely one-sided - in God's favour; if this theological account of the sufficiency of divine actions and the consequent redundancy of human effort is transposed onto questions of causality, it is not surprising to find the Cartesians arguing in favour of the inefficacy of all secondary causes. When challenged to provide 'non-theological' reasons for occasionalism, the Cartesians seem to have relied on two assumptions: (i) the principle of simplicity, and (ii) , Descartes' theory of substances and modes. The com­ bined result of these assumptions , together with the theory of God's causality, was to effectively stifle any conceptual change in the Cartesians' . approach to collision problems. The language of occasional causes is first introduced by La Forge, in 1 664, in describing the relation between thoughts and brain-states.6 It is further developed by Cordemoy (in an argument which is discussed below), and finally defended on simplicity grounds by Malebranche. Malebranche repeats earlier arguments to the effect that the constant conj unction of types of brain-states and ideas can only be explained by

)68

DESMOND M . CLARKE

reference to God. However, once God comes into the picture, sensory stimuli no longer have the power, of themselves, to determine what sen­ sation (or idea) we will have. By generalizing this kind of consideration, he concluded that no created cause is efficacious of itself "There is no relation of causality between mind and body . . . there is no such relation between body and body, nor between one mind and another. No created being can, in short, act on any other by an efficacy which it has of itself"? "Everything depends on God because all causes are able to act only through the efficacy of the divine power."g Since true causality is defined as autonomous causality, there is no difficulty in constructing the kinds of argument which Hume later exploited. If we perceive a constant conjunction between two types of event - e.g. the impact of one billiard ball with another and the subsequent motion of the ball which was struck - we tend to assume that the prior event is the true cause of the second one.9 In order to predicate a true causal relation of such pairs of events, we would have to perceive the force of the cause ; but Malebranche, no matter how hard he tries, cannot conceive of an autonomous power or efficacy in bodies. This is not an argument about whether or not the 'force' in question is perceptible ; Malebranche is willing to concede the existence of many things which cannot be perceived, including the efficacy of God's power. 1 O The argument hinges, rather, on the conceivability of such a force independently of God: "I do not even think it a rash j udgment to assert that those who maintain that creatures have force and power in themselves advance what they do not clearly con­ ceive"Y The same point is urged in the Meditations Chretiennes: "it is useless to open one's eyes to judge the efficacy of creatures"; 1 2 only ideas can decide such issues. Thus to decide if one creature has a true causal effect on another, we would have to decide if "there is a natural and necessary relation (rapport)" 1 3 between the assumed cause and its effect. The metaphysical theory of God's agency undermines the apparent relevance of empirical evidence for deciding the question. One is tempted to reply to Malebranche : even if God is the primary cause, why not accept Descartes' view that physical bodies are secondary causes with efficacious forces ultimately derived from God? Why should there be a choice between God and secondary causes ? The only reply one gets is: that secondary forces are redundant. If God acts in the simplest way possible, then he can arrange all the regularities we perceive in nature by relying on two things : (i) his causal power, which is necessary in any theory, and (ii) the laws which express the ways in

CARTESIAN SCIENCE IN FRANCE, 1 660 - 1 700

1 69

which his causal power operates. 1 4 He does not need, besides, to have real forces in secondary causes. The simplicity and pervasiveness of God's agency is otherwise expressed in terms of the passivity or inertness of matter; "properties of extension can consist only in relations of distance" .1 5 The second metaphysical consideration which influenced the Car­ tesians' account of collisions was a theory of substances and modes which was inherited from Descartes. The clearest version of this argument is found in Cordemoy (although it is also used by Malebranche); briefly, since motion is a mode of an individual substance, it cannot be transferred from one substance to another. The argument in Cordemoy runs as followS : 1 6

Axioms 1 . Whatever something can lose without ceasing to be what it is, it does not have of itself. 2. Every body can lose its motion . . . without ceasing to be a body. 3. One can only conceive of two kinds of substance, namely Spirit and Body. 4. To move . . . is an action. S. An action can only be continued by the agent which initiated it. It follows that no body has motion as an essential mode of its substance ; and therefore the 'first mover' cannot be a body. Since there are only two kinds of substance available, the 'first mover' must be a spirit; and by the Sth axiom, it must be the same spirit which continues the motion once it has initially caused it. On first sight ihis contradicts one of the insights which almost defines the new sciences of the seventeenth century, and which is expressed in Descartes' first law: that once a body is moved, we do not require any further (physical) explanation of why it continues to move. 1 7 Only changes in a body's condition of motion or rest require explanation. However, even in Descartes, God's creation and conservation of the universe are only rationally distinct ; creation and conservation are identical in reality. Therefore God's concurrence is constantly required to maintain a moving body in whatever condition it is in. Although no new physical explanation is required to maintain a body in motion, God's creative concourse is necessary. Cordemoy's analysis merges the physics/metaphysics distinc­ tion and argues that bodies cannot cause the motion of other bodies. The constant conjunctions we experience in impact situations mislead us into

170

DESMOND M . CLARKE

thinking that the motion of one body can truly cause another to move. 1 8 One i s tempted to reply: assuming that 'motion' ultimately is caused by God, and even that God's conservation is necessary · to preserve it in existence, why could one moving body not transfer some of its motion to a second body and thereby merit the title of an authentic secondary cause? Cordemoy's atomism is relevant at this stage, together with the Cartesian concept of modes. Cordemoy argued, in his 1 st Discourse, that matter is composed of small, indivisible substances called bodies, and that matter is an aggregation of such bodies. 1 9 The modes of any substance, just like the speed of a given motion, cannot be separated from the substance itself. They are merely the way in which a substance exists. Therefore the mode of one substance cannot be communicated to another. A substance can cease to have a mode, but the individuality of its modes prevents them from being communicated to other substances. 2o Descartes avoids this conclusion by thinking of the whole of matter as one substance; however, if each atom is a substance, then clearly transferring motion from one body to another is transferring the modes of an aggregate of substances to another cluster of them. If Cordemoy's argument works, the impotence of secondary causes is independent of dualism (except insofar as he relies on dualism in axiom 3 ). Not only can a body not causally affect the soul, but one body cannot even cause motion in another body. 21 This metaphysical theory need not affect our usual way of talking about causal relations ; we can attribute to ourselves the power of moving our hands, but only in the sense of an occasional causal power;22 and "one could say that this body is the physical or natural cause of the motion it communicates [to others] , because it acts according to natural laws".23 In fact, in most cases in which we ask for the cause of something, it is the natural cause we are looking for, and it is inappropriate to give the general cause, God, since his causality applies in every case. 24 The implications of these theological and metaphysical considerations are most evident in Malebranche. He worked on the problem of collisions from the first edition of the Recherche in 1 675, until the final (6th) edition of 1 7 1 2, constantly modifying his attempts to calculate the distribution of motion which results from hard-body collisions. However, the changes in formulae were consistent with a fundamental rejection of the concept of force as being in any way an independent explanatory entity; force is reducible to the measured quantity of motion of a moving body. Thus, the final chapter of the Recherche rejects the idea that a body at rest has any

CARTESIAN SCIENCE IN FRANCE, 1 660 - 1 700

171

force at all : I conceive only that bodies in motion have a motor force, and that those at rest have no force for their state of rest, because the relation of moving bodies to those around them is always changing; and therefore there has to be a continuous force producing these continuous changes, for in effect it is these changes which cause everything new that happens in nature. But there need be no force to make nothing happen. 25

This might look as if he is only denying a motive force to bodies at rest ; however, he means to reject any force at all in such bodies, as is clear from the discussion of impact rules in 1 7 1 2 : "Rest has no force to resist motion". 26 This explicitly rej ects Descartes' attempts to calculate the inertia of a body at rest, which Malebranche considered a mistake. Moving bodies alone have force, and this is calculated by the product of what Male­ branche calls the 'mass' (masse) and the speed. Collisions between bodies are the "natural or occasional causes of the distribution of motions" : the only problem remaining, therefore, is to calculate how quantities of motion are redistributed in a way which coincides with our experience of collisions (so as not to end up with false rules of collision as Descartes did!). This very brief consideration of the Cartesians' attitude to the dynamics of 'hard-body' collisions - without any of the relevant calculations - is evidently not enough to establish the general thesis that Cartesians were philosophers first and only physical scientists in a secondary sense. There are two other considerations which support that contention : (a) The structure of Cartesian syntheses : many Cartesian scientists, especially Le Grand and Regis, followed the scholastic pattern of Logic, Metaphysics, Physics and Morals. A good Cartesian got his metaphysics sorted out first and then devised a physics within the constraints of his metaphysics. (b) The Cartesian concept of what could count as an explanation in science was strongly influenced by metaphysical considerations of what kinds of things exist. The consistent reliance on the primary/secondary quality distinction is a symptom of their ontological squeamishness ; Car­ tesians favoured explanations as mechanical models of explananda,27 because the only kinds of things which exist are small parts of matter in motion. Therefore such theoretical entities as 'attractive forces' are pre­ cluded, not only because they purport to be true causes, but also because they don't look 'physical' enough for Cartesians.

1 72

DESMOND M. CLARKE

I return to this point about the dominant role of metaphysics in the conclusion ; I need to say something at this stage about the second issue raised at the beginning, namely, the role of empirical evidence in Cartesian science.

B. R E A S O N AND E M P I RICAL EVIDENCE

Two quotations from Malebranche : "The senses always deceive you"; and: "Be advised, then, once and for all, that only Reason should stand in judgment on all human opinions not related to faith",2 s and this includes science. This is surely the standard Cartesian dichotomy between decep­ tive senses and reliable reason. However, M alebranche also wrote; "we are not deceived by our senses but by our will, through its precipitous j udgments". 29 He even goes so far as to say that reason sometimes misleads us if we fail to acknowledge empirical evidence : . . . reason teaches us that we should not feel pain when our arm is cut off. The soul is superior to the body. and, according to the light of reason, its happiness or unhappiness should not depend on the body. But experience sufficiently shows us that things are not as reason says they should be, and it is ridiculous to philosophise against experience. 3o

To reconcile the apparently contradictory texts in Malebranche the follow­ ing interpretative strategy should be adopted. All Cartesian references to the deceptiveness of the senses should be read, not as a chapter in the history of scepticism, but rather as a corollary of the distinction between primary and secondary qualities. In other words, we cannot legitimately assume that natural phenomena are, objectively, as they seem to be: "there is nothing in objects of our senses similar to the sensations we have of them".31 We cannot transfer our perceptions onto the assumed objective causes of our perceptions, as if there were something called 'heat' in those things which we experience as hot, and so on. Yet this is what we tend to do in our unscientific or unsophisticated judgments which are charac­ teristic both of scholastic philosophy and the beliefs of childhood. Reason intervenes, therefore, not to doubt the veracity of our sensations,32 but to check our inferences from sensory experience to obj ective claims about natural phenomena. M alebranche follows Descartes' usage in using the term 'sensation' to refer to the spontaneous or uncritical judgments so-called 'natural' judgments - which we tend to make on the occasion of

CARTESIAN SCIENCE IN FRANCE, 1 660 - 1 700

173

having a sensation.33 These judgments are invariably mistaken, and in that sense our senses deceive us. Once we have learned our lesson, then we need a method for relying on sensory evidence without being deceived. And there is no way in which we can do physics without empirical evidence being decisive: Reason demonstrates these things: but if reason can be withstood, experience cannot. For since we have discovered through the use of telescopes or large glasses spots in the sun as large as the entire earth, . . . we can no longer deny that it is subject to change a great deal more than the earth we inhabit. What we have proved by abstract arguments must be demonstrated through sensible experiments to see if our ideas are in agreement with the sensations we receive from objects, for it often happens that such arguments deceive us . . . (italics added). There are still some persons . . . so opinionated that they do not want to see things that they could no longer contradict if they would only open their eyes. 34

If Malebranche is so keen on empirical evidence, it might seem as if he should give it a central role in physics. He does. His criticism of experi­ mental scientists match those of Descartes. Briefly: scientists should choose experiments on the basis of theory and not by chance ; they should concentrate on simpler, easier experiments rather than on complex ones which involve a great variety of variables ; they should notice the many factors which affect the outcome of an experiment; they should refrain from drawing too many conclusions from a single experiment; and finally, they should integrate the specific experiments they do into a more general physical and metaphysical theory of nature.35 This all sounds so reasonable that one might wonder why Cartesianism was ever (mistakenly?) viewed as being rationalist at all. What more could one hope for in terms of the integration of empirical evidence into theory construction? Two different kinds of comment are needed at this point: one about the Cartesian use of empirical evidence in science; the other, about the Cartesian (philosophical) theory concerning the extent to which experience (in some sense to be defined) can supply the kind of empirical control over hypotheses which is expected from it in science. On the first point - the use of empirical or experimental evidence in science - one finds both elegant experimental technique and its complete opposite. Jacques Rohault is an able exponent of experimental technique at its best, although one tends to suspect that the experiments were fashioned with a definite bias in favour of a Cartesian conclusion. By contrast, many Cartesians gave themselves a lot of latitude in framing

1 74

DESMON D M. CLARKE

hypotheses and in weaving a likely account of some phenomenon which is almost completely hypothetical. Not surprisingly, the final sentences of such hypothetical accounts agree with experience, or at least' with what they believed to be empirically known. The next step is usually: "and this is all therefore confirmed by experience" . What it should read is perhaps: "the hypotheses do not contradict our ordinary experience of physical objects". It is the ad-hoc character of such explanations which strikes the modern reader as a priori. What one does not find, however, among the Cartesians in this period is a robust, rationalist rejection of empirical evidence in favour of reason, roughly along the lines which some of Descartes' or Malebranche's statements might lead one to expect. Experience, or empirical evidence, is an essential component in physical science for the Cartesians. In fact, it was not so much that they doubted experience; they just were not s uch good scientists. On the question of explaining how experiential evidence can be inte­ grated into our knowledge-claims, there was general agreement on some­ thing like Descartes' distinction of three levels or degrees of sensation. Since Malebranche makes this distinction more carefully than Descartes, his analysis is more reliable. Malebranche indicates that the term 'sen­ sation' can apply to any of the following: 1 . The action of some external object on the sensory organs of one's body. 2. The effect of such on the sensory organs, which is passively received by the relevant sensory organs. 3. The soul's 'perception' of number 2; or the idea, in the mind, which corresponds to 2 (e.g. the perception of heat). 4. Those 'natural judgments' which we tend to make, in error, as a result of number 3. 3 6 Apart from the occasionalist nuances of the link between 2 and 3, there is nothing very new here in Cartesian philosophy. Since ideas arise in the mind in this empirical fashion, one might question the extent to which Cartesians could or should rely on them. How about innate ideas, and mathematics, and Descartes' ideal of a mathe­ matical science of nature? Again, general comments rather than detailed arguments are initially in order. Most Cartesians agreed in rejecting innate ideas as either necessary or helpful. La Forge gives an account which is close to Descartes' response to Regius, which he quotes at length. 37 For him, ideas are only innate in the sense that the mind can generate them without any empirical stimula-

CARTESIAN SCIENCE IN FRANCE, 1 660-1 700

1 75

tion but there is no need to assume that there are any actual ideas in the soul at birth. 3 8 Likewise, Malebranche rejects innate ideas, because if God is needed to communicate ideas to us, he can accomplish the task more simply than by providing us with an almost infinite supply of innate, actual ideas. 39 And Regis is almost hostile to the suggestion of innate ideas: "when I say that the ideas of God, of the soul and of the body are innate, I do not mean that they are independent of the body; I only mean to say that these ideas are always in the soul explicitly or implicitly".4o The lack of commitment to innate ideas is not matched by any corresponding endorsement of the scholastic axiom: "nihil est in intel/ectu quod prius non fuerit in sensu". La Forge rejects this, again for reasons very similar to Descartes', because he could not understand how the motions of particles (to which sensory stimulation is reducible) could adequately explain the fact that we have , for example, the idea of GOd.41 Malebranche notoriously

defends the pure intellect against any dependence on sensory stimulation; but this opinion is rejected in turn by Regis, who returns to a more moderate version of Cartesian epistemology; "the hypothesis of modern philosophers, who admit a pure understanding, that is an understanding which acts independently of the body, can have no solid foundation . . . It is therefore without any basis that modern philosophers assert that there is anything in the understanding which has not passed through the senses".42 However the Cartesians might argue over pure understanding, they are agreed at least in this much ; that empirical knowledge involves the coincidence of sensory stimulation and mental sensation, and that this combination is what is relevant to physical science. CON C L U S ION

The history of early modern science is the history of the interrelationship of what we would now classify as theology, philosophy, and science, together with an indeterminate number of social, cultural and political factors which significantly affected its development at various crucial stages. Any attempt to reconstruct a particular stage of the history of science is doomed to failure if it ignores too many of the details of such a complex relationship. The history of philosophies of science is similarly complex; to understand a particular model of science is to understand the many relevant features of the historical period in which it flourished. This synoptic review of Cartesian science in France, in the second half of the seventeenth century, should support this general thesis and also

1 76

DESMOND M. CLARKE

explain why my account is seriously incomplete. Even so, °a few points emerge which deserve summarising: ( 1 ) The extent to which theological views were a dominant part of the intellectual environment in France, and the extent to which they influenced both science and the philosophy of science. This influence was sometimes external, sometimes internal. Those who were theologically suspect and hence politically embarrassing were simply not invited to belong to the Academy; in fact the dissemination of their views was censored. On the other hand, theological views so influenced both Malebranche and his opponents (such as Arnauld) that one is not surprised to see Malebranche situate his theory of knowledge within a theology of fallen man. (2) A second familiar feature of this historical period is the close con­ nection between metaphysics and physics. One might contrast the Car­ tesians with their contemporary, Newton, on this point, and argue that Newton's interminable difficulties with the ontological status of forces were ultimately fruitful for physics.43 By contrast, the Cartesians in general tried to get their metaphysics straight first, and then construct a physics within the constraints of prior metaphysical clarity. Metaphysics therefore dictated what kinds of physical theories were even worth considering. That might seem like a sound policy if we had some way of establishing a non-empty, a priori metaphysics. The history of science seems to support the value of a different strategy: to do science with minimal a priori constraints, and to sort out ontological difficulties later. I have indicated two points on which Cartesian metaphysics cramped physics: (a) where the concept of substance and its modes re-entered the domain of physics to dictate what kinds of things could or could not happen. A similar attempt to define causality a priori resulted in the exact opposite conclusion to that drawn by Hume about collisions between elastic bodies. The tendency towards positivist philosophy of science was latent in the very first (metaphysical) move; (b) the constraints on theo­ retical entities which resulted from the Cartesian concept of matter. There are no signs of a mathematical physics in this period of Cartesian science along the lines, for example, of Newton's Principia. So that it is a poor argument to claim that only those entities which were mathematically describable were allowed into physics. That may be true; however, the criterion was applied prejudicially in favour of entities which could be imagined, and these in turn were determined by our ordinary experience of

CARTES I A N

SCIENCE IN FRANCE, 1 660-1 700

1 77

everyday, macroscopic bodies. (3) Finally, the voluntarist tendencies of Descartes' philosophy were developed further by Malebranche's concentration on the arbitrariness of causal connections ; as a result, Cartesian science (at least in theory) became more explicitly empiricist than it had been previously. One cannot know how the physical world is without using one's senses ; but, unfortun­ ately, the senses tend to deceive. We are left with a hypothetical picture of the physical world which, minimally, does not contradict our experience.

University College, Cork

NOTES 1 See, for example, Charles J. McCracken, Malebranche and British Philosophy (Oxford: Clarendon Press, 1983); A.A. Luce, Berkeley and Malebranche (London: Oxford University Press, 1 934). 2 For the early history of the Academy, see Roger Hahn, The Anatomy ofa Scientific Institution; The Paris Academy of Sciences. 1 666-1803 (Berkeley: University of California Press, 1971); John M. Hirschfield, The Academie Royal des Sciences 1666-1683 (New York: Arno Press, 198 1 ); Harcourt Brown, Scientific Organizations in Seventeenth-Century France (1620-1680), 2nd edition (New York: Russell and Russell, 1967). 3 Cf. Brown, p. 214, and T. McClaughlin, 'Censorship and Defenders of the Cartesian Faith in Mid-Seventeenth Century France', Journal of the History of Ideas, 40 ( 1979): 563-5 8 1 ; F. Bouillier, Histoire de la philosophie cartesienne (Paris: Delagrave, 1 868), I , pp. 466-485. 4 See Jean-Robert Armogathe, Theologia Cartesiana (The Hague: Nijhoff, 1 977); R. Laymon, Transubstantiation: Test Case for Descartes's Theory of Space', in Problems of Cartesianism, edited by T. Lennon et al. (Montreal: McGill-Queen's University Press, 1982): 149-170. 5 Gerauld de Cordemoy, Oeuvres Philosophiques, edited by P. Clair and F. Girbal (Paris: Presses universitaires de France, 1 968). 6 L 'Homme de Rene Descartes . . . avec les remarques de Louis de la Forge (Paris, 1 664), p. 268; Traitte de {'Esprit de {'Homme et des ses Facultez et Fonctions. et de son union avec Ie corps (Paris, 1 666), in Oeuvres Philosophiques, edited by P. Clair (Paris: Presses universitaires de France, 1 974), p. 96. Cf. Rohault, A System of Natural Philosophy, translated by J. Clarke, with notes by S. Clarke, 2 Volumes (London, 1 723), Volume I, p. 248: "motions of body . . . are the occasions of particular perceptions in it". 7 Dialogues on Metaphysics, translated by W. Doney (New York: Abaris B ooks, 198 1 ), p. 89. 8 Ibid. , p. 257. 9 For this argument, see The Search After Truth. and Elucidations of the Search After Truth, translated by T. Lennon and Paul J. Olscamp (Columbus: Ohio State University Press, 1980), pp. 224-225, 658-660. 10 Cf. Dialogues, No. VII. II Ibid. , p. 658. 12 Oeuvres completes, 2 1 Volumes (Paris: Vrin, et Ie centre national de la recherche scienti-

1 78

DESMOND M. CLARKE

fique, 1 962-1969), Volume X, p. 60. 13Ibid., p. 6 1 . 1 4 Search After Truth, p . 663. 15 Dialogues, p. 147. 16 Cordemoy, Oeuvres Philosophiques, pp. 1 3 5 fT. 1 7 Descartes, Principles of Philosophy, translated by V. R. Miller and R.P. Miller (Dordrecht: Reidel, 1983), p. 59. 18 Cordemoy, Oeuvres Philosophiques, p. 137. 1 9 Ibid. , pp. 95-105. 20 Cf. Le Forge, Traitte, pp. 250-25 1 : "motion is nothing but a mode which can no more pass from one subject to another than the other modes of matter". 21 He rejects causality by human minds in Oeuvres Philosophiques, pp. 140-142. 22 Meditations Chretiennes, in Oeuvres completes, Volume X, p. 92. 23 Ibid., p. 54. 24 Dialogues on Metaphysics, p. 87. 25 Search After Truth, p. 5 1 7. 26 Malebranche, Oeuvres, Volume XVII, Part I, p. 59. 27 Cf. Claude Gadroys, Le systeme du Monde (Paris: Desprez, 1 675), Preface, and pp. 139-140. 28 Meditations Chretiennes, Oeuvres, Volume X , p. 103, and Search After Truth, Preface, p. xxviii. 29 Search After Truth, p. 23. 30 Ibid., p. 342. 3 1 Dialogues on Metaphysics, p. 7 1 . 32 Dialogues on Metaphysics, p . 1 17, and Meditations Chretiennes, i n Oeuvres, Volume X , p . 5 1 : "Ce sont l a des faits, & mes sens a regard des faits sont des temoins irreprochables". 33 Search After Truth, pp. 34, 46-47, 52-53; Dialogues on Metaphysics, pp. 1 17, 283. 34 Dialogues on Metaphysics, p. 257; Search After Truth, pp. 5 17, 9 1 . 35 Search After Truth, pp. 1 59-160. 36 Ibid. , pp. 52-53. Cf. Descartes, Oeuvres (Paris : Vrin, 1964-1974), Volume VII, pp. 436-437, and La Forge, Traitte, p. 249. 37 Trait/e, pp. 1 7 1- 1 73. 38 Ibid., pp. 1 8 1 , 293. 39 Search After Truth, pp. 226-227. 40 Regis, L 'Usage de la Raison et de la Foy, ou ['accord de fa foy et de fa raison (Paris: Jean Cusson, 1 704), p. 27. 4 1 Trait/e, pp. 1 7 1- 1 73. 42 L 'Usage de la Raison, pp. 106-107. 43 See Ernan McMullin, Newton on Matter and Activity (Notre Dame: University of Notre Dame Press, 1978), for a discussion of Newton's ongoing concern with the ontological nature of force.

RICHARD FRANCKS

CARICATURES IN T H E H I STORY OF PHILO S OPHY : THE CA S E O F S P I N O Z A .

I may as well proclaim m y allegiance from the first: I think that Philosophy and its Past l is one of the most important books that have been written for a long time, and that it has a great deal to teach anyone who is interested in the history of philosophy, and therefore anyone who is interested in philosophy. And by that, I suppose, like most people who say such things, I mean that I learned a lot from it, and if ! can learn from it, then obviously so can anyone else. I shall make no attempt here to enlarge on what Ayers, in particular, has to say, and I shall certainly not expand on my relatively minor points of disagreement with the book. All I shall do is to offer a case-study, which I hope will simultaneously exemplify what I take to be the main point of Philosophy and its Past, and also borrow its authority to sell some of my own ideas about Spinoza. The case I shall be arguing is this: that Spinoza suffers more than most philosophers from the current tendency, and in fact the current licence, to trivialise and distort the philosophers of the past. As in most cases, this tendency takes two major forms : first, there is the naive assumption that we can attach to the words of the past the meanings and connotations of the words of the present simply on the basis of lexical similarity; and second, there is the simplistic and as it were Procrustean application to the past of the categories of the present - the presupposition that the philoso­ phers of the past will exemplify certain crudely-defined types, invented by later philosophers, in advance of any serious attempt to ascertain just what those earlier thinkers were trying to say. Spinoza, I shall suggest, is particularly badly served in these respects, in that he made a unique contribution to the philosophy of his time, and one which demands our best attention; but the absurd caricature of his theories which is current in Britain today means that his contribution is undervalued to such an extent that very few people even bother seriously to read him. Spinoza talks a great deal about the power and importance of reason in human affairs. The pure understanding, for Spinoza, cannot go wrong: its ideas are necessarily true, and in fact as certainly true as the ideas of God himself, whereas "knowledge from vague experience" is an example 1 79 A. J. Holland (ed.). Philosophy. Its History and Historiography. 1 79-194. l OR: " 1m n Rp;r!p/ Pun!ichil1u rrnn nnH1J

1 80

RICHARD FRANCKS

of "knowledge of the first kind", which is the only cause of falsity. 2 By way of reason, mankind can escape from the consequences of his finite nature, can transcend his emotions, avoid sin and error, and achieve perfect freedom in the intellectual love of God. 3 This principle is central to Spinoza's teaching - in fact it is virtually the whole of it - and it is reflected in the style of everything he writes. His tendency is always to present mathematical, and especially geometrical, examples wherever possible in his works; and as anyone who has ever tried to read it will know to their cost, the Ethics, his greatest work, is actually written in the axiomatic form, as a series of definitions, axioms, propositions, demonstrations, and so on. Even in his other, less formal writings , the same tendency is detectable : there is a sustained attempt to write clear, impersonal, unemotional, eminently reasonable prose, giving his writing a distinctive feeling of dogged inexorability which some readers have claimed to find delightful, but which I confess to finding merely colourless at best. The point, though, is this : that it can hardly be denied that if one is looking for a slogan to sum up the message of Spinoza's philosophy, it would be to the effect that human reason can in principle attain to a perfect knowledge of all that is real. As a result of this undeniable feature of Spinoza's work, various things have come to be said about him. It is held, for instance, that Spinoza believed that all knowledge could be arrived at by deduction from a priori-derived axioms ; that the only genuine knowledge is mathematical; that all knowledge derived from experience is worthless, and that empirical science, construed as the attempt to derive from observation substantial truths as to the nature of reality, is a chimerical pursuit. Now,just in order to set the record straight, and perhaps encourage you to read a little further, let me say that I believe that none of these things is true of Spinoza, if indeed they have ever been true of any modern philosopher, which I think unlikely. I would maintain in fact that Spinoza believed that no worthwhile knowledge of the world can be arrived at by pure deduction; that the truths of mathematics are no more certain than those of, for example, chemistry; that all human knowledge is grounded in experience, and that empirical science, construed in the same way, can in principle embrace the whole of reality. In this paper, however, I shall not directly argue for any of these points. Instead I shall ask one question: does the undeniable fact of Spinoza's insistence on, one might almost say obsession with, the power and scope of human reason provide any evidence at all for the conclusion that he

THE CAS E OF SPINOZA

181

undervalued the testimony o f experience and was therefore out o f step with the scientific spirit of the age in which he lived? I shall suggest that it does not, and that the reason it is held to do so is based on: first, a mis­ understanding of the concept of reason as Spinoza employs it - a mis­ understanding born of equating his idea with ours of the same name ; and second, a misunderstanding of the time in which Spinoza lived, the problems he was looking to answer and the audience for which he was writing - a misunderstanding which stems from the naive assumption that since the seventeenth century saw the birth of that view of science to which we ourselves subscribe, it follows that anyone who then wished to trumpet its virtues would go about it in roughly the same way as we would ourselves. To a present-day philosopher, reason is the opposite of experience. We make a fundamental distinction between two kinds of truths, the a priori and the a posteriori, and we hold that while the latter are known by experience, the former are known by pure reason. It is often held that this fact means that a priori truths must be analytically true, since reason can know only logical relations between or within concepts. A priori truths are known as truths of reason, and since their truth is independent of any experience, philosophers have maintained that they are empty of any experiential content, and even that they have no factual content at all. If anyone today, therefore, were to maintain that all the truths of the universe can be apprehended by reason, or that man's happiness, his freedom and his salvation are all to be attained by the use of reason, we should be quite justified, in the absence of any clear statement to the contrary, in locating him within this body of theory and concluding that he held that these objectives could be achieved by the philosopher in his armchair, by means of pure thought, unmixed with any facts of experience. But the same was not by any means true for Spinoza. In 1 677, when Spinoza died, all this complex infrastructure of reason, which we have inherited and developed and which plays such a large part in determining just what for us reason is, lay in the future, and the term was employed in very different contexts, and embedded in very different theories. The idea that there is some kind of distinction to be drawn between reason and experience is of course as old as philosophy itself. From the earliest times there had been held to be a division between reason and sense, the latter being construed as both the pleasures of the body and the representations of the senses. Certainly this separation was central to

1 82

RICHARD FRANCKS

orthodox mediaeval theories of man as a rational animal, sharing his senses with the animals, and distinguished from them by his rational soul. Reason, therefore, while perhaps inferior to revelation as a source of knowledge, is the highest and noblest faculty of the human mind, whereas bodily sense is material and corrupt, and knowledge which is derived from it is tainted by man's carnal nature. As for example Bonaventure said: "Sense perception . . . is called the lower light because sense perception begins with a material object and takes place by corporeal light".4 Nor was a division between a priori and a posteriori knowledge unknown in the seventeenth century. Leibniz famously separated truths of reason from truths of fact towards the end of the century, some years after Spinoza's death.5 In Spinoza's own works there is no such explicit division, although I suspect that his distinction between things which are necessary in respect to their essences and things which are necessary only in respect to their cause contains the seeds of much of Leibniz' work.6 The essential point, however, is this: that neither in Spinoza nor in Leibniz is there any suggestion that the role of reason is only the apprehension of a priori truths; and consequently we would not be justified, without very clear evidence to support us, in concluding of either of them that because they stress the importance of reason they are thereby stressing the import­ ance of pure a priori knowledge. Our modern concept of a relatively 'pure' reason, untouched by any experience, has developed gradually over the years since Spinoza's time. In particular, the most important stage of that development was the work of Kant, some 1 00 years after Spinoza's death. By making a division between what is usually called, in English, 'Reason', which "goes out beyond the sphere of a possible experience" in apprehending a priori truths, and 'Understanding', which decides questions in the acquisition of empirical knowledge,? Kant set the tone for this general tendency to restrict the scope of reason to that of pure reason, which is limited to what can be known a priori in advance of any experience, and to regard knowledge which is in any way derived from experience as necessarily lying outside its scope. If the power of reason for Spinoza does not imply a total separation from experience, how should it be interpreted? In Spinoza's day the concept of reason was of something 'impure' in the extreme, to such an extent that anyone who talked of reason was more likely to be referring to empirical science than to logic or mathematics. Thomas Sprat, for instance, in 1 677, in his official history of that great champion of 'experimental philosophy'

THE CASE OF SPINOZA

1 83

the Royal Society, describes its work as "the highest pitch of human reason", and he calls on all the people of Europe to unite their "reasons" against ignorance and false opinion.8 How does this come about? It is interesting in this context to compare the use of the term which is still current in non-philosophical circles today. The politician, for example, who argued that in considering the question of capital punishment we should put aside our emotional reactions and our sectional interests in order to let reason prevail, was not suggesting that we should somehow attempt to deduce government policy from intuitively certain axioms, or that we should try in any other way to settle the question a priori. What he meant, and indeed went on to say, was that we should attempt to solve the problem by the careful and systematic analysis of the facts of the matter, chief among which, of course, was the available empirical evidence. It is in the light of this kind of appeal to reason, it seems to me, that we should seek to interpret Spinoza. In the seventeenth century, the word 'reason' and its equivalents seem to have meant nothing more technical than the intellectual faculty, or the power of thought, as distinct from, for example, the will, or the emotions. As a source of knowledge, therefore, the fundamental division was not between reason and experience, but between reason and faith. This, after ail, was the topic which had exercised Christian philosophers since at least the days of Augustine, and which had been even more central to the Islamic and Jewish traditions with which Spinoza was perhaps better acquainted. And the Reformation and Counter-Reformation, by focussing debate on the nature of the individual's relationship to God, had served only to heighten the conflict. Luther, for example, had been led to say that " . . . reason mocks and affronts God in spiritual things and has in it more hideous harlotry than any harlot . . . Reason must be subject and obedient to faith". 9 Remarks of this kind - and they were not unknown on both sides of the religious divide - become ludicrous if they are interpreted as implying an attack on pure a priori knowledge. What did Luther have to fear from that quarter? Who were the defenders of the a priori against whom he inveighed so violently? The question that is at issue is not the nonsensical one of the divine against the a priori, but the very pressing problem of the gOodly against the human: the light of revealed truth, whether direct or through the mediation of the church, as against all other, purely human, sources of knowledge, here bracketed together under the name of 'reason'. 1 0 Now, I am not suggesting that when Spinoza stresses the power of reason we must understand him as taking a stand on the question whether

1 84

RICHARD FRANCKS

the individual's salvation is to be mediated by the church. What I am saying is that a glance at history will enable us to see that the debate on that issue forms part of the contemporary context into which his remarks must be fitted in a way in which the division between experience and pure reason does not. We are therefore entitled to take him as siding with reason against divine inspiration, but not, without more evidence than I have ever seen offered, with reason against empirical investigation. It is against just such remarks as this of Luther's, for example, that Spinoza complains that in his time "the light of reason is not only despised, but by many even execrated as a source of impiety; . . . human commentaries are accepted as divine records, and . . . credulity is extolled as faith". I I But there is a second, and, for us, more important contrast which was generally implied in seventeenth-century uses of 'reason', namely that between reason and imagination. From the earliest times, both in philo­ sophy and in less formal contexts, the deliverances of reason had been opposed to the impressions of the imagination, or fancy: i.e. it had been recognised that the mind is capable of envisaging far more than 'in strict reason' we are justified in believing. These ideas of the fancy might be purely 'imaginative' inventions, or perhaps dreams, or might equally be deceptive appearances which mature consideration would overthrow. Reason in this sense is the arbiter of good judgment, the discerning faculty which separates the true from the false. As Descartes said: " . . . the power of judging well and distinguishing truth from falsehood . . . is what we properly mean by good sense or reason".1 2 In the context of seventeenth-century science this usage had a particular significance. What we would call the scientific analysis of phenomena was then still quite novel, and its advocates believed that they had discovered a new method of investigation and a powerful new tool for revealing the hidden structure of the world. The conclusions to which that science led them were often very surprising, and in direct opposition to the way things appeared to the casual observer: the solid earth, for example, had been found to be in motion, and the wandering sun to be at rest. What could have been more natural, therefore, but that the supporters of the new science should describe this contrast in terms of the power of reason to correct the fanciful impressions of the imagination? And given that the phenomena with which seventeenth-century science was primarily con­ cerned were the familiar objects of day-to-day experience, foremost among these impressions of the imagination which a reasoned understanding would reveal as illusory were judgments based on simple sense-reports. In

THE C A S E OF SPINOZA

185

precisely this way both Descartes and Spinoza reject as 'imagination' the appearance of the sun, which at first sight seems to be quite close, but which to the eye of reason, i.e. to the man of science, is known to be very far away . 1 3ln the same way it comes about in general that anyone in the seventeenth century who advocated the power of the new empirical science to unlock the secrets of the universe saw it as his task not to praise the power of sense-experience to yield genuine knowledge about the world, but on the contrary to remind his audience of the weakness and unreliability of j udgments based on day-to-day sense-impressions, in order to prepare the ground for his insistence that reality is very often not as it seems. Perhaps the best example of this usage is provided by the astronomical work of Galileo. In Galileo's writings, throughout the controversy over the Copernican system, it is not Galileo himself, but his Aristotelian oppo­ nents who defend the power of sense-experience. In the Dialogue ofthe Two Chief World Systems, for example, he represents those opponents as object­ ing that "in Copernicus' view one must deny one's senses", while "philoso­ phers of every school [agree that1 the senses and experience should be our guide in philosophising". But in the face of arguments of this kind Galileo's own spokesman is quite unmoved, and he praises Copernicus for his sublime intellect on the grounds that "with reason as his guide, he resolutely continued to affirm what sensible experience seemed to contradict". 1 4 To make all this clear, I want now to concentrate on one philosopher whose work I think best illustrates the role of reason in seventeenth­ century science, and thereby best demonstrates what I take to be Spinoza's intention. B acon, as everybody knows, was the great empiricist, the champion of experiment and the guiding light of the scientific revolu­ tion. I want here briefly to examine his account of the nature, scope and correct employment of reason in the light of what I · have been saying. In his article on 'Rationalism' in the Encyclopaedia ofPhilosophy, Bernard Williams defines his subject as "the philosophical outlook or program which stresses the power of a priori reason to grasp substantial truths about the world and correspondingly tends to regard natural science as basically an a priori enterprise"Y He explicitly identifies this approach with the work of Descartes, Spinoza and Leibniz, and he then quotes this remark of Bacon's : "The empiricists are like ants, they collect and put to use ; but rationalists, like spiders, spin threads out of themselves". And Williams goes on, "Bacon, of course, preferred the ants". 1 6 But he didn't. The first thing to say about this is that it is very odd of Williams here

1 86

RICHARD FRANCKS

to equate Bacon's talk of ,rationalists' ("rationales") with the subject of his own article. Bacon was writing, after all, in 1 607, when Descartes was only 1 1 years old, and before either Spinoza or Leibniz was born. Yet if Williams means that their kind of rationalism existed before them, he does not say so. In reality, the object of B acon's attack is not the 'rationalists' at all, but the scholastics, who, incidentally, were also attacked by Descartes, Spinoza, and even Leibniz. This is made clear, for example, in the Advancement of Learning, where B acon again says that The schoolmen . . . their wits being shut up in the cells of a few authors, chiefly Aristotle their dictator, . . . did out of no great quantity of matter spin out unto us those laborious webs of learning which are extant in their books. 1 7

And the same image occurs agam elsewhere, sufficient to make the intention quite clear. More significant than this, though, is the fact that j ust as B acon's rationalists were not those of Williams, neither were his empiricists. The complete passage, from which Williams quotes only the beginning, goes like this (Bettany's translation): The empirics are like the ant, they only bring together and use; the rationalists are like spiders, which spin webs out of their own bowels ; but the bee follows a middle course, for she draws her materials from the flowers of the garden and the field, and yet changes and digests them by a power of her own. Nor is the true process ofphilosophy unlike this . . . 1 8 (My italics).

Bacon, then, is no more in favour of the ants than he is of the spiders : he follows this third course, which we can hardly avoid calling the 'bee-line'. Again the analogy recurs in several places, and its point is the same each time ; indeed I think it would not be an exaggeration to say that the advocacy and explanation of this ·third course, described always as intermediate between the other two, is B acon's whole philosophical pur­ pose. This may seem to be no more than a trivial, if careless, mistake on the part of Williams. My reason for dwelling on it is that I think it provides a very clear, if rather extreme, example of the dangers of too simply reading back into the past the categories of the present, and also of the kind of misrepresentation of the philosophers of the past which such a procedure must of necessity involve. To appreciate this, and to see its relevance to Spinoza, we need to look more closely at B acon's middle course, his 'bee-line'.

T H E C A S E OF S PINOZA

1 87

One of the points of difference between Spinoza and Bacon, as Spinoza himself remarks, 1 9 is that while Spinoza believed, like Descartes, that the human understanding is of itself infallible, and therefore that error is the result of outside influence and impurities in the judgment, Bacon took a dim view of the unaided powers of what he called "the narrow cells of the human understanding" : "The understanding, left to itself, ought always to be suspected". This is his standard complaint against the scholastics, that they have "too great a reverence, and a kind of adoration of the mind and under­ standing of man". As a result, they have attempted to come to conclusions as to the nature of the world solely on the basis of how it seemed to them, in their cloistered isolation, that it must be. They have "tumbled up and down in their own reasons and conceits", and their learning has dissolved into a "number of subtle, idle, unwholesome and, as I may term them, vermiculate questions , which have indeed a kind of quickness, and life of

spirit, but no soundness of matter or goodness of quality". The schoolmen have gone astray because their over-estimation of the powers of the understanding has led them to withdraw "themselves too much from the contemplation of nature, and the observations of expe­ rience" - they have speculated in the absence of observation, which is always disastrous:

For the wit and mind of man, if it work upon matter, which is the contemplation of the creatures of God, worketh according to the stuff, and is limited thereby: but if it work upon itself, as the spider worketh his web, then is it endless, and brings forth indeed cobwebs of learning, admirable for the fineness of thread and work, but of no substance or profit.

This is the Bacon with which we are all familiar. But the other side of his work is too often, as by Williams, overlooked. His attacks on the weakness of the unaided reason, for instance, are matched by a more than equal number of remarks as to the frailty of the senses, worthy of any so-called 'rationalist' : "By far the greatest hindrance and aberration of the human intellect proceeds from the dullness, incompetency and fallacies of the senses". Just as he attacks as spiders those who, from too great a reliance on reason, draw conclusions in the absence of experience, so he also attacks as ants those who follow experience without correctly submitting it to the test of reason:

1 88

RICHARD FRANCKS

Men of learning, (but supine and easy), have taken up, for the construction and constitution of their philosophy, certain rumours and, as it were, reports and breezes of experience, and have allowed to them the weight of legitimate testimony.

Such people - among whom for some reason he seems to count Gilbert seek knowledge not from observation, but from 'vague experience', and employ not true induction, but induction 'by simple enumeration'. Neither of these two approaches, neither the spider's nor the ant's, has in Bacon's view any value ; yet until his own day, he believed, they were the only ones to have been tried. His famous optimism as to the power of science is therefore entirely based on this belief that he is offering a new method from which "great hopes may be entertained", to wit, "a true and legitimate union between the experimental and rational faculty". Philo­ sophy is therefore like the bee because It does not rely either exclusively or principally on the strength of the mind, nor does it lay up in the mind materials supplied from natural history and mechanical experiments in their raw state, but stores them in the intellect, after having altered and digested them.

The details of the method by which B acon aims to achieve this are complex and to us rather eccentric, but the outline is clear enough - it begins from the collection of data and proceeds by induction to discover the underlying 'forms' of phenomena, and the 'axioms' which govern them. But it would be naive of us and unfair to Bacon to suggest that this was all that the union of the experimental and rational faculties consisted in: the two are interwoven throughout. Observations are made not in a haphazard or casual manner, but according to a carefully worked-out programme, and great care is taken at every point to eliminate as far as possible the weakness of the senses. From such observations tentative hypotheses are drawn according to a strict method, and further experi­ ments are thereby devised to test them, or to reveal further information, and so on. In theory, therefore, the judgments of reason are always to be constrained by the facts of experience, and the observations of experience directed by the dictates of reason. In this way we may hope to climb "by a continuous path, through the forests of experience to the open lands of axioms". From all this, then, it is quite clear that the 'power of its own' by which philosophy, like the bee, 'changes and digests' the facts of experience, is none other than the power of reason - the intellectual faculty from the union of which with experience so much is to be expected. Hence Bacon's

THE CASE OF SPINOZA

1 89

repeated and seemingly puzzling remarks, whenever he is summarising his message, to the effect that "to conquer the difficulties and obscurities of nature", we need "a more perfect use of reason". Of course, none of this insistence on reason implies that Bacon was really a rationalist, or was less scientifically-minded than we usually think. Nor does the juxtaposition of such remarks with his more famous demands for experimentation mean that he was somehow 'confused'. Precisely the opposite is the case: it is j ust because B acon insisted on the importance of reason in 'natural philosophy' that we are justified in regarding him as the inspiration and chief prophet of the scientific revolution, because it is just these remarks which mark out Bacon as no mere empirically-minded dabbler or Renaissance collector of curiosities, but as a true 'scientist', in the sense of a man who sought to discover truths about the world by the careful and systematic application to it of the power of thought, or the intellect, or reason. Now, if all that is true of B acon, why is it not also true of Spinoza? Although in his formal theory Spinoza describes three kinds of knowledge, in practice he uses much more often a simpler division between reason, or the understanding, and imagination.2o The point of this paper is to say that, given the context within which Spinoza lived and wrote, we must take this as primarily a division, not between a priori and a posteriori, but between scientific understanding and casual appearance. Although there is considerable textual evidence to support this con­ clusion, I do not propose to go into it here; more significant for our purposes is the absence of any real evidence to the contrary. Spinoza, it is true, was writing some 50 years after B acon, but nevertheless the world in which he wrote was still a great deal closer to B acon's than it was to our own. When Spinoza says that the nature of the world can be discovered by the correct use of reason, therefore, we should take him as meaning something similar to what B acon had meant by the very similar remarks quoted above. When Spinoza talks of the need to discover axioms in our investigations, we should take him not to be suggesting some kind of a priori science, but to be advocating an experimental procedure similar in its most basic essentials to that which B acon was describing. And when Spinoza says that sense-experience is a cause of falsity, we should take him as meaning, not that science is a worthless pursuit, but as meaning, like B acon, that a scientific understanding will often contradict the appear­ ances of sense - particularly must this be so, since Spinoza explicitly restricts his remarks in this context to B acon's own category of 'knowledge

190

RICHARD FRANCKS

, from vague experience . 2 1 I am not arguing here that just because B acon meant these things, it follows that Spinoza must have meant them also; but given that Bacon did say these things, it at least follows that if Spinoza had intended something as radically different as is usually maintained, he would have needed, if he was to have a hope of being understood by his contemporaries, to make it very clear that he was using the terms in such an eccentric way, and then to set out very clearly the odd theory of science and of knowledge which he was proposing. B ut he does not, and we have nothing which we can set against such very Baconian-sounding comments as "the knowledge of nature is sought from nature", or "the interpretation of nature consists in the examination of the history of nature . . . ", and "philosophy is based on common notions which must be sought from nature alone". 22 As a result of such remarks, Spinoza's own contemporaries at least were in no doubt as to what he intended.23 Although modern biographies tend to present him as a kind of saintly, other-worldly recluse, there is no evidence that he was perceived as such in his own lifetime. He was , after all, a contemporary of Huygens, Boyle and Newton. He lived at the centre of the most civilised, and the most politically and economically advanced, state in Europe, at the height of its success not only economically, but also in art and in science. He himself was actively engaged in what has been called the 'nursery of Dutch science', the lens-making industry, as were all the great men of science of his time, including Huygens and Leeu­ wenhoek.24 While he is said to have lived a relatively quiet life, and his intimates were drawn chiefly from the extreme protestant 'Collegiant' circle, it would be absurd to picture him as cut off from the intellectual developments of his day. He is known to have been well acquainted with Huygens, and he also corresponded regularly with Oldenburg, the secreta­ ry of the new Royal Society in England, and through him with Boyle. He was well known to men like Leibniz and Tschirnhaus, and was a close friend of Jan DeWitt. In general, while he can hardly be said to have been in the forefront of the scientific developments of his day, he was cl.early quite at home with the men of the new, self-confident, secular, scientific world of the late seventeenth century. If he had really been the rationalist and mystic he is said to have been, what could he have found to say to them, and what could they have replied? If he had really believed that all truths could be discovered a priori, would he have been likely to be invited to visit Conde, or to lecture at Heidelburg? It is, I suppose, possible that some of these people might have been prepared to overlook such an

THE CASE OF SPINOZA

191

eccentric theory for the sake o f his mystical intuition o f the Unity o f Being, but it is unlikely that they all should. In fact, of course, there is no indica­ tion that any of them did. He discusses optics with Huygens and chemistry with Boyle ; he wrote a treatise on the rainbow and a treatise on probabili­ ty; he is suitably sceptical about ghosts and astrology and alchemy and new Messiahs; he denounces final causes and teleological explanation with the best of them, and his letters contain experiments and pieces of appara­ tus and observations of various kinds. He was, in other words, a typical man of reason, or man of science, of his time, except that he was rather more uncompromising than most. And he paid a fairly typical price - he was derided as the 'reformer of the new philosophy', he was excommuni­ cated as a heretic, and he was universally condemned as a materialist and an atheist. The basic problem here, I think, which determines our oversimplifica­ tion of B acon just as it does our misinterpretation of Spinoza, is a mis­ understanding of the nature of the general change in the intellectual climate which was taking place in the period in which they both lived. The seventeenth century, as we all know, was the age of the scientific revolu­ tion. The great transformation of outlook we know by that name consisted essentially in the discovery, or perhaps the rediscovery, of the power of empirical investigation to reveal the nature of reality. It seems to follow from this that the key discovery of the seventeenth century was that of the power of sense experience, and so we are led to expect the philosophers of the seventeenth century who were interested in promoting this change to emphasise that power, and we classify them as rationalists or empiri­ cists according to the extent to which they appear to do so. Spinoza stresses the power of reason, and so is classified as a rationalist who had no interest in the new science. What I am suggesting, by contrast, is that precisely the opposite is the case - that if we want a slogan to sum up the vast political, religious, social and philosophical change which was taking place in the seventeenth century, then we must say that the philosophers and scientists of the seventeenth century discovered, or perhaps redis­ covered, not the power of experience, but the power of reason. They discovered, that is, the power to reason about experience instead of reasoning in isolation from it, exactly as B acon described. Empirical observations in Spinoza's day, we must remember, were two a penny. People had been making empirical discoveries at an ever­ increasing rate for some 200 years. Copernicus had been dead for 1 00 years, Tycho for 60. Kepler's laws were already a full 50 years old.

192

RICHARD FRANCKS

Galileo's amazing telescopic observations were even older, and by Spinoza's time telescopes and microscopes had become the playthings of the fashionable world. (Pepys, for instance, had both, and used binoculars to look at women during church services.) In the 1 660's and '70's there simply was no question as to whether or not empirical observation could produce knowledge of the world - the world was reeling from the impact of all the knowledge which had been so produced. From ancient Greece and from the New World, from the rings of S aturn to the wings of the flea, new facts had been discovered at such a rate that the common boast of the seventeenth century was that more had been revealed in a few recent years than in all the previous 2,000.2 5 The great problem which had to be solved, therefore, was not the patently absurd question, then as now, of whether or not empirical investigation was necessary to knowledge, but rather that of how all this wealth of empirical knowledge could be systema­ tised and accommodated and related to what had seemed to be the knowledge of previous ages. And the answer which was arrived at was that it would only be achieved by the use of reason. On the scientific level, therefore, the characteristic discoveries of the seventeenth century were not further observations , but rather reasoned analyses of observations, the greater part of which had been inherited from previous generations. Newton, for instance, was no great observer; Kepler even less so. Galileo, it is true, did make observations of great significance, but his reputation as a scientist rightly rests more firmly on the reasoned method which determined those observations, and on the mathematical models by means of which he interpreted them. To paraphrase another remark by Spinoza, by the application of reason the scientists of the seventeenth century were striving to understand natural phenomena as intelligent beings, and no longer merely to gaze at them like fools.26 If this kind of application of reason was to be shown to be worthwhile, however, there were a number of questions of what we would call a philosophical nature which had to be answered. Broadly, they can be divided into three groups : epistemological questions, concerned with showing how scientific investigation could yield genuine knowledge ; meta­ physical questions concerning the nature of the reality which science was claiming to reveal, and questions in the philosophy of religion as to the legitimacy of attempting to obtain knowledge by any such means. These were the problems which the philosophers of the scientific revolution, such as Descartes and Spinoza, set themselves to solve. To all these questions, Spinoza had answers to otTer. Many of these answers are very difficult, and

T H E C A S E OF S PI NOZA

193

several of them are unique in modern Western philosophy. As such they demand the closest attention from philosophers, but they will never receive the consideration they deserve, and their significance will never rightly be assessed, as long as we continue to ignore the real Spinoza and confine our attention to the gross caricature which is currently presented by the history of philosophy.

NOTES 1 Philosophy and its Past, by J. Ree, M . Ayers and A. Westoby (Sussex: Harvester Press, 1978). 2 See Ethics, 2, Props. 44, 43 Schol., 40 Schol. 1, and below, note 20. 3 Ethics, passim, especially Part 5. 4 From 'Retracing the Arts to Theology . . . '. Quoted from Philosophy in the Middle Ages, edited by A. Hyman and J. Walsh (New York: Harper and Row, 1967), p. 423. 5 E.g. Monadology, 33. 6 See Ethics, I, Prop. 33, Schol. 1 ; De Intellectus Emendatione, Para. 53. 7 See, e.g., Critique of Pure Reason, A229, B282. 8 11istory of the Royal SOciety, edited by J. 1. Cope and H. W. Jones (St. Louis: Washington University Press, 1958), pp. 1 10, 57. 9 Last sermon in Wittenberg, 1 546. Taken from Luther's Works, American edition (Philadel­ phia: Fortress [Concordia] Press, 1 959), Volume 5 1 , pp. 374, 379. 10 Luther, of course, lived some 100 years before Spinoza, but the debate continued. Significantly, by the 1 670's the target for attacks of this kind, in England at least, seems to have shifted away from the Renaissance humanistic learning of the established universities towards the new experimental philosophy, which was coming to replace it as a paradigm of human learning, and as the most successful case of the use of human reason. See R. F. Jones, 'The Humanistic Defence of Learning in the Mid-Seventeenth Century', in Reason and Imagination, edited by J. A. Mazzeo (London: Routledge, 1 962). Il Tractatus Theologico-Politicus, Preface. 12 Discourse, I , Para. I. lJ See, e.g., Ethics, 2, Prop. 35, Schol., and 4, Prop. I , Schol.; Discourse, 4, last para.. 14 Dialogue Concerning the Two Chief World Systems, translated by S. Drake ( Berkeley: University of California Press, 1962), pp. 248, 254, 339. 15 Encyclopaedia of Philosophy, edited by P. Edwards (London: Collier-MacMillan, 1 967), Volume 7, p. 69. 1 6Ibid. , p. 73. 17 This and all subsequent quotations from B acon are taken from Essays Civil and Moral . . . etc. , edited by G. T. Bettany (London: Ward Lock, 1 892). The passages quoted are taken from 'The Advancement of Learning', 1 , pp. 1 16- 1 20; 'Novum Organon', 1 , Aph. 50-105, pp. 259-288, and 'The Great Instauration', Preface, pp. 405-41 I . 18 I have been unable to find a translation of the 'Cogitata et Visa', an early draft of the 'Novum Organon', from which Williams quotes, and have given here the parallel passage from 'Novum Organon', I , Aph. 95, which is virtually word-for-word the same. The 'Cogitata' passage runs: "Empiricos enim formicae more congerere tantum et uti, rationales autem aranearum more tel as ex se conficere. Apis vero rationem mediam ess.e, quae materiam ex •

194

RICHARD FRANCKS

floribus tam horti quam agri eliciat, sed earn etiam propria facultate vertat et digerat. Neque absimile verae philosophiae opificium esse". 1 9 Letter 2. 20 This is shown, for example, in Ethics, 2, Prop. 44, where, as the Corollary makes clear, the term 'reason' is used to include both 'ratio' and 'scientia intuitiva'. Without this interpretation expressions like 'a man of reason' or 'in accordance with reason' make little sense. 2 1 Ethics, 2, Prop. 40, Schol. 2; De. Int. Em. , para. 1 9. 22 Tractatus Theologico-Politicus, Chapters 7 and 14. 23 I am indebted in what follows to Prof. Dr M. J. Petry, of Rotterdam, whose generous help and advice have gone some way towards alleviating my ignorance of intellectual currents in seventeenth-century Holland. 24 See The Dutch Republic, by C.H. Wilson (London: Weidenfeld and Nicholson, 1968), p. 95. 25 And it should not be assumed that difficulties of communication might have significantly reduced the force of these discoveries: in May, 1 665, for instance, we find Spinoza in a letter to Oldenburg commenting on telescopic observations made in Italy only a few months before. 26 Ethics, 1 , Appendix.

STUART BROWN

LEI B N I Z' S B R E A K WITH CART E S IAN 'RATIONAL I S M'·

In a short paper for the Acta eruditorum in 1 694 Leibniz gave the first public intimation of his new metaphysics. In it he lamented the fact that a first philosophy remained to be sought. Descartes, for all that he had contri­ buted "some admirable things" (Loemker, 1969, p. 432; Gerhardt, 1 875-1 890, iv, p. 468), had "failed to distinguish the certain from the uncertain" and in consequence missed the mark. Metaphysicians were in need of a new way of proceeding (proponendi ratio) which would, in one of Leibniz's favourite metaphors, provide them with an Ariadne's thread by which to escape from the maze of problems in which they were lost. What was needed was a procedure that would serve as a check on the truth of metaphysical propositions - a way of avoiding the kind of error into which Descartes and his followers had fallen. The fault that Leibniz saw in Descartes is one which students of Philosophy quickly learn to identify as a general weakness in 'Rationalism'. Rationalism, they are told, is the view that everything worthy of the name 'knowledge' is either axiomatic, i.e. intuitively certain or evident by the light of reason, or else can be derived from such axioms. A stock problem with Rationalism, so described, is that of finding reliable criteria for what is intuitively certain or, in Descartes' terminology, for 'clear and distinct ideas'. That is exactly the problem noted in Leibniz's complaint that Descartes "failed to distinguish adequately the certain from the uncertain". Leibniz's emphasis on what we might call the 'subjectivity' of the appeal to clear and distinct ideas is a repeated one. ' The Cartesians were simply avoiding the need to prove what they advanced "on the pretext that those who will meditate on these ideas will discover in them the same thing as they; that is to say, that those who will accustom themselves to the j argon and way of thinking will be similarly predisposed, which is very true". (Erdmann, 1 840, p. 1 36) Given Leibniz's emphasis on this difficulty it is remarkable that Leibniz himself is so often branded as some kind of disciple of Descartes in epistemological matters and indeed as committed to exactly this form of 'rationalism'. This stock picture of Leibniz is presented, for instance, in Roger Scruton's book on Kant: 1 95 A . J. Holland (ed.), Philosophy, Its History and Historiography, 195-208.

196

STUART B ROWN

Leibniz belonged to a school of thought now generally labelled 'rationalist' . . . Leibniz believed that the understanding contains within itself certain innate principles, which it knows intuitively to be true, and which form the axioms from which a complete description of the world can be derived. (Scruton, 1982, p. 14)

Nor is Scruton alone in considering 'rationalism' as a school or fashion initiated or at least revived by Descartes. J. H. Randall suggested it was a way of thinking about knowledge which pervaded seventeenth-century Philosophy; With Descartes all save those actually making experimental discoveries in physics believed, throughout the century, that the surest foundation of truth was not any appeal to the fallible testimony of sense experience, so often proved wrong, but rather the clear and distinct intuition of geometrical axioms. We know intuitively, with absolute certainty, that the whole is equal to the sum of the parts, and that a straight line is the shortest distance between two points: similarly it was hoped that such intuitive axioms could be discovered in mechanics, in morals, in politics, and in religion . . . Rationalism of the geometrical type was the popular intellectual method of The Age of Reason . . . (Randall, 1940, p. 262)

Randall understates the prestige of Euclidean geometry. Works of physics like Newton's Principia or, come to that, Leibniz's Essay on Dynamics were presented in geometrical form. At the same time Randall overstates the influence of Descartes' interpretation of Euclidean geometry. Euclidean geometry was indeed a paradigm of true science. But it admitted of other interpretations than that of being based on intuitively certain or 'clear and distinct' ideas. In short it was open to someone to acknowledge the scientific prestige of Euclidean geometry without being what in twentieth­ century terms is known as a 'rationalist' or, in nineteenth-century terms, an 'intuitionist'.2 That was Leibniz's position, as is clear from this criticism of Descartes: . . . I t i s astonishing t o m e t o see that that famous philosopher of our times Descartes, who has recommended so much the art of doubting, has so little practised it in good faith on the occasions when it would have been most useful, contenting himself with alleging the self­ evidence for ideas which Euclid and the other Geometers have very wisely refused to begin with; this is also the way to cover up all sorts of illusory ideas and prejudices. (Wiener, 1 95 1 , p . 3 5 ; Gerhardt, 1 875-1890, vii, p . 1 65)

Philosophers should imitate the method of geometry. But that method did not depend on axioms being taken as intuitively certain and as not requiring further proof. Leibniz sided with those, like Giles de Roberval,

LEIB NIZ'S BREAK WITH CARTESIAN 'RATIONALISM'

197

who called for further work to be done on the foundations of Euclidean geometry. Yet in doing so, he took geometry to exemplify, not a fully demonstrated science, but a method of making progress in science without meeting the sceptic's demand for indubitable first principles : I blame Euclid much less for assuming certain things without proof, for he at least established the fact that if we assume a few hypotheses, we can be sure that what follows is equal in certainty, at least, to the hypotheses themselves. If Descartes or other philosophers had done something similar to this, we should not be in difficulty. Moreover, the sceptics, who despise the sciences on the pretext that they sometimes use undemonstrated principles, ought to regard this as said also to them. I hold, in contrast, that the geometricians should be praised because they have pinned down science with such pegs, as it were, and have discovered an art of advancing and of deriving so many things from a few. If they had tried to put off the discovery of theorems and problems until all the axioms and postulates had been proved; we should perhaps have no geometry today. (Loemker, 1969, p. 384; Gerhardt, 1 875-1 890, iv, p. 355)

Leibniz's rejection of a rationalist methodology was a reluctant one. Indeed his initial response to Descartes' difficulties had been to advocate a more austere rationalism. But although the requirements of such a rationalism could be met by arithmetic, they were not met by geometry. Leibniz's austere rationalism provided a model for God's knowledge of the world and perhaps for that to which human souls might attain in some future state of 'beatific vision'. But it was too austere to explain the success of geometry or to recover metaphysical territory from the sceptics. A different method would be needed if metaphysical disputes were to be resolved and progress made in this backward science. In the 1680's and the early 1 690's the method which Leibniz advocated as an alternative to 'rationalism' was modelled on pure geometry. His paper 'On the Reform of Metaphysics . . .' of 1 694,3 however - and his discussions of his 'New System . . . ' in the mid- 1 690's, suggest that he was moving even further from rationalism. The direction was that of a 'mixed' view of metaphysics as modelled on his dynamics, in which 'axioms' and 'hypotheses' both play a part. In the first part of this paper I shall trace these developments in greater detail. I shall try to bring out how those who have taken Leibniz for a 'rationalist' have assimilated his epistemology and methodology too closely to that of Descartes. In the second and third parts I shall develop this theme further, but also consider whether these confusions are rooted in a mistaken model for the history of Modern Philosophy.

198

STUART BROWN

I.

Much modern usage of the word 'rationalist' in philosophical circles can be summed up in the definition of a 'rationalist' as someone who holds that anything worthy of the name 'knowledge' is either self-evident (by the light of reason) or can be demonstrated as following from 'self-evident' truths. The title 'rationalist' would probably be reserved for those who hold that there is metaphysical knowledge of this kind. But it is a virtue of the definition that it can include both 'sceptics' and 'dogmatists' as well as those who are sceptical on some matters but not on others. It can therefore include Descartes himself as well as those like Leibniz's Paris friend, Simon Foucher, who thought Descartes had failed to answer scepticism. Foucher, like Leibniz, regarded Descartes as a restorer of Academic Scepticism.4 Unlike Leibniz, however, Foucher remained faithful to the method of doubt and to the search for 'evident truths'. Leibniz had initially (in 1 675) agreed with Foucher that a metaphysics should be pursued along rationalist lines: I agree with you that it is important once and for all to examine all our presuppositions in order to establish something sound. For I hold that it is only when we can prove everything we assert that we understand perfectly the thing being considered. (Loemker, 1969, p. 1 5 1 ; Gerhardt, 1 875-1 890, i , p . 369)

But Leibniz gradually became disillusioned with the method of doubt and indeed with the project of proving everything asserted in metaphysics. By 1 680 or so he had come round to the view that metaphysics could only make progress by using assumptions which had not been proved. It was on this basis that Leibniz found he was able to engage in the constructive metaphysics of his Discourse on Metaphysics and other writings of the mid 1 680's. In his 'grande lettre' to Foucher of 1 686, Leibniz defended his new method as one licensed by the practice of the Geometers : . . . in matters of the human sciences we must try to advance and even if the only way to do this was by establishing many things on a few suppositions , that remains useful, for at least we should know that all that remained to reach a full demonstration was to prove these few suppositions, and in the meantime we should have some hypothetical truths and escape from the confusion of disputes . . .

LEIBNIZ'S BREAK WITH CARTESIAN 'RATIONALISM'

199

If then we supposed for example the principle of contradiction as well as that in every true proposition the notion of the predicate is included in that of the subject, and several other axioms of this nature, and if from these we could prove many things as demonstratively as the Geometers, would you not find this result of consequence? But we would have to begin this method one day if we were to begin to finish the disputes. It would always be a means of gaining territory. (Gerhardt, 1 875-1 890, i, p. 3 8 1 ff., trans. R. Niall D. Martin)

The inclusion of the principle of contradiction as if it were an axiom of the same 'nature' as the principle of inesse is misleading. For the principle of contradiction can be claimed to be an 'immediate truth' of reason and Leibniz himself was later to claim that it was the only truth in this category.5 Leibniz never claimed that the inesse principle was an immediate truth. He did conclude a defence of it by saying: "Either the predicate is in the subject or else I do not know what truth is". (Gerhardt, 1 875-1 890, ii, p. 56) But it can hardly have escaped his attention that to take the apparent indubitability of the inesse principle as evidence of its truth would be to fall into exactly the kind of error which (he and others like Foucher believed) vitiated Descartes' appeal to 'clear and distinct ideas'. What Leibniz sought, rather, was the right to include the inesse principle and 'several other axioms of this nature' as assumptions on the basis of which to establish 'hypothetical truths'. The basis of this right in metaphysics is grounded by Leibniz in the claim that it is just this method which proved successful in geometry : For example, Archimedes supposes only a few things : that the straight line is the shortest, that of two lines of which each is everywhere concave on the same side, the line included is shorter than the including line, and on this basis he rigorously concludes his demonstrations. (Gerhardt, 1 875-1890, i, p. 3 8 1 ff.)

Leibniz did not specify what other axioms 'of this nature' he had in mind. But there is a p aper written by him within a few months of the 'grande lettre' which affords clues. This is the paper commonly (though mis­ leadingly) entitled 'First Truths', after its opening two words. In 'First Truths' Leibniz lays out his metaphysical system in a deductive form. It is reasonable to infer that he intended as amongst the 'other axioms' just those which are not presented as following from the two with which he begins, namely, the principle of contradiction and the inesse principle. If that is so then another 'axiom' appears to be "all created individual sub­ stances are . . . expressions of the same universe, and of the s ame universal cause". (Parkinson, 1 973, p. 90 ; Couturat, 1 903, p. 5 2 1 ) By reference to

200

STUART B ROWN

other texts such as the Discourse it may be possible to derive this proposi­ tion from other more primitive 'suppositions' about 'the connection between things' and about every substance being 'like a mirror of God', without doing violence to Leibniz's thought. The choice of axioms should indeed be from amongst those propositions which there is no reason to doubt. But otherwise they should be selected for their 'fruitfulness', i.e. because many other things can be clearly shown to follow from them. It was this latter consideration which Leibniz stressed in his 'grande lettre' to Foucher. Around 1 686 Leibniz was particularly impressed with the fruitfulness of his inesse principle. But there is little mention of this principle in Leibniz's later writings. For instance, Leibniz claimed in Discourse, Section 33 that "the very idea or essence of the soul implies that all that appears to it, all its perceptions, must arise in it spontaneously (sponte) from its own nature . . . " . This is claimed as a direct consequence of the inesse principle as applied to individual substances, i.e. "that everything that happens to the soul, and to each substance, is a consequence of its notion". By the time of the New System, however, the spontaneity of substances is advanced as part of a hypothesis : "God first created the soul, and every real unity, in such a way that everything in it must spring from within itself, by a perfect spontaneity with regard to itself, and yet in a perfect conformity with things outside". (Gerhardt, 1 875-1 890, iv, p. 484) It is not the inesse principle but the requirement that substances must be 'true unities' which becomes the basis of Leibniz's later defence of spontaneity. It is because of the sim­ plicity of substances that Leibniz claims, in the Monadology, that there is "no means of explaining how a monad can be altered or changed within itself by any other created thing". (Section 7) It is possible that Leibniz had second thoughts about the truth of his inesse principle as generalised to apply to substances. It was certainly more contentious than he had at first believed, in particular because of the problems it seemed to pose for free will. But there is no evidence that Leibniz thought these problems were unanswerable. The evidence points to the conclusion that Leibniz continued to believe in the inesse principle but ceased to make the claims for it that he was prepared to make in 1 686. Some weight should be given to the consideration that, as it seemed to Leibniz, there was an increasing tendency to discount writings which made use of Scholastic jargon. But Leibniz himself was not deterred from an open, if qualified, defence of Scholasticism by this consideration. A more likely explanation of the fact that Leibniz virtually dropped the inesse

LEIBNIZ'S BREAK WITH CARTESIAN 'RATIONALISM'

201

principle may be that he ceased to regard it as so fruitful and therefore fundamental as he had done in 1 686. He may, in particular, have concluded that he was mistaken in believing that the spontaneity of substances followed directly from the inesse principle.6 In general the inesse principle is replaced in Leibniz's later writings by the principle of sufficient reason. In Leibniz's later writings he is sometimes willing to have his system considered as a "hypothesis which is suitable for explaining phenomena". (Loemker, 1 969, p. 493 ; Gerhardt, 1 875-1 890, iv, p. 5 1 8) In reply to Foucher's critique of the New System he roundly declared that "all systems come after the event to save phenomena or appearances". (Parkinson, 1 973, p. 1 20 ; Gerhardt, 1 875-1 890, iv, p. 496) This does not mean that Leibniz believed he was offering no more than one hypothesis amongst many. But it does betray a further shift on his part away from classical rationalism. For it suggests that metaphysics is in some respects more like empirical science than it is like geometry. Metaphysical principles, like Descartes' allegedly clear and distinct idea of material substance as consisting in extension, could be criticised on a posteriori grounds as well as a priori ones. A system which provided for forces in nature might thus be claimed to save appearances in a way Cartesian mechanics could not. One of the most striking features of Leibniz's mature philosophy is his willingness to support his claims with both a priori and a posteriori argu­ ments.? He sometimes refers to his a priori arguments as proofs or demon­ strations. But they are evidently not proofs or demonstrations in the rationalist sense. For, if they were, nothing further would be added by a posteriori arguments to the same conclusion. If, on the other hand, the a priori arguments are based on principles which are not themselves known intuitively but are put forward as 'suppositions', then those principles may themselves be defended on a posteriori grounds. In the Discourse on Metaphysics Leibniz expressed the belief that "people who are capable of meditation will judge favourably of our principles for this very reason: that they can easily see the nature of the connection between soul and body, which seems· inexplicable in any other way". (Section 33) If this criterion was at that time regarded by Leibniz as no more than a selling-point for principles of which he was personally con­ vinced on other grounds, he changed his tune in his papers on the 'reform' of metaphysics. For the 'reform' he had in mind is exactly that of finding a touch-stone by which to distinguish what is true and what is false in metaphysics. Unlike the truths of mathematics the truths of metaphysics do not carry their own 'proofs' with them. Unlike the truths of mathemat-

202

STUART B ROWN

ics, therefore, those of metaphysics are not known intuitively. Metaphy­ sics, according to Leibniz, requires a 'new method' if proofs are to be available for it. In the French version of Leibniz's paper on the 'reform' of metaphysics he concludes with the remark that he hopes that the 'great problem' of the mutual communication of substances and of the union of soul and body "will be resolved in such a clear manner, that this of itself will serve as a proof that we have found the key to some of these things".8 One way of expressing this reform is to say that Leibniz rejected the Cartesian appeal to our ideas in favour of a pragmatic criterion of meta­ physical truth. We do not have intuitive knowledge of the essence of substance, as Descartes had mistakenly supposed. On the contrary, as Leibniz remarked specifically a propos of the concept of substance : 'The true sign of a clear and distinct notion is one's having the means for giving a priori proofs of many truths about it . . . ". (Remnant and Bennett, 1 98 1 , p. 219) If we had to count on immediate knowledge of our ideas then Locke's conclusion that we have "no clear idea of substance in general" would be the correct one. But "this opinion", Leibniz insists, "springs from a demand for a way of knowing which the object does not admit of'. A false account of substance will show itself either in having consequences that are false or by being unable to give an intelligible explanation of what is known independently to be true. A true account, by contrast, is to be recom­ mended because of its fruitfulness in solving problems and in explaining what is known independently to be true. According to Leibniz's paper on the reform of metaphysics, "the true and fruitful concepts, not only of substance, but of cause, action, relation, similarity, and many other general terms as well, are hidden from popular understanding". It was not surprising therefore that metaphysics remained in a backward state. What was needed was a way of testing metaphysical propositions, taking 'fruitfulness' as the sign of truth rather than merely inspecting the propositions themselves. It is on this basis that Leibniz advertises his concept of substance : "This is so fruitful that there follow from it fundamental truths, even about God and minds and the nature of bodies - truths heretofore known in part though hardly demonstrated, and unknown in part, but of the greatest use for the future in the other sciences". (Loemker, 1 969, p. 433 ; Gerhardt, 1 875-1 890, iv, p. 469) Leibniz's position in relation to Cartesian rationalism may thus be sum­ marized by saying that, in common with the sceptics , he rejected the appeal to intuition of 'clear and distinct' ideas as a criterion of truth. Clear and distinct ideas were not, as Descartes supposed, the starting-point for

L E I B N I Z ' S B REAK WITH CARTE S IA N 'RATI ONALI S M'

203

metaphysics but rather its goal. Leibniz did not reject the rationalist ideal of knowledge, which remained for him the proper model for God's knowledge. To some extent his retreat from rationalist strategies for the advancement of metaphysics can be seen as reflecting a growing scepticism about humans attaining that ideal. But it also reflects a growing confidence in a more modest programme for the advancement of metaphysics. 2.

The persistence o f the view that Leibniz was a rationalist may b e put in a general perspective by saying that Leibniz has been a victim of something like what Jonathan Ree, elsewhere in this volume, refers to as 'integral , history .9 The 'integral history' of Modern Philosophy characteristically represents 'rationalism' and 'empiricism' as two opposed but ultimately erroneous schools of thought whose partial insights can only be grasped from a later standpoint, either that of Kant or perhaps the author of the history himself. Such integral histories of Modern Philosophy are so common as to be the rule rather than the exception. Part of the reason for this is that Kant himself assumed such an integral history and so a historian can mistake his tacit acceptance of some of Kant's assumptions about his own place in the history of Modern Philosophy for a fidelity to the historian's subject matter. C. R. Morris, for example, represents Kant in the following way in his Locke, Berkeley, Hume :

Kant starts from the conviction that both rationalists and empiricists have failed; and their failure he puts to himself in this way. The rationalists thought that all knowledge was the product of pure thinking, of which mathematics is the type; that is to say, it is the product of the mind's own activity, the essence of the mind being to think. Thus perception for them played no part in knowledge at all, except to pull a trigger, as it were, and thereby to be the necessary occasion of the mind's being active in its own essential manner, viz. thinking. This view breaks down because of the difficulties connected with innate principles, and also because nobody really believes that perception plays no contributory part in knowledge at all, especially in view of the emphasis laid on observation and experiment by modern science. The empiricists, on the other hand, sought to maintain that all knowledge was through and through grounded in perception : in their hands this came to mean that all knowing is perceiving; and thus either the existence of thought was denied altogether, or else thinking was regarded as making no contribution to knowledge. This view inevitably resulted in the scepticism of Hume. Having put the situation to himself in this way, Kant inevitably sought to steer a middle way between the two views . . . (Morris, 1 9 3 1 , p. 163, italics added)

204

STUART BROWN

For Morris it seems as if there is a kind of internal dialectic to the history of Modern Philosophy, in which each stage is seen as somehow following on logically from the preceding one, by opposition, development or syn­ thesis. Within such a scheme there is no place for Leibniz except as a disciple of Descartes, at best taking rationalism to greater extremes. Leibniz's view that all our ideas are innate has been read as such a development. So has his espousal of the inesse principle. 1 0 I have already argued that i t is a mistake t o construe Leibniz's epistemology or metaphysical methodology as even broadly the same as that of Descartes. When Leibniz refers to his 'principles' we are not to suppose that these must be known intuitively. When, moreover, he writes of 'demonstration' he should normally be taken to mean 'rigorous deduc­ tive argument' and not what he himself refers to as 'full demonstration'. ! ! In certain cases, such as arithmetic - where reduction to identities is possible - Leibniz did indeed believe that full demonstration (based on intuitively known premises) was possible. But there are no metaphysical arguments for which Leibniz was prepared to make such a claim. Even the ontological argument yielded no more than a moral certainty since it was required to assume that the concept of God was free of contradiction something we are entitled to presume , according to Leibniz, but not some­ thing of which we can claim intuitive certainty. ! 2 These, however, have not been the only points at which Leibniz's philo­ sophy has been assimilated to Descartes'. His view of innateness has been similarly affected, as in the summary quoted from Scruton above : "Leibniz believed that the understanding contains within itself certain innate prin­ ciples, which it knows intuitively to be true . . . ". (Scruton, 1 982, p. 14) Leibniz's views, however, unlike those of Descartes, cannot correctly be stated by an equation of innate principles with those known intuitively to be true. Only the principle of contradiction is admitted by Leibniz as an 'immediate truth' of reason. (The Cogito, for Leibniz , is an immediate truth of experience.) He is happy to acknowledge that the principle is, therefore, innate. But what Leibniz classes as 'innate principles' are propositions of varying epistemological status. In this respect he is quite different from Descartes. If 'rationalism' be taken to be what Scruton says it is, involving commit­ ment to innate principles known by intuition,' then Leibniz largely sided with Locke in opposing it. He could declare himself entirely on the side of Locke in wishing "to fight the laziness and shallowness of thought of those who use the specious pretext of innate ideas and truths, naturally engraved

LEIBNIZ'S BREAK WITH CARTESIAN 'RATIONALISM'

205

on the mind and readily assented to, to avoid serious inquiry into where our items of knowledge come from, how they are connected, and what certainty they have". (Remnant and Bennett, 1 98 1 , p. 75) The reason why Leibniz is committed to saying that all our ideas are innate is not that he is an extreme rationalist but because he held that "nothing naturally enters our minds from without". (Discourse, Section 36) His doctrine of innate ideas is, in short, a special case of his theory that all substances are naturally independent of one another. Everything that happens to a sub­ stance "must spring from within itself, by a perfect spontaneity with regard to itself, and yet in a perfect conformity with things outside". (Gerhardt, 1 875-1 890, iv, p. 484) Leibniz was well aware that there were those who wanted to restrict the word 'innate' to truths "which are accepted straight away, by instinct". (Remnant and Bennett, 1 9 8 1 , p. 90) He was quite willing to accept this usage for the sake of discussion. But the only such truths which "share in the natural light" are identities. These, for Leibniz , are the only immediate truths of reason. Our knowledge of the 'indemonstrable principles' of morality, though innate and instinctive, is by contrast "based on inner experience - confused knowledge". Though morality is indeed, for Leibniz, "a demonstrative science" it is not based on intuitive knowledge, since "its principles are not known in a luminous way". (Remnant and Bennett, 1 98 1 , p. 88ff ) .

3.

My primary concern in this paper has been to argue that Leibniz ceased to be a 'rationalist' in the sense of one who believed that the right method in metaphysics was to begin with intuitively known axioms in the hope of deriving most of the truths that mattered by rigorous deduction from themP I have argued that Leibniz sought, on the contrary, to j ustify adopting principles which are not intuitively known as a means of advanc­ ing a constructive metaphysics and that his own system of the Discourse and later writings was put forward on this new basis. Much of the mis­ understanding of Leibniz has been generated, as I have tried to bring out, by a number of confusions between Leibniz's epistemological and methodological doctrines and those of Descartes. There may be many explanations for these confusions. But their persistence in spite of the trouble Leibniz took to distinguish his philosophy from that of Descartes is remarkable.

206

STUART B ROWN

To suggest that the role of extreme rationalist has been one required of Leibniz by integral histories of Modern Philosophy from Kant onwards is to imply that the confusion with Descartes is motivated. But, if Leibniz is a victim of integral history of philosophy, it is a case of the biter bit. For he himself regarded past philosophy as a series of errors brought about by one-sidedness. It requires his own philosophy to bring out what truth there is in the various philosophical sects : The lack of substantial reality in the sensible things of the sceptics; the reduction of everything to harmonies or numbers, ideas, and perceptions by the Pythagoreans and Platonists; the one and the whole of Parmenides and Plotinus, yet without any Spinozism; the Stoic connected­ ness, which is yet compatible with the spontaneity held to by others; the vitalism of the Cabalists and hermetic philosophers who put a kind of feeling into everything; the forms and entelechies of Aristotle and the Scholastics; and even the mechanical explanation of all particular phenomena by Democritus and the moderns; etc. - all of these are found united as if in a single perspective centre from which the object, which is obscured when considered from any other approach, reveals its regularity and the correspondence of its parts. (Loemker, 1969, p. 496; Gerhardt, 1 875-1 890, iv, p. 523ff.)

The hope of finding a 'single perspective centre' in one's own philosophy from which the main lines of past and present philosophy can be seen as falling into place may be an illusory one. B ut, if so, Leibniz was no freer of it than Kant. If it is not an illusory hope then at all events the integral history of Modern Philosophy, so far as Leibniz is concerned , needs some re-writing.

Open University

NOTES • This paper has been revised in the light of comments made on its presentation at the conference on 'Philosophy and Its History'. I am grateful to Mr. George MacDonald Ross for his comments as chairman of that discussion. A particular debt is due to Dr R. Niall D. Martin with whom I have had several discussions of the topics covered in this paper .and with whom I have been engaged in collaborative research on the Leibniz-Foucher correspondence. I See, for instance, Loemker, 1 969, p. 293. (Gerhardt, 1 875-1 890, iv, p. 425) 2 J. S. Mill, for instance, attributed to what he called "the intuitive school, from Descartes to Dr Whewell" the view that it is an infallible test of the truth of a proposition that its falsehood is "inconceivable". (Mill, 1 849, p. 1 73) 3 His De primae philosophiae emendatione paper was published in the Acta eruditorum, 1694. Leibniz sent a French version of this paper to Bossuet around the same time. The French title 'Reflexions sur l'avancement de la metaphysique reelle . . . ' suggests that 'emendatione' is

LEIBNIZ'S BREAK WITH CARTESIAN 'RATIONALISM'

207

better translated as 'reform' or 'advancement' than 'correction' or 'emendation'. See R. A. Watson, 1 966. Leibniz's assessment of Descartes seems to have been considerably influenced by Foucher. After conceding that Descartes had contributed "some admirable things", he goes on, in his 1 694 paper on the reform of metaphysics: "Above all, he rightly restored the study of Plato by leading the mind away from the senses and there upon also added to it the doubts of the Academy". (Loemker, 1 969, p. 432; Gerhardt, 1 875-1 890, iv, p. 468) 5 See Remnant and Bennett, 1 98 1 , p. 36Iff.. 6 I have argued that this is indeed so in Brown, 1984, Chapter 19. 7 See, for instance, his 'Specimen of Discoveries of General Nature', 1 686, Parkinson, 1 973, pp. 75-86 (Gerhardt, 1875-1890, vii, pp. 309-3 1 7). In putting forward his 'hypothesis of concomitance' Leibniz writes: "Even if this did not have an a priori proof, it would maintain the position of a most plausible hypothesis". (Parkinson, 1973, p. 80; Gerhardt, 1 875- 1 890, vii, p. 3 1 3) Replying to Foucher in a similar vein in 1 696, Leibniz writes: "It commonly suffices for a hypothesis to be proved a posteriori, because it satisfies the phenomena; but when we have other reasons as well, and those a priori, it is so much the better". (Parkinson, 1973, p. 1 28; Couturat, 1 903, p. 496) 8 Urban and Levesque, 1 909- 1 923, Volume 6, p. 527ff.. Leibniz's own French version of his De primae phi/osophiae emendalione paper differs in details from the Latin version published in the Acta erudilorum in 1 694. 9 See above, p. 3ff.. I O Couturat construes the inesse principle as entailing that "All truth is analytic". He goes on : "As a consequence, everything in the world must be intelligible and logically demonstrable in terms of pure concepts, and deduction is the only scientific method . . . Thus the philosophy of Leibniz appears as the most complete and systematic expression possible of intellectual rationalism". (Couturat, 1 90 1 , p. xi) The inesse principle, however, is a model for God's knowledge of the world and it does not follow from it that deduction from 'pure concepts' is even a possible scientific method for humans. Leibniz thought, at least for a while, that many conclusions could be drawn from the 'supposition' that this model of God's knowledge was correct. But this programme, though it assumes a kind of transcendental rationalism, does not itself involve a rationalist methodology. I I For instance, in a paper of around 1 680, Leibniz refers to the basis on which Archimedes "rigorously completes his demonstrations" (Wiener, 1 95 1 , p. 36; Gerhardt, 1 875-1890, vii, p. 1 66) as that of establishing "many things on a few hypotheses", a point made again in his 'grande lettre' to Foucher of 1686. It was a virtue of this procedure that "at least we should know that there remain only these few hypotheses to be proved in order to arrive at a full demonstration . . . " (italics added). 12 See Remnant and Bennett, 1 9 8 1 , p. 438. 4

REFERENCES Brown, S. c.: 1 984, Leibniz, Harvester Press (Philosophers in Context Series), Brighton. Couturat, L. : 1 90 1 , La /ogique de Leibniz, Felix Alcan, Paris. Couturat, L. (editor): 1903, Opuscu/es et fragments inedits de Leibniz, Felix Alcan, Paris. Erdmann, J. E. (editor): 1 840, G. G. Leibnitii opera omnia philosophiae quae extant, 2 volumes, Berlin.

208

STUART B ROWN

Gerhardt, C. 1. (editor): 1 875-90, Die Philosophischen Schrijien von G. W. Leibniz, 7 volumes, Weidmann, Berlin. Loemker, L. E. (translator & editor): 1969, Gottfried Wilhelm Leibniz : Philosophical Papers and Ll!tters, 2nd Edition, D. Reidel, Dordrecht. Mill, 1. S.: 1 849, A System of Logic, Longmans, London. Morris, C. R : 193 1 , Locke, Berkeley, Hume, Oxford University Press, Oxford. Parkinson, G. H. R. (translator & editor): 1 973, Leibniz : Philosophical Writings, 1. M. Dent & Sons, London. Randall, I. H . : 1 940, The Making of the Modern Mind, Houghton Mifflin Co., Cambridge, Mass .. Remnant, P. and Bennett, 1. (translators & editors) : 1 9 8 1 , G. W. Leibniz: New Essays on Human Understanding, Cambridge University Press, Cambridge. Scruton, R: 1 982, Kant, Oxford University Press (Past Masters Series), Oxford. Urban, Ch. and Levesque, E. (editors): 1 909- 1 923, Correspondence de Bossuet, Paris. Watson, R A. : 1 966, The Downfall of Cartesianism ( 1 673- 1 7 1 2), NijhofT, The Hague. Wiener, P. P. (editor): 195 1 , Leibniz: Selections, Charles Scribner's Sons, New York.

EDWIN Mcd\\NN LOCKEAN M E C H AN I S M

Locke subscribed to the Mechanical Philosophy, i n Gassendi's and Boyle's version of it. On this view, all of the powers and qualities of bodies, and all the changes in these powers and qualities which result from the actions of these bodies one upon the other, issue entirely from the "two grand principles of bodies, matter and motion" (Boyle ). 1 The main points of the view, more particularly, were these: (a) all bodies are made up of matter, and only of matter; (b) the essence of matter consists in the qualities of extension and solidity; (c) bodies large enough to be perceived are compounded out of physically indivisible bits of matter too small to be perceived (the so-called minima naturalia) and have no other constituents (in particular, no immaterial constituents); (d) in consequence of being extended and finite, each body has a determinate bulk or size and figure; and finally, (e) any change in the qualities of a body is the result of the alteration of the bulk, figure, relative situation and/or motion of the solid parts of the body, the latter alteration being due to the action upon that body, perhaps through a material medium, of the mechanical affections of the solid parts of some other body or bodies. 2 A very strong case can be made for the ascription of these doctrines to Locke; I will not go into this here as Locke's commitment to mechanism is so widely acknowledged. Nevertheless, I will quote at the outset a sentence (the last sentence, in fact) from Locke's late, posthumously published, Elements ofNatural Philo­ sophy; it is a nice succinct statement of the view: By the figure, bulk, texture, and motion of thes e small and insensible corpuscles, all the phenomena of bodies may be explained. 3

This is the view I will call 'mechanism': Margaret Wilson has recently argued that Locke is not a consistent mechanist; she maintains that other doctrines of his conflict with his Boylean 'official position', on which the qualities and powers of a body 'flow from' the real essence of that body.4 She instances his view that matter might have the power of thought superadded directly to it by God; his insistence that we cannot conceive there to be any connection between 209 A. J. Holland (ed.), Philosophy. Its History and Historiography. 209-231 . (c) 1 Q R ,\ 1m n

Ilbirlol Puh!;"hino rnmnnwl1J

210

E D W I N M cCANN

the primary qualities of (the constituent solid parts of) bodies and the sensations they cause in us, and hence between those primary qualities and the sensible secondary qualities of bodies ; his pessimism about the prospects for an adequate account of the cohesion of bodies; and his concession that gravity cannot be understood as a mechanical pheno­ menon, that is, that gravitational attraction does not seem to involve the transfer of motion through impulse. In this paper I will argue that Locke is in fact a consistent mechanist. He does hold all of the doctrines Wilson mentions, but these doctrines are not inconsistent with mechanism of the sort set out in the quote above, when this is properly understood. I.

I want first of all to identify more precisely the obstacle to a straight­ forward reading of Locke as a mechanist. This is important, as there are a number of issues here that can get tangled up together. We should note first that the lacunae in the mechanist account of the world to which Wilson draws our attention are of quite different sorts. Locke's worry about cohesion, for example, is simply that there are objec­ tions to all the extant proposals for a mechanical resolution of the phenomenon ; he does not suggest that cohesion is inherently non­ mechanical, or that it cannot be explained mechanistically, but only that we do not now know how to do so. In this case, as well as in the more problematic case of gravity, Locke does not cite any particular reasons for thinking the phenomenon to be insusceptible of mechanical explanation.5 In the case of the possibility that certain systems of matter may have the power of thought, the situation is different. Thought and matter as we conceive them are of such different natures that we cannot see how they could possibly be connected; we might put this by saying that thought, sensation, the power of moving the parts of one's body by willing, etc. seem to us to be inherently non-mechanical phenomena. This deep explanatory gap extends to the connections between the primary qualities of bodies and their secondary ones, in view of the fact that these connections depend on the connections between the primary qualities of bodies and the sensations these bodies cause in us. This last problem is especially troubling since, as Locke notes in the chapter on our ideas of substances, most of the powers and qualities we recognize in bodies are secondary qualities, or else powers defined in relation to the sensations they produce in us.6

LOCKEAN MECHANISM

21 1

Although the recalcitr ant phenomena resist mechanistic explanations for different reasons, Locke puts them in one basket in the following passage from IV.iii.29: But the coherence and continuity of the parts of Matter; the production of Sensation in us of Colours and Sounds, etc. by impulse and motion; nay, the original Rules and Communica­ tion of Motion being such wherein we can discover no natural connexion with any Ideas we have, we cannot but ascribe them to the arbitrary Will and good Pleasure of the Wise Architect. 7

The problem, at bottom, for those who wish to read Locke as a consistent mechanist lies in the ascription of the recalcitrant phenomena to God's arbitrary will and good pleasure. Bodies would have the powers and qualities in question not simply in virtue of their mechanical constitutions (what Locke calls their real essences), but instead they would have these powers and qualities in virtue of God's arbitrary action. Thus, Margaret Wilson, discussing the problem of primary/secondary quality connections, writes : . . . at first thought it might seem that Locke could consistently hold that a body's powers to produce ideas flow naturally from its real essence, while also maintaining that the ideas themselves are arbitrarily annexed to whatever motions of matter habitually cause them. But of course this is not really the case. For it follows from Locke's account that a body has its powers to produce ideas only because o/the divine acts of annexation. Therefore, . . . we find conflict with the official position that there is in reality an a priori conceptual connection between a body's real essence and its secondary qualities. 8

We have to be careful how we take Wilson's claim. She might seem to be attributing a sort of occasionalism to Locke, in light of the emphasis on God's action and the consequent arbitrariness of the annexation.9 Wilson does not suggest, however, that Locke thinks that bodies or their powers and qualities have no causal efficacy; indeed, she suggests that Locke's view is that these powers and qualities do produce the effects in question, but are able to do so only by virtue of God's having ordained the requisite general laws connecting the primary qualities of bodies with these effects. But this means that there are no suitable explanatory connections between the mechanical affections of a body and its secondary qualities. The connections that are lacking, on this reading, are connections in rerum natura, and not merely connections perceived or apprehended by us. It would be a short way with the problem to read the passages we have

212

EDWIN M cCANN

been discussing as having only an epistemological import. Then, Locke is saying only that we do not know the explanatory connections that are in fact there, connections apprehended by God and the angels. Ayers takes this approach, drawing attention for example to the fact that in the passage I have quoted from IV.iii.29 Locke talks of the ideas we have, wherein we can discover no natural connexions, so that we cannot but ascribe them to the arbitrary determination of God's will. There are two reasons why we should resist the suggestion that Locke's point is merely an epistemological one. In the first place, on the face of it the appeal to God's will is more than an epistemological place-holder. If it were only this, it would have been much more appropriate for us to simply admit that while we suspect there are connections here, we do not know what they are or even what they are like ; why appeal to God's will, if all we mean to mark is our ignorance of the connections? To ascribe the connections to God's arbitrary will and good pleasure must be to issue an hypothesis , however tentative and unsupported it may be, concerning the causal ancestry of the connections. Ayers himself notes that in some cases Locke's appeal to God's will is meant to carry ontological weight; lO I do not see any clear basis for reading IV.iii.29 and kindred passages in a different way. Second, and most important, Locke would seem to have no good reason to suppose that there are mechanistically intelligible connections here, of which we are simply ignorant. He does go on to say, in IV.iii.29, that The Things that, as far as our Observation reaches, we constantly find to proceed regularly, we may conclude, do act by a Law set them; but yet by a Law, that we know not; whereby, though Causes work steadily, and Effects constantly flow from them, yet their Connexions and Dependancies being not discoverable in our Ideas, we can have but an experimental Knowledge of them.

We conclude that there are lawlike connections of some sort because of the regularities we observe in the powers and qualities of bodies. There is no license given here to suppose that these connections must be mecha­ nistically explicable, however, and Locke's talk of a law set them certainly has in it the suggestion of divine action. It would sort better with Locke's general agnosticism about the ultimate explanation of the qualities and operations of bodies to see him as issuing the least specific hypothesis available as to the source of the connections , rather than dogmatically insisting that they derive from matter and motion in some as yet unknown manner. I I

LOCKEAN MECHANISM

213

The problem for a mechanistic reading o f Locke i s therefore this : because of our inability to conceive a mechanistic explanation of such phenomena as the cohesion of bodies, their mutual gravitational attrac­ tion, their power to cause sensations in perceivers in a regular manner and thus their possession of secondary qualities and other powers defined in reference to sensation, we are forced to ascribe these phenomena to God's action as determined by his arbitrary will. Since this ascription is not merely an epistemological place-holder but is instead an hypothesis about the causal ancestry of the phenomena, it is inconsistent with the mecha­ nist's claim that all the phenomena of bodies can be explained in terms of the bulk, figure, texture and motion of their solid parts. 2.

A t this point I want t o consider what might b e involved i n God's super­ adding powers or qualities to bodies, for by doing so we shall see how we can square Locke's commitment to mechanism with his concession that God's action is required if bodies are to have the powers and qualities in question. When he first comes to speak, in IV.iii.6, of the possibility that God might superadd the power of thought directly to matter, thus making matter capable of thinking, he puts this possibility in the following way: We have the Ideas of Matter and Thinking, but possibly shall never be able to know, whether any mere material Being thinks, or no; it being impossible for us, by the contemplation of our own Ideas, without revelation, to discover, whether Omnipotency has not given to some Systems of Matter fitly disposed, a power to perceive and think, or else joined and fixed to Matter so disposed, a thinking immaterial Substance . . .

His reason for thinking this i s given as follows : For I see no contradiction in it, that the first eternal thinking Being should, ifhe pleased, give to certain Systems of created senseless matter, put together as he thinks fit, some degrees of sense, perception, and thought.

Looked at in the context of our present problem, certain phrases in these passages which might easily be overlooked or treated as throwaways take on importance. It is not simply matter, but certain systems of matter fitly disposed, or put together as God thinks fit, to which God is supposed to have superadded the power of thought. These phrases recur often in Locke's extended defense of his claim that God may endow matter with

214

E D W I N M cCANN

the power of thought against the objections of Stillingfleet, talking there of God's superaddition of thought to matter "ordered as he sees fit", or "so disposed as he thinks fit", and the likeP Perhaps the most interesting passage along these lines is this one, in which Locke is answering Stillingfleet's charge that he will be unable to explain why we are capable of abstract thought, whereas brutes are not, if we and the brutes are, equally, merely material things : . . . if Omnipotency can give Thought to any solid Substance, it is not hard to conceive, that God may give that Faculty in an higher or lower Degree, as it pleases him, who knows what Disposition of the SUbject is suited to such a particular way or degree of ThinkingY

These passages suggest a picture of superaddition on which God super­ adds a power or quality to body by somehow connecting that power or quality with a certain type or types of material constitution, i.e. with a certain disposition of parts. There is an alternative interpretation of these passages, on which the connections between the primary-quality constitutions of bodies and the powers or qualities that are to be superadded are somehow already there, laid out in the nature of things. 14 God would then be seen as superadding powers or qualities to particular bodies by contriving them so that they satisfy the structural descriptions implicit in the antecedently existing connections. This interpretation, it seems to me, involves the attribution to Locke of an unreasonable commitment to the correctness and adequacy of the corpuscularian hypothesis; it sorts ill with those passages in which Locke agnostically stresses the tentativeness of his espousal of the hypo­ thesisY It also fails to fit with those central passages in which Locke insists that even if we knew the real essences of bodies we would be unable to tell what their consequent powers and qualities might be; for this we need to know the connections between primary and secondary qualities as well . 1 6 It is hard to see what this "other and more incurable part of our ignorance", over and above our ignorance of the real essences, would consist in, if the relevant connections were a matter of mechanical necessity laid out in the nature of things. For given access to the real essences of the bodies involved, what would stand in the way of our simply working out the powers and qualities that flow from the real essence, given that the connections themselves are a matter of mechanical necessity? We do better to interpret these passages as implying that God actually forges the connections between types of material constitution and the

LOCKEAN M ECHAN I S M

215

superadded powers and qualities. Taking this route enables us to avoid the difficulties of the alternative interpretation, and further it chimes in well with those passages in which Locke stresses the arbitrariness of the con­ nections and their dependence on God's will. We have already looked at IV.iii.29, in which Locke talks of bodies acting according to a law set them; in the section j ust preceding that one, Locke has set up this claim by arguing that ordinary experience convinces us that thoughts (acts of will) can produce motion in bodies and that bodies can produce thoughts (sensations) in the mind, although we cannot conceive how this should be so: These, and the like, though they have a constant and regular connexion, in the ordinary course of Things : yet that connexion being not discoverable in the Ideas themselves, which appearing to have no necessary dependance one on another, we can attribute their connexion to nothing else, but the arbitrary Determination of that All-wise Agent, who has made them to be, and to operate as they do, in a way wholly above our weak Understandings to conceive.

In IV.iii.6, the passage on superaddition, Locke says something similar, and he also suggests that God superadds the power to produce sensations of a certain sort by annexing the sensations to motion of a certain kind : . . . Body as far as we can conceive being able only to strike and affect body: and Motion, according to the utmost reach of our Ideas, being able to produce nothing but Motion, so that when we allow it to produce pleasure and pain, or the Idea of a Colour, or Sound, we are fain to quit our Reason, go beyond our Ideas, and attribute it wholly to the good Pleasure of our Maker. For since we must allow he has annexed Effects to Motion, which we can no way conceive Motion able to produce, what reason have we to conclude, that he could not order them as well to be produced in a Subject we cannot conceive capable of them [i.e. "in some Bodies themselves, after a certain manner modified and moved"] , as well as in a Subject we cannot conceive the Motion of Matter can any way operate upon?

And finally, in IV.vi . l 4 he says that to establish truths about substances that are universal and certain (and non-trifling), we would have to know . . . what Changes the primary Qualities of one Body, do regularly produce in the primary Qualities of another, and how. Secondly, we must know what primary Qualities of any Body, produce certain Sensations or Ideas in us. This is in truth, no less than to know all the Effects of Matter, under its divers modifications of Bulk, Figure, Cohesion of Parts, Motion, and Rest.

And this, he goes on to say, we can know only by revelation.

216

EDWIN M cCANN

All this points to the following as Locke's position : God superadds a power or quality to body by ordaining that a law holds connecting a certain type or types of material constitution (i.e. a certain arrangement of the mechanical affections or primary qualities of the constituent parts of bodies) with the power or quality. The law is arbitrary in that it is only one of a number of possible but mutually exclusive connections that might hold between types of constitution and resultant qualities, and that it is the one that does obtain is due only to the undetermined action of God. Never­ theless, the connection thus forged between the primary qualities of the body and the secondary qualities and other of its powers is a necessary connection, in Locke's sense, since God is thought of here as decreeing a law connecting the qualities. This will guarantee that every time a body has the appropriate inner constitution it will have the associated powers and qualities, and it will make true many counterfactuals of the form, 'if a body were to have such-and-such a constitution it would have such-and-such a quality, and would do so-and-so in such-and-such circumstances'. In keeping with Locke's injunction not to limit God's omnipotence by our narrow conceptions,17 and his insistence on our ignorance with regard to these matters, we should take this picture of superaddition to be nothing more than our best conj ecture (or rather, the only even provisionally satisfying one we can construct) as to how bodies might have powers and qualities which we cannot see how to connect with their mechanical affections. Even so, the position we have arrived at for Locke is, I will argue in the next section of the paper, compatible with mechanism and can even serve to bolster it . . . Before I leave the topic of superaddition I should comment on a passage from Locke's third letter to Stillingfleet which has come into prominence in the recent discussions of mechanism and superaddition. It runs as follows :

The Idea of Matter is an extended solid Substance; wherever there is such a Substance, there is Matter, and the Essence of Matter, whatever other Qualities not contained in that Essence, it shall please God to superadd to it. For example, God creates an extended solid Substance, without the superadding anything else to it, and so we may consider it at rest: To some parts of it he superadds motion but it still has the Essence of Matter: Other parts of it he frames into Plants, with all the excellencies of Vegetation, Life, and Beauty, which is to be found in a Rose or a Peachtree, &c. above the Essence of Matter in general, but it is still but Matter: To other parts he adds Sense and Spontaneous Motion, and those other Properties that are to be found in an Elephant. Hitherto it is not doubted but the Power of God may go, and that the Properties of a Rose, a Peach, or an Elephant, superadded to M atter, change not

LOCKEAN M ECHAN I S M

217

the Properties of Matter; but Matter i s i n these things Matter still. But i f one venture t o go one step further and say, God may give to Matter, Thought, Reason, and Volition, as well as Sense and Spontaneous Motion, there are Men ready presently to limit the Power of the Omnipotent Creator, and tell us, he cannot do it; because it destroys the Essence, changes the essential Properties of Matter. To make good which Assertion they have no more to say, but that Thought and Reason are not included in the Essence of Matter. I grant it; but whatever Excellency, not contained in its Essence, be superadded to Matter, it does not destroy the Essence of Matter, if it leaves it an extended solid Substance . . . 18

This passage has become the focus of a controversy between Ayers and Wilson over the propriety of the phrase "superadded property". The dispute has , it seems to me, been resolved;19 in fact, I doubt there was ever any deep disagreement on this point. The real issue between them lies elsewhere. It concerns the sense in which superadded qualities are not natural to the bodies that have them. Ayers and Wilson agree that in this passage Locke distinguishes between the properties a body has by virtue of being an extended solid thing and those it has by virtue of being a body of this or that particular sort (peach-tree, elephant, etc.), these latter qualities having to be superadded to the body; they can thus be said to fall outside the 'natural powers' of matter. Ayers takes the passage to show how limited this claim is; in saying that the superadded powers or qualities are not natural to matter Locke is recording only that they are not contained in, or entailed by, the essence or nature of matter, viz. extension and solidity. 2o This would leave it open to Locke to say that the qualities do flow naturally from the body's mechanical constitution, in the sense that it would be a matter of mechanical necessity that a body constituted in a certain way would have these qualities. Wilson, on the other hand, takes the distinction between the properties of matter or body as such and the properties a body has as a member of a particular species of ' body to be inconsistent with a "thorough mechanism about real essences", presumably since she thinks that the "hidden constitution" from which the latter qualities flow is not materia1. 2 I On her view, then, it would seem that the superadded powers and qualities do not flow naturally even from the determinate real essence of the body, if this is taken to consist only in the bulk, figure, texture, and motion of the body's solid parts, which are only modifications of matter. Our previous discussion indicates that both of these construals of the passage are slightly off the mark. In this passage Locke is trying to flatten out the notion of superaddition, so that God's act of superadding the power of thought to m atter is seen to be of a piece with the superaddition

218

EDWIN M c CANN

of motion to matter or that of any of the qualities going into the nominal essence of a species of bodies. To this end he focuses on the strong sense in which a quality might be said not to flow naturally from a body, viz. that it does not follow from the attributes of extension and solidity. But this cannot be the only sense in which qualities such as the power of thought or the power to produce certain sensations are not natural to body. We have already encountered passages in the Essay in which Locke denies that there are any conceivable connections between the mechanical affections of a body and its secondary qualities, so that we must attribute the existing connections to God's action.22 This affords a perfectly good sense in which these powers or qualities are not natural to matter: they do not flow from the body's inner nature, or real essence, alone; left to itself, with no action on God's part, the body would not have those secondary qualities or other of the powers that it does in fact have. This, we have seen, is what we commit ourselves to when we put the connections down to God's arbitrary will. It is not just a matter of what is or is not entailed by extension and solidity, but also of what can and what cannot be supposed to flow with mechanical necessity from the ,determinate real essence or mechanical affections of a body, taken by itself. On the other hand, there is no basis in the passage for taking it as Locke's view that in superadding powers or qualities to bodies God is adding any immaterial or non-mechanical constituent to them, as Wilson seems to do. Indeed, the context of the passage points to quite the opposite view, for Locke goes on to argue that one cannot come up with a reason why a merely material thing cannot have the power of thought if one grants that merely material creatures can have sense and spontaneous motion, as Stillingfleet, no Cartesian on this score, does, while still maintaining that brutes have no immaterial souls, as Stillingfleet, quite orthodox on this score, also does. On the understanding of superaddition that we have come to, in superadding powers and qualities to bodies God does not add anything to them at all, at least in terms of real constituents. By instituting the relevant laws, God gives the mechanical affections of a body - its determinate real essence - dispositions to cause the appropriate effects. Nothing has changed in the body itself, nor have any new constituents of any kind been added to it; the body, or more particularly its mechanical affections, is simply able to produce more, and different sorts of, effects than it was able to do before.23 Thus, if there is a sense in which super­ added powers or qualities are not natural to the bodies endowed with them, there is another sense in which they are natural to them: in the

LOCKEAN MECHANISM

219

course of nature, a s God has created i t (in part b y ordaining the appropriate connections), the powers or qualities flow naturally from, in the sense of being necessarily connected with, the determinate real essence or mechanical constitution of the body. We see, then, that the passage we have been considering poses no problem for the interpretation I have given, and indeed accords quite well with it. 3.

We have arrived at a consistent view for Locke, but is it mechanism? The outstanding reason for doubting that it is stems from the fact that God's action is required for the appropriate connections to hold, and thus cannot be eliminated from any complete explanation of 'all the phenomena of bodies'. This problem is highlighted in both of Wilson's papers ; we have encountered it in the quotation given above (p. 2 1 1), and in her later paper she says that Locke's doctrines about superadded qualities entail that " . . . mechanistic principles, or primary qualities, have limited explanatory power in Locke's considered view: the purposive action of an eternal thinking being is also required to account for phenomena".24 Although Wilson is right in thinking that Locke denies that we can explain how a body has the secondary qualities and other of the powers it in fact has without adverting to God's action, this is no bar to his being a thoroughgoing mechanist. First of all, we should note that God's actions are required only to set the general background for any particular causal interactions ' among bodies, and thus for a particular body's having a certain set of causal powers; he does not directly work any of the effects which proceed from the superadded powers or qualities, nor need he superadd the qualities to particular bodies on a case-by-case basis. If we have to appeal to the actions of God only in such a general way, we needn't see these actions as interventions in the natural order; far from interfering with or supple­ menting the natural workings of matter-in-motion, these actions establish (in part) what are these natural workings. 2 5 At this point we need to remind ourselves of the going alternatives to the mechanical philosophy, as these presented themselves to Locke (and Boyle).26 First and foremost, and most familiar to us, there was the Aristotelian or Scholastic view (in a number of variants). Roughly speak­ ing, on this view natural phenomena are explained by appealing to substantial forms and real qualities as the causal agents in natural change;

220

EDWIN McCANN

these causal agents, if not explicitly immaterial, always operate in the light of final causality, hence non-mechanically. Less well known to us, but of great concern to Boyle, was what he called spagyritic chemistry and what we know as iatrochemistry, the 'philosophers by fire', among whom the most prominent were Paracelsus and, closer to Locke's and Boyle's time, J. B . van Helmont. Their three basic principles (salt, sulfur, and mercury) were held to be essentially active forces, and hence living or vital forces. Finally, and perhaps of more concern to Locke than to Boyle, were the Cambridge Platonists. Henry More and Ralph Cudworth both argued for the need to recognize immaterial, animate causal agents - 'plastic natures' or 'plastic principles' - which determine all natural change, matter being essentially passive. Cudworth even argued for an anima mundi to accom­ plish the aims with which God created the world. 27 Although I have given only very sketchy descriptions of these competing views, it can be seen that they are quite different. They have one thing in common, however, which sets them off from the mechanical philosophy: they all assert the existence in bodies of non-mechanical (and usually immaterial) causal agents which are ultimately responsible for some if not all of a body's natural operations. Seen in this light, it is clear that the distinctive feature of mechanism is its refusal to postulate any causal agents in bodies except for the bulk, figure , texture, and motion of their solid parts. Locke's position, as we have come to understand it, is thoroughly mechanistic. God ordains certain general laws and gives matter its initial disposition and motion; against this general background or 'frame' of the world the mechanical affections of bodies come to have certain causal powers which they otherwise would not have had. But it is the mechanical affections which have these powers ; it is the mechanical affections of body which are the only causally efficacious agents in any natural change. Even if Locke is a mechanist in this sense, it might still be thought that his views on scientific explanation are not consistent with the views we have been discussing. In the characterization of the problem given by Wilson (quoted above), she gives as Locke's (and presumably Boyle's) 'official position' that there is an a priori conceptual connection between the real essence of a body and its secondary qualities. She apparently bases this attribution on the strict analogy she finds in Locke between explan­ ation in natural philosophy and geometrical demonstration. She says in her original paper that "Many passages show that Locke conceives the relation of real essence to derivative properties as analogous to that between the

LOCKEAN MECHANISM

22 1

definition of a geometrical figure and the properties deducible from the definition",28 and in her later paper she glosses the problem she raises for a consistent mechanism in Locke in this way: My claim, however, was just that Locke thought these qualities [secondary qualities, and superadded qualities generally) cannot 'arise naturally' from Boylean primary qualities, in the sense that the former cannot be 'explained' (through something like geometrical demonstra­ tion) in terms of the latter.29

Although there is no doubt that Locke did compare explanations in natural philosophy with geometrical demonstrations, it is important to see that the similarity between them to which Locke points is limited to this, that in each we have (or would have, if we could achieve the relevant explana­ tions) non-experimental knowledge of the properties of the object in question. The knowledge we would have from explanations in natural philosophy, if we were in a position to give them, would thus be a priori; but this does not mean that the connections figuring in such explanations would be conceptual, or j ust like the ones involved in geometrical demon­ strations . Let u s consider the passages which are supposed t o show that Locke thinks of the relation between the real essence of a body and its derivative powers and qualities as analogous to that between the definition of a triangle (which, triangles being modes, is both its real and its nominal essence) and its derivative properties. Wilson cites four such passages, of which two are the ones most commonly cited as support for the attribution to Locke of a 'rationalist' or 'deductivist' conception of natural science. These are from Il.xxxi.6 and IV.iii.25 respectively : The complex Ideas we have of Substances, are, as it has been shewn, certain Collections of simple Ideas, that have been observed or supposed constantly to exist together. But such a complex Idea cannot be the real Essence of any Substance; for then the Properties we discover in that Body, would depend on that complex Idea, and be deducible from it, and their necessary connexion with it be known; as all Properties of a Triangle depend on, and as far as they are discoverable, are deducible from the complex Idea of three Lines, including a Space. I doubt not but if we could discover the Figure, Size, Texture, and Motion of the minute Constituent parts of any two Bodies, we should know without Trial several of their Oper­ ations one upon the other, as we do now the Properties of a Square, or a Triangle.

The first of these passages is concerned not with the relation between the corpuscularian real essence of the body and its qualities, however, but

222

EDWIN McCANN

instead with the relation of the nominal essence to those qualities. Locke is giving a reductio argument against the Scholastic doctrine of substantial forms, which he interprets as holding, in effect, that the real essence of a body is its nominal essence. If this were so, Locke is arguing, then we would have demonstrative knowledge of the properties of substances, just as we have of those of modes; but in fact we have no such knowledge. The reason for this, Locke goes on to say, is precisely that the real essence of a substance is not its nominal essence: " . . . it being nothing but Body, its real Essence, or internal Constitution, on which these Qualities depend, can be nothing but the Figure, Size, and Connexion of its solid Parts". (II.xxxi.6) The other passage, from IV.iii.2S, does not say that if we knew the real essences of bodies we could give demonstrations in natural philosophy strictly analogous to those we give in geometry ; it only says that we could know "without trial" not all, but "several", of the operations of bodies, just as we know without trial or experiment the properties of triangles, in their case on the basis of geometrical demonstration. It is instructive to consider the examples Locke gives of the knowledge we might have if we had ideas of the real essences of substances ; if we knew, he says, "the Mechanical affections of the Particles" of rhubarb, hemlock, opium, and a man, . . . as a Watchmaker does those of a Watch, whereby it performs its Operations, and of a File which by rubbing on them will alter the Figure of any of the Wheels, we should be able to tell before Hand, that Rhubarb will purge, Hemlock kill, and Opium make a Man sleep; as well as a Watch-maker can, that a little piece of Paper, laid on the Balance, will keep the watch from going, till it be removed; or that some small part of it, being rubb'd by a File, the Machin would quite lose its Motion, and the Watch go no more. (IV.iii.25)

What the watchmaker and the geometer have in common is their ability to know, without trial or experiment, what properties the objects of their respective areas of concern have, or what they might be expected to do in certain circumstances. The natural philosopher would have this ability, too, with respect to those operations of bodies that lend themselv�s to the simplest kind of mechanical understanding. We could imagine discovering, for example, that the constituent particles of opium are so shaped as to plug up pores through which normally pass some of our animal spirits, thus reducing our activity and making us drowsy; we could then know in advance of trying it that this particular sample of opium will put Philip Marlowe out of commission, just as Marlowe's locksmith friend could fashion from memory a key which he knows will a fit a certain lock. There

LOCKEAN MECHANIS M

223

is nothing in this of demonstration, or conceptual connections. As regards the other passages Wilson cites, IV.vi. 1 1 just makes the same point as IV.iii.25 : if we had ideas of the real essences of substances, and knew "how those Qualities flowed from hence", we would no more need to make experiments to find out the properties of gold than we do to find out the properties of a triangle. And in III.xi.23, the passage Wilson quotes, Locke says that we cannot conceive "the manner" in which angels know what properties and operations flow from the real essence of a substance; this speaks agaihst the idea that they come to this knowledge by geo­ metrical demonstration, or something very much like it, drawn exclusively from the real essence of the substance. Even more telling, I think, is the fact that Locke explicitly contrasts the connections that figure in geometrical demonstrations with those that figure in natural philosophy, and this in one of the passages that is central to the topic of mechanism and superaddition. At the start ofIV.iii.29 Locke says that in geometry, for example, " . . . there are certain Relations, Habitudes, and Connexions, so visibly included in the Nature of the Ideas themselves, that we cannot conceive them separable from them, by any Power whatsoever . . . Nor can we conceive this Relation, this connexion of these two Ideas, to be possibly mutable, or to depend on any arbitrary Power, which of choice made it thus, or could make it otherwise". The next sentence is the one I quoted on p. 2 1 1 above; the connections that figure in natural philosophy are not like the ones in geometry, but instead depend on the arbitrary power of God: in other words, they are not conceptual connections. That Locke should see a contrast of this kind between the connections involved in geometrical demonstrations and those involved in explanations in natural philosophy is prefigured not only in his general distinction between modes (where the real essence and the nominal essence, or defining idea, are identical) and substances (where they are distinct), but also in the distinction he draws at the beginning of Book I V between knowledge o f relations o f ideas and knowledge o f co-existence or necessary connexions of ideas.3 D A careful consideration of B ook IV, Chapters iii and vi, where Locke discusses the reasons why we cannot obtain knowledge of the necessary connection or co-existence of qualities in substances will make it clear that what is required for such knowledge is (a) knowledge of the real essences or internal constitutions of the sub­ stances, and (b) the connections between these real essences and the powers or qualities of the substances. These latter connections are not conceptual ones, depending as they do on God's arbitrary choice; but they

224

EDWIN McCANN

are such that, if we could come to know them (say through revelation) we would be able to know in advance, and with certainty, what bodies would do in various circumstances (supposing we knew their real essences as well), and this is j ust to have proper scientific knowledge of the powers and qualities of bodies. Even given the high standards Locke set for explanation in natural philosophy, then, there is no inconsistency between his commitment to mechanism and his treatment of secondary qualities, or of superadded qualities in general. I think we have arrived at a picture of Locke as not only a consistent mechanist, but a sophisticated one as well. Or rather, a relatively sophisticated one. For this picture of Locke does not leave him looking so very modern. In this respect I am quite in agreement with Wilson,3 1 who remarks with disapproval the tendency among recent com­ mentators to see Locke as basing his arguments on a scientific theory that is in essentials the correct one. This underestimates both the very large differences between Boylean corpuscularianism and modern day physics and physical chemistry and the nature and extent of the gaps in arguments for Locke's and Boyle's views based on the scientific adequacy or superiority of corpuscularianism. (I think that Locke rarely, if ever, argues from the scientific adequacy of corpuscularianism, even in the case of the distinction between primary and secondary qualities. But that is another story, to be reserved for another occasion.) On the other hand, if this picture of Locke does not make him look very much at home in modern science, it does situate him squarely in the tradition of late seventeenth-century English natural philosophy, and particularly in the tradition of natural religion. Starting with the work of Walter Charleton, and particularly prominent in Boyle, and later, in Newton and his acolytes Samuel Clarke and Richard Bentley (both of whom were Boyle lecturers), there was a concerted effort to rid the mechanical philosophy of its taint of atheism, and even to show that the being and attributes of God could be established by rational arguments drawn from natural philosophy. 32 Central to these arguments was the need to call upon the providence and omnipotence of God in order to under­ stand natural phenomena, and this of course is a feature of Locke's mecha­ nism. Seeing this is crucial to the understanding of Locke's views ; otherwise they will look to be inconsistent, or at least subject to internal tensions. Many contemporary commentators would prefer to do away with, or failing this, overlook Locke's need to appeal to God's providence in

LOCKEAN M ECHANISM

225

squaring his commitment to mechanism with his views about the limited extent of our knowledge. Behind some of this, at any rate, is the view that any appeal to God's providence or omnipotence must be bad philosophy, in that any invocation of a Deus in philosophy must be an invocation of a Deus ex machina. There is nothing wrong with this view, as long as it does not blind us to the fact that, given the constraints on seventeenth-century theology, appeal to God's providence and omnipotence are not a case of 'anything goes'. S uch disputes as those between Leibniz and Clarke show that the resolution of such issues involve recognizably philosophical argu­ ments, and a high level of philosophy at that. One of the things I hope my discussion of Locke has shown is that a proper appreciation of the ways in which God, his attributes, and/or his actions figure in the philosophical positions of some of the philosophers of the seventeenth and eighteenth centuries will not only enable us better to understand their views, but to see them to be (relatively speaking, at least) more philosophically defen­ sible than they otherwise might seem to be. 33

University of Southern California

NOTES I

Robert Boyle, The Origin of Forms and Qualities ( 1 666) (hereafter OFQ) in Works (London, 1 772), Volume III, p. 1 6 ; Selected Philosophical Papers ofRobert Boyle, edited by M. A. Stewart (Manchester: Manchester University Press, 1979), p. 20. All references to Boyle's writings will be to these editions, abbreviated to W and S respectively. All references to Locke's An Essay concerning Human Understanding ( 1 st edition, 1690) will be by Book, chapter, and section number to the edition by P. H. Nidditch, published at Oxford by the Clarendon Press in 1975. 2 Boyle gives a nice summary statement of this view in a passage in the Origin of Forms and Qualities where he is stating his aims in that work: "That, then, which I chiefly aim at is to make it probable to you by experiments (which I . think hath not yet been done) that almost all sorts of qualities, most of which have been by the Schools either left unexplicated, or generally referred to I know not what incomprehensible substantial forms, may be produced mechanically - I mean by such corporeal agents as do not appear either to work otherwise than by virtue of the motion, size, figure, and contrivance of their own parts (which attributes I call the mechanical affections of matter, because to them men willingly refer the various operations of mechanical engines); or to produce the new qualities, exhibited by those bodies their action changes, by any other way than by changing the texture, or motion, or some other mechanical affection, of the body wrought upon." ( W Ill, p. 1 3 ; S, p. 1 7). Notice that these characterizations of mechanism are at variance with that given by Michael Ayers in his 'Mechanism, Superaddition, and the Proof of God's Existence in Locke's Essay', Philosophical Review 90 ( 1 98 1 ): 2 1 0-25 1 ; (hereafter, 'M, S, & P'). He says (p. 2 10) that mechanism "was the view that the laws of physics can be explained, in principle if not by us, by being deduced

226

EDWIN McCANN

from the attributes possessed essentially by all bodies qua bodies ; i.e., from the nature or essence of the uniform substance, matter, ofwhich all bodies are composed". In the first place, as Ayers himself recognizes, mechanists make use of motion in their explanations, but motion is not part of the essence or nature of matter. In the second place, neither Gassendi nor Boyle (nor Locke) speaks of deriving or deducing the laws of physics ; instead, for them the goal of mechanism is to resolve particular phenomena in mechanistic terms. 3 Locke, Works (London, 1 823), Volume III, p. 33. Note that in this work explicit notice has been taken of Newton's account of gravity. 4 Margaret Wilson, 'Superadded Properties: The Limits of Mechanism in Locke', American Philosophical Quarterly 16 ( 1979): 143-1 50. I will refer to this paper as 'Limits'. 5 On cohesion, see ILxxiii.23-27. Interestingly, these remarks on the problematic character of cohesion are located in the chapter on solidity in the 1 685 draft of the Essay known as Draft C, which is in the Pierpont Morgan Library in New York. Locke does not mention the problem of gravity in the Essay; the passages cited by Wilson and Ayers are these, from Some Thoughts concerning Education ( 1 693) and from the third of his replies to Stillingfleet ( 1 699): . . . it is evident, that by mere matter and motion, none of the great phaenomena of nature can be resolved: to instance but in that common one of gravity, which I think impossible to be explained by any natural operation of matter, or any other law of motion, but the positive will of a superiour Being so ordering it. (Sec. 192, Works, Volume IX, p. 1 84) (Note by the way Locke's denial here that it is by any "natural operation" of matter that it attracts other matter.) Tis true, I say, "That Bodies operate by impulse, and nothing else". [ILviii. l l , which Locke altered in the fourth edition to accord with the concession in this passage) And so I thought when I writ it and yet can conceive no other way of their operation. But I am since convinced by the Judicious Mr. Newton's incomparable Book, that it is too bold a Presumption to limit God's Power, in this point, by my narrow Conceptions. The gravitation of Matter towards Matter, by ways inconceivable to me, is not only a Demonstration that God can, ifhe pleases, put into Bodies, Powers and ways of Operation, above what can be derived from our Idea of Body, or can be explained by what we know of Matter, but also an unquestionable and every where visible Instance, that he has done so. (Mr. Locke's Reply to the Right Reverend the Lord Bishop of Worcester 's Answer to his Second Letter (London, 1 699) - hereafter Answer to Second Letter - p. 408; Works, Volume IV, pp. 467-468) A couple ofpages earlier on in the Answer to Second Letter (p. 404; Works, pp. 464-465) Locke had pointed out that to explain gravity in terms of self-motion in matter, or as action at a distance, involves us in difficulties, as we cannot conceive matter to be able to move itself, nor can we conceive action at a distance. He nowhere criticizes attempts to explain gravity mechanistically, as for example Boyle's invocation of an aether composed of "peculiar sorts of corpuscles" (see Of the Systematical or Cosmical Qualities of Things ( 1 67 1 ), W Ill, p. 309, and Cosmical Suspicions ( 1 67 1 ), W Ill, p. 3 16), nor could he have taken from Newton the suggestion that gravity is definitely a non-mechanical phenomenon. Newton certainly did not claim in his incomparable book (Principia, 1 st edition) that it is, but only that we are not in a position to decide among various mechanical and non-mechanical hypotheses as to the physical basis of gravitational attraction, i.e. as to what feature or features of bodies makes

LOCKEAN MECHANI SM

227

them able to influence each other's motion without apparent contact between them. So in the Stillingfleet passage he must be speaking about what we are now in a position to derive from our idea of body or what we know of matter, and, in the Education one, about what we are now in a position to explain. Note that the Education passage, with its reference to "the positive will of a superiour being so ordering it" as an explanation of gravity, gives encourage­ ment to the reading of IV.iii.29 that I have proposed. 6 See II.xxiii.8, 9, 1 0, and 37. That it is the difference in nature between thought or sensation and matter that makes for the problem of conceiving primary/secondary quality relations is made plain in a number of passages; I will quote from IV.iii.28, but see also IV.iii.6, 1 2- 1 3 , 29: These mechanical affections of Bodies, having no affinity at all with those Ideas, they produce in us, (there being no conceivable connexion between any impulse of any sort of Body, and any perception of a Colour, or Smell, which we find in our Minds) we can have no distinct knowledge of such Operations beyond our Experience ; and can reason no otherwise about them, than as effects produced by the appointment of an infinitely Wise Agent, which perfectly surpass our Comprehensions. As the Ideas of sensible secondary Qualities, which we have in our Minds, can, by us, be no way deduced from bodily Causes, nor any correspondence or connexion be found between them and those primary Qualities which (Experience shews us) produce them in us ; so on the other side, the Operation of our Minds upon our Bodies is as unconceivable. How any thought should produce a motion in Body is as remote from the nature of our Ideas, as how any Body should produce any Thought in the Mind. 7 Locke's inclusion of "the original Rules and Communication of Motion" among the problematic phenomena is not remarked by Wilson. Locke might be thinking of the difficulty we have of conceiving how motion is transferred by impulse, a difficulty noted at I I.xxiii.28; but the word "original" suggests that the problem he has in mind has instead to do with God's institution of laws of motion and his originally setting some bodies in motion. We are in the dark concerning exactly what he did in this regard, as well as about how he did it. 8 'Limits', p. 147; see also M. Wilson, 'Superadded Properties: A Reply to M. R. Ayers', Philosophical Review 91 ( 1982) - hereafter 'Reply' - p. 2 5 1 . 9 Ayers, ' M , S, & P', p. 2 1 9, takes Wilson t o b e urging, o n the basis of passages such as IV.iii.28, that Locke was inclined to favor M alebranchean occasionalism, an imputation Wilson rejects ('Reply', p. 249). There is a widespread misapprehension about the status of laws of nature according to occasionalism, so although Wilson and Ayers are free of it it will still be worthwhile to point out here that Malebranchean occasionalism does not deny that there are laws of nature governing phenomena. On the contrary, Malebranche insists as much as anyone ever did that there are such laws, understanding them to be the general intentions with which God created the natural order. The key tenet of occasionalism is instead its denial of second causes, God being the only genuine causal agent (with perhaps an exception made for the wills of created spirits). The chief argument for occasionalism rests on the doctrine of continuous re-creation; the motion of a body, for example, consists in its being re-created from moment to moment in a succession of contiguous positions, and the laws of motion governing the body are simply general rules God follows in placing bodies from one moment to the next. (See the seventh of Malebranche's Dialogues on Metaphysics, translated by W.

228

EDWIN McCANN

Doney (New York: Abaris Books, 1 980), especially pp. 1 53-163.) In seeing this as the main argument for occasionalism I am in agreement with T. M. Lennon, 'Philosophical Com­ mentary', in his and Olscamp's translation of Malebranche's The Search after Truth (Colum­ bus: Ohio State University Press, 1 980), pp. 8 1 6 ff., and in disagreement with L. Loeb, From Descartes to Hume (Ithaca: Cornell University Press, 1 98 1 ), pp. 200-205. IO See 'M, S, & P', pp. 225-226. II I have in mind the agnosticism expressed in passages such as IV.iii. 1 1 and 16 and IV.xii. 1 0, as well as the more specific difficulties concerning cohesion, and so on, that are discussed in the chapter on ideas of substances (II.xxiii). 12 See, for example, Locke's Answer to Second Letter, pp. 406, 409, 41 4, 4 1 8 ; Works, Volume IV, pp. 466, 468, 47 1 , 474. 1 3 Locke, Answer to Second Letter, p. 405, Works, Volume IV, p. 465. 14 I take Ayers' view to be something like this. See 'M, S. & P', pp. 2 1 3- 2 1 5, 222-2 3 1 , 244-246. 15 See for example IV.iii. 1 1 and 16, and IV.xii.9-1 3 . 1 6 See for example IV.iii. 1 2 - 1 3 , IV.vi.7. 11 We have encountered hints of this stance in the quotation from IV.iii.28 given above, and it is also to be found in IV.iii.6; it is a major theme of the discussion of the issue of thinking matter in Answer to Second Letter, pp. 396-4 1 8 , Works, Volume IV, pp. 459-474. 18 Answer to Second Letter, pp. 397-398, Works, Volume IV, pp. 460-46 \ . 1 9 See Wilson, 'Reply', pp. 25 1-252. 2° 'M, S , & P', pp. 226-23 1 (esp. 229), 238-239. For arguments that one or another quality is not essential to body ifnot every body has the quality, see IV.x. 1O and 14. There is a very interesting application of this argument in the wind-up of the passage we are now looking at: In all such Cases, the superinducement of greater Perfections and nobler Qualities, destroys nothing of the Essence or Perfections that were there before; unless there can be shewed a manifest Repugnancy between them; but all the Proof offered for that, is only, That we cannot conceive how Matter, without such superadded Perfections, can produce such Effects; which is, in Truth, no more than to say, Matter in general, or every part of Matter, as Matter, has them not; though we cannot conceive how Matter is invested with them, or how it operates by Vertue of those new Endowments. Nor is it to be wonder'd that we cannot, whilst we limit all its Operations to those Qualities it had before, and would explain them, by the known Properties of Matter in general, without any such superinduced Perfections. (Answer to Second Letter, pp. 400-40 1 , Works, Volume IV, p. 462) 21 'Reply', pp. 250-2S \ ' Wilson indicates that the view expressed here is somewhat altered from the one she held in her earlier paper. 22 See the passages quoted in note 6 above, particularly IV.iii.28 and 29. 23 Note that this model gives us a straightforward solution to another, related problem raised by Wilson ('Limits', p. 146, and see also Ayers, 'M, S, & P', p. 2 1 8), viz. what it means to say that a superadded power or quality belongs to a body, given that it does not derive from the body's real essence alone. Against the background of the laws God has ordained, the power or quality does derive from the real essence alone; it is that that is the sole basis in actuality for the power, and it is that that is the proper causal agent responsible for bringing about the

LOCKEAN MECHANISM

229

relevant effects. 24 'Reply', p. 2 5 1 . 25 See Boyle's interesting discussion of the universal and particular notions of nature, in A Free Inquiry into the Vulgar�y Received Notion of Nature (published probably in 1685 or 1 686, but started in 1 666 - hereafter Nature, W V, p. 177; S, 1 87-1 88. 26 For discussion of this background see E. J. Dijksterhuis, The Mechanization of the World­ Picture (Oxford: Oxford University Press, 1 96 1 ), pp. 433ff. ; Marie Boas (Hall), 'The Establishment of the Mechanical Philosophy', Osiris 10 ( 1952): 412-54 1 ; T. Kuhn, 'Robert Boyle and Structural Chemistry in the Seventeenth Century', Isis 43 ( 1 952): 1 2-36; J. L. Heilbron, Electricity in the 1 7th and 18th Centuries (Berkeley: University of California Press, 1979), pp. 1 9-46; and for the best short introduction see R. Westfall, The Construction of Modern Science (New York: John Wiley and Sons, 1971 - recently reprinted by Cambridge University Press), esp. Chapters 2 and 4. 27 For plastic natures in Cudworth, see the selections from The True Intellectual System of the Universe in The Cambridge Platonists, edited by C. A. Patrides (Cambridge, Mass.: Harvard University Press, 1 970), pp. 288-325. 28 'Limits', p. 143. 29 'Reply', p. 249. 30 See IV.i.3, 5-6; IV.iii, passim. 3 1 'Limits', pp. 143-144 and 1 47, note 1 5. 32 See Charleton's Physiologia Epicuro-Gassendo-Charletoniana, or a Fabric of Philosophy Natural/founded upon the Hypothesis ofA tomes ( 1 654), a work based very much on Gassendi's Syntagma. For general discussions of these issues see R. Kargon, Atomism in England from Hariot to Newton (Oxford: Clarendon Press, 1966), esp. Chapters 8-9; J. Redwood, Reason, Ridicule, and Religion (Cambridge, Mass.: Harvard University Press, 1976), Chapters 2 and 4; and R. Westfall, Science and Religion in Seventeenth Century England (Ann Arbor: University of Michigan Press, 1966), Chapter 5, and esp. pp. 108-1 10. 33 Earlier versions of this paper were given at the Conference on Locke's Philosophy at Rutgers University, June 1983, and at the Royal Institute of Philosophy Conference on the History of Philosophy at the University of Lancaster, September 1983. My thanks to participants at those conferences for their remarks, and special thanks to the organizers of those conferences, James Buickerood and John Yolton for the Rutgers conference and Alan Holland for the Lancaster conference, both for their invitations and the splendid arrange­ ments. I want to thank colleagues and friends whose comments and criticisms helped so much: Michael Ayers, Joshua Cohen, Paul Hoffman, Jeremy Hyman, Thomas Kuhn, Thomas Lennon, John Milton, G. A.J. Rogers, Margaret Wilson and Kenneth Winkler.

APPEND I X

W A S B OY L E A N O CCA S I ON A LI S T ? In this appendix I want to counter the recent suggestion (in J . E. McGuire, 'Boyle's Conception of Nature', Journal of the History of Ideas 33 ( 1 972): 523-542) that Boyle was an occasionalist. I think there is a convincing case to be made that Boyle was no more an occasionalist than was Locke.

230

EDWIN McCANN

McGuire writes : Like Berkeley after him, Boyle adheres to the doctrine that only moral agents, responsible for their behavior, are capable of causative action . . . Thus, though physical objects appear to act so as to bring about change, in reality true causal power cannot be ascribed to them. (pp. 535-536)

(It is hard to be sure exactly how far McGuire intends the analogy with Berkeley's views to go, especially since he later says that Boyle implicitly holds a Humean constant conjunction theory of causality, which latter is of course incompatible with a Berkeleyan theory.) One passage on which McGuire relies heavily is a famous one from A Free Inquiry into the Vulgarly Received Notion o/Nature in which Boyle expresses qualms about loose talk of laws governing bodies: But to speak strictly (as becomes philosophers, i n so weighty a matter), t o say that the nature of this or that body is but the law of God prescribed to it is but an improper and figurative expression. For, besides that this gives us but a very defective idea of nature, since it omits the general fabric of the world and the contrivances of particular bodies, which yet are as well necessary as local motion itself to the production of particular effects and phenomena besides this, I say, and other imperfections of this notion of nature that I shall not here insist on, I must freely observe that, to speak properly, a law being but a notional rule of acting according to the declared will ofa superior, it is plain that nothing but an intellectual being can be properly capable of receiving and acting by a law. ( W V, p. 1 70; S, p. 1 8 1 ) -

I n this passage Boyle i s denying only that laws are the only real causal agents in bodies, or that they are real causal agents at all. This of course leaves it open for the bodies themselves, or their mechanical affections, to be genuine causal agents. Indeed, just two sentences beyond where I left off quoting, Boyle writes: And it is intelligible t o me that God should a t the beginning impress determinate motions upon the parts of matter, and guide them as he thought requisite for the primordial constitution of things, and that, ever since, by his ordinary and general concourse maintain those powers which he gave the parts of matter to transmit their motion thus and thus to one another.

(As Boyle's several references to God's 'ordinary and preserving concourse' make clear (one is at Nature, W V, p. 1 79 ; S, p. 1 90, and see in particular OFQ, W Ill, p. 48), nothing more is involved in this concourse than God's preserving bodies, and their motion, in existence, perhaps only by refraining from annihilating them.) As regards the passage from The Christian Virtuoso ( 1 690) quoted by

LOCKEAN MECHAN I S M

23 1

McGuire on pp. 535 -536, it is to be noted that Boyle is describing the views of 'the Cartesians', and not necessarily his own. There are many passages in which Boyle says that local motion is the principal cause of phenomena; see for example OFQ, W I l l , pp. 1 5 , 47; S, pp. 1 9 , 69-70, to mention just one work, and the claim occurs in any number of Boyle's other works. And let me end by quoting, almost at random, two general statements of Boyle's, the first from About the Excellency and Grounds o/the Mechanical Hypothesis, appended to The Excellency o/ Theology as compared with Natural Philosophy ( 1 674) and the second from the work McGuire draws upon most heavily, A Free Inquiry into the Vulgarly Received Notion 0/ Nature : I plead only for such a philosophy as reaches but to things purely corporeal, and, distinguish­ ing between the first original a/things and the subsequent course a/nature, teaches concerning the /armer not only that God gave motion to matter, but that in the beginning he so guided the various motions of the parts of it as to contrive them into the world he designed they should compose (furnished with the seminal principles and structures or models of living creatures). and established those rules 0/ motion, and that order amongst things corporeal, which we are wont to call the laws a/nature. And having told this as to the /armer, it may be allowed as to the latter to teach that, the universe being once framed by God, and the laws of motion being settled and all upheld by his incessant concourse and general providence, the phenomena of the world thus constituted are physically produced by the mechanical affections of the parts of matter, and what they operate upon one another according to mechanical laws. ( W IV, pp. 68-69; S, p. 1 39) . . . it seems manifest enough, that whatever is done in the world, at least wherein the rational soul intervenes not, is really effected by corporeal causes and agents, acting in a world so framed as ours is, according to the laws of motion settled by the omniscient Author of Things. ( W V, p. 176) ,

l. R. MILTON

LOCKEAN M EC H ANI S M : A C O M M ENT

Professor McCann's interesting paper raises a number of complex questions about the correct interpretation of Locke's philosophy. The first problem is what is meant by 'mechanism' or the 'mechanical philosophy'. If we have no precise specification of what these terms mean, then the question of whether Locke was or was not a consistent mechanist cannot be given a satisfactory answer. The problem is that it is not easy to produce a definition of these two (presumably synonymous) terms that is not either too wide, too n arrow, too vague, or merely tautological.

When Professor McCann states that the one thing which the views opposed to the mechanical philosophy have in common is that "they all assert the existence in bodies of non-mechanical (and usually immaterial) causal agents", I only the clause in brackets saves the characterisation from being a pure tautology. The statement that the powers and qualities of bodies issue entirely from the "two grand principles of bodies, matter and ,, motion 2 is imprecise and very broad. The statement that "the distinctive feature of mechanism is its refusal to postulate any causal agents in bodies ,, except for the bulk, figure, texture, and motion of their solid parts 3 is, on the other hand, too narrow. It could serve as a description of Cartesian mechanism, but it would appear to exclude any form of Newtonianism, an essential feature of which is the use of a concept of inertia which cannot be reduced to extension and its modes in the way that Descartes required. The problems of framing a definition of the mechanical philosophy are intractable precisely because there is no exactly formulated philosophical creed corresponding to this name. What united the adherents of the mechanical philosophy was not a list of common theses but something like a shared Kuhnian paradigm. Nature (or perhaps only inanimate nature4 ) was conceived as one vast machine, and individual bodies as small individual machines. This explains the recurrence of explanatory meta­ phors involving clocks: here was one of the very few kinds of thing whose outward behaviour could be explained, not merely in principle but actually, in fact, in terms ofthe arrangement and movement of its internal parts. The central tenet of the mechanical philosophy was that the inner mechanisms of bodies were essentially similar to those which we can see in clocks, locks 233 A . J. Holland (ed.). Philosophy. Its History and Historiography. 233-239.

234

l. R. MILTON

and other machines. This much was agreed; what concepts might be necessary for constructing complete mechanical explanations of the properties of bodies remained a matter of controversy. For this reason the boundaries of the mechanical philosophy are inherently imprecise and always potentially subject to controversy. This imprecision in no way renders the term useless, but it does carry the implication that debates about whether anyone was or was not a mechanist need to be examined with some care. There are four main areas in which it has been suggested that there is at least a prima facie case for supposing Locke to have departed from any reasonably strict interpretation of the mechanical philosophy : (i) the cohesion of bodies, (ii) gravity, (iii) the link between primary and secon­ dary qualities, and (iv) the possibility of thinking matter. On the first and second of these I find myself in substantial agreement with Professor McCann. In Essay II.xxiii.23-27 Locke's point is not that the cohesion of bodies cannot be explained using the mechanical philosophy, but rather that we have no comprehension whatever of how it might be explained, whether on the basis of the mechanical philosophy or in any other way. The case of gravity, though more complex and difficult, appears to be essen­ tially similar. The difficulties arise because some of Locke's remarks are very difficult to reconcile with others - notably the statement in Some Thoughts concerning Education that "by mere Matter and Motion, none of ,, the great Phaenomena of Nature can be resolved 5 and the statement in the Elements of Natural Philosophy that "by the figure, bulk, texture, and motion of these small and insensible corpuscles, all the Phaenomena of Bodys may be explain'd".6 Nevertheless a fairly clear line of thought does emerge, especially in the Second Reply to Stillingfleet, which contains Locke's fullest, as well as his latest, discussion of the topic. Locke's funda­ mental point, on which he repeatedly insists, is that we simply do not understand what matter is and what it can do. " [The] Gravitation of Matter towards Matter . . . inevitably shows that there is something in Matter that we do not understand."? We c'annot conceive how one body can move another except by impulse, but we know far too little to say what is and what is not naturally possible: "but that a solid S ubstance may not have Qualities, Perfections and Powers, which have no natural or visibly necessary Connection with Solidity and Extension, is too much for us (who are of Yesterday, and know nothing) to be positive in".8 The other two roblem areas require rather more discussion. The "deep explanatory gap" which Professor McCann sees between the primary and

LOCKEAN MECHANI S M : A COMMENT

235

the secondary qualities of bodies certainly exists, but it seems to me to be rather less troublesome than he supposes. Developments in the natural sciences make this rather clearer to us than perhaps it could have been to Locke and his contemporaries. Consider one of Locke's favourite examples : the yellow colour of gold. We can see that there are two quite separate questions. (i) Why does gold reflect a very high proportion of incident electromagnetic radiation of wavelength 500-700 nanometres, and why is the proportion absorbed increasingly great as the wavelength falls below about 500 nm? (ii) Why should electromagnetic radiation with the characteristics just described produce in us the experience which we call perceiving yellow light? The first of these is a question which clearly falls within the province of the physicist, and is one to which quantum physicists have been able to produce increasingly satisfactory answers. The second question has proved much more difficult to answer. One sign of its intractability is that it is not at all clear even whose job it is to answer it, or what a satisfactory answer would look like. The main reason for this rather unsatisfactory state of affairs is that the problem of our perception of secondary qualities cannot be discussed in isolation from the whole deeply intractable problem of the relation of mind and body. The general impression that a fairly casual reader of the Essay would probably receive is that Locke held a dualist theory of mind and body, and that his main point of divergence from Descartes was his belief that we have no knowledge of the essence of either material or immaterial sub­ stances. 10 The problem with this interpretation, notoriously, is that Locke was seriously prepared to consider the possibility that God might give "to some systems of matter fitly disposed, a power to perceive and think". ) ) Here two questions arise. The first is whether the possibility we are con­ cerned with is purely epistemic, as Michael Ayers has claimed,) 2 or whether it is at least to some extent ontological. I agree with Professor McCann against Dr Ayers in choosing the second alternative. Locke's view is not merely that we are ignorant of the nature of matter and therefore that we cannot dismiss the possibility that appropriately constructed material systems might have the power of thought. Instead it is the much stronger claim that because God is omnipotent he can bring about any state of affairs which can be described without self-contradiction. ) 3 Locke's re­ peated appeal to God's omnipotence was very much more than a mere rhetorical device. Because of it he was not confined to arguing that our ignorance of the essence of matter makes us incapable of discerning the boundary between what is and what is not naturally possible. Instead he

236

l . R. M ILTON

was able to place his opponents, notably Stillingfleet, on the defensive, by repeating his claim that God can make all things agree which do not involve a contradiction. His admission of our ignorance of the essence of matter, far from providing the starting point of his argument, actually weakens it, in that it leaves open the possibility, albeit remote, that some­ one might one day prove that the notion of a thinking extended (or solid) substance really did contain a hidden self-contradiction. What has become known as the problem of superaddition has acquired this name because of Locke's suggestion in the Essay that "God can, if he pleases, superadd to Matter a Faculty of Thinking".1 4 What precisely he had in mind when he used this word is by no means clear. There appear to be two possible interpretations. l s One is that God gives such powers and qualities to matter that when material particles are arranged in certain very complex ways, the resulting system has the power of thought. The other is that God can and perhaps does give to certain (selected) systems of bodies additional powers, which enable those bodies to think. The first of these is the view held by such theistic materialists as Joseph Priestley, and it is the view which has been ascribed to Locke himself by Dr Ayers. 1 6 I agree with Professor McCann i n rejecting this interpretation, but I find his arguments based on Locke's agnosticism about the mechanical philo­ sophy less than wholly convincing. 1 7 More substantial evidence in favour of the second interpretation can be found if we examine carefully what Locke himself wrote, especially in the Second Reply to Stillingfleet, and in Chapter x of Book IV of the Essay. In the Second Reply Locke says that we should not suppose that God cannot superadd new qualities to matter "unless it can be proved to be a Contradiction, that God should give to some parts of Matter, Qualities and Perfections, which Matter in general has not ; though we cannot conceive how Matter is invested with them, or how it Operates by Virtue of those new Endowments". 18 Again, when Locke states that "certain parcels of Matter, ordered by the Divine Power, as seems fit to him, may be made capable of receiving from his Omnipotency the Faculty of Thinking",19 there appears to be a clear distinction being made between the ordering of the matter and the superaddition of the power of thought. The evidence provided by the Essay confirms this conclusion. On the first interpretation, the spontaneous appearance of a thinking being in a purely material universe would be immensely improbable, but not in any way logically impossible. In the Essay however Locke states quite unam­ biguously that it is "as impossible, that Things wholly void of Knowledge,

LOCKEAN MECHANISM : A COMMENT

237

and operating blindly, and without any Perception, should produce a Knowing Being, as it is impossible that a Triangle should make itself three Angles bigger than two right ones".20 Locke's reason for this has nothing to do with the agnosticism about the nature of matter which Professor McCann emphasises (and which can certainly be found elsewhere in the Essay), and is indeed dubiously compatible with it. Locke's view was that "unthinking Particles of Matter, however put together, can have nothing thereby added to them, but a new relation of Position, which 'tis impossible should give thought and knowledge to them".21 Particles of matter "knock, impell and resist one another . . . and that is all they can do".22 Without the superaddition of entirely new powers by God, therefore, thought cannot arise. The evidence provided by these passages and by others like them sup­ ports the interpretation which I have described, but the support falls well short of incontrovertible proof.23 Is there any possibility that such proof might be found? One promising line of approach is to ask the question of whether God has the power to give any parcel of matter, however con­ structed, the ability to think. If he has, then the first interpretation can be conclusively rejected. One argument which Locke employs in the Second Reply certainly does suggest that God has the power to give any material substance the power to think. Locke supposes God to have created a solid extended substance and an immaterial substance, both in a state of complete inactivity. "Now I would ask," continues Locke, "why Omnipotency cannot give to either of these Substances, which are equally in a state of perfect Inactivity, the ,, same power, that it can give to the other? 24 The first power which Locke then considers is spontaneous or self-motion. The second is the power of thought. Here Locke is quite explicit: The same is visible in the other Operation of Thinking; both these substances may be made, and exist without Thought; neither of them has, or can have the Power of Thinking from itself: God may give it to either of them, according to the good Pleasure of his Omnipotency; and in which ever of them it is, it is equally beyond our Capacity to conceive, how either of those substances thinks 25

Two further passages in the Second Reply are also relevant. In the first, Locke is replying to Stillingfleet's query as to why, on Locke's view, brute animals lack the power of abstraction which men possess. Locke's answer is that "if Omnipotency can give Thought to any solid Substance, it is not hard

238

1. R. M ILTON

to conceive , that God may give that Faculty in a higher or lower Degree, as it pleases him, who knows what Disposition of the Subject is suited to a particular way or degree of Thinking".26 The use of a hypothetical clause here is not a sign of speculation or uncertainty, as a reading of this passage in its full context makes clear. Here at least Locke clearly supposes that God can give the power of thought to any solid substance. The second passage occurs a few pages later.27 Here Locke is taking advantage of Stillingfleet's admission that God does have the power to change a body into an immaterial substance. Locke therefore considers the implications of the following sequence of actions : 1 . God deprives a material substance of its solidity; 2. God then gives this substance a faculty of thinking; 3. finally God gives the substance solidity, thereby making it material once more. Locke clearly regards all these changes as possible. There is no suggestion whatever in the text either that they are possible only for a limited class of material substances, or that God, when restoring the quality of solidity, is required at the same time to give the substance a particular, highly complex kind of internal structure. If the interpretation of Locke's thought outlined in the last few pages is correct, the question whether Locke's acceptance of the possibility of thinking matter amounts to a genuine departure from the mechanical philosophy can be seen to be fundamentally misconceived. Most of the adherents of the mechanical philosophy regarded its truth or falsity as something to be determined a posteriori; no one, as far as I know, claimed that God simply could not have created a world governed by non­ mechanical principles. Hence the question of whether it is logically possible that a material substance should have the power to think is essentially independent of the question of whether the mechanical philo soc phy provides an adequate basis for explaining the phenomena ofthe world in which we live.

Imperial College. London NOTES

Above, p. 220. p. 209. 3 Above, p. 220. 4 On the limitations of the mechanical philosophy with regard to animals, see Boyle's remarks in the Second Part of The Christian Virtuoso in The Works of the Honourable Robert Boyle (London, 1 772). Volume VI, p. 729. I

2 Above,

LOCKEAN MECHANI S M : A COMMENT

239

5 Some Thoughts concerning Education, Section 1 80. (Third and later editions, Section 1 92). The Elements of Natural Philosophy, Chapter 12, concluding words. 7 John Locke, Mr Locke's Reply to the Right Reverend the Lord Bishop of Worcester's Answer to his Second Letter (London, 1699), p. 404. Cited subsequently as Second Reply. Also in The Works of John Locke, 1 1th edition (London, 1 8 1 2), Volume IV, pp. 464-465. 8 Second Reply, p. 405; Works, Volume IV, p. 465. 9 Above, p. 2 10. to Essay, II.xxiii.28-33, especially 32. Compare also such anti-materialist remarks as I I.xxxiii. 1 7, presumably directed against Hobbes and his admirers. II John Locke, An Essay concerning Human Understanding, edited by P. H. Nidditch (Oxford: Clarendon Press, 1 975), IV.iii.6. Cited subsequently as Essay. 12 M. R. Ayers, 'Mechanism, Superaddition, and the Proof of God's Existence in Locke's Essay', Philosophical Review 90 ( 19 8 1 ) : 2 1 0-25 1 , p. 224. \3 Second Reply, pp. 400, 406, 408, 430; Works, Volume IV, pp. 462, 466, 467, 483. 1 4 Essay, IV.iii.6. 1 5 These do not appear to be quite the same as Professor McCann's two interpretations, (above, Section 2, §§ I & 2). 16 Ayers, op. cit. , p. 246. 1 7 Above, Section 2, §2. 18 Second Reply, p. 400; Works, Volume IV, pp. 462-463. My italics, 1 9 Second Reply, p. 407; Works, Volume IV, p. 468. 20 Essay, IV.x.5. 21 Essay, IV.x. 1 6. 22 Essay, IV.x. lO. 23 A very different interpretation can be found in Ayers, op. cit. , pp. 244-246, 24 Second Reply, p. 403; Works, Volume IV, p. 464. 25 Second Reply, pp. 404-405 ; Works, Volume IV, p. 465. 26 Second Reply, p. 409; Works, Volume IV, p. 468. My italics. 27 Second Reply. pp. 412-4 1 3 ; Works, Volume IV, pp. 470-47 1. 6

PART III

PHILOSOPHY I N THE EIGHTEENTH CENTURY

M . A. STEWART

H UME AND THE " M ETA P H Y S I C A L A R G U M ENT

A

PRIORI"

l . I NTRODUCTION

There is a theistic argument which is discussed at least twice in the Burne corpus, both times rather perfunctorily. This perfunctoriness has carried over to some of his commentators, who are not always clear as to what the argument is or about the force of Hume's comments on it. On page 23 of A Letter from a Gentleman to his Friend in Edinburgh Burne calls it "the metaphysical Argument a priorI"' and in Part 9 of Dialogues concerning Natural Religion simply "the argument a priori". I It is the argument of Demea. In the Letter, Hume cites Samuel Clarke as author of the offending argument, and in the Dialogues Clarke is quoted by name as author of the defence of one particular move in it. Burne had originally criticized Clarke's defence of another move, without identifying the theological context, at Treatise p. 80, and it was his critic's sensing the threat there to the theistic argument that helped prompt the controversy to which the Letter was a contribution? The references are to Clarke's A Demonstration of the Being and Attributes of God of 1 705. Burne's wording comes closest to Clarke's reformulation in the 6th edition of 1 725, which I shall therefore follow here. To readers familiar with the conventional modern distinction between a priori and a posteriori reasoning, Burne's nomenclature may suggest the ontological argument, but that is to ignore the extensive reliance on causal language throughout the argument. McEwen, Kemp Smith and others have suggested that either Burne or Clarke combined together two distinct arguments, the ontological and the cosmological,3 and there were undoubtedly minor polemicists of the period who were guilty of some such conflation.4 Kant was of course to argue later in his Critique ofPure Reason (B620-64 1 ) that the cosmological argument was incomplete unless topped out with the ontological argument, which was then invalid. The gap which Kant found in the cosmological argument was the same gap which Clarke had found before him, and which Clarke tried to plug by that section of his argument which has read to later commentators like the ontological 243 A. J. Holland (ed.). Philosophy. Its History and Historiography. 243-270. A 1 O Q , 1..... n

DA:rI"./ D•• J.. l;.. L;.,.,.. rro._.... .... .,.

.

.

244

M. A. STEWART

argument. But both they and Kant have been wrong to identify this as the ontological argument, which Clarke never endorsed, and which cannot be combined with the cosmological argument without sending us in circles. For the cosmological argument, in the relevant form, takes as one of its premises that God is a necessary being, infers the existence of such a being from the contingency of the world, and goes on to deduce some of the perfections of this being. The ontological argument, on the other hand, concludes that God is a necessary being from premises about his perfections. Hume was writing not only before the Kantian typology of arguments had displaced the traditional simple distinction between a priori and a posteriori reasoning, but at a time when the interpretation of that dis­ tinction was itself in flux. Hume was moving towards identifying apriority with demonstrability, of which in turn he had an unusually narrow con­ ception. But as late as the 1 730's Clarke's opponents Edmund Law and Daniel Waterland had been defending the traditional cosmological argu­ ment as both demonstrative and a posteriori, rejecting only the a priori element introduced by Clarke; and they like Clarke still conceived apriority in its traditional sense of reasoning from cause to etTect. 5 When Clarke had rej ected the ontological argument, he had done so not only on the ground of its general obscurity, but because he could not see the validity of any inference from idea to existence. He did not see how to establish, in the mental proposition 'There is a Self-Existent Being", the "Absolute Impossibility of removing" the idea of an existent thing from the idea of a self-existent one unless there was independent proof that a self-existent being must exist.6 But the a posteriori demonstration of that led on to further a priori enquiry into how things must be if a self-existent being is to exist: namely, that in so far as the ground or foundation of its existence will have to lie within its own nature, whatever that may be, its nature must be such that it cannot consistently with that nature not exist. Whether this attempt not just to establish but to explain or account for the existence of God was j ustifiable, or even intelligible, is what the early 1 8th-century debate was principally about. In the Letter at page 23, Hume contrasts "the metaphysical Argument a priori, which many Men of Learning cannot comprehend, and which many Men both of Piety and Learning show no great Value for" not only with "the Arguments a posteriori from the Order and Course of N ature", but also with other "metaphysical Arguments" - arguments apparently which, though metaphysical, are not a priori in the otTending sense, however Hume construed it. Speaking of the principle of universal causation, a principle

H U M E A N D T H E " M ETAPHY S I CA L ARGUM ENT A PRIORF'

245

construed it. Speaking of the principle of univers al causation, a principle which he thinks that only a fool would deny, Hume remarks : I say further, that even the metaphysical Arguments for a Deity are not affected by a Denial of the Proposition above-mentioned. It is only Dr. Clark's Argument which can be supposed to be any way concerned. Many other Arguments of the same Kind still remain; Des Cartes's for Instance, which has always been esteemed as solid and convincing as the other.

This may be disingenuous, but if Descartes can be supposed to have put forward an argument which is metaphysical but less a priori than Clarke's it must be the cosmological argument of Meditation III; and even Hume is found ascribing a form of cosmological argument - though theistically not a very significant one - to Philo in Part 2 of the Dialogues. I shall take it for granted, then, that the ontological argument, as such, is never con­ sidered by Hume ; or at least that the argument under discussion in both contexts where Clarke is mentioned is a distinctive form (distinctively a priori) of that argument which in its various forms reasons, from the fact that things are a certain way (e.g. contingent, or caused, or moved) in the world, that there must be a being which is not itself that way, and whose being a different way (e.g. necessary, or uncaused, or unmoved) accounts for the way that everything else is. When Hume attacks the a priori argu­ ment it is always a form of a priori argument that deploys causal concepts ; and it is primarily because of the way it deploys causal concepts a priori that he attacks it. I am not concerned here with rehabilitating the a priori argument, and if the only historical point at issue were to get clear about Hume's target it would scarcely warrant a separate paper. Getting the right target is only the first step towards putting the Dialogues back into historical context ; and doing that, I want to argue, is essential to understanding the changing fortunes of the argument of the work as a whole, and to solving what the historically uninformed like to call the 'problem' or 'enigma' of the Dia­ logues. My paper aims to address the general historiographical question of how one should read Hume's Dialogues, by focusing upon some of the particular issues raised by the interpretation of Part 9. 2. DEMEA'S EXPO S ITION

Demea's presentation of the Clarkean argument is deceptively informal and is confined exclusively to the third paragraph in Part 9 of the Dia-

246

M. A. STEWART

logues, apart from certain corollaries hinted at but not pursued in the opening paragraph. Clarke's own argument, extending to the deduction of the divine attributes, had been more than a hundred pages long. It is not just this discrepancy in relative lengths that makes Demea a somewhat distant follower. His argument is accorded little of the characteristic structure of Clarke's ; in particular it fails to identify a cumulative sequence of propositions to which Clarke's premises are directed in turn, and which he seeks to establish by a series of reductio arguments. The first four of these propositions, or intermediary conclusions, need to be noted here for future reference : I. II. Ill. IV.

Something has existed from all Eternity. There has Existed from Eternity, Some One Unchangeable and Independent Being. That unchangeable and independent Being, which has Existed from Eternity, without any external Cause of its Existence; must be Self-Existent, that is, Necessarily-existing. What the Substance or Essence of that Being, which is Self­ Existent, or Necessarily-Existing, is; we have no Idea, neither is it at all possible for us to comprehend it.

Notwithstanding the fourth claim here Clarke goes on to infer in turn that the self-existent being is eternal, infinite, single, intelligent, free, etc., each new proposition being separately argued. His Humean follower reduces this to a small compass and some con­ fusion. The following steps can be detected. [I] [2]

Whatever exists must have a cause or reason of its existence. It [is] absolutely impossible for any thing to produce itself, or be the cause of its own existence.

In the Letter, following Treatise I. I1I.iii, the causal principle was differently phrased: "Whatever begins to exist must have a Cause of Existence". Though that is not equivalent to [ I ] , there are familiar enough ways of adapting a cosmological argument to accommodate this alternative formu­ lation. But neither is [ I ] equivalent to the causal principle on which Clarke's argument depends. The formulation draws on the wording in the 6th edition of Clarke's Demonstration ; but by omitting all but the opening words of it, Demea loses much of the point of the distinction between

H UM E AND T H E "META P H Y S ICAL ARGUM E NT A PRIORI"

247

external cause and internal reason that was central to Clarke's argument :

Whatever Exists, has a Cause, a Reason, a Ground o f its Existence; ( a Foundation, o n which its Existence relies; a Ground or Reason why it doth exist, rather than flat exist;) either in the Necessity of its own Nature, and then it must have been of it self Eternal: Or in the Will of some Other Being; and then That Other Being must, at least in the order of Nature and Causality, have Existed before it. (6th edition, p. 9)

This was for Clarke a premise for a distinct stage in the argument, establishing his Proposition I, a stage missing from Demea's summary. That it is missing, not simply because Hume has Demea condense the argument, but because he is condensing it without a clear grasp of its structure, is shown by his formulation of [2] . 7 That is extracted from a different stage in Clarke's argument, the stage at which Clarke establishes his Proposition III, after the argument for Proposition II has intervened. As the grammar of Hume's text stands ("it being absolutely impossible . . . ", etc.), [2] seems to be presented as a premise for [ 1 ] . Not only does that make no logical sense of Clarke's original argument: it makes none of Demea's either. For the purposes of the argument, [2] is better construed as a supplementary premise, independent of [ 1 ] . From [ 1 ] and [2] together Demea reckons to be able to infer that, i n so far as each "cause or reason" is in turn something that "exists", and we "mount up" recursively from effects to causes : we must either [A) go on in tracing an infinite succession, without any ultimate cause at all, or must [B) at last have recourse to some ultimate cause, that is necessarily existent.

Option [BJ , which is glossed in Clarkean terms as recourse to a "Being, R EA S ON of his existence in himself; and who cannot be supposed not to exist without an express contradiction", is established by defeating [A ] , and [A] is defeated in two further steps : who carries the

[3]

[4]

The whole eternal chain or succession, taken together, . . . requires a cause or reason, as much as any particular obj ect, which begins to exist in time. External causes, there are supposed to be none. Chance is a word without a meaning. Nothing can never produce any thing. ' "

This is an abridgement of part of the argument for Clarke's Proposition II (not Proposition III), without expressly mentioning that proposition, but

248

M. A. S TEWART

it leaves only implicit a principle that is importantly explicit in Clarke, namely that a succession of contingent beings must be a contingent suc­ cession. That is why [3] is a straight application of [ 1 ] ; but the demonstra­ tion in [4] that there are no other options warrants closer study. "External causes, there are supposed to be none." External to what? The question was, originally: what is the cause of the endless succession of causes? If there were no cause of it, we could as well have had Nothing instead of it. "What was it then, which determined something to exist rather than nothing, and bestowed being on a particular possibility, exclusive of the rest?" Now for Demea to say that there is supposed no cause external to the succession is to contradict the whole purport of the argument. Alternatively, to say that there is supposed no cause external to that which determined the eternal being to exist rather than nothing would be true to Clarke's argument, but it is not the natural interpretation of the expression "some­ thing rather than nothing" in the quotation just given. In Clarke's argu­ ment, the supposition that there would be no room left for external causes was part of a reductio argument against the possibility of only the series of dependent causes : for Clarke there has to be an external cause. But in the argument of the Dialogues, the supposition that there is no external cause is offered not as an incoherence or absurdity, but as a constructive step. Clarke and Demea are not on the same wavelength, in their use of the same language. In short, though all the individual phrasing is lifted from Clarke, the total argument bears no more relation to the historical Clarke's argument than the ephemeral pamphlets which circulated in criticism of Hume's philosophy during his lifetime bear to his philosophy. Stringing together somebody's phrases is not the same as reconstituting his argument, a point that Hume and his associates were quick enough to make in their own defence in the controversies of the 1 740's and 1 750's, when they protested vigorously at the iniquity of "broken and partial Citations" (Letter, p. 33). There are two other weaknesses in the argument assigned to Demea. First, Demea's gaffe over "external causes" suggests that he is not fully clear about the distinction between linear and hierarchical causation on which the cosmological argument traditionally depends. "In mounting up . . . from effects to causes, we must either go on in tracing an infinite succession, without any ultimate cause at all, or must at last have recourse to some ultimate cause, that is necessarily existent." Certainly to a casual reader this will tend to suggest that Demea is looking into the possibility

HUME AND THE "METAPHYS ICAL ARGUMENT A PRIORF'

249

of blocking an infinite regress, in either the linear or the hierarchical series. The "infinite succession" in this case is a linear or temporal one, described in the dialogue in terms of precedence and eternity. Whether there is an infinite temporal succession is something on which proponents of the cosmological argument, Clarke included, have traditionally deferred to revelation. The "ultimate" cause that they have sought is compatible with there being an infinite temporal succession, so long as it is outside the succession. This is what is meant in talking about a hierarchical order of causation. For example, we explain the formation of a puddle in terms of a current downpour of rain, the downpour in terms of the present state of the clouds, and that in turn by more general phenomena of nature which combine to determine that state, all these factors being considered as operative at roughly the same time. Sooner or later, concurrent explana­ tion of this kind reaches a natural stopping point, as theist and non-theist alike run out of ever more ultimate principles, whether within the natural order or without. Hume was conscious of this hierarchical conception of causation in writing the first Enquiry, pp. 30-3 1 ; but his talk of "mounting up from effects to causes" in the Dialogues would as likely as not suggest temporal regression to an 1 8th-century reader. Secondly, Demea's argument exploits an ambiguity in the formulation of [3]. When it is argued that the whole "chain or succession" stands in need of a "cause or reason", the point may be [3a]

There is a cause or reason for there being this eternal succession and not another one

or it may be [3b]

There is a cause or reason for there being any succession at all.

The principle at the back of both formulations is no doubt the same, namely some sort of Sufficient Reason principle to the effect that, where there are two or more logical possibilities of which only one can be realized at once, there is some cause or reason to account for the realization of the one and the non-realization of the other(s). But an opponent's tactics will differ, according to whether [3a] or [3b] is meant; and in respect of both the ambiguities that I have j ust explained, Cleanthes in the dialogue readily and consistently avails himself of the more easily refuted interpretation of the argument.

250

M. It-. STEWART

3. CLEANTHE S ' OBJECTI O N S

Cleanthes' attempt to rebut Demea's argument, which takes up para­ graphs 5 to 9 of Part 9, is as undistinguished as Demea's presentation. He starts by conflating two distinct features of the original Clarkean argument - its supposed demonstrativeness, and the apriority claimed for it by its proponents, i.e. its ability to establish a sufficient reason in the nature of things for the existence of a particular sort of being. Cleanthes then offers a supposedly decisive obj ection to a priori-demonstrative reasoning of this kind: Nothing is demonstrable, unless the contrary implies a contradiction. Nothing, that is distinctly conceivable, implies a contradiction. Whatever we conceive as existent, we can also conceive as non-existent. There is no being, therefore, whose non-existence implies a contradiction.

This over-rated argument confuses the plain truth that you cannot con­ sistently reject the conclusion of a demonstration if you accept its prem­ ises, with the plain falsehood that you cannot consistently reject the con­ clusion independently of the premises. If a demonstration were a piece of reasoning in which the conclusion could not consistently be denied no matter what the premises, the premises would be redundant and the demonstration unnecessary: we should have a necessary truth, but a ne­ cessary truth follows from anything whatever. This is not the sort of thing that those who debated the existence of God were talking about. In trad­ itional metaphysics the premises of a demonstration must indeed be self­ evident; but that is a less stringent requirement than that their denial be self-contradictory and is compatible with their being contingent. There is probably some slithering between the two notions of necessity and self­ evidence in Clarke as well as in the Dialogues, because of a tendency to underestimate the role played by supplementary premises in generating the contradictions upon which reductio arguments depend. If the notion of self-evidence is too nebulous for modern logic, with its disdain for any suggestion that the function of argument is to carry psycho­ logical conviction, then modern readers may also balk at Cleanthes' use of "contradiction" as a psychological term. It is part of the continuing Cartesian inheritance of the period that contradiction was seen to derive from an incommensurability between ideas. It was detected in the attempt to render the ideas 'distinctly conceived' or fully articulate in the mind, on the assumption that any inconsistency must show itself in 'confused' ideas.

HUME AND THE "METAPHYSICAL ARGUMENT A PRIORF'

2S 1

So p can be demonstrated only if (i) p is distinctly conceived in this technical sense and (ii) not-p is not distinctly conceivable. The first stage in Cleanthes' attack (para. 5 ) was to claim that this dual requirement is not in fact met, because the second condition is not met, in any judgement relating to existence; but this is a straight denial of, rather than a counter­ argument to, the Clarke an position. The second stage (para. 6) is to claim further that the conditions of demonstration never will be met, at least in respect of the deity, while "our faculties remain the same as at present" and our articulate ideas do not extend to the "whole essence or nature" of the deity. But thirdly (para. 7) we are in a like ignorance with regard to the nature of matter and cannot show a priori that the claims made on behalf of the deity do not apply to matter. The infirmity of our ideas of the divine nature is not anything that Clarke - or, witness Part 2 of the Dialogues, Demea - would deny. They had not been trying to derive the existence of the deity from any idea of its nature but rather from a recognition of the contingency of the material world; and although it was inferred that the being in question would have a nature which was inconsistent with its ever having come into existence or passing out of existence, we are not ourselves expected to understand how this is possible. It was a commonplace that - quite generally - we do not have ideas of essences, and what should have been at issue instead is whether we have ideas adequate to the argument. Cleanthes makes the same mistake as before, of trying vainly to conceive of a necessary being in a mental vacuum, rather than in relation to a prescribed sequence of thought; and when, side-stepping Demea's formulation and going back to Clarke's original, he cites the supposed parallel of the impossibility of conceiving that 2 + 2 #- 4, he fails to notice that Clarke had used this as an analogy, not for the conclusion in isolation from the argument, but for the cumulative argument of Propositions I to III. Clarke is explicitly Cleanthes' target when he argues against Demea in para. 7 that, for all we know a priori - without the evidence of Design and without adequate ideas - the material world may be the necessary being; and once again the criticism is levelled without regard to the preceding argument :

I find only one argument employed to prove, that the material world is not the necessarily­ existent Being; and this argument is derived from the contingency both of the matter and the form ofthe world. "Any particle of matter," 'tis said, "may be conceived to be annihilated; and any form may be conceived to be altered. Such an annihilation or alteration, therefore, is not

252

M. A. STEWART

impossible." But it seems a great partiality not to perceive, that the same argument extends equally to the Deity, so far as we have any conception of him; and that the mind can at least imagine him to be non-existent, or his attributes to be altered. It must be some unknown, inconceivable qualities, which can make his non-existence appear impossible, or his attributes unalterable: And no reason can be assigned, why these qualities may not belong to matter.

Clarke does indeed allow, in the defence of his Proposition III, that the form and matter of the material world can be conceived differently from the way they actually are, and may therefore possibly change; but at least part of his argument for this is that it is a transparently observable fact that they do change. If this is so, then they do not meet his criteria for having "an Absolute Necessity in the Nature of the Thing it self' ; and that there must be something that has such an "Absolute Necessity" is judged to have been shown by the preceding argument. While Cleanthes has denied that anything could have shown this, he has not identified any actual errors in Clarke's argument since he has not identified the strict argument. There was no need for Clarke to have invoked any dubious conception of the deity, and he has not done so, by this stage. Later, by the time he has argued not only that there is an unchangeable being, but that it is self­ existent, single, intelligent, free, wise, etc., then we can consider whether the lengthy cumulative argument, in the face of nothing more than doctrinaire posturing from Cleanthes, has built up into something that looks like the picture of a deity or not. A more sympathetic view of Cleanthes' activity would be to see him as constructing a sceptical dilemma for the proponent of non-experimental reasoning. Just as our inability to conceive of necessary existence carries no ontological implications, so our inability to conceive the non­ annihilation of the world may just be a function of our ignorance. But if that is so, it was misleading of Cleanthe& himself to put any weight upon what we can conceive, viz. the non-existence of anything if we can conceive its existence, when the fact of the matter must be that in both cases what we can conceive as possible is neither here nor there. Cleanthes is not, however, a sceptic when it comes to experimental reasoning; and what we really needed here, to get an effective engagement between Cleanthes and Demea, was a serious discussion instead of unexamined assumptions on both sides, of how far our ideas are a measure of what is possible and impossible. The two remaining steps in Cleanthes' criticism of Demea (paras. 8 and 9, marking Hume's resumption of the manuscript after breaking for a time at para. 7) continue to reflect this lack of engagement, although they are

HUME AND THE "METAPHYSICAL ARGUMENT A PRIORF'

253

presumably intended to enhance the plausibility of the view that, for all we know a priori, the material world may be the necessary being. First, he says, "it seems absurd" to look for a first cause of an eternal succession, since the eternal succession would then have to have a beginning and the cause would have to precede it. It is now Cleanthes who is offering to establish a matter of fact from what it makes sense to conceive! But 'absurd' was one of Clarke's favourite terms of logical appraisal, so perhaps this is j ust part of an attempt to catch Demea with his own petard. Cleanthes assumes, however, that whether or not the chain of causation has been eternal, the debate has been solely about linear causation ; and there has been no effort to rebut the orthodox idea that an eternal succession might have a coeval cause. Cleanthes concludes with the objection that if you can account for the causation of each item in a series there is nothing further involved in explaining the cause of the series. I pointed out above that there was an ambiguity in Demea's original argument. In so far as Demea was defending [3a]

There is a cause or reason for there being this eternal suc­ cession and not another one

then Cleanthes' counter-argument will work. Ifeach stage in the succession determines its successor then by explaining the formation of the series you have in fact accounted for the whole. B ut in so far as Demea is defending [3b]

There is a cause or reason for there being any succession at all

- explaining, not why there is everything, but why there is anything - then the artificiality of a collection, on which Cleanthes' objection depends, is beside the point. Now much of Cleanthes' criticism has been a variation on standard themes in the literature on the cosmological argument, themes which are not original with the author of the Dialogues. The first, supposedly decisive, objection was expressed more circumspectly in A Letterfrom a Gentleman, pp. 22-23, as a thesis attributable to Archbishop Tillotson, "That the Being of a God is not capable of Demonstration, but of moral Evidence".8 It is commonplace in other writers on natural religion in the same period as Tillotson, e.g. John Wilkins, Edward Stillingfleet and John Ray, and throughout the 1 8th century. Since it was this (by then) new orthodoxy that Clarke had set out consciously to confute by his Demonstration of 1 705, the

254

M. A. STEWART

doctrinaire restatement of that orthodoxy by Cleanthes in the Dialogues can hardly have amounted to a refutation : it is more a digging in of heels. That the essences of things, including God, are unknown to us, and that speculation on those essences is futile, is a part of the same orthodoxy, and a part that Clarke did not actually challenge. Cleanthes and Demea are alike parties to a tradition which holds that there are truths beyond reason; they are in different but still established camps on the matter of whether we can actually know what some of these truths are. The question of whether the material universe may be the necessary being is as old as the cosmological argument itself, and most proponents of the argument at least since Philoponus in the 6th century have addressed themselves to it. If the substance of much of Cleanthes' argument is not distinctively Humean it nevertheless clearly has a Humean form. That is, it is expressed in terms of Hume's philosophical categories, and reflects a Hume's-eye view of how the objections would go . There is nothing very surprising in that. Hume did, after all, write the book. 1. I argued that Cleanthes' initial objection involved a logical blunder, about demonstrability and contradiction. That logical blunder was one of Hume's own blind spots. See, for example, Treatise pp. 79-80, 89, 463, Enquiry pp. 25-26, 1 63 - 1 64, typified by this extract from the first of the Enquiry passages : That the sun will not rise to-morrow is no less intelligible a proposition, and implies no more contradiction, than the affirmation, that it will rise. We should in vain, therefore, attempt to demonstrate its falsehood. Were it demonstratively false, it would imply a contradiction, and could never be distinctly conceived by the mind.

2. Cleanthes' follow-up move, denying "meaning" to any talk of "neces­ sary existence", is modelled on Hume's argument about the idea of power or efficacy and the idea of energy at Treatise pp. 1 6 1- 1 62, 1 67-1 68, 266-267, Enquiry pp. 76-77, though it does not follow that a real-life Cleanthes would accept the parallel argument about power. Hume himself, while frequently granting the need to assume unknown or secret powers, insisted we could have no ideas of them, but only dummy ideas "drawn from something extraneous and foreign". 3. On whether the material world might be the necessary being Hume is as open-minded as Cleanthes. We undoubtedly believe in the necessity of causal connections, and can explain the nature and origins of this belief, but we cannot justify it. We cannot justify it, not because it is false, but

HUME AND THE "METAPHYSICAL ARGUMENT A PRIORr'

255

because our ideas are chronically inadequate ; and were they adequate we might well see the necessity of those principles (and more) which at present are mere projections from our experience of observable regularities, in the sense that we would actually be able to infer an effect from its cause. As far as experience alone goes : As the actions of matter have no necessity, but what is deriv'd from these circumstances [constant union and the inference of the mind], and it is not by any insight into the essence of bodies we discover their connexion, the absence of this insight, while the union and inference remain, will never, in any case, remove the necessity. ( Treatise, p. 400)

If we let the non-rational processes of the mind take their course, we shall be led to believe, as we all do believe, that the operations of external bodies are necessary, and that in the communication of their motion, in their attraction, and mutual cohesion, there are not the least traces of indifference or liberty . . . The actions, therefore, of matter are to be regarded as instances of necessary actions; and whatever is in this respect on the same footing with matter, must be acknowledg'd to be necessary. (pp. 399-400)

Whether they are objectively necessary is undiscoverable, because the actual causal processes are undiscoverable : I t has been observ'd already, that i n n o single instance the ultimate connexion o f any objects is discoverable, either by our senses or reason, and that we can never penetrate so far into the essence and construction of bodies, as to perceive the principle, on which their mutual influence depends. (p. 400)

4. Hume like Cleanthes held that any cause is prior in time to its effect ( Treatise, p. 76), though this probably has to be read consistently with the possibility that the end of a cause may be co-terminous with the beginning of its effect. Priority in time is quite different from what Clarke had meant by priority in "the order of Nature". 5. The final paragraph of Cleanthes' criticism seems to be modelled on Hume's adaptation in the Treatise (p. 30) of a passage from the French geometer Nicolas de Malezieu. The Malezieu original can readily be con­ sulted in Kemp Smith's The Philosophy ofDavid Hume (p. 34 1 ), but the two Hume passages are worth setting out in j uxtaposition.

256

M. A. STEWART

Dialogues But the WHOLE, you say, wants a cause. I answer, that the uniting of these parts into a whole, like the uniting of several distinct counties into one kingdom, or several distinct members into one body, is performed merely by an arbitrary act of the mind, and has no influence on the nature of things. Did I show you the particular causes of each individual in a collection of twenty particles of matter, I should think it very unreasonable, should you afterwards ask me, what was the cause of the whole twenty. That is sufficiently explained In explaining the cause of the parts.

Treatise Twenty men may be said to exist; but 'tis only because one, two, three, four, &c. are existent; and if you deny the existence of the latter, that of the former falls of course . . . (T]hese twenty men may be con­ sider'd as an unite. The whole globe of the earth, nay the whole universe may be con­ sider'd as an unite. That term of unity is merely a fictitious denomination, which the mind may apply to any quantity of objects it collects together; nor can such an unity any more exist alone than number can, as being in reality a true number. But the unity, which can exist alone, and whose existence is necessary to that of all number, is of another kind, and must be perfectly indivisible, and incapable of being resolved into any lesser unity.

Note in both passages the use of the arbitrary number 20 to represent a potentially infinite series, the analogy between this series and the whole universe, and the parallelism between the phrase "merely , . . an arbitrary act of the mind" and "merely a fictitious denomination, which the mind may apply to any quantity". That such Humean-Iooking manoeuvres are ill suited to refuting the Clarkean argument is not a new thesis, but I hesitate to ally myself with such other recent critics as Rescher and Stove, whose ahistorical counter­ manoeuvres to Cleanthes' criticism offer little more than a reductio ad absurdum of the analytic method.9 There are lessons here for anyone who still thinks philosophical argument exists in a timeless vacuum. Nicholas Rescher's short notelet is a case study of Cleanthes' first obj ection, intended to demonstrate that "the utility and fruitfulness of the application of the tools and concepts of modern formal logic to the study of current philosophical issues" is something which "also extends in an historical dimension, and can afford a useful means for the analysis of the conceptual content of historic philosophical texts" (p. 1 7 1 : italics original). As such, it is a failure, His symbolization equates does not assert existence with asserts non-existence, and the conceivability of a claim that is made with conceivability that the claim is made (p. 1 73). Rescher's analysis takes no account of the traditional meaning of the expression "distinctly con-

HUME AND THE "METAPHYSICAL ARGUMENT A

PRIORF'

257

ceive", and he offers the following remarkable defence of "Hume's con­ clusion" that there can be no satisfactory a priori proof of a "matter of fact or existence" (p. 1 72): "a valid deductive argument can not carry us beyond its premisses to establish something not logically contained (however cryptically) within them. Any demonstrative conclusion relating to existence must consequently be based upon premisses in which this existential thesis is actually contained, be it overtly or covertly". That, of course, is what any defender of the cosmological argument would cheer­ fully accept in opposition to Hume; but then, Rescher has not noticed that it is the cosmological argument that is under discussion. D. C. Stove's paper seems to me to be no closer to grasping 1 8th-century modes of thought. He re-writes most of the argument, Demea's as well as Cleanthes', in the language of 20th-century positivism : "Demea made it clear in paragraph 3 that he regards the Deity as 'a necessarily existent being'. The conclusion of his argument was, 'there is a Deity'. He regards this conclusion, therefore, as being a necessarily true proposition". (p. 303) Stove is then dumbfounded when Cleanthes goes on to contemplate the a priori possibility that the world could, after all, be a necessary being, particularly when this comes only a paragraph after the claim that there is "no meaning; or, which is the same thing, none that is consistent" to the term "necessary existence". Apparently not recognizing that "no meaning" in Hume means only an inability to frame Humean ideas, 1 O he contends that p aragraphs 6 and 7 of Part 9 are inconsistent with each other ; and having apparently forgotten Hume's two-edged discussions of liberty and necessity, he argues that paragraph 7 and the seemingly associated specu­ lations of Philo in Parts 6 and 9 and elsewhere are so out of step with "the philosophy which Hume published in his lifetime" that we are to think he shifted his opinions dramatically in the middle of writing the Dialogues and that "metaphysics never was hisforte anyway". This attempt to convert incomprehension into a priori exegesis is effort wasted. Two minutes' examination of the manuscript or photographs would show that para­ graphs 6 and 7 were part of a single piece of continuous writing and that the rest of Part 9 belongs to the same period of composition ; aI,ld five minutes' comparison of the textual changes which Hume made to the ten editions of the first Enquiry between 1 750 and his death will show that he did not modify his views in any relevant way during the period of composi­ tion of the Dialogues. But this is only surface skirmishing. There is a more basic reason why Rescher, Stove and many other commentators here are not even beginning -

258

M. A. STEWART

to make sense of what is going on. They are assuming without question that Cleanthes in Part 9 is speaking for Hume. If the Humean parallels I have cited above seem to have reinforced this impression, then the remainder of this paper is intended to dispel it. 4. T H E CHARACTERIZATION OF T H E DIA L O G UES : P H I LO AND CLEANTHES

Even if Cleanthes' final objection, in Part 9 paragraph 9, carries echoes of the Treatise, it is not an objection that Hume could consistently have supported in the Enquiry or Philo in the Dialogues. In Part 2 Philo repeats the argument of Enquiry p. 148 that it is the very singularity of the universe that precludes our engaging in causal reasoning about it. That can hardly be squared with Cleanthes' objection that it is the artificiality of the singularity of the universe that precludes . . . etc .. Writing to James Balfour of Pilrig on 1 5 March 1 75 3 about the dialogue appended to the second Enquiry, Hume said "In every Dialogue, no more than one Person can be suppos'd to represent the Author"," but I do not think that anyone has taken this to mean that the other person in that dialogue - the moral sceptic Palamedes - represents another particular individual. Hume wrote the letter at a time when he was engaged inter alia on his Dialogues concerning Natural Religion. An earlier letter, to Gilbert Elliot of Minto on 10 March 1 7 5 1 , and the relatively late re-casting of the end of the prologue to the Dialogues, both suggest that Hume saw the personae of this work as being distinct philosophical types, or in his term "characters", rather than individuals. It is a suggestion that has now been fleshed out with a good deal of suggestive textual detail in a very stimulating recent discussion by Michael Pakaluk.' 2 There is however no real inconsistency in the dialogue-writer's identifying himself with one of the types he has created, and indeed the whole work must be an arid academic exercise unless the other types also have some historical basis. The point is well enough conveyed in Hume's letter to Elliot : 1 3 I have often thought, that the best way of composing a Dialogue, wou'd be for two Persons that are of different Opinions about any Question of Importance, to write alternately the different Parts of the Discourse, & reply to each other. By this Means, that vulgar Error woud be avoided, of putting nothing but Nonsense into the Mouth of the Adversary: And at the same time, a Variety of Character & Genius being upheld, woud make the whole look more natural & unaffected. Had it been my good Fortune to live near you, I shou'd have taken on me the Character of Philo, in the Dialogue, which you'll own I coud have supported naturally

HUME AND THE "METAPHYSICAL ARGUMENT A PRIORF'

259

enough: And you woud not have been averse to that of Cleanthes.

That Philo is Hume's mouthpiece is, though disputed, not seriously disputable on the biographical evidence. It was the obvious construction put on the Dialogues by contemporary critics as irreconcilable as Beattie and Priestley, and by the reviewer in the Monthly Review, and has been subsequently challenged only by those who do not understand literary method. Obviously the three philosophical types depicted in the work are not at variance with one another the whole of the time. At any one time, and for any one issue, at least two of them may be in agreement, or not overtly in disagreement; and any dialogue aims to proceed by finding ways to build upon common ground. But to allow that Cleanthes, for example, may well say things on occasion with which Philo would not disagree is not

the same as saying that Cleanthes may an occasion stand in for Philo,

whether for purposes of literary variety or as a mark of authorial irony. That is just witless floundering. Each character in the Dialogues has some sort of philosophical and historical integrity. Let us apply this first to the character of Philo, the Humean sceptic, in Part 9. Philo comes in at the end, as an umpire, only after Demea and Cleanthes have had their say, and I have not yet said anything about this stage of the argument. I noted at the beginning of this paper that when Hume wrote the letter from which A Letter from a Gentleman was derived he identified "the metaphysical Argument a priorI" as an argument "which many Men of Learning cannot comprehend, and which many Men both of Piety and Learning show no great Value for". In his initial draft of Part 9 of the Dialogues Hume confined Philo to this same evaluation, only adding a brief psychological gloss : the argument a priori has seldom been found very convincing, except to people of a meta­ physical head, who have accustomed themselves to abstract reasoning, and who finding from mathematics, that the understanding frequently leads to truth, through obscurity, and contrary to first appearances, have transferred the same habit of thinking to subjects, where it ought not to have place. Other people, even of good sense and the best inclined to religion, feel always some deficiency in such arguments, though they are not perhaps able to explain distinctly where it lies. A certain proof, that men ever did, and ever will derive their religion from other sources than from this species of reasoning.

Here we have the beginnings of an explanation as to why Hume's discussion of the argument, through the mouths both of Demea and of

260

M. A. STEWART

Cleanthes, is so feeble : he did not think it worth his serious attention, and probably did not claim to understand the argument. A certain degree of superficiality was built into Hume's method of writing from the outset, since, with rare exceptions, the length of each Part of the Dialogues was determined not by the demands of the subject-matter, but by what could be got on to a single folded (i.e. 4-sided) manuscript sheet. But in Part 9 he did not even use up his allowance of paper. Then he came back to it and on the spare fourth side first added, then cut, then restored what we now know as the penultimate paragraph assigned to Philo. It is as though he was of two minds whether he had anything useful to add to the debate or not; but it is worth noting what he did add. It is a two-edged argument designed to discomfort both parties if they care to listen. If you are not content, as Cleanthes had been formerly, with the "wonderful regularity" that strikes a "superficial observer", and you start to speculate about delving more deeply, then it is possible that, if we did really understand the basis of the order we admire in "the whole oeconomy of the universe", we would find a necessity in it which would be counter-productive to the argument, like the necessities in algebra which are attributable neither to chance nor design but are simply a function of the way things must be. This is pushing C1eanthes' opposition to Demea further than Cleanthes could comfortably go ; for Cleanthes was trying to show that any argument one could patch up to support belief in a necessary being isn't enough to establish the conclusion that this being is the Deity (a conclusion that needs the Design argument), whereas Philo wants to show it might undermine that conclusion. Their arguments are superficially similar but they are not in fact the same ; similarly in Part 4 and Part 8 Philo gave an interpretation of the theory of active matter which would make the Deity redundant. Let us now turn to Cleanthes and take seriously the question, why is it the liberal theologian and not the sceptic who leads the attack on the a priori argument? Here we need to distinguish the cosmological argument in general from the a priori and demonstrative' turn given to the argument in the earlier decades of the 1 8th century. Cleanthes, like any liberal theologian of the mid-1 8th century, would presumably accept Philo's a posteriori expression of the cosmological argument in Part 2, without per­ ceiving that it is somewhat stronger on rhetoric than on substance : But surely, where reasonable men treat these subjects, the question can never be concerning the Being, but only the Nature of the Deity. The former truth, as you well observe, is

HUME AND THE "METAPHYSICAL ARGUMENT A

PRIORF'

26 1

unquestionable and self-evident. Nothing exists without a cause; and the original cause of this universe (whatever it be) we call GOD; and piously ascribe to him every species of perfection. Whoever scruples this fundamental truth, deserves every punishment, which can be inflicted among philosophers, to wit, the greatest ridicule, contempt and disapprobation.

But in rejecting the a priori demonstrative form of the argument Cleanthes is perfectly typical of the leading Scottish intellectuals of the period. 14 Francis Hutcheson as a young man had found Clarke's a priori argument unsatisfactory and sent him some objections ; and whatever Clarke may have replied, Hutcheson retained his deep suspicion of a priori arguments for the rest of his life and thought the baillement they induced did no service to religion. 1 5 Colin Maclaurin, in his paraphrase of Newtonian theology, first held up Descartes as an awful warning of the scientific error that can follow from attempts to do a priori theology; he then went on, though without naming Clarke in criticism, to insist that the necessary existence of the First Cause was inferrable only a posteriori, "from our existence, and that of other contingent beings around US". 1 6 Henry Home of Kames, soon to be Lord Kames, in Part II, Essay VII (later renumbered VIII) of Essays on the Principles of Morality and Natural Religion ( 1 75 1 ) expressed similar distrust to Hutcheson's , which he too had nurtured since he had engaged in a youthful correspondence with Clarke, still extantP When John Witherspoon published the 2nd edition of his witty satire Ecclesiastical Characteristics against the Shaftesburians in 1 754, he added a footnote on an unnamed university where, about 7 years previously, every 1 6-year-old could be relied upon to dismiss the name of Dr Clarke with scorn : that would fit the Glasgow of Hutcheson and Leechman, and Witherspoon had lived a few miles outside Glasgow for the previous 10 years ; but it would perhaps also fit the Edinburgh of William Wishart, where he had been trained and where there had been strong Hutchesonian influence. All these critics of the a priori argument were strong devotees of the Design argument, and the character of Cleanthes as a philosophical type clearly falls somewhere within this tradition. Hurlbutt started to ask the right questions when, rightly dubious at the efforts of a recent literary critic to identify Cleanthes with the person of Butler, he convincingly tracked down verbatim echoes of Cheyne, Maclaurin, etc., in Cleanthes' turns of phrase. But he jumped to too quick a conclusion in supposing that Cleanthes is always the mainstream 1 8th-century Newtonian. Cleanthes' Newtonianism is at times too mechanical, his epistemology too heretical, -

262

M. A. STEWART

and his flirtation with necessitarianism too dangerous, to suit that image. In particular, there is too much sheer Humeanism in Cleanthes' discussion of the causal principle and the empirical basis of causal reasoning (in Part 1 0 as well as Part 9) and in his treatment of conceivability and dem­ onstrability. If therefore Cleanthes is to represent a credible type (and why else would Hume have asked Elliot's advice on the portrayal ?), there must be people who accepted these sizeable chunks of Humean philosophy but were not prepared to go the whole way - people who resisted the sceptical tendencies of that philosophy and endorsed the Design argument in a fairly naive form, but based their resistance to Hume's logic not on rational refutation but on the assurances of sense and feeling. My claim is that the historical personage who most obviously fits the character is also the one personage in Scotland in the early 1 750's who was still bothering to argue in print against Clarke's a priori argument - its long-standing opponent Lord Kames. Kames and Elliot were close friends and discussed philosophy together, as did Kames and Hume, and we know from correspondence that themes that were to be worked into the 1 75 1 Essays were already known to Hume in 1 746 and to Elliot in 1 750. 1 8 In the words of Kames' Advertisement: A plan is prosecuted, in support of the authority of our senses, external and internal; where it is occasionally shown, that our reasonings on some of the most important subjects, rest ultimately upon sense and feeling. This is illustrated, in a variety of instances; and from these, the author would gladly hope, that he has thrown new light upon the principles of human knowledge: - All to prepare the way, for a proof of the existence and perfections of the Deity, which is the chief aim in this undertaking.

Given the interchange of information among his acquaintance Hume would have known something of this even before publication, just as Kames and at times Elliot, Blair, et al. had previews of Hume's work. I do not mean that Cleanthes is modelled exclusively on Kames, rather than Cheyne or Maclaurin, any more than I mean that the Humean elements in either Cleanthes or Kames go very deep or go much beyond the principles needed to rebut the a priori argument. But Cleanthes represents the theologico-scientific interest of which Kames in the early 1 750's was typical. We may say that, while Cleanthes speaks up in general for the new experimental theism of the Enlightenment divines, the portrayal is coloured by Hume's personal closeness to one of their prominent secular associates. Cleanthes' reduction of everyday "reason" to "experience" in Part 2

HUME AND THE "METAPHYSICA L ARGUMENT A

PRIORF'

263

para. 25 of the Dialogues, his appeal to the evidence of "feeling" in Part 3 para. 7 and excuse for the "ignorant savage" in para. 9, the challenge to his "feeling and experience" in Part 10, and his denial of both the misery and the wickedness of man also in Part 1 0 , are all characteristically Kamesian. The following is offered as documentary evidence that analogues of Cleanthes' 'Humean' objections to Demea are to be found in Kames' 1 75 1 Essays. The first passage comes immediately after Kames has rejected the attempts of Locke and Clarke to demonstrate the causal principle. In short, there does not appear to me any contradiction in the above proposition, that a thing may begin to exist without a cause: and therefore, I dare not declare the fact to be impossible. But sense and feeling afford me a conviction, that nothing begins to exist without a cause, tho' reason cannot afford me a demonstration of it... Fond of arguments drawn from the nature of things, we are too apt to apply such arguments without discretion ; and to call that demonstration, which, at bottom, is nothing but a conviction from sense and feeling. (pp. 275-276) But as we cannot find any inconsistency in the latter supposition, we cannot justly say that it is demonstrably false. (p. 3 1 9) We are not so much acquainted with the nature or essence of any one thing, as to discover a necessary connection betwixt it and its powers, that the one subsisting, the other must also subsist. There is nothing so easy to conceive, as that the most active being, shall at once be deprived of all its activity: and a thing that may be conceived, can never be proved inconsis­ tent or impossible. (p. 30 I) Admitting that something has existed from all eternity, I find no Data to determine a priori, whether this world has existed of itself from all eternity, in a constant succession of causes and effects; or whether it be an effect produced by an Almighty Power. It is indeed hard to conceive a world eternal and self-existent, where all things are carried on by blind fate, without design or intelligence. And yet I can find no demonstration to the contrary. If we can form any obscure notion of one intelligent being, existing from all eternity, it appears not more difficult to form a notion of a succession of beings, with or without intelligence; or a notion of a perpetual succession of causes and effects. (pp. 3 1 7- 3 1 8) In matters so profound, it is difficult to form ideas with any degree of accuracy. I have observed above, that it is too much for man, to grasp, in his idea, an eternal Being, whose existence, upon that account, cannot admit of the supposition of a cause. To talk, as some of our metaphysical writers do, of an absolute necessity in the nature of the Being, as the cause of his existence, is mere jargon. For we can conceive nothing more clearly, than that the cause must go before the effect, and that the cause cannot possibly be in the effect. (pp. 343-344) There is nothing in the whole universe that can properly be called contingent, that may be, or may not be; nothing loose and fluctuating in any part of nature; but every motion in the natural, and every determination and action in the moral world, are directed by immutable laws: so that, whilst these laws remain in their force, not the smallest link of the universal chain of causes and effects can be broken, nor any one thing be otherways than it is. (pp. 1 8 1-182) It is true, the production of one thing by another, even in a single instance, implies a power; and this power is necessarily connected with its effect. (p. 300)

264

M. A. STEWART

In his 1 723 correspondence with Clarke, Kames had argued that every caused effect was a necessary being and to that extent the whole causal chain was necessary, so that Clarke should have confined his argument to establishing God's self-existence and not necessity. Clarke countered with a distinction between relative and absolute necessity which seems to have had no impact on Kames. It is hardly surprising that Kames rejected the argument from the contingency of the world ifhe denied that the world was contingent. His particular brand of necessitarianism had one of its fullest expressions in his Philosophical Society paper 'Of the Laws of Motion', published in the 1 754 collection of that society of which Hume was co-editor. 1 9 He there defended the view that active power is essential to matter and scandalized more orthodox opinion by his claim that belief in God's active involvement in the day-to-day causal process was "whimsical". John Stewart, in leading the scientific establishment against Kames, defended Clarke's reading of Newton's physical theology ; and although this did not extend to a defence of Clarke's a priori argument, the controversy was exploited by those contemporary clerics who wished to make such a defence against Kames. I have so far found nothing in Kames directly analogous to Cleanthes' concluding objection, though that objection does of course simply put another perspective on something Kames does say, namely that he can see no logical incoherence in the possibility that an endless chain of causes and effects constitutes the sum-total of things - it is only our "sense or feeling" that counts against it. But the logical move that Cleanthes purports to be countering (that "the WHOLE . . . wants a cause") is a move that undoubt­ edly had been made against Kames after the initial publication of his Essays. So whether or not the objection can be traced further back than the Dialogues, it can still be seen as Hume's attempt to formulate a defence of a Kamesian position. Kames, nicknamed Sopho by his critic, had been reprimanded for leaving open the possibility of an uncaused chain of caused effects : This eternal chain of causes and effects is made up of a finite or an infinite number of productions. An infinite number is an absurdity: and a finite number leads unto, and lands in a first cause; and therefore this cause is necessary and self-existent. How SOPHO comes to make up a chain uncaused, and consisting of caused productions, is, what I do not pretend to understand, but as so many inconsistencies as there are links. It is true, that, from a distributive to a collective sense, the consequence will not always hold. In a collection of things there may be some particular properties, and some particular defects, that cannot be

HUME AND THE "METAPHYSICA L ARGUMENT A

PRIORr'

265

attributed to the whole: for instance, in a nation some are learned, and some are ignorant; but then learning and ignorance are but accidents in respect of man. But, in essentials, what is said of the whole is applicable to every individual person, except the totality or the collection itself abstractly considered. Ten thousand men is a number or a whole, every one of which is a reasonable animal. It is of the essence of man to be such; and it is of the essence of a production to be an effect, and the essence of an effect to have a cause: and therefore there cannot be an uncaused chain, consisting of caused effects or productions. I desire to be excused from giving my assent to SOPHO's discovery, "Though every link be supposed a production, the chain itself exists without a cause, as well as one eternal being does." Nor do I thank him for his inference, 'Therefore an eternal succession of beings is not a more natural supposition, than one eternal self-existing being." The disparity is great. The one is a demonstration, and the other is an absurdity.

If it is this piece of contemporary reasoning against Kames that Cleanthes is trying to come to grips with in his final objection, then I think it is possible to see how the reading of it may have triggered in Hume's mind some recollection of his old discussion of Malezieu. It is to this polemical source that I now turn. 5. T H E CHARACTERIZATION OF T H E DIA L O G UES : D E M EA

I have so far argued that there is no need to deviate from the view that Philo is the Humean philosopher in that part of the Dialogues where the attack on Demea is led by Cleanthes. Philo's avowed boredom with the a priori argument is a reflection of the same lack of interest that led Hume to write so cliche-ridden a debate on it ; but another factor in explaining this is the state of the debate when Hume was writing. The link with Kames is not difficult to establish, but the historical thesis gains in credibility if it can be shown that there was more going on at the time than just Kames re-living the battles of his youth - if in the same period there were others, opposing the opposition, fighting a rearguard action in defence of Clarke in particular or demonstrative arguments in general. Then we shall at last have the historical key to the characterization of the Dialogues that has so often eluded the commentators : Hume, through Philo, is adjudicating the debates of his contemporaries. Hume, let us remember, did not choose to criticize his own contemporaries in general, and his friends in particular, by name - for all that he was mortified if they did not publicly discuss him - and the dialogue format was the obvious means remaining to him to engage in their debates. Although I can suggest here the line to explore, it has to remain speculative until we can see whether the manuscript study that is now

266

M . A. STEWART

planned for the Princeton Hume edition will be sufficient to pinpoint the date when Part 9 of the Dialogues was written. We can infer from Hume's correspondence with Elliot that he had got somewhere past the end of Part 3 by March 1 75 1 , and from stylistic evidence that he had reached an ending (not the present ending) by the end of the 1 750's; and for reasons that are too complex to go into here, it is possible to show from the physical state of the manuscript that there was a major break in the composition of the work after Part 4, and another midway through part 12, and that some of the substantial revisions to the earlier parts (notably Part 3) came only after Hume was into Part 12. It is natural to suppose, but not yet certain, that Parts 5-12 were written 'it some time over the period 175 1-1 759. For now, all I can do in support of my thesis is point to the undoubted fact that a declared opponent of Kames (and Hume) was writing in defence of Clarke, i.e. this was a topical debate, in and after late 1 753. The writer was the elderly Rev. George Anderson ( 1 676-1 756), the first Master or administrative head of the newly-founded George Watson's Hospital (boys' school), whose An Estimate ofthe Profit and Loss ofReligion, Personally and Publicly Stated (anon. , Edinburgh, 1 753) was an extensive and at times not un-witty attack on Kames' Essays, with side-glances at Hume. My quotation in the previous section was from pp. 1 8 1- 1 82 of this work. The same critic sought to have Kames' publishers censured by the Edinburgh presbytery in 1 756 (Kames himself not having yet publicly acknowledged his authorship), and to this end published a further 20-page pamphlet The Complaint Made to the Presbytery of Edinburgh, Verified. The second half of the pamphlet restated Anderson's objections to Kames' discussion of the arguments for the existence of God by picking out the Kames texts he found obnoxious. Between these two publications there had appeared in London the posthumous Works ( 1 754) of Bolingbroke, whose own extraordinarily rambling criticisms of Clarke caused more outcry than they were worth and were not altogether appreciated by those who were fighting the same cause; Anderson in his final book before he died (A Remonstrance against Lord Viscount Bolingbroke 's Philosophical Re­ ligion, cited in note 4) coupled Kames and Bolingbroke as infidel writers. The criticisms in An Estimate of the Profit and Loss of Religion had mean­ while been taken up by the Rev. John Bonar in An Analysis of the Moral and Religious Sentiments contained in the Writings of Sopho, and David Hume, Esq. , a pamphlet issued at Edinburgh prior to the attempt to have Kames and Hume censured by the General Assembly in 1 755. In his represen­ tation of Kames' views Bonar drew in part, as had Anderson, on texts

HUME AND THE "METAPHYSICAL ARGUMENT A PRIORF'

267

where Kames had written in criticism of Clarke; but unlike Anderson in 1 753, he did not cite Clarke by name. These debates continued. However, the further one goes into the 1 750's, the harder it is to feel sure that these various writings actually antedated the writing of Part 9 of the Dialogues. I put most of my weight on the fact that Anderson's Estimate was out by November 1 753, that Hume was aware of Anderson's activities, and that Kames studied the book closely enough to make a good many tactical changes to his argument (notably a deletion of most of the references to feeling) in the 1 758 reissue of his Essays. If there is good reason to think that Cleanthes has a Kamesian streak in him, then there is good reason to think that the opposition of Cleanthes to Philo and Demea reflects in some degree the contemporary conflict between the Kames circle and their different critics. Without some such explanation, it is difficult to see why Hume ever wrote another

extensive work - the Dialogues - when all his important criticism of natural religion had appeared already in the Philosophical Essays (the first Enquiry) of 1 748 and he also had The Natural History ofReligion underway; and given the strong similarity between the latter work and William Robertson's subsequent writing on the religion of India it is as likely as not that those ideas were also the topic of discussion in the circles in which Hume was currently moving. A reasonable hypothesis would then be that the Dialogues, having started life around 1 750 as Hume's answer to the watered-down Humeanism which Kames was preparing against him, developed as it progressed into a commentary on the controversy which Kames' work itself induced; this would mean that Hume found a more positive role for Demea as time went on, namely as defender of an argu­ ment which Kames' attack brought back to public attention. The final leaf of the manuscript of Pamphilus' prefatory remarks - with the famous passage in which the characters of the three participants are drawn, and the "accurate philosophical turn" of Cleanthes is contrasted with the "rigid inflexible orthodoxy" of Demea - is a relatively late insertion, replacing an earlier ending to the prologue which was in the manuscript at the time (in 1 75 1 ) when Hume showed the sample to Elliot. Did he revise his view of the roles of his characters after nearing the end of the work? Anderson was neither a great nor an original thinker, but neither was he quite the zealous fool that those who take at face value Hume's customary belittling of his opponents would have us think. He had come to public notice in the 1 730's for an attack on public entertainments, which he seems to have regarded as being clearly outlawed by Pauline texts

268

M. A . STEWART

against ribaldry and carousal; and in the pulpit he was doubtless very strait-laced and bibliocentric. But that was not the basis for his attack on Kames. He had a fair grounding in standard logic, but was opposed to syllogistic disputation. He had absorbed Locke's epistemology, read Bayle extensively if not with admiration, had a nodding acquaintance with the work of Shaftesbury and Hutcheson, and was able to quote classical and medieval sources as readily as Locke and Clarke. His scholarship was far from impeccable, but his attack on Kames was , throughout, a philo­ sophical and not a bibliocentric one. He was a conservative intellectual, not an anti-intellectual in the way that extreme evangelicals like Duncan Forbes of Culloden were. Indeed he goes to some lengths to make out Kames to be the intellectual philistine, who has no conception of the nature and operations of reason a fa Locke. In his attack on Kames' theology , Anderson concentrated on three things which are related: ( 1 ) Kames' talk about conceivability and his use of appeals to conceivability to rebut what he (Kames) abusively called "abstract" reasoning; (2) Kames' constant resort to "sense and feeling" on matters which others attribute to reason and understanding; (3) his refusal to find any demonstrative certainty in any principle that is necessary either to prove the existence of God or to understand the operation of causes in general. Anderson was thrown by two things in particular: the extreme narrowness of Kames' (more Humean than Hume) conception of reason; and the emphasis on the contingency from the reasoner's point of view of the relation between an effect and its cause and on the a posteriori character ofthe causal principle. The latter neither he nor many others of his contem­ poraries ever began to understand and it is not surprising that his views on these matters should get short shrift from a Kamesian Cleanthes. But there was just as little comprehension on the part of Hume and Kames, who pretty certainly did not realize how much mystification they them­ selves had introduced when by their strict definitions of reason and demonstration they had changed the terms of the a priori argument. If Anderson and his associates kept stumbling over these it was not alto­ gether their fault, and his intuitions were right when he sensed that some of the difficulties and peculiarities in Kames' position lay hereabouts. The clash between Cleanthes and Demea is a clash between two philo­ sophical traditions neither of which understood the other's terms, neither of which therefore was in a strong enough position to refute or even engage fruitfully with the other. The futile debate is, moreover, more metho­ dological than substantive, in that both parties are arguing wide of the -

HUME AND THE "METAPHYSICAL ARGUMENT A PRIORF'

269

point that the cosmological argument is ultimately about, namely the existence of a necessary being in the hierarchical and not the temporal order of causation. In so far as this confusion on both sides can be documented in the historical debates of the period when Hume was writing, it both supports my thesis that Hume's discussion was a reflection of a more immediately topical debate than we are apt now to assume, and explains why that discussion is a mess. The topical debate was the debate that Hume himself was brought into by his friendship with Kames' circle at the time (before Kames' alliance with Reid) when the partial overlap in their philosophical views threatened both their careers and was being exploited by those who were out to embarrass and curtail the influence of their liberal clerical friends. If this line of argument can be sustained in the particular case of Part 9 of the Dialogues, it suggests a potentially fruitful way of going about the understanding of other Parts as wel1.2o

University of Lancaster

NOTES I References to A Letter from a Gentleman are to the first edition (Edinburgh, 1 745), which has been reproduced with the original pagination in all recent editions; it was not Hume's own work but it incorporates the text of a letter which was. In quoting from the Dialogues I normally follow the text of the posthumous first edition of 1 779, citing Part and if necessary paragraph numbers for identification in other editions. References to the Enquiries and A Treatise ofHuman Nature are to the texts of L. A. Selby-Bigge as corrected by P. H. Nidditch (Oxford: Clarendon Press, 1975 and 1978). 2 See M. A. Stewart, David Hume and the Edinburgh Chair, forthcoming in the Journal of the History of Philosophy, monograph series. ' 3 See the editions of Hum e's Dialogues concerning Natural Religion edited by Bruce McEwen (Edinburgh : Blackwood, 1907), p. xlv, and by Norman Kemp Smith (2nd edition, Edinburgh: Nelson, 1 947), p. 1 1 5. The argument is correctly identified by E.J. Khamara and D. G. C. Macnabb, 'Hume and his Predecessors on the Causal Maxim', in David Hume: Bicentenary Papers, edited by G. P. Morice (Edinburgh: Edinburgh University Press, 1977), p. 153, in spite of my apparent stricture in a review of the Morice collection published in Hume Studies 7 ( 1981), p. 99. A line of the review was lost from the typescript through my own fault, creating a reversal of the sense. 4 For example, George Anderson, A Remonstrance against Lord Viscount Bolingbroke 's Philoso­ phical Religion (Edinburgh, 1 756), Section 6. 5 Edmund Law's An Enquiry into the Ideas of Space, Time, Immensity, and Eternity (Cambridge, 1 734) has a section answering John Jackson's defence of Clarke against previous criticisms by Law. Waterland's A Dissertation upon the Argument a Priori for Proving the Existence of a First Cause: In a Letter to Mr. Law was appended anonymously to Law's work. 6 A Demonstration of the Being and Attributes of God (6th edition, London, 1 725), pp. 1 9-20.

270

M . A. STEWART

7 This point was picked up by Khamara and Macnabb, art. cit. . 8 John Tillotson, The Wisdom o f Being Religious', i n Works (London, 1 696), sermon I. Hume is summarizing, not quoting. 9 Nicholas Rescher, 'An Argument of Hume's', in Essays in Philosophical Analysis (Pittsburgh : University of Pittsburgh Press, 1 969; repr. Washington : University Press of America, 1 982), pp. 1 7 1-179; D. C. Stove, 'Part IX of Hume's Dialogues', Philosophical Quarterly 28 ( 1 978): 300-309. There is more substance in a supplementary paper by James Franklin, 'More on Part IX of Hume's Dialogues', Philosophical Quarterly 30 ( 1 980): 69-71 ;more still in D.E. Stahl, 'Hume's Dialogue IX defended', Philosophical Quarterly 34 ( 1 984): 505-507. 10 This point has most recently been made by John P. Wright, The Sceptical Realism of David Hume (Manchester: Manchester University Press, 1 983), Sections 1 3-14. 11 National Library of Scotland: ms. 3582, foL 2 1 4. 12 Michael Pakaluk, 'Philosophical "Types" in Hume's Dialogues', in Philosophers ofthe Scottish Enlightenment, edited by V. Hope (Edinburgh : Edinburgh University Press, 1 984), pp. 1 1 6-1 32. 13 The Letters of David Hume, edited by J. Y. T. Greig (Oxford: Clarendon Press, 1932), Volume I, p. 1 54. 14 Cf. James McCosh, The Scottish Philosophy (London: Macmillan, 1 87 5 ; repr. Hildesheim : Olms, 1966), p. 7. There are minor exceptions in the first half of the century: the idiosyncratic Andrew Baxter and William Dudgeon, and the expatriate George TurnbulL 1 5 See William Leechman's preface to Hutcheson's posthumous System of Moral Philosophy (Glasgow, 1 755), Volume I, pp. iv-vi. Hutcheson seems to have been following his mentor Gershom CarmichaeL See Carmichael, Synopsis Theologiae Naturalis (Edinburgh, 1 729); McCosh, op. cit., pp. 4 1-42. 16 Colin Maclaurin, An Account ofSir Isaac Newton 's Philosophical Discoveries (London, 1 748), Book IV, Chapter 9. Robert H. Hurlbutt III, in Hume, Newton, and the Design Argument (Lincoln, Nebr.: University of Nebraska Press, 1 965), Chapter 9, demonstrates that Hume read some pages of Maclaurin's lectures after their publication in 1 748 and while he was working on the Dialogues. But there has never been any evidence to support the frequently heard claim that he heard Maclaurin lecture on Newton in his youth. 1 7 Scottish Record ot1ice: GD 24/ 1/548. 18 See Hume to Kames, in New Letters of David Hume, edited by R. Klibansky and E. c. Mossner (Oxford: Clarendon Press, 1 954), p. 20; and Elliot to Hume, in J. H. Burton, Life and Correspondence of David Hume (Edinburgh: Tait, 1 846), Volume I, p. 324. Elliot like Kames trained as a lawyer but he subsequently became a prominent Scots politician. Richard H. Popkin has the quite mistaken idea that he was a "liberal Protestant clergyman". See Popkin's introduction to Hume's Dialogues (Indianapolis : Hackett, 1 980), p. viii. 19 Essays and Observations, Physical and Literary, Read before a Society in Edinburgh, and Published by them (Edinburgh, 1 754), Volume !, pp. 1-69. On the Hume-Kames-Stewart debate, see Wright, op. cit. , and Michael Barfoot, James Gregory (1 753-1821) and Scottish Scientific Metaphysics / 750-1800 (Ph. D. dissertation, Edinburgh 1983), Chapter 2. Wright's interpretation has been debated by Barfoot in a paper presented to the Hume Society in August 1984. 20 This paper is based on work completed during a period as a Visiting Research Fellow at the Institute for Advanced Studies in the Humanities, University of Edinburgh, supported in part by a grant from the Leverhulme Trust. A primitive version of it was read to the Hume Society in 1 982.

NICHOLAS CAPALDI

T H E H I S T O R I C A L A N D P H I LO S O P H I C A L S I G N I F I CANCE O F H U M E' S T H EO RY O F T H E S E L F

I NTRODUCTION

The aim of this paper is to rescue Hume's theory of the self from the distortions to which it has been subj ected by Hume's analytic readers. This will permit us to rediscover some fundamental insights about the self, philosophy, and temporality. H UME'S ACCOUNT OF THE S ELF

Hume believes in the existence of a self. As he makes clear in Book II of the Treatise, 'Of the Passions', the self is "that individual person, of whose I actions and sentiments each of us is intimately conscious" (p. 286), and he further describes the self by reference to "the qualities of our mind and body" (p. 303). If there is, therefore, a self in Hume's view, then of what does it consist? As the previous quotations indicate, the self consists of a mind and a body. If the self consists of a mind and a body, we must at some point raise the question of the relationship between the two. But for the moment we want to stress that there are two parts to the self, a mind and a body, and we cannot assume that what Hume says about one of these parts, e.g. the mind, applies to the other part or to the whole which they compose. So important is this point that Hume warns his readers in Book I that there are going to be two different discussions of the self. . . . we must distinguish betwixt personal identity, as it regards our thought or imagination, and as it regards our passions or the concern we take in ourselves. The first is our present subject . . . (p. 253)

He also warns us that what he says in Book I applies only to the first part of the self, namely the mind. (ibid. ) What does Hume tell us about the mind? The mind is described as : ( 1 ) "a bundle 'Of collection of different perceptions" (p. 252); (2) "a kind of theatre, where several perceptions successively make their appearance" 271 A . J. Holland (ed.), Philosophy, Its History and Historiography, 271-285. :� L.:� � r��A _ _ '_"

272

NICHOLAS CAPALDI

(p. 253); and (3) "a system of different perceptions or different existences, which are link'd together by the relation of cause and effect .." (p. 26 1 ).2 So it is the mind which is a set of perceptions , not the self. If Hume does believe in a self which consists of a mind and a body linked in some way, then it is clear that we can have an idea of the self. If there is an idea of the self we shall want next to distinguish the idea of the self from the self, and then to ask just what kind of idea it is. In H ume's system there are two kinds of ideas, simple and complex. For Hume, the idea of the self is not a simple idea. Hume denies that we have a simple idea of the self. The main point of the discussion in Book I is to expose the belief in a simple self and in the simple idea of the self as a myth. There are some philosophers, who imagine we are every moment intimately conscious of what we call our SELF; that we . . . are certain, beyond the evidence of a demonstration, both of its perfect identity and simplicity . . . Unluckily all these positive assertions are contrary to that very experience, which is pleaded for them, nor have we any idea of self, after the manner it is here explain'd . . . there is no such idea. (pp. 25 1-252)

The same point is reiterated in the Appendix . . . . we have no impression of self or substance, as something simple and individual. We have, therefore, no idea of them in that sense. (p. 633)

It is clear from the whole context that what Hume is denying is that we have either the experience or the idea of a simple and identical self. He is not denying that there is a self, nor is he denying that we have an idea of the self. He is denying that we have a simple idea of a simple self. If we have an idea of the self, but it is not a simple idea, then it must be a complex idea. The idea of the self is complex because the idea of the self "represents" (p. 278) (a) "a succession of related ideas and impres­ sions" or (b) "a connected succession of perceptions" (p. 277). Note that the referent of the complex is both to ideas and to impressions, not to impressions alone. There are two questions we must raise about this complex. First, how is the idea acquired, and, second, what are the relationships integral to the parts of the set? The answer to the first question is that the complex idea of the self emerges originally in action as the obj ect of the indirect passions of pride and humility. The passions of pride and humility are impressions, and they

SIGNIFICANCE OF HUME'S THEORY OF THE S ELF

273

have a preceding idea which causes or produces them. The indirect passions in turn produce or cause another idea. The second produced idea, called by Hume the object of the passions, is in the case of pride and humility none other than the idea of the self. The ideas which cause the passions are memories of past pleasures and pains. As an idea, the complex idea of the self must be preceded by an impression. No idea can arise without a preceding impression. The preceding impression is the passion either of pride or of humility. But some readers are bound to ask, shouldn't this produce the idea of pride or the idea of humility? Shouldn't an idea of the self come from an impression of the self? The answer to these questions is that it is only in the case of simple ideas that there is an original simple impression of sensation. Complex ideas do not have a one-to-one correspondence ( Treatise, p. 3 ). No com­ plex idea, including and especially the complex idea of the self, need have a direct analogue in experience. Complex ideas, moreover, are formed through the creative activity of the mind's association of ideas. Many readers are so accustomed to the simplistic and empiricist or positivist 1 930's rendition of Hume that they tend to ignore Hume's own stated position. What is more important is that action is a fundamental category in Hume's analysis. Just as pure epistemology is incoherent and leads to scepticism when we try to understand ourselves as disembodied minds, so the concept of ourselves cannot be obtained in pure thought. It is the reference to action that underscores the importance of the body as part of the self. This rootedness in action is so fundamental that it cannot be explained. It is the ground of explanation. "Unless nature had given some original qualities to the mind, it cou'd never have any secondary ones; because in that case it wou'd have no foundation for action, nor cou'd ever begin to exert itself" (p. 280). 3 There is, thus, in Hume's theory a pre­ cognitive context of all analysis, and it lies in bodily action. We turn, now, to the second question about the complex idea of the self, namely, what relationships hold the set together. The specific natural relations integral to the set are resemblance and causation. Both ' re­ semblance and causation rely upon memory.

(a) . . . the memory not only discovers the identity, but also contributes to its production, by producing the relation of resemblance among the perceptions. (p. 26 1 ) (b) . . . .our identity with regard to the passions serves to corroborate that with regard to the imagination, by the making our distant perceptions influence each other, and by giving

274

N I C H O L A S CAPAL D I

u s a present concern for our past o r future pains o r pleasures . . . . Had we no memory, we never shou'd have any notion of causation, nor consequently of that chain of causes and effects, which constitute our self or person . In this view, therefore, memory does not so much produce as discover personal identity, by shewing us the relation of cause and effect among our different perceptions. (pp. 26 1-262) . .

These two functions of memory must be clearly distinguished. The memory that produces is the memory of ideas of past pleasures and pains that give rise to the indirect passions. The indirect passions are simple impressions of reflection caused by ideas. The memory that discovers the idea of the self is the memory of how the indirect passions give rise to the idea of the self as their object. This shows clearly that Hume's discussion of personal identity with regard to thought presupposes what he says about personal identity with regard to the passions. It shows, as well, that the complex idea of the self or personal identity is known only in retrospect. One cannot directly confront one's self. One can think about its past or its future, but even in the latter case what we have is a causal projection of an idea drawn from the past. What memory does not reveal is an unchanging self. Paradoxically, the idea of the self can be re-identified but never identified. Re-identification does not presuppose identification. This should not sur­ prise us, for in Hume's metaphysics there are no universals, only the resemblances within an historical series. PAST PLEASURES & PAINS ..... INDIRECT PASSIONS ..... IDEA OF SELF t

(MEMORY)

PRODUCES

t

DISCOVERS

Let us sum up this account of the complex idea of the self by reference to the chart below. A. The idea of the self is a complex idea of the imagination and of the memory. B. The imagination, in Hume's scheme, is the ideational analogue to sensory reflection. There are, however, no complex impressions of re­ flection to which complex ideas of the imagination can directly correspond. There is, according to Hume, no conception of how such complex im­ pressions could be held together. We are not dealing with a mirror image. Hence, memory cannot produce such a copy. C. There is no simple impression of sensation of the self. D. The complex idea of the self, therefore, must come either from complex impressions of sensation or simple impressions of reflection. E. It cannot come from complex impressions of sensation alone because

S IGNIFICANCE OF H UM E'S THEORY OF THE S ELF

275

these are produced by external obj ects. At most, they would only be impressions of the physical body. F. The complex idea of the self, therefore, must come from simple impressions of reflection (i.e., the passions).

�-

sensation (produced by external objects and bodily processes: T, 73, 2 1 1 , 246, 287)

Simple reflection passions (produced by simple remembered* ideas of pleasure and pain: T, 7, 84, 276) -

{

IMPRESSIONS

sensati�n (produced by external object, e.g. Pans: T, 3, 2 1 1 , 297)

Complex reflection (empty no notion of how they could be held together: T, 1 90, 25 1 , 259, 307, 320, 636) -

�

PERCEPTIONS

memory (from simple impressions both of sensation and reflection: T, 3, 23 1 )

Simple imagination (empty - n o innate ideas: T, 7, 33, 74, 1 58, 1 6 1 ) IDEAS 23 1 )**

.

Complex (relations, imagination modes, and (T, IO) substances)

[

philosophical relations natural relations (resemblance, contiguity, and causation: T, 260, 636)

* This is the memory that produces the idea of the self (T, 253, 261 , 277, 280) ** This is the memory that discovers the idea of the self.

H UME'S M I S G IV I N G S

The most remarkable thing about this remarkable theory is that Hume is dissatisfied with it. The . notorious dissatisfaction is expressed in the Appendix. We must ask with what was he dissatisfied, and what does this

276

NICHOLAS CAPALDI

dissatisfaction imply. Philosophers begin to be reconcil'd to the principle, that we have no idea of external substance. distinct from the ideas of particular qualities. This must pave the way for a like principle with regard to the mind, that we have no notion of it. distinct from the particular perceptions. So far I seem to be attended with sufficient evidence. But having thus loosen'd all our particular perceptions, when I proceed to explain the principle of connexion, which binds them together, and makes us attribute to them a real simplicity and identity; I am sensible, that my account is very defective, and that nothing but the seeming evidence of the precedent reasonings cou'd have induc'd me to receive it. Ifperceptions are distinct existences, they form a whole only by being connected together. But no conn ex ions among distinct existences are ever discoverable by human understanding. We only feel a connexion or determination of the thought, to pass from one object to another. It follows, therefore, that the thought alone finds personal identity, when reflecting on the train of past perceptions, that compose a mind, the ideas of them are felt to be connected together, and naturally introduce each other. However extraordinary this conclusion may seem, it need not surprize us. Most philosophers seem inclin'd to think, that personal identity arises from consciousness; and consciousness is nothing but a reflected thought or perception. The present philosophy, therefore, has so far a promising aspect. But all my hopes vanish, when I come to explain the principles, that unite our successive perceptions in our thought or consciousness. I cannot discover any theory, which gives me satisfaction on this head. In short, there are two principles, which I cannot render consistent; nor is it in my power to renounce either of them, viz. that all our distinct perceptions are distinct existences, and that the mind never perceives any real connexion among distinct existences. Did our perceptions either inhere in something simple and individual, or did the mind perceive some real conn ex ion among them, there wou'd be no difficulty in the case. For my part, I must plead the privilege of a sceptic, and confess, that this difficulty is too hard for my understanding. I pretend not, however, to pronounce it absolutely insuperable. Others, perhaps, or myself, upon more mature reflexions, may discover some hypothesis, that will reconcile those contradictions. (pp. 635-636)

What Hume claims to be defective is his account of the unity of conscious­ ness. Twice in his statement of the problem, he identifies the unity of consciousness as the problem. ( 1 ) . . . when I proceed to explain the principle of connexion, which binds them together . . . I am sensible, that my account is very defective. (p. 635) (2) But all my hopes vanish, when I come to explain the principles, that unite our succes­ sive perceptions in our thought or consciousness. (p. 636)

Why does Hume believe that his account of the unity of consciousness is defective? In order to answer this question, let us distinguish anew between the self and the idea of the self. What Hume realizes is that he has explained the idea of the self. To this account he still subscribes when

S I GN I F I CANCE OF H U ME'S THEORY OF THE S ELF

277

he reiterates that "the thought alone finds personal identity, when reflect­ ing on the train of past perceptions" (p. 635). But the idea of the self is not the same as the self understood in terms of the unity of consciousness. It is because Hume distinguishes between the self and the idea of the self that he perceives a problem. If there were no difference between the self and the idea of the self, ifthe subj ect were just another obj ect, there would be no problem. What does Hume think would be a kind of account for the unity of consciousness ? He suggests two possibilities. First, if there were a real connection among perceptions we would have an explanation for the unity of our consciousness. Second, if perceptions inhered in a simple substance, we would have an explanation. Either inhesion or real connection could explain the unity of consciousness. The difficulty for Hume is that he can accept neither. Hume denies real connections. He had refused to consider seriously the possibility of in­ hesion in substances. Instead, he had insisted upon the distinctness of every perception. So a conflict exists between the belief in the unity of consciousness and Hume's commitment to two principles. This explains the inconsistency Hume notes in his own theory. The two principles Hume mentions do not conflict with each other, so they must conflict with a third. What they jointly conflict with is a belief in the unity of consciousness.4 There is additional support for my claim that Hume's misgivings arise because of his recognition that he has not accounted for the unity of consciousness. Hume himself draws a parallel between the conflict or inconsistency in his account of personal identity and " . . . those contra­ dictions , and absurdities, which seem to attend every explication, that human reason can give of the material world" (p. 633). The contradictions that Hume wished to avoid with regard to the material world were the contradictions, or more accurately, difficulties of explaining the substance of external physical objects. He thought that he had circumvented that issue by focusing on how the mind organized the data without committing himself on the Source or origin of the data.

The intellectual world, tho' involv'd in infinite obscurities, is not perplex'd with any such contradictions, as those we have discover'd in the natural . . . wou'd we hearken to certain philosophers I am afraid 'tis at the hazard of running us into contradictions . . . concerning the material or immaterial substances, in which they suppose our perceptions to inhere. (p. 232) . . .

278

N ICHOLA S CAPALDI

So what Hume means by 'contradictions' is the issue of inhesion. Initially, Hume thought he faced no such problem with regard to the mental world. s But the problem re-emerged. Hume does not despair of identifying the principles that unite our successive perceptions. He has already identified them as resemblance and causation. His problem is to explain or to give the origin of those principles. It is one thing to explain the belief in the identity of material objects by reference to the mind without raising the question of the inhesion of matter. But after having assumed and asserted early in the Treatise that there were material foundations to the relations which unite,6 accounting for such principles still raised the issue of mind-body interaction and inhesion. Hume realized that some account must be forthcoming when he came to deal with the unity of consciousness. He admits as much in the Enquiry: Is there not here, either in spiritual or material substance, or both, some secret mechanism or structure of parts, upon which the effect depends, and which, being entirely unknown to us, renders the power or energy of the will equally unknown and incomprehensible. (EHU, pp. 68-69)

To sum up, then, Hume believed in the unity of consciousness but realized that he had failed to account for it. This led him to reconsider the doctrine of inhesion, but in any case he came to believe that no conceptualization of that unity was possible. The unity of consciousness as the preconceptual condition or context of conceptualization emerges as a fundamental philo­ sophical discovery in Hume. H U M E' S D I S COVERY

Let me attempt now to summarize in a different but analogous vocabulary what is both Hume's theory of the self and his great discovery. 1 . We must distinguish the selffrom the body. Although we are a body and we cannot exist without a body, it is also the case that we have a body. As selves, we have both a mind and a body. 2. Although the selfcan know itself as an object. the body cannot know itself. The knowledge the self has of itself develops temporally. We may always have been conscious, but we were not always self-conscious. Knowledge of the self is something which comes to be. S elf-knowledge emerges in retrospect.

S I G N I FI CA NCE OF H UM E ' S THEORY O F TH E S ELF

279

3. Since the self can know itself as an object, the self can respond to its own activity just as others respond to its activity. That is, I can come to view myself from a social perspective. I can come to know myself through others. This characteristic of self-awareness is exactly what H ume stresses in his discussion of the indirect passions of pride and humility which give rise to the idea of the self, and it is as well the foundation of Hume's theory of sympathy and morality. 4. This capacity to respond to its own activity, permits us to distinguish within the self between the '[' and the 'me '. In Hume, it is the distinction between 'self' and 'idea of the self'. The T is different from the 'me', but there is also an interaction between the two. 5. What is this 'F which is aware of the social 'me'? It is, first of all, rooted in bodily activity. This is why among other things the 'me' is the object of the indirect passions, but the T is the mysterious "original and natural instinct" (p. 286). This is also the locus of the problem of inhesion. 6. The '[' cannot know itself as an object. This is Hume's discovery. It can never be discursively characterized. Yet it is a necessary precondition of all knowledge and action.7 7. Although the 'I' cannot know itself, the 'I' can know the 'me'. I can know my self The self as me, the idea of the self, may be an object of analysis. The 'I' cannot be an object of analysis. No analysis of the self as 'me' should be confused with an analysis of the 1 Hume never expressed dissatisfaction with his analysis of the self as 'me'. What he despaired of, and rightly so, is giving an analysis of the T. One of the reasons that the T cannot know itself is that it is embodied. If the T were a disembodied mind, then we could have ultimate selfconsciousness in which the distinc­ tion between subject and obj ect disappeared. There would then be no difference between 'looking at' and 'looked at'. Ironically, it is embodiment that limits understanding. 8. The 'J' reveals or shows itself in unfolding action. It undergoes develop­ , ment. There is thus always more of the '! to come. This is another reason why it cannot be an object of analysis. On the other hand, the developing , or evolving of the '! allows for the self to know itself (as 'me'). It also explains why self-knowledge is historical. If the T were a disembodied teleological unfolding as opposed to an evolving embodiment then the process would be finite, and we would come , finally to know the '! . Hegel would be right. But Hume is not Hegel. If Hume is correct, then there would be a unique function for philosophy as the historical (but non-teleological) awareness of the evolving '1'. '

'.

280

N I CH O L A S CAPALDI

9. If the 'I' cannot know itself, then there is a permanent paradox: How can 'I' recognize that it is 'I' of whom 'I' am conscious? The only answer would be that 'I' would have already to know myself in advance. But the 'I' cannot know itself. The 'I' can only know the social 'me' that has evolved temporally. Therefore, the 'J' is forever beyond conceptualization, beyond being an object. 1 0. Contemporary analytic philosophy is an extreme form of concep­ tualization, i.e., it advocates the possibility of the total conceptualization of the world (scientism). This necessitates that the subject be assimilated into the object. The absorption of the subject into the object, whether in terms of crude behavioristic psychology, or depth grammar, or sophisti­ cated philosophy of mind, is precisely what Hume would deny. So, Hume is in direct conflict with all versions of analytic philosophy. For Hume, the self as 'I' is not an objective idea. It emerges temporally and is not only independent of conceptualization but the ground of conceptualization. 1 1 . If Hume is right, we can appreciate why the so-called philosophy of mind is sterile. The philosophy of mind has transformed philosophy into the ahistorical analysis of propositions about the mind. Despite its best intentions, it ends up becoming behavioral. The mind, however, is not a logical structure. 1 2 . The mind, the '1', is the unity of consciousness. It is revealed to us only in the activity of a living, embodied individual. It is only in action (with reference to the passions) that the unity of consciousness is most vivid. All of Hume's direct statements about the unity of consciousness are in Book II, Of the Passions : " . . . that individual person, of whose actions and sentiments each of us is intimately conscious" (p. 286). It is only when we separate the mind from action and try to conceptualize its activity that we detach ourselves from it, and it becomes a mystery.8 1 3 . If the T is not amenable to analysis or conceptualization, can we say anything about it? The answer is yes, as long as we do not insist that all rational discourse is analytical discourse. The alternative way of speaking rationally about the 'J' isfrom the perspective ofcommon sense. Common sense presupposes both the unity and the certitude of the individual conscious­ ness. It is this common sense certitude and unity which is the ground of all intelligibility and all arguments. This common sense ground is neither alien nor peripheral to Hume's philosophic,!l enterprise. On the contrary it is central. The grounding in common sense is the announced Copernican Revolu­ tion in Hume's philosophy, it is the new foundation, it is why it is "pre-

SIGNIFICANCE OF HUME'S THEORY OF THE S ELF

28 1

sumptuous and chimerical" for philosophers to seek "the ultimate original qualities of human nature" (p. xvii), it is why it is impossible to explain "ultimate principles" (p. xviii). It is the common-sense perspective that links action with the refutation of scepticism ; that permits Hume to say that "'tis in vain to ask, Whether there be body or not? That is a point, which we must take for granted in all our reasonings" (p. 1 87); that explains why the vulgar are correct and why false philosophy is parasitic upon the vulgar ; that explains why Hume does not explain the social world but presupposes it;9 and finally, why in the Enquiry he can say that philosophy is "nothing but the reflections of common life, methodized and corrected" (EH U, p. 1 62). In order to make sense of the ground of analysis we require a kind of philosophical psychology, a philosophical anthropology, a philo­ sophical sociology, and a philosophical history. To mistake this for psy­ chologism or the genetic fallacy is the fundamental ideological distortion of analytic philosophy. Understanding is, for Hume, an organic, social, and historical activity. As such it itself cannot be understood analytically and incrementally. Organic activities, unlike mechanisms , grow and exist in a state of exchange with their environment. That is why it is only through action, not thought alone, that we discover who we are and add to what we are. "Understanding, when it acts alone, and according to its most general principles, entirely subverts itself." (p. 267) In the course of such ex­ changes, organic activities or entities transform themselves. They do not exist j ust in space, but qualitatively transform themselves through time . 1 O Thus, even momentary logical structures are intelligible only by reference to the history of structural transformations. The historical origin and de­ velopment of any part is always a relevant and crucial condition of its meaning. Time, as Hume says, is the source of identity (p. 200). H I S TORICAL S I G N IF I CANCE

We can draw from this discussion a lesson about philosophical scholar­ ship. Hume has been seriously distorted by his analytic readers. Analysts accept the idea of the self, but they deny the pre-cognitive self. Analysts want to claim that Hume belongs to them. They want to see him as someone who believes that philosophy can proceed without reference to a subject or knowing mind. 1 1 If one wanted to attribute this position to Hume then one would have to find in Hume (a) the denial of an extra­ conceptual self, (b) an insistence that the self and the idea of the self are

282

NICHOLAS CAPALDI

the same, and (c) the denial that self-consciousness emerges in a temporal process. Attributing the bundle theory to Hume served all of these pur­ poses.12 As we have seen, all three of these ascriptions are false. Rather than denying an extra-conceptual self, Hume discovers it. What he denied was a simple, continuous, disembodied Cartesian self. It is Hume's analytic readers who fail to distinguish when Hume is speaking about the self and when he is discussing the idea of the self. These are the same readers who ignore Book II because it is there in the discussion of the passions that Hume insists on the temporal and emerging character of the idea of the self as well as the unity of consciousness. Curiously, the pre-cognitive self as the unity of consciousness is not used by Hume for any purpose suspect to analysts. Hume does not argue for a soul which survives the death of the body, he does not appeal to intro­ spection as the privileged source of incorrigibly clear and distinct ideas, and he certainly did not think that the pre-cognitive self was the unambiguous moral conscience of Protestantism. So why does he need it? Analysts are at a complete loss to comprehend it. 1 3 What they cannot see is that Hume has a different conception of philosophy. P H I L O SOPH I CAL S I GN I FICANCE

The historical distortion of H ume is the reflection of a fundamental conflict over what philosophy is. Analysis is committed to total conceptualization. This means that the self becomes a self-conscious concept. How have they attempted to show this? They have tried three approaches : behavioral reduction, transcendental argument, and the criteriological approach. Let us look briefly at each of these. It is now admitted that the behavioral approach is a failure because we cannot give an extensional or empirical analysis. The transcendental approach, both Kantian and Strawsonian, fares no better. Transcendental arguments are themselves intelligible only from within a common sense perspective. Moreover, any attempt to articulate definitively the objective necessities within which our linguistic options are exercised presumes that such logical structures are impervious to change. There are no such consensual analyses, and if there were it would remain an unexplained intellectual miracle. The criteriological approach is the most instructive. Criteriologists 1 4 seem to agree that if one criterion for a concept is dropped, then the concept changes. On the other hand, they allow that if new or additional criteria automatically change the concept, then no analysis is possible. So

S IGNIFICANCE OF H UME'S THEORY OF THE S ELF

283

they have compromised on the view that some additions are acceptable without changing the concept. But here is precisely where the serious philosophical issues appear. If some changes are acceptable, then how do we understand the process of addition? This is just a linguistic version of Hume's original question about self-identity. We have changed the idiom but not the problem. If the addition process is incremental or closed, then we should be able to specify criteria for criterion change. Only Hegel claimed to be able to do that. If the process is open, then we cannot conceptualize the process of conceptualization. Hume's genius lay in making precisely this point. The process can only be explained historically. The philosophical significance of Hume's theory of the self is that it discloses what philosophy can and cannot be. If Hume is correct, then the history of philosophy is integral to philosophy itself. If our conception of ourselves is historical then this necessitates a continuous reinterpretation and reintegration of the history of philosophy. It really is important to get the history right, for whoever prevails in the historical debate decides both what we think we are doing and ultimately what we do. S UM MA R Y

The unity of consciousness is the recognition of ourselves as a single person with diverse experiences. Such a unity is a necessary condition of personal identity. The recognition of this unity emerges in time, that is it comes to be, as the result of our ability to relate present experience to past and future experience. This ability, in turn, is revealed to us in action. Our unity of consciousness is given to us only in our role as agents not as cognizers. Self consciousness in the sense of the unity of consciousness is never consciousness of a self. The unity of consciousness is therefore never formalizable. Personal identity is our recognition of ourselves as persons. A person is a being conscious of its identity through time, that is a self-conscious being. A self-conscious being is one who grasps the unity of his consciousness. Since this unity is not formalizable, we can never state the sufficient conditions of our identity. Every temporal process is intrinsically incom­ plete. It cannot be captured by a static logic that is merely spatial in character. This is what accounts for the specious intractability of the mind-body problem. Our unity of consciousness exists in time, but there is no exact space in which to locate it.

284

NICHOLAS CAPALDI

The only way in which adequate criteria of personal identity can be achieved is artificially, in Hume's sense of the fictional. Such criteria are the result of negotiation about our past history. Only persons possess a history which enables them to compare what they are with what they have been. The issue of personal identity is thus historical, social, and moral.

Queens College, The City University of New York

NOTES I All page references are to the Selby-Bigge edition of the Treatise, 2nd edition, revised by P. H. Nidditch (Oxford: Clarendon, 1978), and, where noted, of the Enquiry Concerning Human Understanding, 3rd edition, revised by P. H. Nidditch (Oxford: Clarendon, 1975). 2 �What is the soul of man? A composition of various faculties, passions, sentiments, ideas; united, indeed, into one self or person, but still distinct from each other", Dialogues, Part IV, p. 1 59 of the Kemp-Smith edition (Indianapolis : Bobbs-Merrill, 1 947). 3 See also Treatise, p. 286. No act has a structure. Intuition or the noting of resemblances is an act. Therefore, intuition has no structure. Whatever lacks a structure cannot be analyzed. Therefore, intuition and the noting of resemblances cannot be analyzed. This is Hume's argument in favor of the pre-cognitive self. 4 For an excellent history of the discussion of the unity of consciousness prior to Hume and by persons wIth whose work Hume was familiar see Ben Lazare Mijuskovic, The Achilles of Rationalist Arguments (The H ague: Nijhoff, 1974). 5 �We must separate the question concerning the substance of the mind from that concerning the cause of its thought" (p. 248); �" . nor have we the most distant notion of the place, where these scenes are represented, or of the materials, of which it is compos'd" (p. 253). Hume twice scored out the following passage from the Dialogues: �it is perfectly indifferent, whether we rest on the universe of matter or on that of thought; nor do we gain any thing by tracing t.he one into the other" (op. cit., pp. 1 60- 1 6 1 ). 6 . the relations of resemblance, contiguity, and causation, as principles of union among ideas, without examining into their causes . . . 'Twou'd have been easy to have made an imaginary dissection of the brain . . . " (p. 60). 7 This is a point that Hume makes repeatedly at the beginning of his discussion of the passions. See pp. 276, 280, 286-287 of the Treatise. S The unity of consciousness is mentioned on pp. 277, 286, 307, 320, 339, 340, 354, and 427. 9 The relation between the social world and the pre-cognitive nature of the self can be seen in the following: When I am at a loss to know the effects of one body upon another in any situation, I need only put them in that situation, and observe what results from it. But should I endeavour to clear up after the same manner any doubt in moral philosophy, by placing myself in the same case with that which I consider, 'tis evident this reflection and premeditation would so disturb the operation of my natural principles, as must render it impossible to form any just conclusion from the phenomenon. We must therefore glean up our experiments in this ""

SIGNIFICANCE OF H UME'S THEORY OF THE S E L F

285

science from a cautious observation of human life, and take them as they appear in the common course of the world . . . " (p. xix). 1 0 Hume rejected complete materialism, and claimed that some things exist without a place. Russell and Carnap are well known examples. See also Karl Popper, 'Epistemology without a knowing subject', in Objective Knowledge (Oxford: Clarendon, 1972). 12 Those who insist upon seeing Hume as a phenomenalist will never get beyond this point. For an antidote read Robert F. Anderson, Hume 's First Principles (Lincoln: Nebraska University Press, 1 966) and Nicholas Capaldi, David Hume: The Newtonian Philosopher (Boston: Twayne, 1 975). 1 3 A.J. Ayer, Hume (Oxford: Oxford University Press , \980), p. 5 3 ; John Bricke, Hume's Philosophy of Mind (Princeton: Princeton University Press, 1980), p. 9 5 ; T. Penelhum, 'Personal Identity', in the Encyclopedia of Philosophy (London: Collier-Macmillan, 1 967), Volume VI, p. 100; P. F. Strawson, Individuals (London: Methuen, 1 959), p. 1 32; B. Stroud, Hume (London: Routledge, 1 977), pp. 1 24, 1 27. 1 4 By criteriologists I mean Wittgenstein in the Investigations, Malcolm in Dreaming, and Shoemaker in Self-Knowledge and Self-Identity. II

ECKART F O R STER

KANT' S R E F UTATION OF I D E A L I S M

The second edition of the Critique of Pure Reason contains , according to Kant, only "one addition, strictly so called", namely a "Refutation of Idealism". (Bxxxix) 1 It is this 'addition' which I want to examine in the present essay. Prima facie its very existence must astonish. For as every student of Kant knows, Kant's solution to the "proper problem" (B 1 9 ) of the Critique, that is, of how synthetic judgments are possible a priori, itself depends on an endorsement of idealism. On this point, there is no obscuri­ ty in Kant: " . . . the possibility of synthetic knowledge a priori . . . was properly the problem on the solution of which the fate of metaphysics wholly rests and to which my Critique . . . was entirely directed. The ide­ alism . . . was . . . taken up into the system as the sole means of solving that problem (although it then also received confirmation from other grounds)".2 Why, then, should Kant include in the second edition of that work, indeed as its only addition, a proper refutation of idealism? It is this question I want to pursue. But not only it. For both the full force and the strangeness of the argument of the Refutation can only be really appre­ ciated, I believe, if it is compared with Kant's earlier attacks on idealism, and if its development is also followed through some of his later and perhaps less well-known writings. In Section 1 of this essay I shall thus try to sketch the kind of problems that led to the formulation of the Refutation. Section 2 will be concerned with Kant's attempts to integrate its conclusion into his critical theory. I . A REFUT A n O N OF I D E A L I S M ?

Although Kant's earliest attacks on idealism date from his pre-critical period, I shall ignore his treatment of the subject prior to the first Critique 3 , and begin with the argument set forth in the fourth Paralogism in A. An idealist (empirical) is here presented as someone who, although not denying the existence of external objects of the senses, contends that we can never, by way of any possible experience, be completely certain as to their reality. (Cf. A368f.) For he holds that we only immediately perceive 287 A . J. Holland (ed.). Philosophy. Its History and Historiography. 287-303. © 1 985 by D. Reidel Publishing Company.

288

ECKART FORSTER

what is in us, namely our ideas or representations. The existence of external things can thus only be inferred from our inner perceptions, that is, if we take these perceptions to be the effects of some external causes. However, since an effect may be due to more than one cause, the inference from a given effect to a determinate cause must always remain doubtful. We can never know for certain, Kant's idealist alleges, whether "all the so-called outer perceptions are not a mere play of our inner senses, or whether they stand in relation to actual external objects as their cause". (A368) Kant here presents Descartes as the typical representative of this philo­ sophical position - a classification which already struck his contem­ poraries as eccentric. Descartes hardly believed that knowledge of the external world must remain uncertain. After all, he had offered, in the Meditations, proof to the contrary. It must have been Kant's conviction that the proof was utterly useless (because dogmatic) which made him treat it as if it were non-existent. At any rate, it is clear that Descartes held that all knowledge of external objects depends on prior proof or inference, and it is precisely this allegation which Kant here attempts to rebut. As he puts it : "In order to arrive at the reality of outer objects I have just as little need to resort to inference as I have in regard to the reality of the objects of my inner sense, that is, in regard to the reality of my thoughts". (A37 1 ) To reach this result, Kant only needs to draw on the Transcendental Aesthetic and its conclusion that space (and time) are merely sUbjective forms of sensibility, nothing in themselves , and that "what we call outer objects are nothing but mere representations of our sensibility, the form of which is space". (A30; cf. A38) Since I am immediately conscious, as the Cartesian concedes, of all representations, and since the objects of outer sense are no less representations than those of inner sense, their reality, their 'immediacy', is on a par. Being representations, these 'obj ects' cannot exist apart from our consciousness. Their reality hence is not inferred, but immediately perceived. Thus, the transcendental idealist or dualist, as Kant calls himself here, may admit the existence of matter without going outside his mere self-consciousness, or assuming anything more than the certainty of his representations , that is, the cogito, ergo sum . Matter is with him . . . only a species of representations (intuition), which are called external, not as standing in relation to objects in themselves external, but because they relate perceptions to the space in which all things are external to one another, while yet the space itself is in us. . .

(A3")O)

KANT'S R EFUTATION OF IDEALISM

289

This argument of the fourth Paralogism, it will be noticed, is not a refutation of idealism in any strict sense. For if Kant succeeds in under­ mining sceptical idealism, he only does so by invoking a stronger version of idealism, and by denying what Descartes doubts , namely the indepen­ dent existence of matter. Appearances in outer sense, Kant concedes, have indeed "this deceptive property that, representing objects in space, they detach themselves as it were from the soul and appear to hover outside it". But by overlooking that "the very space in which they are intuited is itself nothing but a representation" (A385), the Cartesian succumbs to the deception and attributes an independent reality to what is merely a modifi­ cation of our mind, namely, matter. Interestingly, Kant concludes his discussion of the fourth Paralogism by contrasting with the sceptical idealist who doubts the existence of matter, the 'dogmatic' idealist who denies its existence. Since in Kant's subsequent writings Berkeley always appears as the dogmatic idealist, it will be instructive briefly to contrast their respective positions on the point at issue.4 The similarities are indeed striking. Both philosophers deny the reality of matter as an existent independent of the mind. Berkeley's esse est percipi is nicely echoed in Kant's claim "that there is nothing in space save what is represented in it" - a "paradoxical but correct proposition", as he eagerly adds, to which "we must give full credence". (A374a) Consequently, when Berkeley insists that what we call "real things" is not something without the mind but what is "perceived by ,, my senses 5 , Kant confirms this by s aying that all outer perception not only yields proof of something real in space, but "rather is the real itself'. (A375) Both philosophers also agree that, since matter is nothing but an idea or representation, its reality need not be inferred but can be known immediately. In fact, when Kant summarizes his position on this point by saying that "external things exist as well as I myself, and both, indeed, upon the immediate witness of my self-consciousness" (A37 1 ), he is in effect restating a point Berkeley had already made in the third Dialogue: "I do . . . assert that I am as certain as of my own being, that there are bodies or corporeal substances (meaning the things I perceive by my senses)". (II, p. 238) There are, of course, important respects in which Kant and Berkeley differ. Whereas Berkeley, for instance, attributes outer perceptions to God as their source, Kant assumes unknowable things-in-themselves to be their causes. In a transcendental sense, these may thus be said to be really outside us. However, Kant is quick to point out that these are not the

!

290

ECKART FORSTER

objects of which we are thinking in the representations of matter and of corporeal things. (Cf. A372f.) But since "the expression 'outside us' is thus unavoidably ambiguous" (A373), Kant goes out of his way to distinguish "empirically external objects", of which alone he is here speaking, from those which may be said to be external in the transcendental sense, by "explicitly entitling the former 'things which are to befound in space"'. (A373) But exactly what reasons does Kant have for assuming things-in-them­ selves in the first place? He has available a number of arguments for the ideality of space and time ; e.g., the argument of the Aesthetic, of the �ntinomies, or the argument from incongruous counterparts. But what !�rguments are there for the existence of things-in-themselves? As far as I !tan see, there are none. None, that is, if we are not to take as an argument Kant's assertion that it "follows" from the concept "of an appearance in general" that there must also exist things-in-themselves which appear.

I

(A25 1 ; cf. A30) Although later repeated many times (cf., e.g., Prolegomena, § 32; Bxxvif. ; Reflection 5652; etc.), this contention will not do. For, as Kant himself pointed out, the backward inference from an effect (i.e. the appearance) to its cause cannot be certain. More precisely: it cannot establish that the cause is external (cf. A368), nor that it must be a thing (pace Berkeley). The point Kant had in mind can of course be put more sympathetically. For if, as he argues, all empirical knowledge is the outcome of a co­ operation of sensibility and understanding, and if the former faculty is to be regarded as completely passive, a mere "receptivity for impressions" (A50), then it does indeed seem to "follow" that we must also posit a "non-sensible cause of these representations", if "merely in order to have something corresponding to sensibility viewed as a receptivity". (A494) But it is clear, I think, that this line of argument, by itself, would not convince a sceptic regarding Kantian things-in-themselves, regardless of whether he was of Cartesian, Humean, or Berkelean persuasion. I shall come back to this point. For the moment it will suffice to notice that Kant's polemic against idealism, in the A-edition of the Critique, centres around empirical obj ects in space, not things-in-themselves. What might be called the 'second stage' of this polemic takes place in the Prolegomena. By now, the situation has changed considerably. Suddenly Berkeley has become the main target of attack, and idealism is presented as the doctrine which (a) denies the existence of things-in-them­ selves (cf. Prol. , p. 288f.), and (b) asserts that all knowledge through the senses is mere illusion (p. 374). And we find Kant now eager to show that

KANT'S REFUTATION OF IDEALI SM

29 1

his own critical position is the very 'opposite' of this idealism. ( Cf. pp. 288, 374 ) The external reason for this shift of emphasis is well known. While working on the Prolegomena, Kant received a first, eagerly awaited review of his Critique 6, which characterized his book as a "system of higher idealism"; an idealism, that is, which "equally comprises spirit and matter", and which "transforms the world and us into representations". The review was bound to leave the uninformed reader under the impression that Kant denied all reality behind representations and merely proposed a variation of Berkelean immaterialism. ? Kant was not amused. To liken his doctrine to the "mystical and visionary idealism of Berkeley", that "chimera of the brain [Hirngespinst]", was to Kant an "unforgivable and almost deliberate misunderstanding". (Prol. , pp. 293 , 290) In three Notes (§ 1 3 ) and an Appendix to the text of the Prolegomena, Kant tried to set things into perspective again. How does he dissociate himself from Berkeley? As for (a), the denial of things-in-themselves, Kant explicitly states that "there are bodies outside us, i.e. things which although wholly unknown to us as to what they may be in themselves , we know through the representations which their influence on our sensibility provides for us and to which we give the name of bodies". (p. 289) Again, there is no real argument for the existence of things-in-themselves; in fact, Kant here concedes that it "has never entered my mind" to doubt their reality. (p. 293) The admittance of things-in-them­ selves is simply put forward as the distinguishing mark of his own 'critical' brand of idealism. Kant's dealing with (b) is less clear. His objective is to counter all allegations that his idealism reduces experience to mere illusion, by showing that it is Berkeley who cannot distinguish truth from illusion, or dream from reality, and that it is his own transcendental idealism which, on the other hand, is the sole means to secure the employment of these all-important allied distinctions. Berkeley, we are told, has "no criteria of truth". (p. 375) This is, of course, false. In fact, when we compare the views of both philosophers on this subject, we again notice a striking similarity. Berkeley, like Kant, not only emphasizes the importance of the distinction; he also insists that it can be drawn within the realm of ideas : "There is a rerum natura, and the distinction between realities and chimeras retains its full force . . . But then they both equally exist in the mind, and in that sense are alike ideas". (Principles, § 34) It is not in the reference of ideas to something

292

ECKART FORSTER

outside the mind that their objective reality consists, but in their non­ random connectedness, their lawlike succession "according to the fixed order of Nature". For herein, says Berkeley, "consists that constancy and truth of things , which secures all the concerns of life, and distinguishes that which is real from the irregular visions of the fancy". (Third Dialogue, II, p. 258) Now this is precisely the view Kant holds, too: "The difference between truth and dreams", he explains, "is not decided by the nature of the representations that we refer to objects, for they are the same in both, but by the connection of these representations according to the rules that determine the combination of them in the concept of an object, and how far they can stand together in an experience". (Prol. , p. 290) Real, for both philosophers, is not what can exist independently of the mind, but what stands in connection with a perception in accordance with the laws of empirical advance. (Cf. A493; Principles, § 35) Why, then, should Kant believe that Berkeley's idealism, unlike his own, cannot distinguish truth from illusion, if they both agree that it is not in mind-independent external objects that this distinction has to be grounded? The explanation occurs in the 'Appendix' to the Prolegomena; in view of its significance, I quote in full. Berkeley regarded space as a merely empirical representation which, like the appearances in it, only becomes known to us, together with all its determinations, by means of experience or perception; I on the contrary first show that space . . . with all its determinations can be known a priori because it, as well as time, is present in us before all perception or experience as pure form of our sensibility, and makes possible all intuition of sensibility, and hence all appearances. From this it follows : that as truth rests on universal and necessary laws as its criteria, experience with Berkeley can have no criteria of truth because nothing was laid (by him) a priori at the ground of appearances in it, from which it then followed that there was nothing but illusion; whereas for us space and time (in conjunction with the pure concepts of the understanding) prescribed their law a priori to all possible experience, and this yields at the same time the sure criterion for distinguishing truth in it from illusion. (Prol. , p. 374f.)

This argument is hardly very clear as it stands, and I doubt that Berkeley would have been impressed by it. Why should 'merely' empirical, or hypo­ thetical, lawlikeness and illusion be one and the same thing? In the earlier note III to § 1 3 Kant had stated in all clarity that illusion arises if space and time are taken as properties that belong to things-in-themselves ; and B erkeley certainly is not guilty of that confusion. After all, Kant explicitly admits that he is "of one confession" with Berkeley on this point. (p. 374)

KANT'S REFUTATION OF IDEALISM

293

Their disagreement concerns the origin of space and time onl)'. Whereas Berkeley contends that it is em!llilcar-Kint--is- confide�t that he has demonstrated its a priori nature in the Aesthetic. Yet even if he were right on this point, would it follow that Berkeley can have no criterion to distinguish truth from illusion? It is difficult to see why this should be so. I shall return to this point later. For the moment we may notice that, in the second edition of the Critique, Kant returns to his claim that illusion arises from taking space and time as properties of things-in-themselves (cf. B70), and again accuses Berkeley of degrading bodies to mere illusion. (Cf. B 7 1 ) I take this as an indication that Kant, at this stage, has not yet found a satisfactory way of distinguishing his position from that of Berkeley.s In any event, Kant's efforts in the Prolegomena to dissociate himself from material idealism were in vain. In spite of his disclaimers, subsequent interpretations or criticisms largely followed the tenor of the Garve/Feder review and interpreted Kant idealistically, as denying the reality of things­ in-themselves. The doctrines of the Aesthetic and Analytic, thus the prevailing opinion, rule out all knowledge of things other than appear­ ances, hence also knowledge of their existence. On the other hand, their existence, and especially their causal influence on the subject, are pre­ supposed. The famous complaint of Kant's contemporary Jacobi, viz. "that without this presupposition [of things-in-themselves] I cannot enter the ,, [Kantian] system, and with this presupposition I cannot remain in it 9 , IO nicely summarizes a widely shared reaction to Kant's Critique. The need for a second edition thus gave Kant a welcome chance to clarify his views and to counter misunderstandings. The common element in all alterations of the text is, not surprisingly, a playing down of idealistic overtones, and the emphasis of realism in his system. These efforts culminate in a proper 'Refutation of Idealism', now placed (more aptly) as an annex to the second 'Postulate of Empirical Thought in General'. (B274ff.) This argument, to which I shall now turn, is remarkable in many respects. Given the reactions Kant had encountered, we might expect him to clarify his relation to Berkeley, and to argue for the reality of things-in­ themselves. This is not really the case, however. Descartes, not Berkeley, is again the opponent; the existence not of things-in-themselves but of "objects in space outside me" is the target ; and again Kant tries to show that "outer experience is really immediate", not inferred. (B276, n. 1 ) But unlike in the A-edition, Kant no longer draws on the results of the Aesthetic; he produces a novel argument, focussing entirely on the condi-

294

ECKART FORSTER

tions of time-determination. The argument goes as follows. The Cartesian sceptic or idealist maintains that I have certain and immediate knowledge of my own states of consciousness, of my own existence in time, but that the existence of things 'outside me' is doubtful and uncertain. To which Kant replies; Consciousness of my own existence in time is inseparable from the con­ sciousness of its determinability in time. (Cf. B276, Bxl) Now all deter­ mination of time presupposes something permanent in perception. This permanent cannot be something in me, since it is through this permanent that I can determine my existence in time. Consequently, determination of time "is possible only through a thing outside me and not through the mere representation of a thing outside me". Thus, "the mere, but empirically determined consciousness of my own existence proves the existence of objects in space outside me". (B27 5 ) This argument calls for some elaboration. First w e must notice that the Refutation is an indirect proof. It is, more precisely, a reductio ad absurdum of Descartes' position, who gets the game he played turned against himself, "and with greater j ustice". (B276) This must astound us, for in the 'Doctrine of Method' Kant had strongly insisted that in transcendental philosophy it can never be permissible, "so far as synthetic propositions are concerned", to justify assertions by disproving their opposites. Since all transcendental proofs are carried on in what he calls the domain proper to dialectical illusions, where what is merely subjective in our represen­ tations often presents itself as being objective, indirect forms of proof are ruled out as impermissible. (Cf. A789fT./B 8 1 7fT.) Yet the Refutation, no OUbt, is an indirect proof. Moreover, its validity does not depend on the ruth of transcendental idealism. Indeed, this argument can perfectly well tand on its own feet. Unlike its ancestor in the Paralogism of the A-edition, this argument does not presuppose the doctrines of the Aesthetic but focusses entirely on the conditions of time-determination. And it is for precisely this reason, namely, that it only presupposes what Descartes himself had asserted, that this argument is so much more potent than its A-edition ancestor. But now it begins to look as though the argument is perhaps too potent to be handled easily by Kant. For if, as he argues, the determination of my existence in time "is possible only through a thing outside me and not through the mere .representation of a thing outside me", then this refutes Cartesian idealism. B ut it equally seems to refute Kant's own trans­ cendental idealism which requires "that what we call outer objects are

[

KANT'S REFUTATION OF IDEALI S M

295

nothing but mere representations of our sensibility". (A30/B45, my italics) In other words : either the Refutation of Idealism is also a refutation of transcendental idealism, or it does not refute idealism at all. And another point of great interest now begins to emerge. For if it is correct that the Refutation does not depend on, or presuppose, transcen­ dental idealism, then Kant has, unwillingly and against his own expressed conviction, shown that it is possible to establish synthetic a priori con­ clusions in philosophy without endorsing transcendental idealism. We have here, I take it, the first (and only Kantian) instance of a transcendental argument in the modern sense. 1 1 Of course, these conclusions were not Kant's. Transcendental idealism was not to be brushed away ; it had been supported by many arguments. Yet the realistic implications ofthe Refutation are not easily brushed away either. Or rather, the argument is not easily integrated into the Kantian system. What does it mean, for instance, if what we call outer objects are mere representations of our sensibility, that the permanent in perception, which is required for time-determination, cannot be "something in me since it is only through this permanent that my existence in time can itself be determined"? In view of this apparent contradiction, commentators have usually adopted one of the following two interpretations. The first group has interpreted the Refutation 'ontologically', as requiring that the 'permanent in perception' be ontologically distinct from any representation of a thing in me. The obvious problem with this interpretation seems to be that only things-in-themselves could be distinct in this sense. And they cannot, in any way, be encountered 'in perception'. Thus the second group, more subtly, has interpreted the Refutation 'phenomenologically', pointing out that 'representation' and 'object' can be taken, and are taken by Kant, in two ways, viz., as 'mere' modification of our consciousness, or as the intentional object of cons_<::iO\lsness, the phenomenal object as constituted6yThe categorial pri�ciples of the mind. On this view, it is suggested, there is no difficulty in distinguishing between a 'thing outside me' and a 'representation of a thing within me' without evoking things-in-themselves; the apparent contradiction disappears. This latter reading is undoubtedly in better accord with the general spirit of the Critique. However, it is one of the main theses of the present essay that Kant realized the ultimate inadequacy even of this interpretation of the Refutation - a realization which forced him to re-examine his argument several times. His reasoning will be discussed in Section 2.

296

ECKART FORSTER

That Kant was not happy with his own argument is clear from the Preface to the second edition where, in a footnote much longer than the actual proof, he again struggles with this argument. But it is clearer even from a number of short essays, the so-called Reflections 565 3-4 and 63 1 1 -6, which were written soon afterwards and which again are devoted to the problem of idealism. 12 Let us now see how the Refutation can be integrated into the Kantian system. 2. A KANTIAN PHENOMENOLOGY OF T H E CON S CI O U S N E S S OF REA L I TY

What exactly is the permanent in perception which is said to be pre­ supposed in all determination of time? Clearly, it cannot be a permanent representation. We do not have such a thing. And if, with Kant, we take matter to be the permanent, its representation is as transitory and variable as all our other representations. (Cf. Bxli) The permanence which we attribute to matter is consequently not obtained from outer experience but, Kant maintains, "presupposed a priori as a necessary condition of deter­ mination of time" (B278); it is "simply the mode in which we represent to ourselves the existence of things (in the field of appearances)". (A I 86/B229) In the Schematism chapter, consequently, permanence is presented as the schema of the category of substance. Like all schemata it is thus not a presupposition of time-determination but itself an "a priori determination of time in accordance with rules". (A I45/B 1 84) Yet how can the permanent be an a priori mode of determination, an act of the subject's own spontaneity, and at the same time be something outside the subject in relation to which it must determine its own existence in time ? 1 3 Whatever this permanent may be, i t is clear that i t must b e distinguished more properly than Kant has done in the first Critique from the subject's spontaneity in the determination of its own existence. In the Reflections , this problem is at the centre of Kant's interest. First he points out that only my transcendental self-consciousness in the repre­ sentation 'I am', which accompanies the spontaneity of all acts of thought but does not amount to knowledge of myself in time, is really immediate. Empirical self-consciousness requires consciousness of outer objects for its determination. Inner and outer experience are thus on a par; the determi­ nation of my own existence in time must be simultaneous with the determination of the existence of objects in space. (Cf. pp. 306, 6 1 4) Now in so far as I determine myself for the purpose of self-knowledge,

•

KANT'S REFUTATION OF IDEALISM

297

Kant argues, I can become empirically conscious of myself as determined. I can also, however, non-empirically, become conscious of myself as determining myself. 14 Yet if the determination of inner and outer experience must be simultaneous, it must likewise be possible to become conscious, non-empirically, of a permanent as outside me. This conse­ quently cannot be a representation of something in space, but only the consciousness of a determinability of my subject, in which I am not deter­ mining, not spontaneous, because what is determining is not in me. Or, as Kant puts it, "the possibility of representing in intuition things in space is based on the consciousness of a determination through other things, which means no more than my Original Passivity, in which I am not at all active". (p. 307) When Kant now considers the already familiar objection that the notion of a "mere passivity", and with it that of an object outside me, might be wrongly inferred "because we do not perceive the cause of the existence of a representation", we find that he can now deal with it easily: Now the permanent cannot merely be thought in the determination of time, and belong to the spontaneity of determination of self, for in that case it would not lie at the ground of time-determination. Consequently, it must be represented in relation to the mere receptivity of the mind, that is, in relation to something affecting which is different from me. And this representation cannot be inferred, but must be original. (p. 309) (my italics)

The subject's original passivity or receptivity, which in the first Critique played such a vital role without being argued for at all, is here as it were deduced from the 'highest point' of transcendental self-consciousness. The determination of inner sense, Kant maintains, requires for its very possibility a special sense, distinct from, and not itself part of, inner sense. It now begins to become clear, I think, at least in part, that Kant's peculiar argument against Berkeley which we encountered in the Pro­ legomena, might be rendered effective. Berkeley argues for an empirical origin of our representation of space, suggesting that it is derived from our ideas. All he allows for is, in Kantian terms, inner sense and the manifold representations it contains. Under this premise, however, empirical self­ knowledge would be impossible. It is irrelevant that Berkeley thinks some ideas are caused by God because he is not conscious of producing them himself. All this belongs to inner sense. Moreover - and this I take to be Kant's main contention - without an originally different sense, no genuine distinction can be drawn between the subject's own productions and some-

298

ECKART FORSTER

thing independent, between imagination and truth, dream and reality. For this reason, Berkeley can indeed have "no criteria of truth". Whether or not he is conscious of his spontaneity can no more count as a sufficient criterion than the orderliness of his ideas. "No one can merely have inner sense in order to know his inner states, and yet this is what idealism asserts." (p. 6 1 6) This becomes even clearer when Kant confronts the question of how we can know of the co-existence of things, given that our representations succeed one another in our apprehension. (Cf. pp. 6 1 2, 6 1 4, 6 1 6) The answer is familiar from the third Analogy : because in this case we can represent the manifold forwards and backwards. But whereas the Analogies were exclusively concerned with the rules for the relation of appearances in time, Kant now builds on this argument to show that it is impossible to derive spatial representations from inner sense alone. For if the apprehension can proceed not only from A to B, but also backwards , from B to A , it is necessary that A continues to exist. In inner sense, however, in which everything is successive, no backward apprehension can take place. "Hence, in order to determine my own existence in time, an intuition must be given in which the manifold is represented external to one another, and beside one another. An intuition, that is, which makes possible the representation of space." (p. 6 1 6) And it is in this represen­ tation of space, Kant now contends, that permanence "inheres intrin­ sically". (p. 308, cf. p. 6 1 4) It would be a mistake to assume, however, that Kant is here finally crossing the barrier of his transcendental idealism. This is not the case. Although he stresses that we must be conscious of appearances as of "real things" (p. 6 1 6), they are, after all, only appearances. And although space is regarded as the "real representation" of an external relation, it merely "proves", as Kant puts it, "a representation which is not related to the subject". That something is "outside me" means that there is something "which I have to represent in a different manner than myself" . (p. 309) Nevertheless, Kant's reflections have pushed him considerably nearer that barrier. For, instead of the two-fold distinction between transcendental and empirical self-consciousness from which we started, we now find Kant insisting on a three-fold distinction between ( 1 ) transcendental conscious­ ness of my existence in general; (2) empirical consciousness of myself; and (3) empirical knowledge of myself as a being determined in time. (Cf. p. 6 1 5) Self-knowledge, Kant declares here, can only be knowledge of "myself as a being that exists in a world". To which he adds, "I posit my

KANT'S REFUTATION OF IDEALIS M

299

own existence" in a world "for the sake of empirical consciousness and its possibility". (My italics) What does this mean? One of the following Reflections supplies further elucidation: "I myself am an object of my outer intuition in space. Without this I could not know my position in the world". (p. 620) And, in another Reflection: "First we are an object of outer sense to ourselves, for other­ wise we would not perceive our place in the world and could not intuit ourselves in relation to other things". (p. 6 1 9) In other words, if I am to determine my own existence in time, I have to constitute myself as a corporeal being which can stand in spatial relation to something outside me. Kant's Refutation of ldealism, once some of its dazzling consequences are thought through, not only shows the need for the existence of "things in space outside me", but equally for my own existence as a corporeal being in space, i.e., as an object for outer sense to which things can be external. With these Reflections we have virtually reached the end of Kant's explicit polemic against idealism. We have not yet, however, quite reached the end of our story. First of all, we may ask, how is the self-positing of the subject as an object of outer sense to be understood? What is the relation between transcendental and empirical self? How is self-affection related to affection through outer objects? The problem of perception suddenly emerges with new urgency. And space itself has to be recon­ sidered, for, as something which has to be 'outside' the empirical self, it cannot merely be thought of as a form of intuition. Fortunately, there is no need to speculate what views Kant would have taken regarding these problems, for he considers them (together with many other problems) in what has become known as Kant's Opus postumum. 1 5 The context there i s o f course a different one. I n the Opus postumum, Kant attempts to work out a transition from the metaphysical foundations of natural science to physics. He starts, as it were, from the other end, i.e., from matter as a system of moving forces, and works his way backwards, towards the subject. In the present context it would be impossible even to summarize Kant's procedure in the Opus postumum without intolerable superficiality. I shall thus merely take up those points which connect with our problem at hand, and which may help to bring out what price Kant has to pay for an integration of the Refutation into his system of transcen­ dental idealism. We start with the concept of space. Space, we have noticed, must be something external and sensible. As such it must be full, as Kant insists, for empty space is not a possible object of experience. The space we know

300

ECKART FORSTER

from the first Critique, however, the mere form of intuition, "is neither empty nor full". (Opus postumum, I, p. 588) Like all obj ects of sense, space thus has to be taken in a two-fold sense, as something thinkable [denkbar, cogitabile] , and as something sensible or perceivable [spiirbar, dabile] . And because experience of outer objects is possible only if space is also sensible, Kant now declares: "The realization of space itself as a single object of sense, i.e., of empirical intuition, is the principle of the possibility of all experience". (I, p. 564, my italics) To become an object of sense, space must thus be thought of as filled with a manifold of moving forces which affect the subject and make per­ ceptions of outer sense possible. The definition of perception as an intuition combined with consciousness, which Kant had provided in the Critique (cf. A l 1 9), no longer suffices. From the subject's point of view, p erceptions of outer objects are effects: "Outer experience itself depends on the moving forces of matter which affect the subject". (I, p. 573 ) But this apparent realism must not mislead us. For Kant, objects and their moving forces are not given empirically. It is the central theme of his transcendental theory, indeed the condition of its very possibility, that experience is not given but has to be made, has to be constituted from the discrete elements found in an empirical intuition. To have experience of an object and to make the obj ect in the combination of its manifold are for Kant one and the same thing. (Cf. II, pp. 345 , 390) In the act of combina­ tion, the subject represents the object as an appearance, with the moving forces which are the cause of the perception directed towards the subject. As Kant puts it : "the perception of outer objects is nothing other than the act of the subject through which it affects itself'. (II, p. 392) For "the receptivity of appearances is based on the spontaneity of combination in the intuition of oneselF. (II , p. 535) We have here, it seems, reached an important point of contact between the argument of the Refutation, and the investigations of the Opus postu­ mum. For if spontaneity and receptivity are but two sides of the same coin 1 6 , we might expect that the synthetic acts the subject imposes on the manifold will be just those acts required for the empirical determination of its own existence. And indeed, we now find Kant describing "the empirical consciousness of oneself as the co-ordination of perceptions in accordance with the form of possible self-made experience". (II, p. 3 50) And, in another passage: "We cannot extract from our sense represen­ tations anything else but what we have put into them for the empirical representation of ourselves". (II, p. 343, my italics)

KANT'S REFUTATION OF IDEALI S M

301

But, a s w e have noticed earlier, i n order t o have experience, the subject must not only posit outer objects, it must also posit itself as an object of outer sense. And it is this doctrine of the subject's original act of self­ positing which marks the core of Kant's 'new' transcendental deduction in the Opus postumum. Schematically, the following picture emerges from the text. I start from the transcendental self-consciousness of my own existence. I then posit a pure manifold in the forms of space and time. Next I posit something as outside me through which I am affected. The understanding then combines the manifold to a unity and proceeds to its thoroughgoing determination. It is in this act of synthesis of the manifold that I constitute myself as an object in a unified world (cf. II, p. 4 1 8): "The thinking subject creates for itself . . . a world as an object of possible experience in space and time . . . In this world, moving forces, e.g. attraction and repulsion, are put without which there would be no perceptions". (I, p. 23) The subject is thus, with two Kantian terms, not only the "owner" of its representations, but also its "originator". (II, p. 82) The representations are not produced through objects, since objects are themselves the offsprings of our power of repre­ sentation and its synthesis : " Our sensible intuition is not at first per­ ception, as this presupposes a principle to posit oneself and to become conscious of the position". (II, p. 670) In the light ofthis revised theory, several fundamental concepts undergo substantial modification. Space and time, for instance, are now seen as "the actus of our power of representation positing itself' (II, p. 88); not things, but "a product of the power of imagination" (II, p. 76) which the "subject itself generates when it affects itself'. (II, p. 37) The following statement may almost be regarded as a summary of Kant's final position: "Apart from the logical consciousness of myself, I am objectively dealing with nothing but my faculty of representations. The position of something external originates from me in the forms of space and time in which I myself posit the objects of outer and inner sense". (II, p. 97) We can now, finally, come back to the question of things-in-themselves in the Kantian system. Not surprisingly, they are no longer thought of as real things of independent existence. The thing-in-itself is not a peculiar object, Kant explains, but only a negative determination of the thinking subject. (Cf. II, p. 36) It is a mere thought-entity without reality, "an ens rationis in the representation of which the subject posits itself'. (II, p. 36) Its sole function is to represent the obj ect of intuition as appearance. (Cf. II, p. 37) The only remaining thing-in-itself is the active subject: "The

302

ECKART FORSTER

subject is here the thing-in-itself because it has spontaneity". (II, p. 4 1 4f.) And : "The correlate of the thing in the appearance, the thing-in-itself, is the subject which I make into an object". (II, p. 4 1 2) We can here, I think, end our excursion into the realm of pure specu­ lation. There is no need, in the present context, to engage critically with any of Kant's final results. What has become clear, I hope, is that there is a direct route, not indeed the only route but nevertheless a route, which leads from Kant's original conception of a transcendental theory to the full-fledged idealism of the Opus postumum. Moreover, this form of idealism seems unavoidable, if the Refutation of Idealism which Kant presented in the B-edition of his Critique, is to be integrated into his transcendental theory. 1 7

Stanford University

NOTE S I References are to Kemp Smith's translation of the Critique, (London : Macmillan, 1 929), and are given with the usual AlB numbering. 2 I. Kant, Prolegomena, Prussian Academy edition (hereafter: Ak.), Volume IV, p. 377. I have used P. G. Lucas' translation of this work (Manchester: Manchester University Press, 1953). J Cf. I. Kant, Principiorum primorum cognitionis metaphysicae nova dilucidatio ( 1 755), Ak. I , p . 4 1 1 ; and D e mundi sensibilis atque intelligibilisforma etprincipiis ( 1770), Ak. lI , p. 397. Cf. also I. Kant, Vorlesungen aber die Metaphysik (Darmstadt: Wissenschaftliche Buchgesellschaft, 1975), pp. 99-1 03. 4 It is not clear that Kant had Berkeley in mind when writing this passage (A377). It may well be, as G. W. Miller has suggested ('Kant's First Edition Refutation of Idealism', Kant-Studien 62 ( 1 97 1 » , that Leibniz is the "dogmatic idealist" here. (Cf. also A. Kalter, Kants Vier/er Para/ogismus (Meisenheim am Glan: Anton Hein, 1975), pp. 206-208. In Kant's subsequent writings, however, this label is unfailingly reserved for Berkeley. 5 Second Dialogue, in The Works of George Berkeley. Bishop of C/oyne, Volume II, edited by A. A. Luce and T. E. Jessop (London: Thomas Nelson and Sons Ltd., 1 949), p. 2 1 5 . 6 By Garve and Feder, i n Gottingischen Gelehrten Anzeigen, 1 9 January 1 782, 40-48. 7 Also in 1 782, the Allgemeine Deutsche Bibliothek, Volume 52, 1 60-163, printed a review of a German translation of Berkeley's Dialogues which had appeared, like the Critique, in the previous year and had attracted considerable attention (cf. E. Stiibler, George Berkeley's A uffassung und Wirkung in der Deu/schen Philosophie bis Hegel (Zeulenroda: Bernhard Sporn, 1935), pp. 4 1 -43. Both reviews are very similar in tone, and in their (negative) evaluation of the idealistic system under consideration. In fact Feder, who apparently had read the Dialogues (cf. his Insti/utiones logicae et metaphysicae, ed. altera emendatior, Gottingae, 1 78 1 , p. 1 0 1 ), explicitly make s the connection with Berkeley in his review of Kant's Critique: "One of the pillars of the Kantian system is based on these concepts of sensations as mere modifications of ourselves (on which Berkeley, too, mainly builds his idealism) . . . ". Op. cit.,

KANT'S REFUTATION OF IDEALISM

303

p. 4 1 . 8 Kant's familiarity with Berkeley's works has been the subject o f much speculation. What seems at least clear now is that he came to Berkeley through Siris. Herder, who attended Kant's lectures in 1 762-1 764, took the following notes: " . . . Origin of idealism: the truth that the body without thoughts does not make a world [keine Welt ausmacht]. Thus Bishop Berklei, in the treatise On the Usefulness of Tar-water For Our Body, doubted whether there be bodies at all. He maintains (gibt vorl that all bodies are mere appearances of bodies in our soul. And that with much illusion". ( Vorlesung uber Metaphysik (Herder Nachschrifi), Ak. XXVIII, p. 42) Since the two German translations of Siris. both published in 1 745, only contained the medical sections, Kant, who did not read English, must have got hold of the French translation, Recherches sur les vertus de /'eau de goudron, published in Amsterdam in 1 745. ° F. H. Jacobi, Werke, Volume II (Leipzig: G. Fleischer, 1 8 1 5), p. 304. 1 0 For a survey of the period between the tirst and second editions of the Critique, cf. B. Erdmann, Kants Kriticismus in der I . und 2. A uflage (Leipzig: L. Voss, 1 878). 11 I assume here without argument that modern transcendental arguments aim at establishing synthetic a priori conclusions without any endorsement of transcendental idealism. An elaboration of this point can be found in my 'How are Transcendental Arguments Possible?', in Kant and Transcendental A rguments, edited by E. Schaper and W. Vossenkuhl (Cambridge: Cambridge University Press, forthcoming). 1 2 Cf. Ak. XVIII, pp. 306-3 10, 3 1 2- 3 1 3 , 6 10-623. The first two Reflections, probably dating from 1 788-89, may have been written in connection with Kant's intended reply to Eberhard. The rest is from autumn 1 790, and written for Kant's discussion with Kiesewetter of various important points of his system. (In what follows I shall, however, ignore Refl. 63 1 1 as its authorship is not certain. Cf. the editor's comments, pp. 607-6 10.) 1 3 Cf. also G. Lehmann, 'Kants Widerlegung des Idealismus', Kant-Studien 50 ( 1 958/59): 348-362, pp. 360-36 1 . 1 4 cr. Also B 1 53, where the synthesis of the understanding i s characterized as "the unity of an act, of which, as an act, it is conscious to itself, even without sensibility [Einer Handlung, deren er sich als einer solchen auch ohne Sinnlichkeit bewIij3t ist]". 1 5 Published as Ak. XXI, XXII. 16 "Positing and perception, spontaneity and receptivity, the objective and the subjective relation are simultaneous . . . and given a priori in the very same actus . . . " (II, p. 466). 17 I am grateful to Michael Ayers, Alan Montetiore, Peter F. Strawson and Mary Tiles for helpful comments on earlier versions of this paper.

P. B. WOOD

T H E H A G I O G R A P H Y OF C O M M O N S EN S E : D U GALD S T EWART' S ACCO UNT O F T H E L I F E AND W R I T I N G S O F T H O M A S R E I D

I n recent years the role o f the eloge in the cultural world o f eighteenth­ century France has been scrutinized by a number of intellectual and social historians. Developing his work on the provincial learned academies of the French Enlightenment, Daniel Roche has shown how the eloge was used by the Societe Royale de Medecine to create a sense of professional identity and to celebrate the value of dedication and self-improvement. Focusing on the eulogies written by the noted naturalist and Permanent Secretary of the Academie des Sciences Georges Cuvier, Dorinda Outram has analysed Cuvier's utilization of the eloge to convey moral lessons concerning the problematic relations between knowledge and power. Finally, Charles Paul has studied the eloges produced by successive secre­ taries of the Academie des Sciences in the eighteenth century, and has described how these furthered various didactic and ideological ends. I In this paper, I want to apply the historiographical approach of these scholars to a different national context, that of Enlightenment Scotland, and to examine in detail a specific text, namely Dugald Stewart's Account of the Life and Writings of Thomas Reid, which was read before the Royal Society of Edinburgh in 1 802 and published in 1 803. Stewart's A ccount invites such an examination for a variety of reasons. First, the institutional setting in which it was delivered was comparable to that of the French eloges. The Royal Society of Edinburgh self-consciously emulated the French academies in many respects, not the least of which was paying official tribute to deceased Fellows.2 Secondly, Stewart greatly admired the eloges of Fontenelle and Cuvier, and apparently modelled his own eulogies on French exemplars. 3 Hence the approach to the genre of the eloge sketched above suggests itself as an appropriate one to explore. Thirdly, and more importantly, the image of Reid created by Stewart conflicts with both the perceptions of contemporaries and the evidence of Reid's surviving manuscripts. Whereas Stewart pictured Reid as an epistemologist and moralist, the then Professor of Mathematics at Glasgow, James Millar, wrote in 1 795 that Reid was "peculiarly dis­ tinguished by his abilities and proficiency in mathematical learning" , while the Edinburgh Professor of Natural Philosophy, John Robison, referred to

305 A . J. HoI/and (ed.), Philosophy, Its History and Historiography, 305-322. © 1985 by D. Reidel Publishing Company.

306

P. B . WOOD

Reid as a "mathematician and naturalist of the first rank".4 These remarks are borne out by Reid's papers, for these demonstrate that he worked on a range of mathematical problems throughout his life, and that he also did research in electricity, optics, astronomy, natural history, and chemistry at various stages in his career. 5 Drawing on the insights of Roche, Outram, and Paul I shall argue in what follows that the discrepancy between Stewart's portrait and the biographical evidence available from other sources is explicable once we take into account Stewart's institutional environment, and his preoccupations as a leading academic and Whig in Edinburgh around the turn of the nineteenth century. Stewart intended his A ccount to serve two main didactic ends. First, his exposition of the scope and method of Reid's philosophy was to serve as an introduction to the study of Reid's works.6 Secondly, his biography of Reid was intended to convey a moral lesson. Reflecting on his efforts as a biographer in the conclusion of the Account, Stewart wrote: . . . I should not look back on the past with regret, if I could indulge the hope, that the facts which it has been my province to record - by displaying those fair rewards of extensive usefulness, and of permanent fame, which talents and industry, when worthily directed, cannot fail to secure - may contribute, in one single instance, to foster the proud and virtuous independence of genius ; or, amidst the gloom of poverty and solitude, to gild the distant prospect of the unfriended scholar, whose laurels are now slowly ripening in the unnoticed privacy of humble life.?

For Stewart, Reid's life provided a model of Smilesian self-improvement. It illustrated how a talented young scholar - the proverbial Scottish 'lad 0' pairts' - could achieve true success and j ust rewards through personal discipline and individual industry. Stewart conveyed his moral lesson by exploring the tensions between solitary study and social involvement. 8 According to Stewart, Reid had been "strenuously devoted to truth, to virtue, and to the best interests of mankind" ; yet he had also spent his life "in the obscurity of a learned retirement, remote from the pursuits of ambition, and with little solicitude about literary fame"Y Reid's success owed nothing to self-seeking attempts to gain patronage. His appointments were due to the "unsolicited favour of the two learned bodies who succes­ sively adopted him into their number [i.e. King's College Aberdeen and the University of Glasgow]", and his status amongst the literati was the "well­ earned reward of his own academical labours". Because Reid had remained aloof from the charms of polite society, he had been able to maintain "more eminently [and] more uniformly the dignity of philoso-

THE HAGIOGRAPHY OF COMMON SENSE

307

phy". \ 0 Stewart's ideal philosopher, as represented by Reid, was thus an isolated figure, pursuing his studies as a form of retreat from the realm of social and political relationships. This characterization of the ideal philosopher may have been the result of Stewart's political estrangement from the Royal Society of Edinburgh, and indeed from most sectors of Edinburgh society more generally, during the period 1 790 to 1 804 when the French Revolution affected every aspect of Scottish life. As Henry Cockburn recalled: "Everything rung, and was connected, with the Revolution in France ; which, for above 20 years, was , or was made, the all in all. Everything, not this or that thing, but literally everything, was soaked in this one event". I I Those Whigs who remained true to their principles became increasingly isolated both socially and intellectually, as the Tory political manager of Scotland Henry Dundas fanned fears of revolution amongst the landed and middle classes and thereby consolidated his control of patronage and places. Whig academics such as the Professor of Greek, Andrew Dalzel, were regarded with some suspicion by the public, but it was Stewart who aroused the deepest misgivings because of his lectures on political theory and political economy. Cockburn observed that although Stewart was "too spotless and too retired to be openly denounced, [he] was an object of great secret alarm", and Stewart's Whiggery meant that "for several years [he was] not cordially received in the city he adorned". 1 2 Stewart's continued approval of the writings of Tom Paine and Condorcet in 1 793-1 794, which were years of particularly intense ideological conflict, was, as we shall see, challenged by a senior member of the j udiciary, to whom Stewart found it expedient to apologize. Stewart's Whig principles and his ambivalence towards the French Revolution ill-suited the majority of his fellow members of the Royal Society of Edinburgh, since the Society had been dominated by Dundas and his retinue of the propertied elite of Edinburgh from its foundation in 1 78 3 . 1 3 Through the medium of a public discourse, Stewart was seemingly attempting to distance himself from the arena of political controversy, by emphasizing the apolitical aspects of .Reid's career, and hence of the philosopher's role.1 4 Two key passages in the A ccount support this interpretation. Stewart began his biography by presenting Reid's life as a welcome respite from the political turmoil in Europe. l s Later, he portrayed himself in terms of the idealised image of the philosopher which he painted in the A ccount. When paying tribute to his teachers in Edinburgh, Stewart wrote:

308

P. B. WOOD

It is with no common feelings of respect and of gratitude, that I now recall the names of those to whom l owe my first attachment to these studies [philosophy 1, and the happiness ofa liberal occupation superior to the more aspiring aims of a servile ambition. ' 6

It was precisely this opposition between philosophy as a liberal occupation and the servile ambition motivating men in society which Stewart else­ where employed in characterizing Reid's life. By drawing this parallel between himself and his subject, Stewart was therefore declaring his social and political disinterestedness. One might also see in his distaste for 'servile ambition' a thinly veiled attack on Dundas' creatures within the Royal Society, and on the nepotism rife during the regime of 'Harry the Ninth'. l 7 Indeed, Stewart's dislike of the patronage system may partly explain why he made no mention of the machinations of Lords Deskford and Kames to obtain the Glasgow Chair of Moral Philosophy for Reid in 1 764. This reading of the text gains further support from the fact that Stewart had earlier responded to political currents within the Royal Society of Edinburgh in his biographies of Adam Smith and William Robertson. In his life of Smith, which was read to the Society in early 1 793 and first published in its Transactions for 1 794, Stewart had deftly argued that the theories of political economy propounded by Smith and by the physiocrats in France "aimed at the improvement of Society . . . by enlightening the policy of actual legislators". Such speculations, Stewart assured his audience, "have no tendency to unhinge established institutions, or to inflame the passions of the multitude".18 Stewart later remarked that: . . . at the period when this memoir was read before the Royal Society of Edinburgh, it was not unusual, even among men of some talents and information, to confound, studiously, the speculative doctrines of Political Economy, with those discussions concerning the first prin­ ciples of Government which happened unfortunately at that time to agitate the public mind. '9

Thus Stewart's exposition of the doctrines contained in the Wealth of Nations in his portrait of Smith was intended as an oblique reply to those Tories who believed that the theory of free trade represented a 'revolu­ , tionary tendency .2o This exposition proved to be controversial. One of the Lords of Session, Judge Abercromby, expressed the view that by discussing the doctrines of political economy, Stewart had given countenance to, . . . opinions which, in these times, and under the circumstances in which we are now unhappily

THE HAGIOGRAPHY OF COMMON SENSE

309

placed, tend directly to destroy the peace and happiness of society, and to deprive us of everything that is valuable and dear to us in life.21

In reply, Stewart regretted that he had referred to the wntmgs of Condorcet, but pointed out that he had been at pains in his memoir not to imply that he approved of the revolutionary politics of Condorcet or his fellow philosophes, and that in his own lectures he had always warmly recommended the merits of the English Constitution.22 It is not surprising, therefore, that Stewart's life of William Robertson, which was read to the Royal Society of Edinburgh in 1 796, was a model of studied neutrality, a necessary expedient for a man of Stewart's political sympathies in the repressive years following the first sedition trials in Edinburgh in 1 793. 23 As leader of the Moderate Party in the Kirk and the Principal of Edinburgh University, Robertson was a well-known Whig, and

this posed a delicate biographical problem if Stewart wished to avoid any suggestion of political partisanship. Stewart may indeed have been alluding to this problem when he wrote to Archibald Alison that he "scarcely [knew] whose life [he] would not rather have written than Dr Robertson's".24 Consequently Stewart prefaced selections from Robert­ son's correspondence on the American Revolution by disclaiming any sympathy for the Principal's political sentiments as expressed in the quoted letters. 25 This disclaimer was well-advised since attitudes to the American Revolution were still of political significance. Robertson too was known as an initial supporter of the French Revolution. Alexander Carlyle recalled that Robertson had been "so much Dazzled by the Splendour of the French Revolution, That even his S agacity was Impos'd on, and he Could not Listen to the Ravings of Burke as he Call'd them". 26 On this subject Stewart maintained a discreet silence. Given that Stewart had been put on the defensive, it is therefore plausible to read his apolitical characterisation of Reid's life and career as a response to the politically hostile environment of Edinburgh and its Royal Society during the 1 790's and early 1 800's. In charting the evolution of Reid's thought, Stewart faced an intractable biographical problem. On the one han<;l he believed that "the history of [Reid's] opinions and speculations on those important subjects to which he dedicated his talents" would have been, biographically speaking, "the most interesting of all".27 Yet Reid had left no sustained autobiographical statement concerning the origins and development of his various intellec­ tual pursuits. Stewart's apparently straightforward solution to the problem of reconstructing his subject's intellectual biography introduced a syste-

3 10

P. B . WOOD

matic bias, for he restricted himself largely to evidence taken from Reid's published works, rather than exploiting the extensive collection of manuscripts to which he had access.28 Consequently Reid's creative work in mathematics , for example, was seen as being confined to the period prior to the publication in the Philosophical Transactions for 1 748 of his 'An Essay on Quantity', which Stewart interprets as marking a transitional stage in Reid's intellectual development. Of the 'Essay' Stewart wrote: "although [Reid] had not entirely relinquished the favourite researches of his youth, he was beginning to direct his thoughts to other objects".29 Stewart's narrative implied that from 1 748 until Reid's removal to Glasgow in 1 764, Reid was preoccupied with the epistemological problems dealt with in his An Inquiry into the Human Mind on the Principles of Common Sense ( 1 764), and that his scientific interests manifested themselves primarily in his teaching at King's College, Aberdeen. Stewart therefore spoke of a revival of Reid's "early passion for the mathematical sciences" after his move to Glasgow, and again after his retirement in 1 78 1 . 30 Thus although Stewart was able to document Reid's study of chemistry, mathematics, physics, physiology, and political economy from other sources, his concentration on the published works left him unable to relate Reid's studies in these areas to the problems of epistemology and moral philosophy discussed in print, and to assess the significance of these studies within the overall structure of Reid's thought. This narrative bias was reinforced by Stewart's use of the concept of 'progress' - a concept which should alert us to the fact that his biographical reconstruction is akin to conjectural history. As such, the reconstruction of the progress of Reid's thought bears an ambivalent relationship to Reid's actual intellectual development. A passage in Stewart's life of Adam Smith states this ambivalence : In most cases, it is of more importance to ascertain the progress that is most simple, than the progress that is most agreeable to fact; for, paradoxical as the proposition may appear, it is certainly true, that the real progress is not always the most natural. It may have been determined by particular accidents, which are not likely again to occur, and which cannot be considered as forming any part of the general provision which nature has made for the improvement of the race. 3 l

Thus conjectural history reconstructs the 'natural' or logical progress, rather than the inherently 'real' or historical progress, of the human mind. Moreover, Stewart conceived of the 'progress' of Reid's thought in terms of a strictly linear development. The years at New Machar ( 1 737-1 75 1 )

THE HAGIOGRAPHY OF COMMON SENSE

31 1

and King's College, Aberdeen ( 1 7 5 1 - 1 764) serve as the gestation period for the Inquiry, and as the prelude to the Glasgow years when Reid syste­ matised and extended his earlier discoveries and produced his Essays on the Intellectual Powers ofMan ( 1 785) and the Essays on the A ctive Powers of Man ( 1 788). Hence the relationship which Stewart saw between Reid's major works reflected a progressivist framework, for Stewart character­ ized the Inquiry as an 'introduction' to the two Essays.32 Reid's early studies in mathematics and the physical sciences thus fit neatly into Stewart's model of his subject's intellectual progress. Because Stewart regarded mathematics and Newtonian natural philosophy as an "early passion", he saw Reid's work in these areas as providing the methodological foundations for his speculations in moral philosophy, and Stewart repre­ sented Reid's later scientific studies as diversions from his central philoso­ phical concerns, occasioned by the stimulus of his colleagues, or, during his retirement, the need to keep his faculties active. 33 Widely held assumptions about the growth of the human mind may also have informed Stewart's reconstruction of Reid's intellectual career. In 1 755 Alexander Gerard defended the recently reformed curriculum at Marischal College on the grounds that it corresponded more closely to "the natural openings and progress of the human mind". He argued that because "the mind receives first of all impressions and ideas of those sensible things with which it is surrounded" and only later reflects on its own operations, it was necessary to invert the scholastic order of subjects and teach natural history and experimental philosophy before logic and pneumato!ogy. 34 Reid himself claimed that "there is a strong analogy between the progress of the body from infancy to maturity, and the pro­ gress of all the powers of the mind". Like Gerard, he believed that the external senses were the first of our faculties to mature and that reason was the last. Hence Reid thought there was "a natural order in the progress of the Sciences", beginning with the philosophy of body and culminating in the study of the mind. 3 5 Stewart's description of the progression of Reid's interests from natural philosophy to pneumatology and morals suggests that his narrative was partly structured by this conception of the unfolding of the powers of the human mind. 3 6 Inevitably, Stewart's academic role as a moral philosopher and his own philosophical preoccupations determined his assessment of the relative significance of the various facets of Reid's diverse intellectual pursuits. As a moral philosopher, he naturally identified Reid's 'favourite objects' of study as closely allied to his own. Striking too is the parallel between

312

P. B . WOOD

Stewart's career as, first, Professor of M athematics ( 1 775-1 785), and then Moral Philosophy ( 1 785-1 8 1 0) at Edinburgh, and his view of the shift in Reid's interests from mathematics to moral philosophy.37 Through his lives of Reid and Smith, Stewart defined his place within the Scottish philoso­ phical tradition by establishing an historical continuity between their phi­ losophies and his own. Moreover Stewart portrayed his own system as representing the next stage in the progress of Scottish philosophy, since he was systematising Reid's methodological contribution and synthesising the achievements of Hume, Smith, and Reid in the philosophy of mind. Biogra­ phy and history, therefore, served an important legitimatory function for Stewart.38 Furthermore, Stewart's Account provided him with the occasion for a re-assertion of the educational significance of the philosophy of mind. Although George Davie and Richard Olson have rightly stressed the centrality of moral philosophy within the Scottish universities of the eighteenth century, they have tended to overlook a number of institutional factors specific to Edinburgh which affected Stewart's pedagogical practice and conception of moral philosophy. 39 As J. B. Morrell has argued, in the competitive world of the Edinburgh professoriate, individual professors often had to combine pedagogical flair with utilitarian apologia for their subjects in order to attract students.4o Because of the dominance of the Medical Faculty in the university during the late-eighteenth century, and the flexibility within the degree programme, Stewart would have had to compete with his colleagues for sizeable classes and adequate remunera­ tion. Stewart's elegant rhetoric, moral earnestness, and avoidance of abstruse technicalities in his lectures can thus be seen as pedagogical devices appropriate to his institutional environment.4 1 So too can his vindication of the utility of the philosophy of mind, a subject which he discussed at some length in the introduction to the Elements of the Philoso­ phy of the Human Mind, published in 1 792.42 Moreover, the educational programme of the eighteenth century to which most Scots subscribed was itself under attack by members of the young Whig coterie closely asso­ ciated with Stewart, who questioned his view of the status of the philoso­ phy of mind and its honoured place within the university curriculum.43 Stewart commented on a number of these themes in the Account. Throughout he asserted the utility of the philosophy of mind, and quoted from Berkeley and Burke to substantiate his claim that the philosophy of mind was an invaluable preparation for other pursuits.44 Underlying Stewart's assertions was his conception of the cognitive structure of the

THE HAGIOGRAPHY OF COMMON S EN S E

313

sciences, a conception shared by eighteenth-century Scottish moral philo­ sophers.45 He argued that the philosophy of mind bore a relationship to the study of such subjects as logic and ethics which was analogous to that between anatomy and the study of physiology and pathology. He wrote: as a course of medical education naturally, or rather necessarily, begins with a general survey of man's animal frame; so, I apprehend, that the proper, or rather the essential preparation for those studies which regard our nobler concerns, is an examination of the principles which belong to man as an intelligent, active,�ocial, and moral being.46

While Reid himself, according to Stewart, failed to apply the philosophy of mind to its practical uses, his importance as a philosopher lay in his attempt to carry out the preliminary analysis of our intellectual and active powers in accordance with the canons of the strict inductive method.47 Stewart also included a vindication of the moral philosopher's role in society in his concluding encomium on Reid's character. He remarked that Reid's efforts as an educator would "in the judgment of the wise and good, have ranked him in the first order of useful citizens", and quotes from Seneca's Moral Essays to the effect that the man who instills virtue in the hearts of the young is as useful to the state as those who engage in its political affairs.4 8 Biographical description again seems intertwined in this passage with self-justification, for Stewart was here urging the social utility of a role fulfilled by both Reid and himself. As we have seen, in defining this role Stewart was trying to create a particular social niche for himself in response to his political and social isolation in Edinburgh during the 1 790's. The general debate carried on during the second half of the eighteenth century between Scottish savants and radical dissenters in England forms the background for Stewart's extended defence of Reid's philosophy against the criticisms of Joseph Priestley, and his own critique of Erasmus Darwin's Zoonomia. Under the shadow of the French Revolution, this debate intensified, and the doctrines of the English philosophes were tarred with the brush of J acobinism. The figure who articulated these anti-Jacobin attitudes within the University of Edinburgh was the Professor of N atural Philosophy, John Robison, who singled out Joseph Priestley for particular attack.49 The content of Stewart's criticisms of Priestley and Darwin paralleled that of Robison's, although his style was distinctly less inflam­ matory; while Stewart did not share his colleague's rabid anti-Jacobinism, like Robison he was resolutely opposed to the materialism of Priestley and

3 14

P. D. WOOD

Darwin and he too feared the potentially subversive implications of their doctrines. Stewart presented Reid's inductivism as an effective methodological weapon against the materialism of Hume, Helvetius, Hartley, Darwin and Priestley. Stewart argued that Reid's method distinguished him from both scholasticism and the "physiological metaphysicians of more modern times" insofar as Reid had confined himself to the examination of mental phenomena and the discovery of the laws governing these phenomena, whereas the scholastics and 'physiological metaphysicians' had speculated on the nature and essence of the mind, and had confused facts and hypo­ thetical theories. 5o The fundamental opposition between Reid's philosophy and materialism was, for Stewart, not metaphysical but methodological. He wrote : It is, in truth, much more from such examples of sound research concerning the Laws of Thought [i.e. Reid's works] , than from any direct metaphysical refutation, that a change is to be expected in the opinions of those who have been accustomed to confound together two classes of phenomena [i.e. mental and material), so completely and essentially different. 5 1

Again, Reid's exemplification of the rules of a strict inductive method was taken to be his signal contribution to philosophy. Stewart characterised the hypothetical method of the 'physiological metaphysicians' as being a retrograde step in the direction of scholasticism, whereas Reid's induc­ tivism was a progressive development in the philosophy of mind. 52 In his attack on materialism the philosophical and ideological purchase of Stewart's characterisation of Reid in terms of the B aconian virtues of industry, caution, and patience becomes apparent. 5 3 Materialism was presented as a "crude and visionary metaphysics", the product of super­ ficial learning and verbal artifice. According to Stewart its advocates had been deceived by the B aconian idols of the cave, misled by specious linguistic analogies between matter and mind, and seduced by the love of simplicity.54 Reid's 'Baconian' inductivism and the values it embodied are, on the other hand, shown to be the best antidote for these aillictions of the understanding. Hence Reid's inductivism could also be seen as antithetical to the political radicalism associated with the materialism of Darwin, Priestley, and the French philosophes. ss Stewart's aJlusion to, and criticism of, Laplace in his discussion of the misapplication of mathematical methods to the empirical sciences provide another parallel with Robison's attack on the ' Theophobia Gallica'. Unlike Newton and his foJlowers, Laplace and other continental mathematicians

THE HAGIOGRAPHY OF COMMON S EN S E

315

had, according to Stewart, aped the Euclidean method and attempted to deduce all the truths of physics from as few principles as possible. The tendency of their work was, he believed, to withdraw the attention from that unity of design, which it is the noblest employment of philosophy to illustrate, by disguising it under the semblance of an eternal and necessary order, similar to what the mathematician delights to trace among the mutual relations of quantities and figures. 56

Against the atheistic tendencies of Laplace's method, Stewart juxtaposed the consequences of Reid's inductivism, which underlined the differences in logical structure between mathematics and the empirical sciences. Reid and others had sharply distinguished between pure mathematics, which begins with definitions and investigates necessary connexions, and the empirical sciences, which begin with facts and investigate the laws of nature. For Stewart one of the most important consequences of Reid's methodology was that it challenged "that false logic, which while it throws an air of mystery over the plainest and most undeniable facts , levels the study of nature, in point of moral interest, with the investigations of the Geometer or of the Algebraist".57 Ultimately, Laplace's misuse of mathe­ matics derived from the love of simplicity. Another instance in which Stewart saw this principle leading to heterodox conclusions was Erasmus Darwin's attempt in his Zoonomia to explain away instincts as a form of learned behaviour. Such an attempt was unacceptable to Stewart who, like Reid before him, saw instincts as evincing God's providential design in nature. Hence he questioned Darwin's conclusions, arguing that the Englishman had invoked mechanisms which were essentially the same as instincts.58 Stewart's dismissive remark that Darwin's materialism, com­ pared with Hume's, was "the best calculated to impose on a very wide circle of readers",59 reveals something of the social dimension of his cri­ tique. For this apparent fear of the spread of materialism suggests that Stewart, like Robison, associated it with the rise of atheism, and, perhaps, even political radicalism. Stewart's reaction to the materialism of Priestley and Darwin and the atheism of Laplace thus structured his defence of Reid's philosophy, and his description of Reid's method as a weapon against these heterodoxies possessed a specific ideological thrust in the light of the widespread fears of sedition and impiety to which Robison had given expression in his Proofs of a Conspiracy Against All the Religions and Governments of Europe ( 1 797). Stewart's discussion of Reid's 'mode of philosophizing' and the 'scope

316

P. B . WOOD

of [Reid's) researches' contains the clearest statement of the programmatic relationships which he saw between their philosophies, and a defence of what he took to be their shared aims.6o According to Stewart, Reid had been the first philosopher of the mind to carry out his investigations "in strict conformity to the rules of inductive philosophizing". 6 1 While Hume had attempted to introduce experimental reasoning into morals, Stewart (following Reid) argued that Hume's whole system was in fact based on an hypothesis (the theory of ideas), and that Hume had not carried out a systematic analysis of the phenomena of the mind before developing syn­ thetic explanations of them. Unlike Hume, Reid had been deeply influenced by the writings of Bacon, Newton, and subsequent natural philosophers, and possessed an empirical frame of mind. Because of these two factors, Stewart argued, it was Reid who had successfully introduced the inductive method into the study of our intellectual and active powers.62 Yet Stewart j udged that Reid had not provided a systematic discussion of the logic of the science of the mind; like Newton, he had exemplified in his practice the methodological rules contained in the Novum Organum. 63 Stewart believed that the exposition of the proper method for the philoso­ phy of mind was of crucial importance since this project had been left uncompleted by B acon, D'Alembert, and Reid. Given this problem situation, Stewart indicated that it was his intention to provide such an exposition in his Elements of the Philosophy of the Human Mind.64 As his numerous references to the writings of the philosophes suggest, Stewart's understanding of Reid's philosophy was shaped to some extent by his awareness of philosophical currents in France during the eighteenth century. His emphasis on the analytical method and on the methodological categories of analysis and synthesis can perhaps be seen as reflecting the influence of Condillac and D'Alembert, since Reid made neither extensive nor significant use of the analysis-synthesis distinction. 65 Stewart also seems to have been influenced by the reception given to Reid's Inquiry by his Edinburgh teachers Adam Ferguson and James Russell , for he recalled that in their lectures on moral and natural philosophy they particularly welcomed Reid's methodological approach.66 Thus, although methodology was undeniably the key element in Reid's critique of the theory of ideas and his attacks on materialism and necessitarianism, Stewart's assessment of Reid's achievement was greatly influenced by his own reading of the history of moral and natural philosophy since B acon, a reading derived from a variety of Scottish and Continental sources. Reid's self-assessment of his philosophical merit should be borne in mind when evaluating

THE HAGIOGRAPHY OF COMMON SENSE

317

Stewart's view, for Reid did not see his achievement i n Stewart's exclu­ sively methodological terms. Reid wrote to Dr James Gregory: "It would want candour not to own, that I think there is some merit in what you are pleased to call my Philosophy; but I think it lies chiefly in having called in question the common theory of Ideas". 67 In characterising his philo­ sophical legacy, Reid therefore placed more emphasis on his substantive theoretical contribution to the science of the mind than on the metho­ dological reform of that science which Stewart highlighted. Stewart's estimate of Reid's two Essays reveals a second programmatic connexion which he drew between their work. Stewart contended that whereas Reid had seemingly disavowed any attempt to complete an exhaustive analysis of the intellectual and active powers of the mind in the Inquiry, 'the progress of his researches' suggested the feasibility of such a project. That Reid did in fact adopt this aim in his later publications was important for Stewart since he wanted to justify his own philosophical programme in terms of Reid's practice. Alluding to criticisms of philoso­ phers such as himself who attempted to analyse all of the powers of the mind rather than concentrating on one particular faculty or sense, Stewart wrote: DR REID, in . . . his Inquiry, might have been supposed to give some countenance to this opinion; ifhis own subsequent labours did not so strongly sanction the practice in question. 68

He then went on to defend Reid's aims on methodological, moral, and social grounds, but it is clear from his remarks that these arguments also functioned as self-defence.69 Stewart valued both of Reid's Essays as indis­ pensable resources for his own histories of our mental powers, and considered himself to be completing the project which Reid had begun. More importantly, the two Essays, as interpreted by Stewart, provided historical justification for the philosophical programme he was attempting to carry out in his Elements. A third and more specific programmatic connexion to which Stewart drew attention concerned the explication of Reid's contentious appeal to 'common sense'. Stewart readily admitted that Reid's notion of common sense was imprecise, and he announced his intention to undertake in subsequent writings the necessary clarification and refinement of this central doctrine of Reid's epistemology. In the Account Stewart accom­ plished the more modest but nonetheless strategically important aim of distinguishing Reid's notion of common sense from those of James Oswald

318

P. B . WOOD

and James Beattie, enabling him to defuse at least some of Joseph Priestley's criticisms in the Examination ofDr Reid's Inquiry into the Human Mind ( 1 774). He argued that Reid had appealed, not to the beliefs of the vulgar or "the creed of a particular sect, nor the inward light of enthusiastic presumption", but to what Stewart had come to call the 'Fundamental Laws of Belief. His explication and defense of Reid's appeal to common sense was thus based on his own epistemological analysis of the laws of belief, which he later published in the second part of the Elements.7o Moreover, Stewart provided a moral j ustification for the completion of this analytic task when he wrote that by clarifying and supplementing Reid's views savants would vindicate "the authority of truths intimately and extensively connected with human happiness". 7 1 Stewart's discussion of the notion of common sense is particularly significant historiographically, as he opposes his own epistemological programme to that of "some Meta­ physicians in Germany". 72 Stewart's intense dislike of Kant and Kan­ tianism provides a sharp contrast with later Scottish philosophers of the nineteenth century like Sir William Hamilton, and this contrast is reflected in their respective historiq.1 images of Reid. For Stewart, Reid's philosophy was unrelated, and indeed vastly superior to that of Kant. Hamilton and his successors, on the other hand, explored the parallels which they perceived between the philosophies of Reid and Kant in great detail. 73 Stewart's later Dissertation : Exhibiting the Progress of Metaphysical, Ethical, and Political Philosophy fixed the image of Reid initially sketched in the Account. Less a disinterested history of philosophy than an extended polemic, the Dissertation was a nationalistic celebration of the achieve­ ments of what Stewart was the first to call the 'school' of Scottish philoso­ phy, and a vindication of Reid's originality and merit as compared with the pretensions of Kant and his German followers. While significant historio­ graphical innovations occurred in the Victorian era, these two works, more than any other, have shaped our understanding of Reid's career and writings. Thus Stewart's historiography, as modified in the nineteenth century, continues to dominate the study of the life and work of Thomas Reid. As I have argued, Stewart's portrait of Reid reflects his academic and political predicament in Edinburgh at the turn of the nineteenth century, and his own philosophical preoccupations.

University of Toronto

THE HAGIOGRAPHY OF COMMON S E N S E

319

NOTES I

Daniel Roche, Talent, Reason and Sacrifice: The Physician during the Enlightenment', in Medicine and Society in France; Selections from the Annales: Economies, Societes, Civilisations, edited by R. Forster and O. Ranum, translated by E. Forster and P. M. Ranum (Baltimore: Johns Hopkins University Press, 1980), Volume 6: 66-88; Dorinda Outram, The Language of Natural Power: The "Eloges" of Georges Cuvier and the Public Language of Nineteenth­ Century Science', History of Science 16 ( 1978): 1 53-178 ; Charles B. Paul, Science and Immor­ tality: The Eloges of the Paris Academy of Sciences (1699-1791) (Berkeley, 1 98 1 ), passim. 2 Like some of the French academies, the Royal Society of Edinburgh also had two classes of membership, the Physical and the Literary. J In his Dissertation: Exhibiting the Progress of Metaphysical, Ethical, and Political Philosophy Stewart remarked that the eloges of Cuvier had revived the artistry of Fontenelle; see The Collected Works of Dugald Stewart, edited by Sir William Hamilton, I I Volumes (Edinburgh, 1 854-1860), Volume I, p. 328, hereinafter referred to as Stewart, Works. 4 Millar's comment is to be found in A Mathematical and Philosophical Dictionary, edited by Charles Hutton, 2 Volumes (London, 1 795-1796), Volume I, p. 558. Robison's description comes in his article on 'Impulsion' in the Supplement to the Third Edition of the Encyclopaedia Britannica, 2nd edition, 2 Volumes (Edinburgh, 1 803), Volume I, p. 802. I should like to thank Geoffrey Cantor for this reference. 5 Reid's scientific and mathematical papers are surveyed in my Ph. D. thesis Thomas Reid, Natural Philosopher: A study of Science and Philosophy in the Scottish Enlightenment' (University of Leeds, 1984). I am currently preparing an edition of these manuscripts. 6 Dugald Stewart, Account of the Life and Writings of Thomas Reid, D. D. F. R. S. Edin. Late Professor of Moral Philosophy in the University of Glasgow (Edinburgh, 1 803), pp. 94-95, hereinafter referred to as Account. I have used the 1 803 edition of Stewart's Account in preference to that reproduced in Hamilton's edition of Reid's Works because the text of the latter is corrupt. Apart from numerous changes in punctuation and capitalisation, Hamilton made significant textual alterations. 7 Account, pp. 205-206. 8 Cuvier similarly explored these tensions; see Outram, op. cit. , fn. I. 9 Account, pp. 1-2. Stewart here develops a theme found in the dedication of the Inquiry, where Reid writes that he had pursued his studies in the "leisure of an academical life, disengaged from the pursuits of interest and ambition"; The Philosophical Works of Thomas Reid, edited by Sir William Hamilton, 8th edition, 2 Volumes (Edinburgh, 1 895), Volume I, p. 96. 1 0 Account, p. 1 83. II Henry Cockburn, Memorials ofHis Time (Edinburgh, 1 856), p. 80; see also p. 45. The period 1 790-1 804 has been taken for discussion on the basis of Cockburn's remarks on p. 148. 1 2 Ibid., pp. 46, 8 1 -82, 85, 1 03. See also the remarks by a German traveller who stayed in Edinburgh during the winter of 1 794-1795 recorded in 'On the Stile [sic) of Society in Edinburgh', Scottish Register 6 ( 1 795), pp. 1 38-1 39; I should like to thank Geoffrey Cantor for this reference. It was rumoured in Edinburgh during the late 1 790's that the publication of the second volume of Stewart's Elements of the Philosophy of the Human Mind was delayed because of government disapproval; see Alexander Gillies, A Hebridean in Goethe 's Weimar. The Reverend James Macdonald and The Cultural Relations Between Scotland and Germany (Oxford: Blackwell, 1 969), pp. 80-8 1 .

320 13

P. B . WOOD

The social composition of the Royal Society of Edinburgh is carefully analysed in Steven Shapin, 'Property, Patronage, and the Politics of Science: The Founding of the Royal Society of Edinburgh', The British Journalfor the History of Science 7 ( 1974): 1-4 1 . 1 4 For the notion o f a public language i n which such strategies can b e pursued see Outram, op. cit. , fn. I . 1 5 Account, p . 2. 1 6 Ibid., p. 45, italics added. 17 This was James Boswell's nickname for Henry Dundas. 8 1 Stewart, Works, Volume 10, pp. 55-56. 19 Ibid., Volume 10, p. 87. Compare Cockburn, op. cit., fn. I I , p. 1 75. 20 Stewart, Works, Volume 10, p. 87. 2 1 John Veitch, 'A Memoir of Dugald Stewart', in Stewart, Works, Volume 10, p. lxxii. 22 Ibid., Volume 1 0, pp. lxxii-Ixxv. 23 On the sedition trials see H. W. Meikle, Scotland and the French Revolution (Glasgow: Maclehose, 1 9 1 2), Chapter VI. 24 Stewart, Works, Volume 10, p. lxxv. 25 Dugald Stewart, Account of the Life and Writings of William Robertson. D. D. F. R. S. E. . (Edinburgh. 1801). pp. 120-12 1 . 26 Alexander Carlyle. 'A Comparison of Two Eminent Characters Attempted after the Manner of Plutarch', in Anecdotes and Characters of the Times, edited by J. Kinsley (London : Oxford University Press, 1973), p. 28 1 ; see also Meikle, op. cit., fn. 23, pp. 49, 52. 27 Account, p. 7. 28 Stewart describes some of Reid's manuscripts from the Glasgow period in the Account, pp. 169-1 70. It is unclear why Stewart failed to use Reid's manuscripts, as even a cursory examination of them would have indicated their usefulness in reconstructing the development of Reid's thought. 29 Account, p. 19. 30 Ibid., pp. 49, 1 70-1 7 1 . 3 1 Stewart, Works, Volume 10, p. 37. 32 Account, p. 54; see also pp. 85-86 on the relationship between these works. 33 Ibid., pp. 46-49, 1 72-173. 34 Alexander Gerard, Plan ofEducation in the Marischal College and University ofAberdeen. with the Reasons of it (Aberdeen, 1 755), pp. 23-26; see also pp. 6-7, 3 1-32. 35 Intellectual Powers, pp. 4-5 ; Active Powers, p. 379. 36 Mathematics did not fit neatly into this scheme, and Gerard justified the teaching of that science in the first years of the philosophy course on the grounds that it was a necessary preliminary to the study of natural philosophy and that it taught the students how to reason accurately; Gerard, p. 19. 37 Stewart says of Reid's move to Glasgow that it enabled him "to concentrate to his favourite objects [i.e. moral philosophy] , that attention which had been hitherto distracted by the miscellaneous nature of his academic engagements"; Account, p. 46. 8 3 Although the concept of legitimation is currently much debased, I rely here on the discussion of this concept by Quentin Skinner in 'Some Problems in the Analysis of Political Thought and Action', Political Theory 2 ( \ 974): 277-303, esp. pp. 292 If.. Skinner rightly stresses that an individual or group seeking to legitimate their ideas or actions must appeal to the values of the individuals or groups from whom they are seeking legitimation. Without such an appeal, it seems meaningless to talk about a process of legitimation occurring.

THE H A G I O G R A P H Y OF COMMON S E N S E

321

39 G. E. Davie, The Democratic Intellect: Scotland and Her Universities in The Nineteenth Century, 2nd edition (Edinburgh: Edinburgh University Press, 1 964); R. Olson, Scottish Philosophy and British Physics 1 750-1880 : A study in the Foundations of the Victorian Scientific Style (Princeton: Princeton University Press, 1975), Chapter I. 40 J. B. Morrell, 'The University of Edinburgh in the late Eighteenth Century: Its Scientific Eminence and Academic Structure', Isis 62 ( 1 970): 1 58-1 7 1 , esp. p. 1 6 1 . 4 1 Cockburn's description o f Stewart's lecturing style i s the best known and the most suggestive in the context of my argument here; see Cockburn, op. cit., fn. I I , pp. 22-26, 1 74-1 76. 42 Stewart, Works, Volume 2, pp. 57-90. Veitch emphasised this aspect of Stewart's teaching in his 'Memoir'; ibid., Volume 10, pp. xxxix, xli, xliv. 43 See for example Francis Jeffrey's anonymous review of Stewart's life of Reid in the Edinburgh Review 3 ( 1 803-1804): 269-287. Jeffrey's criticisms are discussed in G. E. Davie, The Social Significance of the Scottish Philosophy of Common Sense (Dundee: Dundee University Press, 1973), pp. 14-16. 44 Account, pp. 1 66-167. 45 I have borrowed the notion of 'cognitive structure' from J. R. R. Christie. 46 Account, p. 76. 47 Ibid., pp. 8 1-82. Stewart here curiously ignores Reid's lectures on logic, rhetoric, and ethics, which are mentioned elsewhere in the Account. 48 Ibid., pp. 203-204. The distinctions between the educator and the politician, and between public and private life in the passage quoted from Seneca parallel those drawn by Stewart in his description of Reid's life. 49 On Robison's attack see J. B. Morrell, 'Professors Robison and Playfair, and the Theophobia Gallica: Natural Philosophy, Religion and Politics in Edinburgh, 1 789- 1 8 1 5', Notes and Records of the Royal Society of London 26 ( 197 1 ) : 43-63, pp. 47-48. 50 Account, pp. 96-97. 5 1 Ibid., pp. 99-100. 52 Ibid., pp. 108-1 10. 53 Ibid., pp. I I , 38-39, 1 84. 54 Account, pp. 1 06, 109-1 10, 1 14, 1 25-1 27, 145. 55 Robison likewise used inductivism against Priestley; Morrell, op. cit., fn. 46, p. 48. 56 Account, p. 1 4 1 . See also Stewart's remarks on Laplace in his Dissertation; Stewart, Works, Volume I, pp. 386-387, 467-468, 6 1 3. 57 Account, p. 142. S8 Ibid., pp. 145-1 52. For Reid on instincts see the Essays on the Active Powers of Man (Edinburgh, 1 788), pp. 103-1 1 7. Insofar as Darwin appeared to be denying providential design, his work could be read as having atheistic implications. Stewart makes this charge in his Dissertation; Stewart, Works, Volume I , p. 472. 59 Account, p. 1 06. 60 Indeed, Stewart indicates elsewhere that his discussion of a number of aspects of Reid's career was limited so that he could enlarge on what he saw as Reid's philosophical legacy; see Account, p. 203. 61 Ibid., pp. 34-35 ; see also pp. 56, 86. 62 Ibid., pp. 37-38. 63 Ibid., pp. 38, 56-58, 7 1 , 75, 86. 64 Ibid., pp. 57-58, 7 1 -75.

322

P. B . WOOD

65 My suggestion concerning the possible influence of Condillac and D'Alembert on Stewart is highly tentative, as Stewart's reaction to these savants seems to have been complex. This question deserves further study. 66 Account, pp. 43-44. 67 Ibid., p. 1 22. 68 Ibid., p. 87. 69 Ibid., p. 88. 70 Ibid., pp. 1 52-163; Stewart, Works, Volume 3, pp. 23-69. 7 1 Account, p. 163. 72 Ibid., p. 1 64. 73 See for example Hamilton's remarks on Reid's 'Essay on Quantity' in Hamilton, op. cit., fn. 9, Volume 2, p. 7 1 5. This interpretation of Reid's life was taken to its extreme in A. Campbell Fraser's Thomas Reid (Edinburgh: Oliphant, Anderson & Ferrier, [ 1 898]).

I N D EX

priori, cosmological, design, ontological, transcendental Aristotelianism, scholastic 149- 1 5 1 , 1 57-160, 2 19; and Aristotle 1 62n.; see also scholasticism Aristotle 5, 7, 8, 20, 22n., 1 50- 1 5 1 , 165 Armogathe, Jean-Robert 1 77n. Arnauld, Antoine 34, 1 26, 141n. Arnold, Paul 1 12n. atheism 95, 1 1 8, 1 20, 1 23, 126, 1 29, 224, 315 atomism see corpuscularianism Aubrey, John 1 1 5n. Augustine, St. 183 Avicenna 1 12n. Ayer, A. J. 32, l IOn., 285n. Ayers, Michael R. 25n., 4 1 -44, 47, 1 79, 1 93n., 2 1 2, 2 1 7, 225n., 226n., 227n., 228n., 235, 239n.

Abelard, Peter 5 Abercromby, Alexander 308 abnormal discourse see incommensurable abstraction(ism) 29, 1.50 Academie des Sciences 1 65, 1 76, 305 action, bodily 273, 279, 28 1 , 283 Adam 8, 1 2 1 Ady, Thomas 1 l in. aether 98, 1 38, 226 alchemy 42-44, 95, 99, 108, I I In., 1 1 5, 131 Algazel 1 1 2n. Alison, Archibald 309 American Revolution 309 analytic tradition 28, 3 1 ; method 256, 3 16; philosophy 280-281 Anaximander 6 Anderson, George 264-268, 269n. Anderson, Robert F. 285n. angels 96, 1 02, 104, 1 30, 1 32, 1 50, 1 56, 212 anti-canon, philosophical I I , 22 anti-classics 1 1 , 22 anti-foundationist(s) 34,36 Apel, Karl-Otto 58-62, 64-66, 67n., 68., 79, 8 1 a posteriori argument: in Descartes 1 54-1 57; in Hume 243-244, 260-263; in Leibniz 201 a posteriori knowledge/truth 1 8 1 - 1 83 a priori argument 32-33; in Descartes 1 5 3- 1 54, 1 56 ; in Hume 243-269; in Leibniz 201 a priori history 1 8-20, 3 1 , 34, 43 a priori knowledge/truth 1 8 1- 1 83 argument: the backbone of philosophy 49; see also a posteriori, a

Bacon, Francis 12, 1 07, 1 12n., 1 85 191 , 1 93n., 3 14, 3 16; his experimental method 1 88- 1 89 Baillet, Adrien 1 14n. Baker, Thomas 24n. B alfour, James 258 Barbeyrac, Jean 24n. Barfoot, Michael 270n. B axter, Andrew 270n. B ayle, Pierre 1 00, 1 4 I n., 268 Beattie, James 259, 3 18 Belaval, Y. 141n. Bellarmine, St. Robert 53 Bennett, J.A. 133, 142n. Bentley, Richard 1 37, 224 Berkeley, George 1 7, 24n., 37, 4On., 230, 302n., 303n., 3 1 2 ; -

323

324

I NDEX

on abstraction 1 1 2n . : and common sense 10- 1 1 ; against hidden qualities 1 02; v. Hume 30; idealism distinguished from Kant's 289-293; Kant v. 297; against the New Philosophy 29 Bernoulli, Johann 1 38-\ 39, 143n. Blair, Hugh 262 Bloch, Marc I I I n. Bloor, David l IOn. body, as part of self 271 , 273, 278-279 Bohr, Niels 53 Bolingbroke, Lord (Henry S1. John) 266 Bonar, John 266 Bonaventure, S1. 1 8 2 Bossuet, Jaques-Benigne 260n. Boswell, James 320n. Boullier, F. l 77n. Boyle, Marjorie O'Rourke 23n. Boyle, Robert 99, l IOn., 1 2 1 , \ 33- \ 36, 140n., 1 4 1 n., 142n., 220; aether of 226n.; on animals 238n.; and fairies 96; and magic 1 1 1 n., 123; mechanical philosophy of 209, 224; not an occasionalist 229-23 1 ; scientific aims of 225n.; on signs of spirit and divine will in nature 1 29- \ 32; and Spinoza 1 90- 1 9 1 ; o n universal and particular notions 229n. Bradley, F.H. 4 1 , 48 Brahe, Tycho 1 9 1 B ricke, John 285n. Brown, Harcourt l 77n. Brown, Peter 23n. Brown, Stuart C. 207n. Browne, Thomas 1 2 1 , 1 24, 1 32, 140n. Brucker, Johann Jakob 1 2 , 1 7 , 20, 24n. Buchdahl, Gerd 1 6 1 n., 1 63n., 1 64n. Burke, Edmund 309, 3 1 2 B urnet, Thomas \ 36 Butler, Joseph 261 Butts, R.E. 142n.

cabalism 96, 105 Cambridge Platonists 1 2 1 - 1 22, 220; see also Henry More, Ralph Cudworth

Camisard refugees 1 24 Campanella, Tommaso 8, 100 canon, philosophical 4, 1 1 , 20, 22, 42, 48 Capaldi, Nicholas 285n. Capp, Bernard 1 1 5n. caricature, in philosophy 20-2 1 , 3 1 , 47, 1 22, 1 79-193, 1 95-196, 203-206, 27 1 , 2 8 1 -283; see a/so integral history Carlyle, Alexander 309, 320n. Carmichael, Gershom 270n. Carnap, Rudolf 285 Carneades 6 Cartesian(ism) 44, 195-197, 250, 288-290; cosmology 1 27; epistemology 35, 1 72- 175; use of experimental evidence 1 72- 1 75 ; mechanics 1 27, 20 1 , 233; metaphysics and theology 1 66- 1 72, 1 76; physics 1 66- 1 72; rationalism 195-206; scepticism 294; science 1 65-1 77; see also Descartes Casaubon, Meric l IOn., 1 19, 1 24, 1 40n. catholic(ism) 100, 145 Caton, Hiram 1 1 5n. causal mechanism(s) 50, 97, 155, 1 57-1 58, 1 60- 1 6 1 causation, causality 98, 1 67- 1 70, 230, 243-249, 253-255, 262-265, 268-269; principle of 245-246, 26 1 , 263 ; linear v. hierarchical 248-249, 253 Cavendish, Margaret 1 19, \ 39n. Charleton, Walter 1 20, 122, 140n., 224 , . 229n. Chauvin, Etienne 1 12n. chemistry, and alchemy 1 08, 1 1 5 Cheyne, George 26 1 , 262 Chomsky, Noam 59 Christ 5, 7-8, 96, 98, 1 1 8 Christian Grammar Schools 5 Christianity, and philosophy 6-9, 1 83 Christie, J.R.R. 1 09n. Cicero 5, 9, 23n. Clarke, Desmond M. 1 4 1 n. , 1 61 n., 1 63n., 1 64n. Clarke, S amuel \ 36, \ 38, 143n., 224, 225, 243-269 classics, classic authors 1 1 , 1 7, 22

INDEX 'Cleanthes' 249, 250-265, 267-268 clear and distinct perception/ideas 1 53-1 54, 195- 1 96, 199, 202, 282 Cockburn, Henry 307, 3 1 9n., 320n., 321n. Cohen, I. Bernard 1 4 I n., 143n. cohesion 2 1 0, 226n., 234 Coimbra, university of: commentaries 149, 1 6 I n., 162n., 1 64n. Coker, Matthew 1 24, 140n. Colbert, Jean-Baptiste 165 Coleridge, Samuel Taylor 1 04 Collegiant circle 190 Comenius, John Amos 1 00 commensurable discourse see normal common sense 9-1 2, 280-28 1 , 305, 3 10, 3 1 7-3 1 8 ; philosophy of I I Comte, Auguste 2 1 conceivability 250-254, 256, 262-263, 268 conceptual scheme(s) 3 1 , 35, 52-58, 79 Condillac, E.B. de 24n., 3 16, 322n. Condorcet, Marquis de 307, 309 Conti, Antonio 1 37, 143n. contingency 244-245, 248, 25 1 , 26 1 , 264, 268 contradiction 250, 254; principle of 1 99, 204 Conway, Anne I I I n., 1 35, 140n. Conway, Edward 140n. Copernicus, Nicolaus 1 85, 1 9 1 Cordemoy, Gerauld d e 1 66, 1 67, 1 69-1 70, 1 77n., 1 78n. corpuscularianism 5 1 , 103- 1 04, 108, 2 1 4 ; v . modern physics and chemistry 244; see also mechanical philosophy correspondence, with reality 33, 53-54, 60, 78; of new theory to old 53-54 cosmological argument 243-269 cosmology, Cartesian 1 27- 1 28; magnetical 1 3 3 Cotes, Roger 1 36, 1 37, 143n. Coudert, A. 140n. conterparts see incongruous Counter-Reformation 183 Couturat, Louis 207n.

325

Cowles, T. 1 42n. Cowley, Abraham 1 39n. criteriologists 282-283 criticism 1 3, 1 5- 1 6, 1 9 Croll, Oswald 105, 1 14n., 1 1 5n., 1 17 Cudworth, Ralph 24n., 100, 220, 229n. Culverwel, Nathanael 140n. Cuvier, Georges 305, 3 19n. D'Alembert, Jean Ie Rond 24n., 316, 322n. Dalzel, Andrew 307 Daniel, Gabriel 1 14n. Darwin, Erasmus 3 1 3-3 1 5, 321n. Davidson, Donald 56-58, 79, 82n., 1 00n. Davie, George E. 3 1 2, 32 1 n. da Vinci, Leonardo 103, 1 1 3n. Decker, Johann Heinrich l I On. deconstruction 37, 43 de Duillier, Fatio 1 24 Dee, John 1 03, 1 1 3n. Defoe, Daniel 1 23, 14On. degrees, academic 9 della Porta, Giambattista 107, 1 14n. 'Demea' 243, 245-254, 257, 259-260, 263, 265, 267-268 Demon of Macon 1 23, 140n. demonstration, demonstrability 253, 257, 260-263, 265, 268; in Descartes 1 59; geometrical 197, 1 99, 221-223; in Hume 244, 250-25 1 , 254; in Leibniz 204 Dennett, Daniel 67n. Derrida, Jacques 26n., 41 Descartes, Rene 8, 1 00, 1 0 1 , l iOn., 1 12n., 1 14n., 1 22, 1 32, 140n., 1 4 1 n., 149-1 6 1 , 1 65-175, 1 77, 1 77n., 1 78n., 1 84- 187, 1 92, 1 98, 199, 201 , 202, 207n., 233, 245, 261 ; on the a priori 163n; on demonstration 1 59; empirical methodology of 1 57- 1 6 1 , 1 6 1 n. ; and God 1 1 3n. ; o n hidden reality 103- 1 06; Kant's criticism of 288-289, 293-294; contrasted with Leibniz 195- 1 97, 204-206; on matter and extension 1 62n., 1 63n. ; his metaphysics 149- 1 6 1 ; method of, and

326

INDEX

place in the history of philosophy 7, 9-1 1 , 2 1 , 1 19-120; his physics 149- 1 6 1 ; his physics, empirical justification of 1 54- 1 57, metaphysical grounding of 153-154, 1 56 ; his quest for certainty 149, 1 6 1 , 164n.; representative theory o f 34, 44; his revolution of the sciences 160; Rorty's view of 36, 69, 85-86, 86n.; and scholastic Aristotelianism 149- 1 5 1 , 1 57-1 60; secrecy of 1 1 5n.; view of soul 97-98; see also Cartesianism Desgabets, Robert 1 66 design argument 25 1 , 260-262 Deskford, Lord (James Ogilvy) 308 deviance 107-109, 146-147 Dewey, John 5 1 , 86 De Witt, Jan 1 90 dialectic: in the history of philosophy 29-30, 33, 204; scholastic v. true 7 dialectical illusion 1 6 dialogue form, interpretation o f 258-259, 265 Diderot, Denis 24n. Digby, Everard 7 Dijksterhuis, EJ. 229n. Dilthey, Wilhelm 5 1 Diogenes Laertius 5-8, 20, 24n. distortion see caricature Dobbs, Betty Jo Teeter l IOn. dogmatism, dogmatist 13, 18, 1 9, 24n., 30, 198 dualism, dualist 50, 97, 102, 149, 1 66, 1 70, 288 Dudgeon, William 270n. du Moulin, Peter 140n. Dundas, Henry 307, 320n. Eberhard, Johann August 303n. Eckhart, Meister I \ 3n. eclectic philosophy 12, 3 3 ecstasy 104 edification 33, 39, 49, 70 Edwards, John 1 2 1 , 140n. Einstein, Albert 53

Elizabeth, Princess Palatine 1 25 , 1 26, 1 4 1 n. Elliot, Gilbert 258, 262, 266, 267, 270n. etoge 305, 3 1 9n. empirical: consciousness 298; self 299; self-consciousness 296, 298, 300 empiricism 3 1 , 47, 102, 106- 1 07, 1 14n., 147, 1 85-1 86, 1 9 1 , 203 Enfield, William 24n. Enlightenment 9, 20, 50, 60, 6 1 , 68n.; French 305; Scottish 261-262, 305 enthusiasm 105, 1 1 8- 1 20, 1 24-125, 129 Epicurus 6, 20 epistemology 30, 33-38, 49-68, 73, 82n., 85-86, 1 54, 273, 3 1 0, 3 1 8 Erasmus, Desiderius 6 , 8 essence: as form 1 50- 1 52, 1 54, 1 58, 1 60- 1 6 1 ; or nature 25 1 , 254-255, 263 ; nominal 2 1 8, 221-223; real 102, 209, 2 1 4, 2 1 7-224, 228n. eternity, eternal being 246-249, 253, 263-265 ethics 56, 58, 64, 66, 1 06, 1 7 1 , 3 \ 3 Euclid 7; Euclidean geometry 196- 197 ; Euclidean method 3 1 5 Eustace of Saint Paul 149, 1 6 I n., 1 62n., 1 64n. experiment 27-28, 1 07, 1 20, 1 25 , 1 57, 1 73- 1 74, 1 89, 225n. experimental philosophy 1 1 8-1 19, 1 22, \ 3 1 - \ 33 , 1 82 external history, and internal 20, 1 14n., 1 46-147 extra-sensory perception 104, 1 1 5n. fairies 96, 97 fairy-tale see caricature Farquhar, Helen I I I n. Farrar, F.W. 22n. fascination 1 0 1 , 1 04, 1 12n., 126, 1 28 Feder, Johann Georg Heinrich 293, 302n. feeling, assurances of 262-264, 267-268 Ferguson, Adam 3 1 6 fiction(s) 1 9 , 284; in Descartes 1 52 ; in Hume \3-14 Field, Hartry 86

INDEX

Findlay, J.N. 4 1 , 48 Fisch, H. 142n. Fludd, Robert 100, 104, I Bn., 146 Fontenelle, Bernard Ie Bovier de 305, 3 19n. Forbes, Duncan 268 Forster, Eckart 303n. Foucher, Simon 198-20 1 , 207n. foundations, foundationalism 32, 34-37, 39-40, 49-53, 58, 60, 64, 68n., 77, 80, 85-86 Frank, Robert G. 1 33, 142n. Franklin, James 270n. Fraser, A. Campbell 322n. Freemasonry 1 08 freewill 200; and determinism 63 French Revolution 1 8 , 307, 309, 3 1 3 Freud, Sigmund 59 Frick, Karl RH. 1 I 5n. Frommann, Johann Christian 1 12n. functionalism 38 Gabbey, Alan 140n. Gadamer, Hans-Georg 87, 9 1 Gadroys, Claude 178n. Galen 7 Galilei, Galileo 24n., 53, 1 0 1 , 1 12n, 1 85, 192 Garve, Christian 293, 302n. Gasking, E. 143n. Gassendi, Petrus 23n., 1 0 1 , 1 12n., I Bn., 209, 226n., 229n. Gauquelin, Michel 1 1 5n. Geisleswissenschaften 50, 5 1 , 64, 70 geometry 1 32, 1 54, 1 60, 1 80, 1 96- 197, 199, 220-223 ; non-Euclidean 3 1 Gerard, Alexander 3 1 1 , 320n. ghost(s) 96-98 Gilbert, Neal W. 23n. Gilbert, William 1 3 3 Gillies, Alexander 3 19n. Gilson, Etienne 23n. given, naturally 32; v. direct knowledge 37-38 Gianvil!, Joseph 99, 1 1 In., 1 19, 1 22-123, 1 26, 140n. Goclenius, Rudolf 8

327

God 49, 100, 1 05, 125, 153, 1 67- 1 70, 1 79- 1 80, 1 83, 209, 2 1 1 -225 passim, 235-238, 289, 297, 3 1 5; arguments for existence of I Bn., 243-269 ; divine sensorium, understanding 1 26, 1 28, 1 34; immutability of, in Descartes 1 58 Golinski, J.V. 1 09n. Gouhier, Henri 1 14n. Gouk, P. 142n. gravitation, gravity 98, 1 33, 1 3 5 , 210, 226n., 227n., 234; as occult quality 126; as analogy 1 26 Greatrakes, Valentine 99, 1 1 1 n., 1 24, 1 40n. Gregory, David 1 37, 138 Gregory, James 3 1 7 Guerlac, Henri 142n. Gunther, R 142n. Habermas, Jiirgen 58-6 1 , 64-66, 67n., 68, 79, 8 1 Hahn, Roger 1 77n. Hall, A.R 1 38, 1 4 1 n., 142n., 143n. Hall, M.B. 1 38, 1 4 I n., 142n., 143n., 229n. Hall, Thomas 1 1 1 n. Hamilton, William 3 1 8 , 3 19n., 322n. Hannaway, Owen 1 1 1n. Hartley, David 3 14 H artlib, Samuel 1 3 1 , 140n. Hartsoeker, Niklaas 1 36, 143n. H atfield, Gary 1 63n. Hegel, Georg Wilhelm Friedrich 4, 2 1 , 4 1 , 46n., 48, 59, 279, 283 Heidegger, Martin 5 1 , 86 Heilbron, J.L. 229n. Helmont, F.M. van 1 0 1 , 1 2 1 , 1 22, 1 35 , 220 Helvetius, Claude-Adrien 3 14 Herder, Johann Gottfried 303n. heresy see deviance Herivel, J. 141n. hermeneutics 49, 52, 58-67, 69-76, 78, 8 1 , 82n., 86, 87-92 hermetic(ist) 1 08, 1 1 5n., 1 2 1 Hesse, Mary 69-70, 75-8 1 , 8 1 n., 8 5 , 9 1 , 1 39n.

328

heterodox 73-74, 79, 87-92 Heyd, M. 1 41 n. Hirschfield, John M. 177n. historical materialism 60 history of science 32, 42-43, 5 1 , 53, 95, 147, 1 75- 1 76 Hobbes, Thomas 8, 23n., 1 00, 1 03, 106, l IOn., 1 12n., 1 19, 1 23, 1 25, 1 40n., 239n. Hooke, Robert 1 32- 1 34, 142n. Horn, Georg 7-9, 12, 20, 23n. Hume, David 1 7, 20, 24n., 47, 1 76, 243-269, 269n., 270n., 27 1 - 1 84, 284n., 285n., 3 1 2; in conflict with analytic philosophy 280; and the a priorija posteriori distinction 243-245 ; contrasted with Berkeley and Kant 30-3 1 ; v. Samuel Clarke 243-244; and the criticism of fictions 1 3- 1 5 ; and the characters of the Dialogues 258-259, 265; and the historical context of the Dialogues 245, 260-269; and the manuscript of the Dialogues 252, 257-258, 260, 265-267; influence of Malebranche on 1 65, 1 68; materialism of 3 14-3 16; his conception of philosophy 282-283; and logical positivism 28, 32; his account of the self 27 1 -275; and the unity of consciousness 276-278 Humfrey, John 1 4 1 n. Hunter, Michael 1 39n., 140n. Hurlbutt, Robert H. 261 , 270n. Husserl, Edmund 2 1 Hutcheson, Francis 261 . 268, 270n. Hutchinson, F. 1 23, 140n. Hutchison, K. 1 19, 135, 140n. Huygens, Christian 1 19, 1 36, 1 37, 1 39n., 140n., 142n., 165, 1 66 , 190, 1 9 1 hypnotism 104 hypothesis: in Bacon 1 888; Cartesian 173- 1 74, 177; in Descartes 152, 1 57; in Leibniz 1 98-199, 201 , 207n.; in Locke 2 1 2-21 3 ; in Newton 98, 1 37- 138

INDEX

iatrochemistry 1 33, 220 idea(s): agreement/disagreement of 106; distinct v. confused 250-252, 254, 256-257; and impressions 272-273, 3 1 1 ; naked 29; v. quality 29; of the self 272, 276-277; simple, complex 272-275 ; see also clear and distinct ideas, innate ideas idealism 52-53, 56, 287-302, 302n., 303n.; Cartesian 288, 294; dogmatic 289, 302n.; sceptical 289; see also transcendental idealism Ignatius Loyola, St. 1 1 3n. illusion, in philosophy 32-34, 290-294 litis, e. 142n. imagination 30-3 1 , 1 26, 1 84- 185 impressions see ideas incommensurable discourse 63, 70-75, 79, 82 incongruous conterparts 290 inductive method 188- 1 89, 3 1 3- 3 1 6 inesse principle 199-20 1 , 204, 207n. inhesion, of perceptions 277-279 innate ideas 1 06, 1 74- 175, 204-205 inner sense 297 integral history 3-4, 9 , 12, 20-2 1 , 22, 47-48, 203, 206; as definition (prescriptive) 2 1 intentionality, intentional 6 1 -62; v. phenomenal 36 internal history 20, 1 09, 109n., l I On., 146-147; see also external history Jackson, John 269n. Jacob, J.R. 1 39n., 1 40n. Jacob, M.e. 1 39n., 1 4 1n. Jacobi, Friedrich Heinrich 293, 303n. Jacobinism 3 1 3 Jacobitism 135, 145 Jaeger, Werner 23n. James II l IOn. Jansenism 1 66- 167 Jeffrey, Francis 321n. lobe, T.H. 140n. Johnson, Samuel 1 14n. Jones, G.P. 1 1 5n. Jones, R.F. 193n.

INDEX Jung, Carl Gustav I l l n . Kalter, A. 302n. Kames, Lord (Henry Home) 261 -269, 270n., 308; as 'Cleanthes' 262ff. Kant, Immanuel 25n., 32, 34, 45, 47-48, 5 1 , 60, 80, 206, 287-302, 302n., 303n.; his a priori history of philosophy 1 5-21 , 30-3 1 ; v. Berkeley's dogmatic idealism 289-293; on our consciousness of reality 296-302; and the cosmological argument 243-244; v. Descartes' sceptical idealism 288-289, 293-295; on inner and outer sense 297-298; his integral history 203; his 'magical' world view 104; his restriction of the scope of reason 1 82; and Scottish philosophy 3 1 8 ; his modified theory of space and time 301 ; transcendental argument of 295 Kargon, Robert 229n. Keill, John 1 36, 1 38, 1 42n. Kemp Smith, N. 243, 255 Kenny, Anthony 97, l IOn. Kepler, Johannes 1 0 1 , 103, 1 1 2n., I Bn., 1 9 1 , 192 Kermode, Frank 22n. Keynes, John Maynard 1 1 2n. Khamara, E.J. 269n., 270n. Kiesewetter, J.G.K.c. 303n. Kim, Jaegwon 82n. Kircher, Athanasius 100 Kirk, Robert 97, I IOn. Knight, Gareth 1 1 3n. Knoop, D. 1 1 5n. Koyre, Alexandre 1 4 I n., 143n. Kristeller, P.O. 23n. Kuhn, Thomas 5 1 , 67n., 229n. labelling theory 145- 146 Lacy, John 14 1n. La Forge, Louis de 1 67, 1 74-1 75, 1 78n. language-game 33, 6 1 , 64, 65; transcendental 60 Laplace, Pierre Simon de 3 1 4-3 1 5 , 32 1 n. Lamer, C.J. 1 1 5n.

329

Law, Edmund 244, 269n. laws, governing 230 Laymon, R. 1 77n. Leechman, William 26 1 , 2m70n. legitimation 44, 3 1 2, 320n. Ie Grand, Antoine 1 65, 1 7 1 Lehmann, G . 303n. Leibniz, Gottfried Wilhelm 100, 1 06, 1 14n., \ l 5n., 1 4 I n., 142n., 143n., 1 65, 185, 1 86, 1 90, 195-206, 206n., 302n.; his use of a posteriori justification 201 ; as critic of Descartes 1 95- 1 96; and the criterion of fruitfulness 200, 202; as historiographer l iOn., 125, 206; on the hypothesis of concomitance 207n.; on impact laws 127- 1 28; as victim of integral history 203; on the miraculous 1 36- 137; on Newton, and occultism 98, 1 0 1 , 1 28ff. ; political interests of 1 35, 1 37-1 39 ; on primary and secondary qualities 1 1 2n.; on truths of reason v. truths of fact 1 82. Lennox, J.G. 1 42n. Lerner, Robert E. 1 1 3n. Libavius, Andreas \ l 5n. light of grace, compared with light of nature 105 Lilly, William I IOn. Locke, John 17, 27, 3 1 , 69, 100, \ l2n., 1 24, 128, 1 32, 1 38, 14 In., 142n., 202, 204, 209-225, 229, 233, 239n., 268; as analyst of corpuscularian science 5 1 ; his commitment to the corpuscularian hypothesis 209, 2 14; on God's role in nature 220; on the problem of gravity 226n., 234; seduced into magic 106, 1 30; stresses our narrow conceptions 2 1 6, 226n . ; on the naturalness of superadded qualities 2 1 7-219; his comparison of natural philosphy with geometry 220-224; on primary/secondary quality relations 227n . ; on qualities v. essence of matter 228n.; on the superaddition of thought to matter 209-225 passim, 235-238

330

INDEX

Loemker, Leroy 142n. logic 27, 32, 49, 55, 1 7 l , 3 1 3 logical positivism 28, 35, 89, 257 Lorenz, Konrad 61 Louis XIV 1 66 Luce, A.A. 1 77n. Luther, Maritn 1 83- 1 84, 193n. McCann, Edwin 233-237, 239n. McClaughlin, T. 1 77n. McCosh, James 270n. McCracken, Charles J. 1 77n. Macdonald, Graham 83n. MacDonald, Michael 124, 141 n. MacDonald Ross, G. I l l n., l l 2n., 1 1 7, 123- 1 25, 129- 1 30, 1 35 McEwen, Bruce 243 McGuire, J.E. 14In., 142n., 143n., 229 Maclaurin, Colin 26 1 , 262, 270n. McMullin, Ernan 1 78n. Macnabb, D.G.C. 269n., 270n. McTaggart, John McT. Ellis 4 1 , 48 magic, magician 98-99, 104, 1 l In., 1 1 2n. Malcolm, Norman 285n. Malebranche, Nicolas 37, 40n., 1 00, 1 65 - 1 77, 1 77n., 1 78n.; on collisions 1 70- 1 7 1 ; fuses theology and physics 1 67- 1 68, 1 76; influence of, on Hume 30, 1 6 5 ; on innate ideas 1 75 ; on laws of nature 277n., 228n.; o n the senses and sensory evidence 1 72- 1 74; his appeal to simplicity 167- 1 68 Malezieu, Nicolas de 255, 265 Malinowski, Bronislaw l i On. Manser, Anthony 25n., 48 Mariotte, Edme 165, 166 Marrou, H.1. 22n. Marx, Karl 2 1 , 45, 59 materialist(s}, materialism 13, 285n., 3 1 3- 3 1 6 mathematics 2 7 , 1 06, 149- 1 50, 1 74, 1 76, 1 80, 201 , 305, 3 1 1 , 3 1 5 matter, material world/substance 30, 1 34, 149-1 52, 1 62n., 209, 25 1 -252, 254-255, 260, 264, 277, 288-29 1 , 296, 299; superaddition of thought to 209-210; 2 1 3-219, 235-238

Mayow, John 1 32-133, 142n. meaning: analysis of 32; in Hume 254, 257 mechanical philosophy 98, 1 00, 1 03 , 1 08, 1 1 7 , 1 19, 123, 146, 233-234, 236, 238; and animals I I On., 238; problems of defining 233; limits of 1 30- 1 34; of Locke 209-225 Meikle, H.w. 320n. memory 273-274 Merchant, Carolyn 142n. Mersenne, Marin I IOn., 1 53, 1 54, 1 63n., 1 64n. Mesmer, Franz 1 04, 1 1 3n. Mesnard, Pierre 1 1 3n. metaphysics 3-25, 27-40, 41-46, 47-48, 105, 126, 149- 1 50, 1 53- 1 54, 1 56, 1 76, 195, 197- 199, 20 1-205; see also philosophy method of doubt 198 middle ages 4, 20 Mijuskovic, Ben Lazare 284n. Mill, John Stuart 206n., 208n. Millar, James 305, 3 19n. Miller, G.W. 302n. Milton, John 1 0, 24n. mind, and body 50, 6 1 -62, 97, 1 24- 1 27, 149, 1 67-168, 20 1 -202, 235, 27 1 , 283; Hume's account of 27 1 -272 miracle, miraculous 96, 98-99, 1 02, 1 24, 1 29-1 30, 1 36- 1 3 7 'modern' 5 Moliere, Jean-Baptiste I I I n. Montaigne, Michel de 9, 23n. Moore, G.E. 47 Moore, John 1 2 1 , 140n. More, Henry 97, 99, 1 00, I IOn., I l ln., 1 20- 1 24n 126, 1 27, 1 3 1 , 1 34, 140n., 1 4 1 n., 220 Morin, Jean Baptiste 1 5 5 , 1 59, 1 63n. Morrell, J.B. 3 1 2, 321n. Morris, C. R. 203-204, 208n. Multhauf, Robert I I I n. myths, philosophical 28; see also caricature, integral history naturalism 57, 62, 64-65, 67n., 76-77; in

I NDEX Rorty and Apel 64 ; compared with scientism 76-77 natural kinds 54-55, 58 necessary being 244, 246-248, 2 5 1 -254, 257, 260-26 1 , 263-264, 269 necessity, logical 250, 257, 260 necromancy 96, I I On. neo-idealism 3 1 -35, 38 Neoplatonism 96 'new philosophy' 8 Newton, Isaac 5 1 , 1 08, 1 24, 1 28, 1 34, 14In., 142n., 143n., 190, 1 92, 196, 224, 226n., 264, 270n., 3 14, 3 1 6 ; and alchemy 1 3 1 ; on the problem of gravity 98, 1 76; as magician 1 12n. ; on matter and spirit 1 25 - 1 26; on miracles 1 36- 137; accused of occultism 1 35- 138; his political interests and allies 1 3 5 , 1 3 8 ; public pronouncements v. private convictions 98; the role of spirit in his mechanics 1 28- 1 29 Newtonianism 233, 26 1 , 3 1 1 Nietzsche, Friedrich 45 Noah 8 nominalism, nominalis(s) 1 3, 24n. normal discourse 5 1 , 70-73, 82n. observation(s) see sense perception Occam, William of 1 62n. occasionalism 1 67- 1 69, 227,.n 228n., 229-23 1 ; see also Malebranche occultism, and Catholicism 145; demarcation of from science, religion and philosophy 95-1 09 passim; as deviance 107-109; neglect of 95-96; and philosophical abuse 1 3 5- 1 39 ; and philosophy 100-107; and reason 1 1 7- 1 39; and religion 96- 1 00 ; rhetoric of 1 07- 1 09 ; socio-political significance of 145- 147 Oldenburg, Henry 1 3 1 , 1 90, 194n. Olson, Richard 3 1 2, 321n. Ong, Walter J. 22n., 23n., 24n. ontological argument 243-245 orthodoxy 73, 107- 1 09, 1 1 8, 146 Oswald, James 3 1 7

Outram, Dorinda 305, Overton, Richard 1 1 8,

331

306, 319n., 320n. 1 39n.

pagan, paganism 4-6, 20 pain 36-37, 1 72 Paine, Tom 307 Pakaluk, Michael 258, 270n. Paracelsus (Theophrastus von Hohenheim) 220 Parkinson, G.H.R. 142n. Pascal, Blaise 1 00, 1 5 3 passions, o f pride and humility 272-273, 279 patronage 306-307 Paul, Charles B. 305, 306, 3 19n. Paul, St. 267 Peirce, Charles S anders 60 Penelhum, Terence 285n. Pepys, Samuel 192 perception, theories of 10 1- 1 04, 1 30 percentual knowledge 37-38, 1 72- 1 74, 177, 1 84- 1 85, 192, 290; see also sense perception perennial philosophy 7, 35, 39, 49, S I permanent (in perception) 294-296 Perrault, Claude 1 65 person, concept of 59; see also self personal identity 274, 277, 283-284 Petrarch 8 Pettit, Philip 83n., 85, 87-89 Petty, William 1 20 Peuckert, Will-Erich lOS, 1 12n. 'Philo' 245, 257-260, 265, 267 Philoponus, John 254 philosopher: as entrepreneur 70-7 1 ; as interpreter 7 1 -72; see also hermeneutics; moral philosopher useful 3 1 3 ; as priest of nature 1 3 1 , 1 34; aas under-labourer 45, 70 philosophes 309, 3 1 3-314, 3 16 philosophia Christi 6, 9, 1 2 philosophy: as antique paganism 4-6; as arbiter of knowledge 50, 70; bad v. good 32, 34; barbarian 5; as a battlefield 4 1 -42; post-war British 45; and Christianity 6-9; as loss and recall of common sense 9-1 1 ; as

332

INDEX

critiscism of metaphysical systems 1 1 -20; as cryptic Christiantiy 6-9; as error 20-22, 32-33, 47-48, 1 95, 206; as explanation of error 2 1-22, 30; as hermeneutics 52, 58-67, 69-81 , 83n.; as interpretation 7 1 -72, 74; connection with political interests 1 1 8, 1 35 ; professionalisation of 50; progress in 1 7- 1 9, 27-29, 3 1 , 33, 41-42, 48, 197, 3 12; the right to exist of 44-45; as science 27, 32, 4 1 , 48; the teaching of 5, 45-46; see also perennial philosophy philosophy of mind 78, 280, 3 1 2-3 1 3, 3 1 7 Piaget, Jean 5 9 Pico della Mirandola, Giovanni 8 Piltz, Anders 24n. Plato 6, 7, 8, 1 7 , 20, 1 0 1 , 1 05 pneumatology 1 26, 3 \ \ political radicalism 3 1 5 Popkin, Richard H . 23n., 24n., 270n. Popper, Karl R. 285n. Power, Henry 1 20- 1 2 1 , 1 32, 140n., 142n. power(s) 1 24, 209-221 , 235-238, 254, 263-264; power-relations, social 66-67, 68n. pragmatic conflict 75-76, 89, 9 1 pragmatism 32, 53-54, 68n., 78 Priestley, Joseph 236, 259, 3 1 3-3 1 5, 3 1 8 , 321n. prisca sapientia 95 Proclus 24n. proof see demonstration Protestant(ism) 282; and Catholic 145 Pufendorf, Samuel von 24n. Putnam, Hilary 52-54, 56, 86 Putscher, Marielene I l in. Pyrrho of Elis 6 Pythagoras 6, 7 Pythoness of Endor 96, l IOn. qualities, primary and secondary 29-30, 102- 103, 1 7 1 - 1 72, 2 10-2 \ \ , 2 14, 216, 2 1 8-220, 224, 227n., 234-235 Quine, Willard van O. 32, 36

Ramus, Peter 7, 8, 1 0 Randall, J.H. 1 96, 208n. rational agent model 62 rationalism 47, 6 1 , 1 06, 173, 1 85-187, 1 89, 1 9 1 , 1 95-206 Rattansi, P.M. 1 1 I n., 1 1 5n., 1 39n., 143n. Ravetz, J.R. 109n. 'raw feels' 36-37 Ray, John 253 realism, realist(s) 1 3 , 24n., 52-56, 60 reason 1 8 , 106; v. cause 246-247, 249, 253; the danger of 1 1 7- 1 1 9 ; and empirical evidence 1 72- 175; and faith 183; and imagination 1 84; in Kant 1 82; occultism and 1 1 7-139; pure 1 8 1- 1 83 ; and science 1 84, 1 89, 1 9 1 - 192; and sense 1 8 1 - 1 82; in Spinoza 1 79- 1 8 1 ; union of, with experience 1 88 Reaumur, R.A.F. de 1 39 reduction(ism), reducibility 50, 62-64, 77-78, 1 66, 282 Redwood, J. 229n. Ree, Jonathan 22n., 23n., 29, 4 1 , 42, 44, 97, l iOn., 193n., 203 reference 54-55 Reformation 1 83 Regis, Pierre-Sylvain 1 65, 1 66, 1 7 1 , 1 75, 1 78n. Regius, Henricus \ \ 5n., 1 56, 1 74 Reid, Thomas 1 1 , 1 3 , 24n., 34n 27 1 , 305-3 1 8 , 3 19n., 320n., 321n., 322n.; intellectual progress of 3 10-3 1 1 relativism 56-58, 65-67, 78-8 1 religion, religious 95- 1 00, 107, 1 2 1 , 1 24; see also Christiantiy Renaissance 6, 20 renascence, the concept of 5, 6, 1 2 1 Rescher, Nicholas 256-257, 270n. revelation see truth, revealed Rhine, J.B. 1 1 5n. Ricoeur, Paul 22n. Robertson, William 267, 308-309 Roberval, Giles de 196 Robison, John 305, 3 1 3-3 1 5, 3 19n., 3 2 1 n. Roche, Daniel 305, 306, 3 19n. Rohault, Jacques 165, 1 73, 1 77n.

INDEX

Roman Empire 4 Rorty, Richard 33-40, 40n., 4 1 , 47, 49-59, 6 1 -67, 67n., 68n., 69-8 1 , 8 I n., 82n., 83n., 85-87, 89, 9 1 , 92n. Rosen, G. 140n. Rosicrucian(s) 105, 108 Rossi, Paolo 1 12n. Royal Academy of Sciences see Academie des Sciences Roayl Society of Edinburgh 305, 307-309, 3 1 9n., 320n. Royal Society of London 1 19, 1 2 1 , 123, 1 27, 1 34, 1 39, 183, 190 Russell, Bertrand 2 1 , 32, 47, 85, 285n. Russell, James 3 1 6 Ryle, Gilbert 97, l iOn. Sabine, G.H. 1 39n. Samuel 96 S antinello, Giovanni 24n. Saul, King 96 scalar motion 127-128 sceptici(ism) I I , 1 3 , 1 8-19, 24n., 29-30, 34, 55-56, 1 02, 1 06, 1 20, 1 234, 1 72, 1 97-1 98, 2 8 1 , 289-290; Academic 1 98 ; Cartesian 294 Schaffer, Simon 145, 147 schematism 304 Schmitt, C.B. 23n. scholastic(ism) 7-8, 12, 34, 98, 1 07-108, 1 86- 1 87, 200, 2 1 9, 3 1 1 , 3 1 4 'schools' of philosophy 5, 1 3, 20, 24n., 203, 3 1 8 Schur, E.M. 1 1 5n. Schwartz, Hillel 1 4 1 n. scientific revolution 95, 1 89, 1 9 1 -1 92 scientism 76-78, 280 Scot, Reginald l IOn. Scruton, Roger 1 95- 1 96, 204, 208n. sectarianism 8, 12, 1 1 8 Selden, Jo, 1 23 self: Hume's theory of 27 1-284; v. idea of self 272, 279, 281-282; transcendental 299 self-consciousness 283 ; transcendental 296-298, 301 self-correction 72-73, 87, 9 1

333

self-evidence/intuitive certainty 1 95-1 99, 201-202, 250, 261 self-existent being 244, 246, 263-265 self-expansion 72-73 self-knowledge: in Hume 278-279; in Kant 296, 298 self-positing 299-30 I self-understanding 53, 65-66 Sendivogius, Michael 1 33 Seneca 3 1 3, 32 1 n. Sennert, Daniel 1 22n. sense perception 37-38, 1 0 1 - 1 04, 1 72- 1 74, 1 8 1 , 1 84-185, 1 87, 1 89, 1 9 1 -192, 300 Sergeant, John 140n. Sextus Empiricus 1 06 Shaftesbury, third Earl of 26 1 , 268 Shapin, Steven 142n., 320n. Shapiro, B arbara J. 1 39n. Shoemaker, Sydney 285n. Skinner, Quentin 320n. Smith, Adam 24n., 308, 3 10, 3 1 2 Soal, S.G. 1 1 5n. Societe Royale de Medicine 305 sou see mind, spirit South, Robert 1 2 1 space 283, 288-290, 292-293, 297-301 Spiller, M.R.G. 140n. Spinoza, Benedict (Baruch) 1 00, l iOn., 1 79-193, 193n., 194n.; his contemporaries 190- 1 9 1 ; his use of geometrical examples and axiomatic form 1 80 ; on the role of reason 1 79-1 82; his scientific interests 190- 1 9 1 spirit(s) 1 17, 1 20; -beliefs 96; disembodied 96; embodied 97; and experiments 1 19-1 2 1 ; and matter 1 24- 1 30, 1 32- 133; -testimonies 1 2 1- 1 25 Sprat, Thomas 1 30, 1 4 I n., 1 82 Stabler, E. 302n. Stahl, D.E. 270n. Stanley, Thomas 23n. Staudenbauer, C. 140n. Steneck, N.H. 140n. Steuchus, Augustinus 23n.

334 Stevens, Wallace 3, 22n., 47, 48, 48n. Stewart, Dugald 24n., 305-318, 3 1 9n., 320, 321n., 322n. Stewart, John 264, 270n. Stewart, M.A. 269n. Stillingfleet, Edward 129, 140n., 1 4 I n., 214, 2 1 6, 2 1 8, 236, 237, 238, 253 Stove, D.C. 256-257, 270n. Strawson, P.F. 285n. Stroud, Barry 285n. Stubbe, Henry I l l n. Suarez, Francesco 1 12n. substance(s) 150; see also matter superaddition 209, 2 1 3-219, 223-224, 236-237 superstition 96-97, 1 00, 1 06, 1 08 supervenience 77-78, 83n. supra-mechanical action 132 Sylla, Edith D. 162n. syncretism 8 synthetic a priori 32, 35, 287, 295 system 13, 24n. Tamny, Martin 1 4 1 n. Tandler, Tobias I I I n. textual interpretation 4, 2 1 Thales 6, 20, 2 1 theology 96-100, 1 06, 1 2 1 , 1 66-167, 175-176 Theophrastus 6 Thiers, Jean-Baptiste 1 12n. things-in-themselves 289-293, 295 , 301 -302 Thomas, Keith I I I n., 140n. Thomas Aquinas, St. 149, I SO, 1 5 1 , 1 62n. Thomasius, Gottfried 1 1 5n. Thorndike, Lynn 1 12n. Tillotson, John 253, 270n. time 283, 288, 290, 292-299, 301 Tory, Tories 307, 308 , transcendental: argument 8 1 , 83n., 282, 295, 303n.; consciousness 298; philosophy 294; proofs 294; self-consciousness 296-298, 301 ; theory 302 transcendental idealism, idealist 34, 6 0 ,

INDEX 288, 294-295, 298-299, 303n. Trenchard, John 1 4 1 n. truth, true analytic 32, 1 8 1 ; a priori, a posteriori 1 8 1 - 1 83 ; as consensus 60; criterion of 291-293; and dreams 292, 298; fruitfulness, a sign of 202; hypothetical 1 98- 1 99; and illusion 291 -293; as an island 1 5 - 1 7 ; metaphysical 201 -202; necessary 106, 1 79, 250, 257; objective 76-77; parallel of, with good 56-57; propositional 54-56; revealed 49, 2 1 5 , 224; theoretical 5 7 ; transcendental 56-57 Tschirnhaus, E.W. von 190 Tuck, Richard 24n. Turnbull, George 270n. understanding: descriptive v. participative 6 1 ; in Kant 1 82 unity; arbitrary 256; artificial 253, 258, 265 ; of consciousness 276-278, 280, 282-283, 284n. universities 5, 45, 1 07 van Fraassen, Bas C. 58, 67n. van Leeu\Venhof�k, A nton 1 90 van Straaten, Zak 83n. Vatier, Antoine 1 55 , 1 56, 1 63n. Vaughan, Henry 146 Veitch, John 320n., 321n. verificationism 35-36, 56-57 via antiqua 24n. via moderna 24n. Vico, Giambattista 24n. visual stream see perception vitalism see spirit (and matter) Vives, Juan Luis 8 WagstafTe, John I l ln., 140n. Wallis, John 133, 1 39n. Ward, Seth 1 39n. Waterland, Daniel 244, 269n. Watson, Richard A. 207n, 208n. Webster, Charles 1 32 , 1 39n., 140n., 142n. Webster, John 1 02, l I On., I l in., 1 1 3n.,

INDEX 1 23 Westfall, R.S. 1 14n., 1 4 1 n., 229n. Westoby, Adam 1 93n. Whig(s) 306, 307, 309, 3 1 2 Whiggish interpretation 72 Whiston, William 1 38 Wier, Johann 1 1 1 n. Wiggins, David 8 1 Wilkins, John 1 1 3n., 1 29, 1 33 , 1 39n., 1 4 I n., 253 Williams, Bernard 83n., 1 85 - 1 87, 1 93n. Wilson, C.H. I 94n. Wilson;Margaret D. 86n., 209, 2 1 1 , 2 1 7-22 1 , 223-224, 226n., 227n., 228n. Winstanley, G. 1 1 8, 1 39n. Wishart, William 261 witches, witchcraft 99, 102, l IOn., 1 22-123 Witherspoon, John 261 Wittgenstein, Ludwig 2 1 , 32-33, 36, 5 1 , 76, 8 1 , 86, 285n. Wood, Paul B. 3 1 9 world spirit 98 Wren, Christopher 1 33, 1 38 Wright, John P. 270n. xenodox 73-74, 79, 87-90, 9 1 -92 Yates, Frances A. 1 1 2n. Yolton, John W. 34

335