FROM CAUSE TO CAUSATION
PHILOSOPHICAL STUDIES SERIES VOLUME 90
Founded by WilfridS. Sellars and Keith Lehrer
Editor ...
47 downloads
1190 Views
20MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
FROM CAUSE TO CAUSATION
PHILOSOPHICAL STUDIES SERIES VOLUME 90
Founded by WilfridS. Sellars and Keith Lehrer
Editor Keith Lehrer, University ofArizona, Tucson Associate Editor Stewart Cohen, Arizona State University, Tempe Board of Consulting Editors Lynne Rudder Baker, University of Massachusetts at Amherst Radu Bogdan, Tulane University, New Orleans Marian David, University of Notre Dame Allan Gibbard, University of Michigan Denise Meyerson, University of Cape Town François Recanati, Institut Jean-Nicod, EHESS, Paris Stuart Silvers, Clemson University Barry Smith, State University of New York at Buffalo Nicholas D. Smith, Michigan State University
The titles published in this series are listed at the end of this volume.
FROM CAUSE TO CAUSATION A Peircean Perspective
by
MENNO HULSWIT Heyendaäl Institute, University of Nijmegen, The Netherlands
KLUWER ACADEMIC PUBLISHERS DORDRECHT / BOSTON / LONDON
A C.I.P. Catalogue record for this book is available from the Library of Congress.
a
METU LIBRARY
19^5.
P ^
...>,ııııı
H85 XKJOI
0050343792
ISBN 1-4020-0976-3 (HB) ISBN 1-4020-0977-1 (PB)
Published by Kluwer Academic Publishers, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. Sold and distributed in North, Central and South America by Kluwer Academic Publishers, 101 Philip Drive, Norwell, MA 02061, U.S.A. In all other countries, sold and distributed by Kluwer Academic Publishers, P.O. Box 322, 3300 AH Dordrecht, The Netherlands.
Printed on acid-free paper
All Rights Reserved © 2002 Kluwer Academic Publishers No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands.
CONTENTS
Acknowledgements
ix
Note on References
xi
Preface
xiii
CHAPTER 1 : SOME KEY MOMENTS IN THE HISTORY OF THE CONCEPT OF CAUSATION
1
1
Causation in Ancient Greece
2
1 Aristotle: Four Types of Explanation
2
2 The Stoics: Causation, Exceptionless Regularity, and Necessity
5
Causation in the Middle Ages
8
1 Thomas Aquinas
8
2
3
4
Causation in Modern Philosophy
15
1 The Metaphysical Systems from Descartes till Leibniz
17
2 Critical Philosophy from Locke till Mill
27
Conclusion: Important Changes in the Meaning of Cause
41
CHAPTER 2: CONTEMPORARY APPROACHES TO CAUSATION
47
1
The Contemporary Debate
47
1 Necessary and/or Sufficient Conditions
47
2 Causes and Counterfactual Dependency
54
3 The Instrumental Approach: Causes as Means-to-Ends
56
4 Probabilistic Causation
58
5 The Singularist Approach
60
V
CONTENTS
Acknowledgements
ix
Note on References
xi
Preface
xiii
CHAPTER 1 : SOME KEY MOMENTS IN THE HISTORY OF THE CONCEPT OF CAUSATION
1
1
Causation in Ancient Greece
2
1 Aristotle: Four Types of Explanation
2
2 The Stoics: Causation, Exceptionless Regularity, and Necessity
5
Causation in the Middle Ages
8
1 Thomas Aquinas
8
2
3
4
Causation in Modern Philosophy
15
1 The Metaphysical Systems from Descartes till Leibniz
17
2 Critical Philosophy from Locke till Mill
27
Conclusion: Important Changes in the Meaning of Cause
41
CHAPTER 2: CONTEMPORARY APPROACHES TO CAUSATION
47
1
The Contemporary Debate
47
1 Necessary and/or Sufficient Conditions
47
2 Causes and Counterfactual Dependency
54
3 The Instrumental Approach: Causes as Means-to-Ends
56
4 Probabilistic Causation
58
5 The Singularist Approach
60
VI
CONTENTS
vii
2 Natural Kinds and the Uniformity of Nature; Peirce's Earliest Discussion of Natural Kinds
105
3 Peirce's Baldwin Definition of Kind
107
4 The PRE-Character
108
4
Kinds and Classes
109
5
Classification According to Final Causes
112
6
Criteria of Demarcation
116
7
Recapitulation: Definition of Peircean Natural Classes
119
8
Why Believe in Natural Classes?
121
9
Examples of Natural Classes
122
1 Examples from the Realm of Human Experience: Social Classes, the Sciences, and Man-Made Objects
122
2 The Chemical Elements
123
3 The Biological Species
126
10
Was Peirce a Pluralist Regarding Natural Classes?
127
11
Conclusion: Natural Classes and Causation
131
CHAPTER 5 : THE RIDDLE OF SEMEIOTIC CAUSATION
133
1
Some Fundamental Conditions of Signs as Such
134
1 Early Basic Insights
134
2 Later Developments
136
2
T.L. Short
139
3
Joseph Ransdell
144
4
Some Problems Generated by Short's and Ransdell's Views
147
5
The Causal Role of the Dynamic Object
148
1 Positive Evidence
148
2 Negative Evidence
150
6
Icon, Index, and Symbol
153
7
The Meaning of'Determines'
160
viii 8
CONTENTS Conclusion
164
CHAPTER 6: A SEMEIOTIC ACCOUNT OF CAUSATION
167
1
Criticism of the Received View
167
1 Contemporary Approaches to Causation
168
2 Two Mutually Incompatible Conceptions of Cause
170
3 The Inadequacy of the Received View
171
4 Two Mutually Incompatible Categoreal Frameworks
176
5 Conclusion to Part 1 : Criticism of the Received View
179
2
Necessary Conditions for a Theory of Causation
180
3
Peirce on Causality and Causation
181
1 Peirce's Critique of the Principle of Causality
181
2 Peirce's Conception of Causation
187
3 Causality and Causation: Facts versus Events
190
4 Events and Processes
192
5 Conclusion to Part 3: Causality and Causation
194
A Semeiotic Approach to Causation
195
1 Some Formal Characteristics of Semeiosis
195
2 The Problem of Semeiotic Causation
197
3 Semeiosis Provides the Formal Structure of Causation
198
4 A Semeiotic Approach to Causation
199
Conclusion: a Peircean Approach to Causation
213
4
5
Notes
219
Bibliography
233
Index
243
ACKNOWLEDGEMENTS
I am happy to express my gratitude to a number of people who in some way or other have contributed to the completion of this book. Prominent among them are Guy Debrock, Herman de Regt, Andre De Tienne, Carl Hausman, Nathan Houser, Kenneth Laine Ketner, Peter Kroes, Doede Nauta, Jaime Nubiola, Joseph Ransdell, John F. Sowa, Jaap van Brakel, Shekar Veera, and Rob de Vries. I am especially grateful to Mijke Jetten, who prepared the camera-ready version of the book and who helped making the index, and to Marianne Stienen, who patiently supported me during most of the writing process. I thank the department of Philosophy of Harvard University for permission to quote from the unpublished manuscripts of Charles Sanders Peirce, housed in the Houghton Library. I thank "The Institute for Studies in Pragmaticism" (Texas Tech University, Lubbock) and the "Peirce Edition Project (Indiana University, Indianapolis) for giving access to various sorts of resources. I owe a special word of thanks to Kenneth Ketner (Lubbock), as well as to Nathan Houser and the other members of the Peirce Edition Project (Indianapolis) for their assistance and wonderful hospitality. One of them I wish to mention by name: Andre De Tienne. Andre not only incessantly stimulated me, but he became the zealous promoter of my work. It is a great pleasure and privilege for me to have such an overseas ambassador. Finally, I would like to thank the editors of the Transactions of the Charles S. Peirce Society: A Quarterly Journal in American Philosophy for their permission to use four of the articles that I published in their journal. I also thank the Foundation for Research in the Field of Philosophy and Theology, which is subsidized by the Netherlands Organization for Scientific Research (NWO), for funding my research, and the Heyendaal Instituut Nijmegen for giving me the opportunity to prepare my work for publication.
IX
NOTE ON REFERENCES
All references to sources are given in the body of this study. References to writings other than those by Charles S. Peirce are given as follows: author, date of work cited, and page of text cited - for example (Mackie 1974, 32). The complete bibliographical data are given in the 'Bibliography' at the end of this study. References to writings by Peirce himself are also given in the body of the text, following the standard procedures in Peirce scholarship. References to the Collected Papers of Charles Sanders Peirce (Volumes I-VI, Ch. Hartshorne and P. Weis, eds., and Volumes VII-VIII, A. Burks, ed., Cambridge Mass., Harvard University Press, 1931-1935 and 1968) are given as follows: CP, volume number, decimal point, and paragraph numbers; for example, "CP 4.321" refers to paragraph 321 in volume 4 of the Collected Papers. References to the Writings of Charles S. Peirce: A Chronological Edition (Volume 1: 1857-1866; Volume 2: 1867-1871; Volume 3: 1872-1878; Volume 4:1879-1884; Volume 5: 1884-1886; ed. "The Peirce Edition Project", Bloomington and Indianapolis, Indiana University Press, 1982, 1984, 1986, 1986, 1993) are given as follows: W, volume number, colon, and page number; for example, "W4:321" refers to volume 4, page 321 of the Writings. References to The Essential Peirce. Selected Philosophical Writings (volumes I, 1867-1893; eds. N.Houser and C. Kloesel; Volume II, 18931913; ed. "The Peirce Edition Project"; Bloomington, Indiana University Press, 1992 and 1998) are given as follows: EP, volume number, colon, and page numbers; for example "EP 11:300" refers to volume II, page 300 of The Essential Peirce. References to Reasoning and the Logic of Things (1998, Cambridge Mass., Harvard University Press, 1992; ed. K. Ketner) are given by RLT followed by the relevant page numbers. References to The New Elements of Mathematics (I-IV, ed. C. Eisele, The Hague, Mouton, 1976) are given as follows: NEM, volume number, colon, and page number; thus NEM IV:200 refers to volume IV, page 200 of The New Elements. References to Peirce's letters to Lady Welby, in Hardwick, ed., Semiotic and Signifies: The Correspondence between Charles S. Peirce and Victoria xi
xii
NOTE ON REFERENCES
Lady Welby (Bloomington, Indiana University Press), are given by LW followed by the page numbers. Finally, references to unpublished manuscripts are given by MS, colon, followed by the number assigned to it in Richard S. Robin, Annotated Catalogue of the Papers of Charles S. Peirce (The University of Massachusets Press, 1967). Where relevant, the pages of the unpublished manuscripts are designated by the five-digit numbering system of the Texas Tech Institute for Studies in Pragmaticism; for example, MS 318:00024 refers to manuscript 318, page 24 according to Kenneth Ketner's five-digit numbering system.
PREFACE
Human discourse, whether ordinary, scientific or moral, is characterized by an abundant use of causal terminology. We constantly use causal words - not only explicit references to causes and effects, but also verbs that imply the concept of cause. Some verbs, like 'producing,' 'generating,' 'determining,' 'bringing about,' are blatantly causal. Many transitive verbs, such as 'breaking,' 'dropping,' 'cooking,' implicitly refer to causal relations. Breaking a twig is causing it to break; dropping something is causing it to fall, and so forth. In science, explanations often refer to causes or to causal connections. Indeed, the predominant objective of science may be characterized as the systematic description of or the search for causal relations. In moral discourse, the concept of moral responsibility, which is involved in all moral reasoning, inherently refers to the idea that individuals in some way are the cause of their actions. Similarly, debates between lawyers before a court of law concern the question whether a particular human action (or the omission thereof) was the cause of some particular damage or harm. Lawyers must show beyond the shadow of a reasonable doubt that there is a real causal link between some action and some damage. The fundamental importance of the concept of causation, both to our ordinary practical concerns and to our scientific descriptions of the world, has led some philosophers to the extreme view that causation is so basic as to refuse all further analysis (Scriven 1966; Taylor 1966; Anscombe 1971; Fales 1990). Others, such as Bertrand Russell (1912), took to the other extreme, by arguing that the concept of causation better be abandoned in scientific discourse because it ought to be seen as the unfortunate result of an anthropocentric projection of our confused conception of human agency and power. According to Russell, "... the word 'cause' is so inextricably bound up with misleading associations as to make its complete extrusion from the philosophical vocabulary desirable [...]. The law of causality [...] is a relic of a bygone age" (Russell [1912] 1991, 119). Most contemporary philosophers hold a position that lies somewhere between these two extremes. They respect the central role of the concept of causation in our understanding of the world, and they share the conviction that, problematic though the concept may be, it can be analyzed. I too share that belief and, though I think all philosophical views regarding causation
xiii
xiv
PREFACE
suffer from a basic misconception, I too will try to elucidate the nature of causation. My primary objective in this book is to clarify the ontological question, which in my view concerns causation - or the production of an effect by its cause - rather than causality, which is the relationship between cause and effect. The distinction between causation and causality is of paramount importance in this study. More specifically, I hope to show that many of the unsatisfactory aspects of the existing theories regarding causation are related to the failure to respect this distinction. Breaking a glass is something quite different from the relationship between the 'cause' of the shattered glass and the shattered glass. Our experience of causation is closely related to a certain type of our experience of signs. Our experience of fire causing smoke is connected to our experience of smoke being the sign of fire. The search for causes is always related to the experience of possible symptoms, which are both the effects and the signs of some cause. When the first plane crashed in the World Trade Center, most witnesses saw the horrible event as a sign of some dreadful human error or some fatal mechanical failure; but when the second plane crashed, no one doubted that the crash was the symptom of some terrorist plan. Given the major role of signs in our experience of causes, the scant attention paid to this close relationship between signs and causes in contemporary discussions may itself be regarded as the symptom of a rather remarkable blind spot in the prevailing views of causation. And thus, an approach to the problem of causation from the point of view of an adequate theory of signs is long overdue. My hypothesis is that Peirce's semeiotic provides the best available instrument for working out such approach. Though C.S. Peirce (1839-1914) has never written a work devoted to causation, his innumerable papers and manuscripts contain many valuable insights on the subject, most of which have not yet sufficiently been recognized by the contemporary philosophical community. Thus, my first objective is to make a critical analysis of Peirce's conception of causation. My second objective is to develop a novel approach to causation, based upon the semeiotic of Peirce. I will briefly explain how the various chapters contribute to these objectives. Philosophical theories are always answers to questions that have arisen within a certain historical context. Understanding a philosophical problem therefore requires some historical perspective. Chapter 1 therefore will consist of a review of some important moments in the historical evolution of the concept of causation, which will enable us to better understand the contemporary debates on causation. It will be shown that the complex
PREFACE
xiv
history of the concept of cause is marked by two decisive milestones: (I) the Aristotelian-scholastic conception, according to which causes are the active initiators of some change, and (II) the scientific conception, according to which causes are inactive nodes in a law-like chain of implication. The modern concept of cause will be shown to be the unfortunate result of the interplay of these two conceptions. According to the first conception, 'A is the cause of B' means 'A is the active initiator of a change in B.' According to the second conception, the same expression means: 'Given the occurrence of B, A must necessarily have occurred.' Any analysis of causation must take into account the tension between these two aspects of our concept of causality. In the chapters 3 - 6 1 hope to show that C.S. Peirce provides us with some basic insights for an alternative theory which may resolve this tension. Chapter 2 provides an overview of the contemporary accounts of the concept of cause. The function of this chapter is twofold: first, by offering a clearer view both of the qualities and the shortcomings of the contemporary approaches, it will allow us to better appreciate the relevancy of Peirce's conception of causation; secondly, the conceptual analysis as put forward in this chapter will function as a rough guide to the ontological analyses provided in the later chapters. The contemporary views on causality may roughly be reduced to five different approaches. According to (I) the 'necessary and sufficient conditions approach,' a cause may be described as follows: an event A is the cause of an event B if and only if A is a necessary part of a complex set of conditions C, while C is a sufficient but non-necessary condition of B. For instance, a short-circuit is said to be the cause of a certain fire because though by itself it would not be a sufficient condition of the fire - it is the decisive factor in a set of circumstances (including, for instance, the presence of combustible material) which, while not being a necessary condition, is a sufficient condition for the fire to occur. The fire could indeed have been caused by a different set of circumstances; but given this set of circumstances, the short circuit, which is a constituent element of that set, was the decisive factor.' According to (2) the 'counterfactual approach,' a cause is simply an event without which another event would not have occurred. Thus, saying that a nail was the cause of my flat tire, is saying that, were it not for the presence of the nail, my flat tire would not have occurred. According to (3) the 'instrumental approach,' a cause is an event or state of affairs that can be manipulated so as to bring about or to prevent another event or state of affairs. Saying that a defective gene is the cause of a certain tumor is saying that the illness can in principle be cured by repairing the
xvi
PREFACE
defective gene, or that we could bring about the illness by damaging the gene. According to (4), the 'probabilistic approach,' an event A is the cause of an event B if and only if, given an event A, the probability that B occurs is greater than if A had not occurred. Thus something is the cause of cancer inasmuch as it increases the probability of the occurrence of cancer. This approach may be seen as a 'soft' version of (1), which in effect states that the occurrence of event A involves the probability value of 1 that event B will occur. The 'singularistic approach' (5) is characterized by the idea that causal relations are irreducibly individual, rather than instantiations of universal relations. Thus, one may describe the cause of an event B as the one single event (A) that occurs closest to and immediately preceding event B. For instance, the cause of the fire was the short-circuit because the short-circuit was the one single event immediately preceding the fire. Moreover, the short-circuit, in combination with other circumstances, was the sufficient condition for the outbreak of fire. It will be shown that, though each of these approaches highlights important aspects of the concept of causality, none is ultimately quite satisfactory. Moreover, a number of the most stubborn problems plaguing the modern analytical approach will be discussed and, by the same token, an attempt will be made to list the requirements that must be met by an adequate theory of causation. This will involve a discussion of issues such as (1) the question of continuity in time and space between cause and effect, (2) the question of the probabilistic nature of causal laws, (3) the relationship between causation and time, (4) the direction of causation, (5) causation and 'agency,' (6) teleology, (7) the relationship between uniformity and causation, and (8) the universality of causation. The first two chapters will provide the framework in which Peirce's approach to causation is particularly promising. Chapter 3, which is devoted to Peirce's theory of final causation, is a key chapter of this book. In view of the generally anti-teleological attitude of most scientists, Peirce's defense of teleology seems outdated and even reactionary. But Peirce does not merely defend teleology, he revolutionizes the concept, and by doing so, he revolutionizes the entire concept of causation. In the first part of chapter 3, Peirce's conception of final causation will be analyzed. An important aspect of that conception derives from his conviction that the rejection of final causation by most contemporary philosophers is due to a grave misunderstanding of the concept. The second part of the chapter illustrates the latter point by providing a Peircean critique
PREFACE
xvii
of Ernst Mayr's conception of teleology, which is widely accepted among philosophers of science in general, and philosophers of biology in particular. The basic point of Peirce's critique is that final causes are not future events, but general possibilities, and that the hallmark of final causation is its being a basic characteristic of every process. More specifically, Peirce's theory explains (1) that the final state of a process may be reached in different ways, and (ii) that a process is irreversible. Peirce argues that, far from being two different types of causation, efficient and final causation complement each other in as much as each act of causation has both an efficient and a teleological component. Likewise, he argues that the distinction between purely mechanic and teleological processes is misguided because each and every process has a more or less important teleological aspect. Moreover, in his view, all causation involves an aspect of irreducible novelty, which is related to 'objective chance'. Peirce's theory is helpful in laying bare the fallacies inherent in three of the most important premises of nearly all contemporary discussions regarding teleology, including Ernst Mayr's. Final causes are often rejected because they are, erroneously, thought (i) to be individual events that (ii) retroactively influence the present (backward causation), and (iii) do so in a strictly determinate manner. Thus, if my analysis is correct, Ernst Mayr's rejection of final causation is based on a fatal misunderstanding of the nature of teleological processes. Moreover, if I am right then Mayr's distinction between genuinely enddirected or teleonomic processes and seemingly end-directed or teleomatic processes is essentially without foundation, for both require an explanation by final causation. Chapter 4 deals with Peirce's account of natural classes, a topic which at first may seem irrelevant to the problem of causation. To understand its relevance, one must know that Peirce's analysis of the concept of natural classes is entirely related to the concept of final causation. Peirce holds that causes and effects cannot be mediated by final causes (or 'laws') unless they belong to a natural class. Thus, an adequate explanation of Peirce's conception of causation requires an account of his conception of natural class. Conversely, Peirce's conception of natural class must be considered within the perspective of final causation. Peirce's theory of natural classes involves at least two important insights: (1) all classification is related to some purpose, and (2) natural classifications differ from artificial classifications in as much as the former are related to the final causes of the things themselves. Peirce's conception may be summarized as follows: all things belonging to a particular class do so in virtue of some 'essence' and some additional
xviii
PREFACE
class properties. Thus, for instance, the class of chairs is a natural class because they share an essence that consists in their being for the sake of 'being sat upon', while the property of 'having legs' is a class property. The 'essence,' therefore, is a general principle in virtue of which the members of the class have the same final cause, which is implied in their tendency to behave themselves in a specific manner. Natural classes are non-existing entities; they have the status of possibility, not of actuality. Though the natural classes most relevant to Peirce are those encountered in scientific taxonomy - such as elementary particles in physics, gold in chemistry, and species in biology - they also do involve man-made objects and social classes. Chapter 5 provides an elementary account of Peirce's semeiotic and a critical discussion of the role of causal elements in semeiosis (the activity of signs). The primary function of this chapter is to pave the way for my new, semeiotic approach to causation as proposed in the last chapter of this book. My secondary objective is to solve the so-called problem of 'semeiotic causation,' which according to many Peirce scholars is the most difficult problem in Peirce's semeiotic. I hope to show that any attempt to analyze semeiosis in causal terms is highly problematic. There is a consensus among scholars that Peirce regarded semeiosis as a teleological process. If this interpretation is correct, and if it is true that, as Peirce maintained, teleology involves final causation, efficient causation and chance, then it should in principle be possible to point out the exact role of these causal elements within semeiosis. But, so far, no satisfactory solution to this problem of 'semeiotic causation' has been given. A number of Peirce's later texts strongly suggest that, in spite of zealous attempts, he failed to find a single, encompassing account of semeiotic causation. I will argue that such general account is not possible in principle, and that Peirce's very own classification of signs into icons, indices and symbols expresses three different forms of semeiotic causation. Chapter 6 consists in an attempt at reformulating the problem of causation in the light of Peirce's semeiotic. A review of the reasons for my conviction that the existing theories of causality are inadequate (1), of the historic roots of this inadequacy (2) and of Peirce's approach to causality (3) will be followed by my proposal for a new approach to the problem of causation (4), which is based upon Peirce's semeiotic. (1) The chapter opens with a discussion of the distinction between causality, which is the relation between cause and effect, and causation, which is the production of a certain effect. I will argue that causation is more fundamental than causality, because causes and effects are aspects that have
PREFACE
xix
been abstracted from a continuous process of causation. If this view is correct, as I think it is, then it follows that thinking causation in terms of the relationship between these abstracted elements, which is the case for most contemporary theories, is basically flawed, because it confuses conceptual abstractions and the concrete reality from which they have been abstracted. By focusing on causality, the contemporary theories fatally neglect the more fundamental problem of causation. (2) Next, I try to show how the history of the concept of cause is intrinsically related to the historical evolution from a substance ontology (Aristotle) to a fact ontology (as illustrated by the scientific worldview of the seventeenth century). The inconsistency that characterizes contemporary conceptions of causality is largely due to the meshing of two incompatible systems of categories that form, respectively, the basis of these two ontologies. This is particularly obvious in the discussions regarding the nature of causal relata (the question whether causes and effects are events, facts, substances, or possibly something else). It is suggested that the aporia of contemporary analyses of causality are largely due to an insufficient appreciation of the lack of a consistent set of categories. Peirce's doctrine of the categories provides exactly such consistent framework, and thereby offers a perspective that is conducive to an adequate analysis of causation. (3) The advantage of Peirce's discussion of causality is that it involves both a critique of the principle of causality and a constructive theory of causation. Peirce convincingly shows that (i) cause-effect relations are irreversible, (ii) effects are only partially determined by their causes, (iii) cause-effect relations are mediated by final causes ('laws'), and (iv) causes precede their effects. Peirce viewed causation as a productive event that is part of a teleologically determined sequence (a process). Though every event is produced by a preceding event (the efficient cause), it is also part of a continuous sequence that is characterized by a certain direction (which is determined by the final cause). Moreover, each event is characterized by an aspect of irreducible novelty (objective chance). (4) The ultimate purpose of the last chapter is the development of a new approach to causation, based upon C.S. Peirce's semeiotic. I argue that the seemingly stubborn character of the problem of semeiotic causation as described in chapter 5 is due to a misrepresentation of the problem. This misrepresentation is partly due to the nearly universal but wrong acceptance of Peirce's own misleading suggestion that semeiosis implies the production of events. If it is granted that semeiosis does not concern the production of events, but merely the dynamic relationship between certain aspects of events, it follows that semeiosis does not
xix
PREFACE
necessarily presuppose causation, whereas all causation presupposes semeiosis. Thus, the underlying assumptions of chapters 5 and 6 are radically different. Whereas the basic assumption of chapter 5 is the commonly accepted view among scholars that Peirce conceived semeiosis as a teleological process that somehow involves final causation, efficient causation and chance, the underlying assumption of chapter 6 is that such analysis is impossible on principle. I will argue that if one accepts the commonly accepted view, the analysis of semeiosis in terms of causal concepts should go along the way that is suggested in chapter 5. If, on the other hand, one accepts my argument that causation presupposes semeiosis rather than the reverse view, then the road is free to a new, semeiotic approach to causation as such. Next, the anatomy of (Peircean) events will reveal that events not only produce other events, but that they always produce events of a certain type. More specifically, each event (i) involves a decision selecting one of many possibilities, (ii) is conditioned by a cluster of preceding events and the relevant final cause pertaining to such cluster, (iii) conditions and limits the possibilities regarding future events. Each new event assimilates a past event from a certain perspective. Though the past event stands to the present event as an efficient cause to its effect, this relation is mediated by a final cause, which is the perspective within which cause and effect are related to each other. Each perspective reflects a conditional necessary relation, which has the structure of a material implication (if p, then q). Moreover, Peirce shows that causation involves the transmission of forms. But, contrary to Aristotle's formal causes, Peircean forms are relational structures that have the logical structure of a material implication. Though both Aristotle's formal causes and Peirce's forms are meant to explain the stability of the world, the all important difference between them is that the former pertain to the structure of things, while the latter pertain to the dynamic relation between events. It is shown that the transmission of forms, which is fundamental to causation, is a semeiotic process. Effects are always the signs of their causes, and some event-causes function as signs of future effects. In both cases the interprétant (the mediating sign between a sign and its object) is the mediator of the transmission of forms. Thus, while the symptoms of an illness refer to their cause, the relevant medical law is the interprétant in virtue of which the symptoms (sign) are related to their cause (object). Similarly, when clouds are the sign of rain to come, the interprétant is the law-like relation between cloud-signs and the possible rain.
PREFACE
xxi
The theory thus presented satisfies a number of important conditions that must be met by an adequate theory of causation: (i) it is based upon a single categoreal framework; (ii) it adequately explains the relationship between events and processes, (iii) it explains the relation between causation and continuity, (iv) between causation and time, and (v) between causation and teleology. Moreover, (vi) it explains the status of causes and effects as such. An important aspect of the theory presented is that causes and effects are, though objective, also dependent upon a certain perspective. Because they are aspects that have been abstracted from the continuous flow of events, they are not only determined by concrete reality as such, but also by the abstracting power of our thought. They are not events, but facts. The author of a book is, no doubt, the cause of his book. But the cause and the effect are abstractions derived from the process of writing. Both the author and his book are born from the writing.
Chapter 1 SOME KEY MOMENTS IN THE HISTORY OF THE CONCEPT OF CAUSATION
Those who make causality one of the original uralt elements in the universe or one of the fundamental categories of thought, - of whom you will find that I am not one, - have one very awkward fact to explain away. It is that men's conceptions of a cause are in different stages of scientific culture entirely different and inconsistent. The great principle of causation which, we are told, it is absolutely impossible not to believe, has been one proposition at one period in history and an entirely disparate one [at] another is still a third one for the modern physicist. The only thing about it which has stood [...] is the name of it. (C.S. Peirce, RLT, 197, 1898)
Philosophical theories, obviously, are always answers to questions that are raised within certain historical contexts, which involve the common presuppositions of an era. A thorough insight into a particular philosophical problem therefore requires a historical perspective. In order to better understand the contemporary approaches to causation, and the problems they raise, some important historical moments in the evolution of the concept of causation will be briefly considered in this chapter. I will restrict the exposition primarily to four aspects of the problem of causation: (i) necessity, (ii) the nature of the causal relata, (iii) teleology, and (iv) the distinction between primary and secondary causes. The reason for selecting necessity is obvious: not only is it the most important issue in the contemporary debate, but the concept of causation has, in some way or another, always been associated with necessity. The issue of the relata has been selected because it is of decisive importance for a proper elucidation of this problem. My selection of the problem of teleology may seem strange at 1
2
CHAPTER ONE
first, because according to the current view, the idea of final causation is obsolete. However, my selection is motivated by the fact that Peirce's conception of teleology is of permanent importance to his theory of causation. Finally, I choose to select the issue of primary and secondary causes, not only because of its importance for medieval philosophy, but also because the modern idea of (secondary) efficient causes as involving determinism was rooted in the medieval idea of God's omniscience and omnipotence. Thus, the distinction between primary and secondary causes had a profound and lasting influence on the evolution of the concept of cause. I shall focus my attention upon the conception of cause in, successively, Ancient Greek Philosophy (Aristotle and the Stoics), the Middle Ages (Thomas Aquinas), and the Modern period (Descartes, Hobbes, Leibniz, Locke, Newton, Hume, Kant, and Mill).
1.
CAUSATION IN ANCIENT GREECE
The philosophical concept of causation has a long history. The early Milesians, for example, were concerned primarily with material causes, believing they could explain the world by identifying some basic stuff water, air, or the like - which all things are composed of. The first intimation of an explicit need for the concept of an efficient cause is to be found in Empedocles's notion that besides the elements of earth, air, fire and water, two further elements, 'Love' and 'Strive,' were needed for drawing the elements together or keeping them apart. Plato was probably the first to state the principle of causality: "everything that becomes or changes must do so owing to some cause; for nothing can come to be without a cause" (Timaeus 28a). But, Plato emphasized the causal importance of formal causes. Nothing can be unless there be a changeless pattern of which the individual sensible phenomenon is a mere appearance. Aristotle, who considered Plato's approach highly one-sided, said that forms were put forward by Plato to be what in Aristotle's terms are 'efficient causes,' sources of change or movement.1 Since Aristotle was the first philosopher to give an extensive account of causes, I will start with a discussion of his theory.
1.1
Aristotle: four types of explanation
Though Aristotle discussed his theory of 'causation' in many places, the most important passages are to be found in his Posterior Analytics, his Physics,
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
3
and his Metaphysics. The context always concerns both a certain being and the conditions of knowledge of that being. Thus, Aristotle said, for example, in his Posterior Analytics that knowing a thing involves knowing its aitiai.2 Until recently, Aristotle's theory of the four aitiai was commonly understood as a theory of causation. But many of his examples of X's standing in a relation of aitia to Y's cannot be called causes in the sense of being entities bringing something about or producing a change. We cannot say, for example, that bronze produces a statue, or that the ratio of 2:1 produces the octave (see quote further on). It is therefore important to realize that aitiai are 'prerequisite conditions' rather than causes. Aristotle introduced his theory of aitiai as manners of answering the question, dia til, why? He recognized four ways of answering the question why something is the case. These quite different modes of explanation he called respectively, the material, formal, efficient, and final aitia: In one way, then, that out of which a thing comes to be and which persists, is called an aitia, e.g. the bronze of the statue, the silver of the bowl, and the genera of which the bronze and the silver are species. In another way, the form or the archetype, i.e. the definition of the essence, and its genera, are called aitiai (e.g. of the octave the relation of 2:1, and generally number), and the parts in the definition. Again, the primary source of the change or rest; e.g. the man who deliberated is an aitia, the father is aitia of the child, and generally what makes of what is made and what changes of what is changed. Again, in the sense of end or that for the sake of which a thing is done, e.g. health is the aitia of walking about. ('Why is he walking about?' We say: 'To be healthy', and, having said that, we think we have assigned the aitia.) The same is true also of all the intermediate steps which are brought about through the action of something else as a means towards health. All these things are for the sake of the end, though they differ from one another in that some are activities, other instruments. (Physics II.3, 194b23-195a3) 3
Thus, the material aitia of a statue, for instance, is that from which (hyle) it is built, say marble; 4 the moving or efficient aitia is the sculptor, more exactly, the form in the builder's soul. The formal aitia is its pattern or form (eidos); and the final aitia is its purpose or end (telos)\ for instance, the possession of a beautiful object. The complete explanation of the coming to be of a statue will take into account all of these explanatory factors, because without them the statue would not have existed.
4
CHAPTER ONE
Only the efficient aitia has features we now associate with the idea of causation: first, it produces something that is distinct from itself (the sculptor is distinct from the sculpture); and secondly, it is either simultaneous with its effect, or it precedes its effect (Phys. II.3, 195b 16-21). I will therefore use the terms 'cause' and 'causation' only relative to the efficient aitia. Efficient causation involves a form being transmitted from the efficient cause to the effect. In natural processes, the efficient cause is the same in form as what is produced, as when "man begets man." In artificial productions, the efficient cause is the form in the mind of the artificer: "from art proceed the things of which the form is in the soul" (Metaph. VII.7, 1032al 1-I032b23). The form of his product (effect), which is the same qua form in the efficient cause, comes about from him by means of the motion he sets up. It is his soul, in which is contained the form, that moves his hands in such a way as to make an object with the same form (Generation of Animals I, 21-22). However, how the form is transmitted from the cause to the effect is one of the major riddles of Aristotle's theory of causation. The efficient aitia was defined by reference to some change: it is the "primary source of the change," or "that whence the change comes" (Metaph. V.4, 1014b 18-20). It was thought of as something that produces something. That which was produced was either some new substance, such as ashes from wood, or simply a change in some property of a given substance. Aristotle allowed various kinds of entity to be the efficient cause: sometimes it is a substance, like the sculptor as the cause of a statue {Phys. II.3, 195a34), or the father as the cause of the child (Phys. II.3, 194b31). But sometimes it is an event: hard work is the cause of fitness (Phys. II.3, 195a9), and a raid can be the cause of war (Phys. II.7, 198a 19). However, it appears clearly from Aristotle's Categories that he defended a substance ontology: there can only be relations in the world, of whatever kind, by virtue of substances which can be related. And there can only be events because there are substances behaving in a certain way. Substances constitute 'the basic furniture of the world,' and without them there is blank nothing.5 This allows for two quite different interpretations regarding the causal relata. According to the first (more or less standard) interpretation, all causal claims in terms of states or events are elliptical; the only genuine causal explanatory items are substances. This view entails that, though in some sense efficient causes need not be substances, but may be events, any series of causally related events must have an initiating cause, which must be a substance rather than just another event. The above-mentioned example of 'the hard work (event) as the efficient cause of fitness,' presupposes that there is a person (substance) who does the hard work, and who therefore is the initiating efficient cause. Only this initiating cause could be properly spoken of as the efficient cause of these changes.
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
5
However, recently Annas (1982) put forward the hypothesis that, whereas Aristotle was prepared to mention a wide variety of items as causes (such as people, objects, states and events), he only counted events and states as the proper efficient aitiai. Though Aristotle elliptically cited all kinds of things as efficient causes, properly spoken, the only genuine causal explanatory item is "a state or an event involving some substance" (Annas 1982, 321). By saying, for example, that Polycleitos was the efficient cause of the statue (Phys. II.3, 195b 16-25), Aristotle did not mean to refer to some mysterious relation between Polycleitos and the statue. It was just an elliptical way of saying that some event involving Polycleitos - his actually doing some sculpting - was the proper efficient cause. "For, to be causing some change, the substance must be actualising some disposition which it possesses; the builder is always potentially building, but only as actualising his building capacity can he function as an explanatory cause of the movement of bricks and timber" (Annas 1982, 321). Even so, both interpretations share the idea that there can only be events and causal relations between events because there are substances behaving in a certain way. Thus, even those who defend the view that only events and states can be proper causal items, must admit that Aristotle's theory of causation is completely embedded in his substance ontology. It is a matter of dispute whether Aristotle also defended the modern idea that efficient causes necessitate their effects; there is evidence that he associated explanation by efficient cause not simply with what happens always and necessarily, but with what happens for the most part. Indeed, given a certain man, he must have a father, but given a man, there is nothing that determines him to be a father. In other words, Aristotle defended the view that, given a certain effect, there must be some factors that brought about that effect. But he nowhere inferred from this that given certain conditions, some effect necessarily follows. 6 However, it would appear that there is another kind of necessity involved in the efficient cause. Efficient causation involves the transmission of a form. And precisely this form functions as a boundary condition determining that the behavior of a particular substance cannot exceed certain boundaries. For instance, though the form of man does not determine what a particular man will do, it does determine that he cannot, for example, fly as a bird.
1.2
The Stoics: causation, exceptionless regularity, and necessity
By linking causality with exceptionless regularity and necessity, the Stoics would have a major impact upon the development of the modern accounts of cause. Five of their most basic theses related to causation are: ( I ) the world
6
CHAPTER ONE
is ruled by fate, (2) nothing happens without a cause, (3) causation involves exceptionless regularity, (4) causation involves necessity, and (5) there is a fundamental distinction between external and internal causes. I will consider these theses successively. (1) The Stoic cosmos is an organism the parts of which interact for the benefit of the whole. It is imbued with divine reason (logos), and its entire development is providentially ordained by fate. Moreover, it is invariably repeated from one world phase to the next in a never-ending cycle. The Stoics were the first philosophers to systematically maintain the principle of causality (every event is necessitated by certain causal conditions). Although the principle was the ontological correlate of a logical doctrine, the focus of the Stoic philosophers was the problem of the relationship between moral responsibility and a cosmic script. There was a lively, and sophisticated, debate in which most philosophers tried to reconcile the idea that everything happened by fate (or necessity) with the idea that our actions are somehow in our own power.7 (2) The Stoics rejected the idea that there could be any uncaused events, because that would undermine their basic belief in the coherence of the universe. Chryssipus (c. 280 - c.206) - one of the leading representatives of early Stoicism - formulated this view as follows: "If some event were produced without antecedent cause, it would not be true that all things take place by fate" (Cicero, De Fato, 43). (3) One of the main innovations of the Stoics was that the idea of cause is linked to an exceptionless regularity. However, the Stoic regularities were not laws in the modern sense, for they lacked any reference to a specific content. Examples of such very general laws are 'like is attracted to like,' or 'for every action there is an equal and opposite reaction.' Moreover, this regularity always pertained to particular events, not to types of events (Sorabji 1980,64-5). (4) The Stoic view was that each particular event necessitates its effect. According to Alexander,8 for example, it is necessary that the same effect will recur in the same circumstances, and it is not possible that it be otherwise. The Stoics strictly held to the view that each event has a cause, and that given the same cause and circumstances surrounding the cause (periestekota), the same effect could not fail to occur (Sorabji 1980, 65). The Stoic principle of universal causation - which entails that 'chance' and 'possibility' only refer to our ignorance of the causal connections between events (Long 1996, 164) - is very well expressed in the following passage by an unknown Stoic author: Prior events are causes of those following them, and in this manner all things are bound together with one another, and thus nothing happens in the world
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
7
such that something else is not entirely a consequence of it and attached to it as cause. [...] From everything that happens something else follows depending on it by necessity as cause. (Quoted in Long 1996, 164)
Though this passage could very well have been written by a contemporary philosopher, there is an important difference with the modern conception of cause: contrary to the modern conception, the necessity involved in the causal relation does not pertain to types of events, but only to the relation between particular causes and particular effects. (5) Probably in order to answer the criticism that Stoic determinism entailed the denial of freedom of human action, Chrysippus made a distinction between internal and external causes. Chrysippus considered the example of a spinning-top spinning on a flat surface (Cicero, De Fato, 3944). To explain its spinning we need reference to both an external agency, without which it would not have started to spin, and to the intrinsic nature of the spinning-top, its shape, without which it could not spin at all. The top's spinning is thus the result of two different kinds of causes: an external cause, which was called 'auxiliary and proximate,' and an internal cause, which was called a 'principal and perfect cause.' The internal cause was thought to be more important than the external cause: though there is no action without an external cause, the way a natural substance reacts to an external cause is completely determined by its internal structure or nature. Obviously, the concept of internal cause is derived from Aristotle's concept of formal cause. Chryssipus applied this theory to the explanation of human actions. His problem was to preserve some human autonomy within a universe governed by fate. His solution was that every human action is a response to an external cause. These external causes are an expression of fate, but are in themselves not sufficient to necessitate our actions. The external causes produce the sense-impressions upon which the mind can react by giving or withholding assent. The response to the sense-impressions is determined by the internal cause, which is the nature of the person in question. However, Chryssipus's view seems to involve a paradox: despite his insisting that human actions "are not involved in a nexus of eternal causes but are free from the necessity of fate" (Cicero, De Fato, 38), the reaction of the individual person seems to be completely determined by its nature, just as in the example of the spinning-top. This made A.A. Long conclude that "in spite of the distinction between internal and external causes, the character of the individual falls under the general causal law," for "the capacities with which a man is borne are 'the gift of destiny'" (Long 1996, 168).
8
2.
CHAPTER ONE
CAUSATION IN THE MIDDLE AGES
Although I will restrict my discussion of the medieval concept of cause to the view of Thomas Aquinas (1224/5-74), who may be regarded as one of the most influential representatives of later medieval philosophy, some preliminary remarks are in order. It is well known that in the thirteenth century, most Christian philosophers tried to reconcile the Christian worldview with Aristotle's. The main problem was that there was nothing in Aristotle's scheme of the cosmos that corresponded to the Christian provident God. In the attempt to reconcile Aristotle's philosophy with the Christian idea that God created the world out of nothing, Aristotle's "unmoved mover" was transformed into a "creating cause of existence" (Gilson 1962). More generally, the Liber de Causis - a Neo-Platonic Arabic work of the ninth century, translated into Latin in the twelfth century - had a decisive influence on the concept of cause. Due to that influence, most thirteenth century philosophers, 9 distinguished two quite different sorts of efficient cause: the causa prima and the causa secunda, the former being the originative source of being (God), while the latter, which is to be found in created things only, referred to the origin of the beginning of motion or change. Thus, while the (secondary) efficient cause is the source of motion, there is also an active cause that is the source of being. The First Cause works in all secondary causes, which may be considered as instrumental causes subservient to the first.
2.1
Thomas Aquinas
Like Aristotle, Aquinas not only considered 'causes' primarily as explanatory metaphysical principles, but also thought that four explanatory principles were needed to account for physical bodies. However, Aquinas gave a new twist to the role of these four principles, and, moreover, he gave a new meaning to the notion of efficient causality. A discussion of Aquinas's concept of causation requires some understanding of both his distinction between essence and existence, and his proofs for the existence of God. Whereas Aristotle already had made a distinction between the concept of being and the concept of essence, Aquinas would significantly refine the distinction. The main point behind the distinction is the experience that the concept of something does not necessarily imply that the thing exists. We all know what a unicorn is; in medieval terms, we know its essence. But clearly we also know that no unicorn exists. Therefore it is important to distinguish between what a thing is and the fact that it is. Generally speaking, it may be said that, given any
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
9
existing substance, that substance involves both existence (esse) and essence (essentia). In Aristotle's Metaphysics the emphasis had been on an altogether different distinction between matter and form. Thus Aquinas tried to show how the distinction between essence and existence is related to the distinction between form and matter. Certainly the two distinctions are not equivalent. Indeed, it is tempting to consider form and essence as equivalent because both terms refer to what a thing is. And if form and essence were to be equivalent, it might be tempting to think that matter and existence are equivalent. Aquinas falsifies the notion that essence and form are equivalent by pointing out that the essence of material substances involves both the matter and the form of such substances. And he falsified the notion that existence and matter would be equivalent by pointing out that non-material substances do exist. In the light of this, Aquinas tried to establish a hierarchical structure of reality (see Figure 1).
prime matter essence
J
- material substances J1 I substantial form 1 existence (esse) f essence - spiritual creatures J 1 existence
-God
- essence
form
existence
Figure 1 : essence and existence
The most basic distinction is the distinction between existence and essence. Indeed, for all that exists, the distinction obtains, although it does not obtain for all things in the same manner. Indeed, given the metaphysical unity of God, God is the only substance in which essence and existence coincide. For all created things, the distinction is metaphysically decisive. The distinction between this metaphysical status of God and his creation will be important for Aquinas's conception of causality.
10
CHAPTER ONE
But within creation too, there is a fundamental distinction between existing non-material things and existing material things; for while the former have a form only, the latter are 'composed' of form and matter. In the Summa Theologiae (Ia.2,3), Aquinas formulated five arguments for the plausibility of existence of God. All of these begin with our experience of sense objects. By the light of natural reason, the intellect knows that the existence of sense objects requires a cause, for nothing comes from nothing, ex nihilo nihil fit. The acceptance of the existence of God is based upon the idea that the existence of sense objects requires a finite series of causes, and, ultimately, a First Cause, or God. Although the five ways may be regarded as variations on this basic theme, the second and fifth way are particularly interesting to our topic. The second way, or proof from efficient cause, goes as follows: (1) in sensible things there is an order of efficient causes. (2) It is impossible that this order of efficient causes should be infinite. (3) Thus, there must be a first efficient cause. (4) This first efficient cause everyone calls God. However, this first efficient cause is completely different from the secondary causes. It is a transcendental cause, a cause that does not itself belong to the order of changing things. All the efficient causes in the world, even in their causal activity here and now, depend upon the causal activity of the one transcendent efficient cause. Thus, the word 'first' does not mean first in the temporal order, but first in the ontological order: without the first cause there would be nothing at all. The fifth way, or proof from the order of the world starts from the observation of goal-directed behavior in natural bodies. "This is evident from their acting always, or nearly always, in the same way, so as to obtain the best result." From this Aquinas concluded that they reach their ends "designedly," and not merely by chance. This, in turn, entails that these natural things are directed to their ends by some intelligent being. "Therefore, some intelligent being exists by whom all natural things are directed to their end; and this being we call God." Thus, the second proof makes clear that there are two quite different levels of causation:10 the First Transcendental Cause, and the secondary causes, which operate in the world. The First Cause works in all secondary causes, which are to be considered as instrumental causes subservient to the First Cause. Moreover, the 'proof from the order of the world' shows that God is not only efficient cause, but final cause as well. Furthermore, it suggests that in every order of secondary causes, the final cause must be ontologically prior to the particular secondary causes, for the latter cannot act in an orderly way unless through the power of the final cause. Before discussing Aquinas's conception of final causation, I will first consider his two conceptions of efficient cause.
KEY MOMENTS IN THE HISTORY OF 'CAUSATION' 2.1.1
11
The primary efficient cause
As we have seen above, all created things are marked by the composition of essence and esse. According to Aquinas, all created things derive their esse from that which is Being itself, and their relation to this creating cause is the relation of participation in being: "God is being by His own essence, because He is the very act of being. Every other being, however, is a being by participation" (SCG 11:15.5). Because "the order of causes necessarily corresponds to the order of effects," and because the effect of the primary efficient cause is universal - being - the primary efficient cause must be a universal cause. "Thus, transcending the particular causes of the generation of this or that thing is the universal cause of generation [...]. Now, being is common to everything that is. Above all causes, then, there must be a cause whose proper action is to give being" (SCG II: 15.4). However, this ontological dependence does not involve a temporal beginning. God, as the primary efficient cause, does not precede its effect in duration, but is simultaneous with it. Secondary efficient causes, on the other hand, precede their effects in duration (SCG II: 38.2). Aquinas's conception of the primary efficient cause involves a radical switch in respect of the Aristotelian notion of efficient causality. Whereas in Aristotle, efficient causation was the origin of a change or a motion by means of the transmission of form, in Aquinas, primary efficient causality concerns the production of both matter and form through a creative act. 2.1.2
The secondary efficient cause
The secondary efficient cause is that which induces some form in natural things. Thus, fire causes another thing to burn by communicating its form (fire) to that other thing. Similarly, a real house comes to be as the result of the form of that house, which exists in the builder's mind, being transmitted to the building material. Particular agents necessarily require pre-existing matter, but the effect of their action comes about by bestowing a form upon it: "matter is required by an agent in order that it may receive the action of the agent. For the agent's action, received in the patient, is an actuality of the patient's, and a form, or some inception of a form, in it" (SCG II: 16.7). 2.1.3
Causation and necessity
Aquinas distinguished four kinds of necessitation corresponding to the four Aristotelian causes. Thus the hardness of a saw is a material necessity, while the capacity of learning is a formal necessity in man (SCG II 30.11). But
12
CHAPTER ONE
from our point of view, the most interesting conception of necessity concerns the relationship between final and efficient causality. As it was shown earlier, Aquinas's fifth way concludes from the observation of finality within natural bodies that there must be some intelligent being, God, by which all natural things are ordered to their end. It is important to see that (like Aristotle) Aquinas distinguished between internal and external final cause. Whereas all natural things act according to their own internal final causes, created by God, the ultimate external goal is God himself. For, while the primary goal of created things is self-realization, this striving toward self-realization coincides with the striving toward the ultimate goal, which is God. Thus, in the formation of the world, as well as in all created causality, final causality comes first and works in and through the efficient causes: ... in all ordered efficient causes, where action is done for the sake of an end, the ends of the secondary causes must be pursued for the sake of the end of the first cause; the ends of the art of war, of horsemanship, and of bridgemaking, are ordained to the end of the political art. Now, the issuance of beings from the first being is brought about by an action ordained to an end, since [...] it is accomplished by the causality of an intellect; and every intellect acts for an end. So, if there are secondary causes at work in the production of things, the ends and actions of those causes are necessarily directed to the end of the first cause, which is the last end in things caused. Now, this is the distinction and order of the parts of the universe, which, as it were, constitute its ultimate form. Therefore, it is not on account of the actions of secondary agents that the distinction of things and their order exist; on the contrary, the actions of secondary causes are for the sake of the order and distinction to be established in things. (SCG II: 42.5)
In the order of nature, in each act of causation there is both a final component and an efficient component. The efficient causes are subordinate to the final causes inasmuch as they are means to ends. The final cause is responsible for a twofold necessity in things. The first kind of necessity (I) appears if we consider the cause-effect relationship from the perspective of the efficient cause: In one way, necessity results from that cause inasmuch as it is first in the intention of the agent. And in this regard, necessity derives from the end in the same way as from the agent; for it is precisely so far an agent intends an end that an agent acts. This is true of natural as well as voluntary actions. For in natural things the intention of the end belongs to the agent in keeping with the latter's form, whereby the end is becoming to it; hence, the natural thing necessarily tends to its end in accordance with the power of its form; a heavy
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
13
body tends toward the center according to the measure of its gravity. And in voluntary things the will inclines to act for the sake of an end only so far as it intends that end, although the will, as much as it desires the end, is not always inclined to do this or that as means to it, when the end can be obtained not only by this or that means, but in several ways. (SCG II: 30.15)
In this passage, Aquinas made a distinction between the necessity of socalled natural things (la) and the necessity of voluntary things (lb). In natural things, the necessity is derived from the form of the things. Thus, given the efficient cause, "the natural thing necessarily tends to its end in accordance with the power of its form." This necessity is absolute inasmuch as the way toward the end state is completely determined by the form and the other causal circumstances ("every agent which acts by natural necessity is determined to one effect"). It is interesting to see that, with Aristotle, Aquinas mentioned 'gravity' as an example of final or formal causality, and not as an instance of efficient causality, as has been commonly supposed since the rejection of final causation in the modern period. Contrary to the absolute necessity of natural things (la), there is only relative necessity in voluntary things (lb). For, though voluntary things receive their necessity from the ends or intentions (which were freely chosen by an act of reason (ST IaIIae.l3.1.c)), the end or final cause does not necessarily determine the way toward the end state; often there are different routes leading toward the same end state. Thus, the relative necessity of socalled voluntary things only determines that, given certain circumstances, an action will be initiated to reach the end state, not what the particular action will be. The second kind of necessity (II) appears if the cause-effect relationship is viewed from the perspective of the effect or end state. It concerns the means as necessary condition for the end state. This is called conditional necessity: ... in another way, necessity follows from the end as posterior in actual being; and such necessity is not absolute, but conditional. Thus, we say that a saw will have to be made of iron if it is to do the work of a saw. (SCG II: 30.15)
Similarly, ... the means to an end derive necessity from the end only so far as without them the end either cannot be - life cannot be preserved without food - or cannot well be - as ajourney without ahorse. (SCG II: 31.3)
Thus, the final cause functions as some sort of principle of selection, admitting only those means or ways of actions that contribute to the
14
CHAPTER ONE
realization of the end state. The appropriateness of the means is dependent upon the final cause. However, though Aquinas held that all inanimate things behave according to natural necessity, he made a distinction between two kinds of efficient cause, which, in modern terminology, might be called 'loose causes' and 'tight causes.' Whereas tight causes necessitate their effects independently of any other causal circumstances, loose causes require that other conditions be fulfilled (cfCollingwood [1938] 1991, 153). Aquinas gave the following explanation of 'loose causes': the necessity derived from the efficient cause (la) "has reference both to the action itself and the resulting effect" (SCG II: 30.12). The necessity of the action itself concerns the form of the agent; "for as the agent actually is, so does it act." The resulting effect is also necessitated by the form, though the necessity involved may be only relative, depending on certain other causal conditions: "if fire is hot, it necessarily has the power of heating, yet it need not heat, for something extrinsic may prevent it" (SCG II: 30.12). There are, however, also 'tight causes:' ... the necessity which arises from an efficient cause in some cases depends on the disposition of the agent alone; but in others, on the disposition of both agent and patient. Consequently, if this disposition, according to which the effect follows of necessity, be absolutely necessary both in the agent and the patient, then there will be absolute necessity in the efficient cause, as with things that act necessarily and always. On the other hand, if this disposition be not absolutely necessary, but removable, then from the efficient cause no necessity will result, except on the supposition that both agent and patient possess the disposition necessary for acting. Thus, we find no absolute necessity in those things that are sometimes impeded in their activity through lack of power or the violent action of the contrary; such things then, do not act always and necessarily, but in the majority of cases. (SCG II: 30.14)
Thus, whereas in some cases necessary connection is associated with the efficient cause as such ("the sun's motion, for example, necessarily gives rise to changes in terrestrial bodies;" SCG II: 29.18), and is therefore absolute necessity, in other cases it is relative to both the agent and the patient (for example, in case of the above mentioned fire). However, given both the agent and the patient, the necessity is just as absolute as in those cases in which the efficient cause itself is a sufficient cause. Aquinas therefore concluded that all inanimate things are characterized by natural necessity: "For, as nature is, so is its action; hence, given the existence of the cause, the effect must necessarily follow" (SCG II: 35.4). Whereas man is endowed with free will, "inanimate things, plants, and brute animals" behave
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
15
according to natural necessity (SCG II: 47.3). This natural necessity is responsible for the uniform behavior of natural things: ... the power of every agent which acts by natural necessity is determined to one effect; that is why all natural things happen in the same way, unless there be an obstacle; while voluntary things do not. (SCG II: 23.2)
By saying that "all natural things happen in the same way," Aquinas meant that things belonging to the same type act similarly in similar causal circumstances. By thus relating efficient causality to natural necessity, and natural necessity to law-like behavior, Aquinas would have a major impact on the development of the modern conception of causality. " Basic to Aquinas's view of causation is his substance ontology, according to which there are four explanatory principles to account for physical bodies. In his view, both God and all the secondary efficient causes are conceived as substances (see, for instance, SCG II: 21.5). As in Aristotle, the key to Aquinas's concept of cause is the concept of form as it is related to final causation. The major difference between Aristotle and Aquinas is to be found in his distinction between primary and secondary cause. Moreover, by clearly conceiving efficient causes as means to ends, he has had a major influence on the modern view of nature of people such as Francis Bacon, according to whom knowledge of the efficient cause gives power over nature.
3.
CAUSATION IN MODERN PHILOSOPHY
The rise of modern science in the seventeenth century involved a radical change in the development of the concept of cause. Explanations by formal causation and final causation being rejected, efficient causation alone was considered to provide rational explanations of the phenomena. Moreover, the concept of efficient causation itself had radically changed. Whereas in the Aristotelian and scholastic tradition (a) efficient causation was not restricted to locomotion, (b) did not involve determinism and, (c) efficient causes were conceived as the active initiators of a change, in the seventeenth century the idea took hold that (a) all causation refers exclusively to locomotion, (b) that causation entails determinism, and (c) that efficient causes are merely inactive nodes in the chain of events, rather than active originators of a change. These changes have had a lasting influence on the evolution of our conception of cause, and indeed our entire Western outlook. The history of the development of this outlook is extraordinarily complex, and was influenced by a web of both theological and scientific beliefs. Thus, contrary what is often tacitly assumed, the idea that causation
CHAPTER ONE
16
involves determinism does not have a scientific, but a theological origin. In seventeenth century philosophy the idea of determinism was ubiquitous. In spite of differences in detail and differences in the manner of statement, the arguments for determinism in the writings of Descartes, Hobbes, Spinoza and Leibniz, are very similar. In no case did the conclusion that all things are determined receive its justification from a concern with empirical fact. The validity of the idea that all things are causally determined derived exclusively from the fact that it was considered to be the only idea that was compatible with the doctrine of God's omnipotence and omniscience. If God knows everything and can do everything, whatever is must be. For the same reason, saying that a finite agent is a genuine cause in the sense of being an active initiator of a change, was considered misleading. For only God was believed to be the cause of anything. Hence, the deterministic world scheme had its source in the idea of the dependence of the world upon God; far from being a pronouncement of science, determinism is unambiguously a theological relic. This may perhaps most easily be shown by briefly examining Hobbes's doctrine of determinism. According to Hobbes, there is only one reason why all things happen necessarily: there exists a God with the attribute of foreknowledge: ... from another of God's attributes, which is His foreknowledge, I shall evidently derive that all action whatsoever, whether they proceed from the will or from fortune, were necessary from eternity. For whatsoever God foreknoweth shall come to pass, cannot but come to pass, that is, it is impossible it should not come to pass or otherwise come to pass than it was foreknown. But whatever was impossible should be otherwise, was necessary; for the definition of necessary is, that which cannot possibly be otherwise. [...] And whereas they that distinguish between God's prescience and His decree, say foreknowledge maketh not the necessity without the decree, it is little to the purpose. It sufficeth me, that whatsoever necessary:
but all things were foreknown
were necessary.
was foreknown
by God, was
by God, and therefore all things
And as for the distinction of foreknowledge from decree in
God Almighty, I comprehend it not. They are acts co-eternal, and therefore one. (Hobbes [1656] 1841, 18-19)
Hobbes did not think it necessary to make a distinction between God's foreknowledge and God's decrees. What God decreed happens necessarily; God foreknows what he decreed; therefore, what God foreknows happens necessarily. This straightforward determinism had important consequences for the development of the diverse conceptions of causation in the seventeenth century. In this section I will first discuss the rationalist conceptions of causation of Descartes, Hobbes, Spinoza, and Leibniz - some of the most
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
17
important seventeenth century metaphysicians before discussing the views of the empiricist approaches of Locke, Newton, Hume, Kant and Mill.
3.1
The Metaphysical Systems from Descartes till Leibniz
3.1.1
Descartes: dismissal of substantial forms
Descartes, the 'founding father of modern philosophy,' who said to break with the tradition by starting completely anew, did not for a moment doubt the principle of causation. However, his interpretation of efficient causes as mechanical causes marked an important new development. His mechanistic worldview hinged upon the idea that the principles of nature were identical to the principles of mechanics. Extension, for Descartes (1596-1650), is the only attribute necessary to constitute matter. Extension as such is geometrical. However, matter is divisible and each part has received a certain quantity of motion. From this, all forms and all the properties that we perceive in the world come to be from the local motion of matter (Princ. II: 23-24). According to Wallace, "what was most significant in the eyes of his followers was that [Descartes] had taken the thoroughly complex system of the Scholastics, with its substantial forms, qualities of all types, elements and other principles, and in their place substituted the clear and distinct idea that bodies are composed of particles in motion, so that from this single principle he could give better explanations than those of any of his predecessors" (Wallace 1974, II: 13). This should be understood against the background of Descartes' natural philosophy, the primary aim of which it was to replace the scholastic doctrine of active qualities and substantial forms as causal factors in natural processes by purely mechanical principles of explanation. According to Descartes, the idea of substantial forms and active qualities as causal factors has no basis whatever in our experience of things: Lei another, if he likes, imagine in this piece of wood the Form of fire, the Quality of heat, and the Action which bums it as things altogether diverse; for my part I, who fear I shall go astray if I suppose there to be more in it than I see musi needs be there, am content to conceive in it the movement of its parts. (Descartes, quoted in Miles 1988. 100)
By thus raising a few simple questions regarding the example of the burning of a piece of wood, Descartes put his fìnger on what is perhaps the most important problem with any theory of causation based upon a substance ontology: how can a substantial form be transmitted from a cause to its effect?
18
CHAPTER ONE
The rejection by Descartes (and Galilei and Bacon) of the Aristotelian and Scholastic theory of fourfold causality had a profound influence on subsequent thinkers. By raising matter to the status of substance thereby depriving it of its status of material cause, and by rejecting both formal and final causality as valid principles of explanation, Descartes allows for efficient causality only (Descartes [1644] 1983,1: 28). On the other hand, like Aquinas, he made a distinction between particular causes and one general cause, which is God ensuring the constancy of quantity of motion in the universe: That God is the primary cause of motion; and that He always preserves same quantity
of it in the universe.
the
After having examined the nature of
movement, we must consider its cause, which is twofold: (we shall begin with) the universal and primary one, which is the general cause of all the movements in the world; and then {we shall consider} the particular ones, by which individual parts of matter acquire movements which they did not previously have. As far as the general (and first} cause is concerned, it seems obvious to me that this is none other than God himself, who {being all powerful} in the beginning created matter with both movement and rest; and now maintains in the sum total of matter, by His normal participation, the same quantity of motion and rest as He placed in it at that time. (Descartes [1644] 1983,11: 36)
Interestingly, the particular causes are not the motions of the individual parts of matter, but the general principles or laws of nature: Furthermore, from this same immutability of God, we can obtain knowledge of the rules or laws of nature, which are the secondary and particular causes of the diverse movements which we notice in individual bodies. (Descartes [I644I 1983,11:37)
Descartes listed three of those "natural laws" of motion, as well as seven "rules" which govern the collisions of bodies. In the beginning, God created matter and motion, and he conserves exactly the same quantity of motion for all time. There can be no locomotion - change of place - unless there be some kind of force or power. Since such force was not inherent in matter - 'extension' in no respect involves 'force' God is the efficient cause of any change of motion in an otherwise inert matter. And He does so according to the laws of nature, which became secondary causes. Thus, Descartes attributed some efficient causality to the laws of motion, which determine all particular effects. By doing so they provide causal, mechanical explanations. On the other hand, Descartes maintains that the only 'active initiator of change' is God, the cause of all causes.
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
19
Descartes' theory had the lasting effect of shifting the attention of philosophers and scientists from the primary cause to secondary causes (Sachsse 1979, 32), which he conceived as laws of motion. Descartes' theory entails a radical change in the concept of cause. By identifying efficient causes with deterministic laws, causes are no longer conceived as particulars, but as types. Moreover, instead of being 'active initiators of a change,' they are seen as inactive instruments of God.12 This change had a tremendous impact upon the scientific view of the world. 3.1.2
Hobbes: causation and motion
Like Descartes, the English philosopher Thomas Hobbes (1588-1679) presupposed causal determinism, rejected formal and final causation, and thought that causation was only relevant to motion. He explained all phenomena, even psychological and sociological ones, in terms of causal relations between moving bodies. According to Hobbes, a body is "that, which having no dependence upon our thought, is coincident or coextended with some part of space" ([1655] 1839, 8.2). Thus, even 'endeavors'13 are movements of minute unobservable bodies. However, not the bodies but the accidents of bodies are the causal elements. Hobbes defined a(n) (efficient) cause as "the aggregate of accidents in the agent or agents, requisite for the production of the effect" (DCo 9.4). And he defined an effect as "that accident, which is generated in the patient" (DCo 9.1). Indeed, by 'accident' Hobbes meant a property or characteristic that is not part of a thing but "the manner by which any body is conceived" ([1655] 1839, 8.3). Thus, "the fire does not warm because it is a body [Hobbes, apparently, considered the fire to be a body], but because it is hot; nor does one body put forward another body because it is a body, but because it is moved into the place of that other body" ([1655] 1839, 9.3). However, ultimately, accidents are motions too; they are motions of parts of the body that is changed ([1655] 1839, 9.9). Thus, ultimately, causation consists in motion: a body is a cause "not because it is a body, but because [the body is] [...] so moved" ([1655] 1839, 9.3). Moreover, "rest does nothing at all, nor is of any efficacy; nothing but motion gives motion to such things as be at rest, and takes it from things moved" ([1655] 1839, 15.3). Thus, the causal relata are not particular bodies or substances, but their motions; causation is a relation between the motions of different bodies. Nothing would happen if nothing moved, and the only things that move are bodies. Moreover, all causation occurs by contact, that is to say, it consists in motion of contiguous bodies: "There can be no cause of motion except in a body contiguous and moved" ([1655] 1839, 9.7). There is no action at a
20
CHAPTER ONE
distance. Hobbes made a distinction between mediate and immediate causation. Only in mediate causation is there an intervening body between the body that initiates a series of movements (agent) and the body moved (patient). Properly speaking, all bodies are both agent and patient, for every cause is also an effect of some other cause, and every effect is also the cause of some other movement. 14 Consistent in his attempt to resolve all phenomena to motion, Hobbes rejected the concept of formal or final causes, which in his view are nothing but 'disguised' efficient causes ([1655] 1839, 10.4). Yet, Hobbes maintained a distinction between efficient cause and material cause. Whereas the material cause is just the receptor of the agent's activity, "the aggregate of accidents in the patient," the efficient cause is the aggregate of properties in the agent required for the production of the effect. The material and efficient causes are both part of the entire cause ([1655] 1839, 9.4). Necessity or necessary connection is not associated with the efficient cause as such, but with the entire cause, which entails both the agent and the patient: ... a cause simply, or an entire cause, is the aggregate of all the accidents both of the agents how many soever they be, and of the patients, put together; which when they are supposed to be present, it cannot be understood but that effect is produced at the same instant; and if any of them be wanting, it cannot be understood but that the effect is not produced. (Hobbes [1655] 1839, 9.3)
Thus, (entire) causes are complex conditions (of both agents and patients) that are necessary and sufficient for the occurrence of the effect. A condition that is necessary but not sufficient for an effect, that is to say, "that accident either of the agent or the patient, without which the effect cannot be produced," is called "a cause necessary by supposition" (causa sine qua non) ([1655] 1839, 9.3).15 Thus, when we say that the short circuit was the cause of the fire, we mean 'cause' in the restricted sense of 'cause necessary by supposition.' The entire cause of the fire involves many more factors, which together were sufficient for the fire to occur. In Hobbes's universe, everything happens by necessity: "all the effects that have been, or shall be produced, have their necessity in things antecedent" ([1655] 1839, 9.5). Moreover, given the cause, "it cannot be conceived but that the effect will follow" ([1655] 1839, 9.7; italics mine). This description of cause (involving necessity) corresponds to what Taube thought to be the definition of necessity that was current in the seventeenth century, namely: "that the opposite of which is inconceivable" (Taube 1936, 102). A connection is necessary inasmuch as it is inconceivable, or contradictory that the connection should not obtain. However, this supposed necessity is not based on any matter of fact relation. Like most of his seventeenth century colleagues, Hobbes justified his belief in the necessity
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
21
of causal relationships by postulating God's hand in the causal relations of finite things. To hold that an entity acts in a manner not determined by God was inconceivable. Given this view of necessity, Hobbes meant by chance or contingency "that whereof we do not conceive the necessary cause." Propositions concerning future events are said to be contingent because we do not yet know whether they be true or false. But their verity does not depend upon our knowledge, but on the foregoing necessary causes ([1655] 1839, 10.5). Moreover, something can be called possible, as opposed to necessary, only when we do not know the cause that will produce it. According to Hobbes, 'conceivable things' need not necessarily be 'possible things.' Some being, event or state of affairs is possible only if the necessary conditions for its existence are already satisfied. Thus, every possibility will be actualized at some time: "for if it shall never be produced, then those things will never concur which are requisite for the production of it" ([1655] 1839, 10.4). Closely related to the distinction between necessity and possibility, is the distinction between cause and power. According to Hobbes, the concepts of cause and power are complementary notions. Hobbes gave the following definition of power: "whensoever any agent has all those accidents which are necessarily requisite for the production of some effect in the patient, then we say that agent has power to produce that effect, if it be applied to a patient" ([1655] 1839, 10.1). And because the same accidents that constitute the power of the agent constitute also the efficient cause, it can be said that the power of the agent and the efficient cause are one and the same. However, there is an important difference of perspective: "cause is so called in respect of the effect already produced, and power in respect of the same effect to be produced hereafter; so that cause respects the past, power the future time" ([1655] 1839, 10.1). Hobbes explained that the term 'cause' only makes sense after the effect has occurred. Nothing can be called a cause when there is no effect ([1655] 1839, 9.5). Thus, it is plainly absurd to say that some cause failed to do something or was prevented from acting. And the term 'power' makes sense only before an effect has occurred. Thus, the conception of causes as things having a power of producing certain changes in other things is misleading. 3.1.3
Spinoza: rejection of final causation
Spinoza (1632-77) was strongly influenced by Descartes. The only work of Spinoza that was published during his lifetime under his own name was his Descartes' Principles of Philosophy (published in 1663). It was an attempt to present and recast the first two parts of Descartes' Principles in geometrical order or according to the geometrical method.
22
CHAPTER ONE
The first two parts of Spinoza's posthumously published masterpiece, the Ethics, Demonstrated in Geometrical Order (1677), contain the most oritative exposition of his metaphysical system. It was presented as a deductive system in the manner of Euclid. Each of its five parts opens with a set of definitions and axioms, followed by a set of theorems proved upon the basis of them. Spinoza's view of causation is closely related to his concept of substance. The starting point of Spinoza's metaphysical system is Descartes' definition of substance as "that which requires nothing but itself in order to exist." Whereas Descartes had counted matter and mind as substances, Spinoza concluded from Descartes' definition that there was only one substance, which must be God. There cannot be any more substances, for if there were, God would not be infinite. Spinoza defined substance as "that which is in itself and is conceived through itself; in other words, that the conception of which does not need the conception of another thing from which it must be formed." (Part I, def. 3). This definition has two parts: (i) substance has no external cause but has the cause of itself within itself, and (ii) substance can be fully conceived without reference to any idea outside of it. The very idea of substance includes that it exists. An attribute was defined as "that which the intellect perceives of substance as constituting its essence" (def. 4). Though God has infinitely many attributes, we only know two of them: thought and extension. Whereas Descartes concluded that these two attributes indicate the existence of two substances, mind and body, Spinoza conceived them as different ways of expressing the activity of a single substance: God. Being infinite, God has infinite thought and infinite extension. By a mode Spinoza understood "the modifications of substance, or that which is another thing through which also it is conceived" (def. 5). Accordingly, individual things are particular configurations or modes of God's attributes, expressing these attributes in a determinate way ([1677] 1949, 63). Thus, individual minds and bodies are not substances, but just modifications or modes of the two divine attributes of thought and extension. Spinoza had a unique conception of God, in which he identified God with Nature itself: "By God I understand Being absolutely infinite, that is to say, substance consisting of infinite attributes, each one of which expresses eternal and infinite essence" (def. 6). The world is not distinct from God but is God expressed in various modes of thought and extension. God and the world are not related as an immaterial cause and its material effect. For God is not outside creation, but He is identical with it. The words 'God' and 'Nature' are interchangeable: Deus sive Natura. God is "the immanent, but not the transient, cause of all things. [...] He is the cause of the things which
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
23
are in Himself' (Prop. 18). Moreover, since all truths about God follow necessarily from his essence, and since everything happens 'in' God, everything happens by necessity. Spinoza, therefore, was a strict determinist (of the immanent variety). The problem of necessity and contingency plays on the level of modes. Obviously, in a strictly determined world, there is no room for contingency, every event unfolds in the only possible way in which it can: "In Nature there is nothing contingent, but all things are determined from the necessity of the divine nature to exist and act in a certain manner" (Prop. 29). This rule also applies to the freedom of the will: "The will cannot be called a free cause, but can only be called necessary" (Prop. 32). Again, not unlike Aquinas, Spinoza made a distinction between 'free causes' and 'necessary causes.' Free causes act from the necessity of their own nature (and are therefore the initiators of a change); necessary causes are necessitated by other causes (and are therefore just inactive nodes in a chain): That thing is called free which exists from the necessity of its own nature alone and is determined to action by itself alone. That thing, on the other hand, is called necessary or rather compelled which by another is determined to existence and action in a fixed and prescribed manner. (Spinoza [1677] 1949, Def. 7)
But unlike Aquinas, Spinoza states that God is the only free cause, by which is meant that, for though He simply had to create what He did, He was not forced to do this by some external cause. He alone exists and acts from the necessity of his own nature. Only God is a genuine cause: "God's intellect is the sole cause of things, both of their essence and of their existence" (Prop. 17). The other 'causes' are just the nodes of a chain, completely compelled by previous links. But, whatever the sort of cause, be it God or some secondary cause, the relationship between cause and effect always involves necessity: From a given determinate cause an effect necessarily follows; and, on the other hand, if no determinate cause be given it is impossible that an effect can follow. (Spinoza [1677] 1949, Axiom 3)
Given the reciprocity of the necessary relation between cause and effect, and given that "the order and connection of ideas is the same as the order and connection of things" (part II, Prop. 17), the necessity involved in the causal relationship must be understood as logical necessity. Causes logically
24
CHAPTER ONE
necessitate their effects, and, conversely, effects logically necessitate their causes. Moreover, Spinoza too sees the causal order in terms of the mechanical model. He too radically rejects final causation, which he considers to be an anthropomorphic fiction. The belief that things in the universe have a purpose is due to an invalid extrapolation of our tendency to act with an end in view to all things in the universe, so that they too seem to have a purpose. But it is utterly wrong to look at ourselves and at the universe in this way. "This opinion alone would have been sufficient to keep the human race in darkness to all eternity if mathematics, which does not deal with ends but with the essences and properties of forms, had not placed before us another rule of truth" ([1677] 1949, 74). The truth is that everything just happens from the necessity of God's eternal nature, which simply is. By rejecting final causation and by considering all events as modifications of the eternal substance, Spinoza reduced the causal order to the mechanical order, and the mechanical order to the timeless order of mathematics. By doing so, he came to understand causation as some sort of logical necessitation. 3.1.4
Leibniz: the principle of sufficient reason and the denial of intrasubstantial causality
Gottfried Wilhelm Leibniz (1646-1716) occupies a special place in the history of the concept of causation. Though he was aware of the difficulties inherent to the Aristotelian approach, he also criticized its outright rejection. His approach to the problem of causation must be seen within the context of his doctrine of the more fundamental principle of sufficient reason, which is at the heart of his great metaphysical system. In his paper "First Truths" (ca. 1680-84), Leibniz derived many of his characteristic metaphysical theses from a doctrine about the nature of truth. He held that "in every true affirmative proposition, whether necessary or contingent, universal or particular, the notion of the predicate is in some way included in that of the subject. [...] otherwise I do not know what truth is" ([1686] 1969, 337). From this doctrine, Leibniz inferred his fundamental principle of reasoning, the principle of sufficient reason, which entails that there is nothing without a reason, or that there is a reason for every truth. 'Sufficient reason' refers both to the logical ground and to the real cause of things: "there is nothing without a reason, or no effect without a cause" ([ca. 1680-84] 1969, 268). Leibniz's very peculiar view of causality has its origin in his rejection of the reduction of metaphysical change to locomotion (so characteristic of the approaches of Descartes, Hobbes, and Spinoza). This rejection in turn was
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
25
the consequence of a fundamental critique of the Cartesian concept of matter as extended substance: "I do not think that substance is constituted by extension alone, since the concept of extension is incomplete. Nor do I think that extension can be conceived in itself, but I consider it an analyzable and relative concept" ([1699] 1969, 516). Thus, instead of being an ultimate, unanalyzable quality, extension is an analyzable relation. This view of extension as a relation entails that the final constituents of bodies cannot themselves be extended. For if they were, they would in turn be relations. Thus, Leibniz concluded that the ultimate existents must be non-extended monads and that material bodies have monads as their constituents. The characteristic features of matter - extension, solidity, inertia, etcetera - are derived from the relations between the constituent monads. Thus matter is an entity that may be said to exist only by virtue of the relations between the primary existents (cf Ledere 1986, 87). Leibniz's analysis of matter had significant consequences for the concept of motion. Because motion is a modification of the extensive relations, it too must be secondary to the action of the monad, which is the action of perception: "This is the only thing - namely, perceptions and their changes that can be found in a simple substance. It is in this alone that the internal actions of simple substances can consist" (Leibniz [1714] 1969, 644). Given his view of matter and of motion as a change in extensional relations, Leibniz was forced to forge a new conception of causality that was as original as his view of matter and motion. His view of monads as selfcontained substances entailed that such monads could not be causally related. And this, combined with the view that an adequate conception of a substance involved a conception of all its states, made him conclude that causality is what links all the successive states that constitute the make-up of an individual monad. Indeed, according to Leibniz, each individual substance corresponds to a concept from which everything follows that will ever be true about it: "The complete or perfect concept of an individual substance involves all its predicates, past, present, and future" ([ca. 1680-84] 1969, 268). Therefore, our common-sense belief that substances act upon each other is based on a misunderstanding of the metaphysical origin of the phenomena: It can be said that, speaking with metaphysical rigor, no created exerts a metaphysical
substance
action or influence upon another. For to say nothing of
the fact that it cannot be explained how anything can pass over from one thing into the substance of another, it has already been shown that all the future states of each thing follow from its own concept. What we call causes are in metaphysical rigor only concomitant requisites. (Leibniz [ca. 1680-84] 1969, 269)
26
CHAPTER ONE
Leibniz explains the apparent interaction between substances in terms of his doctrine of pre-established harmony. God has programmed the world in such a way that each monad develops in synchrony with all other monads. Like a good clockmaker who constructs a number of clocks that keep perfect time, God pre-established the harmony of the changes in things at the beginning of the universe ([ca. 1680-84] 1969, 268-9). Thus, all individual created substances are different expressions of the same "universal cause." However, though God caused their existence, their successive states are (normally) produced by their own natures. Indeed, harking back to the Aristotelian notion of substantial form, Leibniz states that every state of every monad is completely determined by its nature or substantial form, which is an internal, active causal principle, so that every simple substance may be said to be "spontaneous," that is to say, "the one and only source of its modifications" ([1712] 1973, 175). In affirming that each individual substance is the cause of its own states, Leibniz rejected Malebranche's occasionalism, which reserves all causal efficacy to the action of God. The doctrine of the spontaneity of substance ensured for Leibniz that created individual substances were centers of activity, a feature he took to be a necessary condition of genuine individuality. Moreover, on the basis of the doctrine that monads are non-extended substances which therefore cannot have bodily properties, Leibniz concludes that the activities of monads must be soul-like or 'mental'. And because mental activities are either epistemic or volitional, and physical forces are will-like, he conceives the latter as the phenomenal expression of the metaphysical activity of appetite. This appetite, which constitutes the volitional aspect of the substantial form of a monad, is a teleological principle in virtue of which the monad tends to evolve from one perception to another in a pre-established way. This aspect of teleological causation, however, does not preclude efficient causality. Each perception follows necessarily from the preceding one. Thus, efficient causality and final causality are complementary. Each efficient cause happens in accordance to a general rule or final cause, which is preordained by God: ... in the whole of nature, efficient causes correspond to final causes, because everything proceeds from a cause which is not only powerful, but also wise; and with the rule of power through efficient causes, there is involved the rule of wisdom through final causes. (Leibniz [1712] 1973, 174)
Apparently, final causation and efficient causation are not different types of causation, respectively functioning in different situations. But in each act of causation there is an efficient and a final component.
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
27
Accordingly, Leibniz also saw a complementarity between explanations through efficient and final causality: ... there are, so to speak, two kingdoms even in corporal nature, which interpenetrate without confusing or interfering with each other - the realm of power, according to which everything can be explained mechanically by efficient causes when we have sufficiently penetrated into its interior, and the realm
of
wisdom,
according
to which
everything
can
be
explained
architectonically, so to speak, or by final causes when we understand its ways sufficiently. (Leibniz [ca. 1696] 1969, 478-9)
However, Leibniz's doctrines of final causality, and of the spontaneity of simple substances, do not imply that nature is creative. On the contrary, each monad behaves in accordance with its original purpose, that is to say, with its nature or substantial form, which it received from the beginning through God's creation. Moreover, Leibniz's idea of pre-established harmony goes hand in hand with his brand of determinism, 'for if the successive states of the monads were not completely determined in advance, their coordination would be purely accidental, which would be incompatible with the principle of sufficient reason. The principle of sufficient reason even entails that God had no real choice in his selecting this world from an infinity of possible words. For, given his moral perfection, he could only choose the best possible world. Leibniz's determinism - which is based on his principle of sufficient reason ("there is nothing without a cause, or no effect without a reason") entails that the necessity involved in the relation between cause and effect is as strong as logical necessity. A complete knowledge of the causes would yield the premises from which by reasoning alone the effects could be concluded.
3.2
Critical Philosophy from Locke till Mill
3.2.1
Locke: causation and power
Descartes, Hobbes, Spinoza and Leibniz tried to build metaphysical systems. More modest on his aspirations, John Locke (1632-1704) merely hoped to discover what kind of things God has fitted us to know, and how we should use our intellect and understanding. His purpose was "to enquire into the original, certainty, and extent of human knowledge; together, with the grounds and degrees of belief, opinion, and assent" (L i, 2). Living in the century that witnessed the birth of modern science - the century of Galileo, Newton, Boyle and others - Locke wished to lay bare and defend its
28
CHAPTER ONE
underlying empiricist ontology against the weight of the philosophical tradition, which was rationalist in temper. It is all the more remarkable that, when it came to his discussion of causation, his approach was basically Aristotelian. Basic to his approach of the concept of causation was the concept of power and the Aristotelian belief that causes are substantial powers put to work. Locke's discussion of causality occurs mainly in chapter 21 ("of Power") and chapter 26 ("Of Cause and Effect, and Other Relations") of book II of his An Essay Concerning Human Understanding (1690). He gave the following explanation of the origin of our ideas of cause and effect: In the notice, that our Senses take of the constant Vicissitude of Things, we cannot but observe, that several particular, both Qualities, and Substances begin to exist; and that they receive this their Existence, from the due Application and Operation of some other Being. From this observation, we get our Ideas of Cause and Effect. That which produces any simple or complex Idea, we denote by the general name Cause; and that which is produced, Effect. (Locke [1690] 1975, II, xxvi, 1)
It is because we notice that the fluidity of the wax (substance) is always produced by the application of a certain degree of heat, that we call the simple idea of heat the cause of the simple idea of the fluidity of the wax, which we call the effect. And when we notice that, by the application of fire, the substance wood (complex idea) is turned into ashes (another complex idea), we consider the fire to be the cause, and the ashes the effect. "So that whatever is considered by us, to conduce or operate, to the producing any particular simple Idea, or Collection of simple Ideas, whether Substance, or Mode, which did not before exist, hath thereby in our Minds the relation of a Cause, and so is denominated by us" (II, xxvi, 1). The idea of power was basic to Locke's approach to the concept of causation. He recognized two types of power: active power or the capacity to produce change, and passive power or the capacity to receive change. Thus, fire has an active power to melt gold, and gold has a passive power to be melted. And the sun has an active power to blanch wax, and wax has a passive power to be blanched by the sun (II, xxi, 4). The mind receives its idea of active power clearer from reflection on its own operations, than it does from any external perception. Man's notion of active power derives most clearly from his experience of the operation of his will on the movements of his body: ... if we will consider it attentively. Bodies, by our Senses, do not afford us so clear and distinct an Idea of active Power, as we have from reflection on the Operations of our minds. [...] A body at rest affords us no Idea of any active
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
29
Power to move; and when it is set in motion it self, that Motion is rather a Passion than an Action in it. [...] The Idea of the beginning of motion, we have only from reflection on what passes in our selves, where we find by Experience, that barely by willing it, barely by a thought of the Mind, we can move the parts of our Bodies, which were before at rest. (Locke [1690] 1975, II, xxii, 11)
Locke viewed causes as substantial powers put to work: Power being the source from whence all Action proceeds, the Substances wherein these Powers are, when they exert this Power into Act, are called Causes; and the Substances which thereupon are produced, or the simple Ideas which are introduced into any subject by the exerting of that Power, are called Effects. The efficacy whereby the new Substance or Idea is produced, is called, in the subject exerting that Power, Action; but in the subject, wherein any simple Idea is changed or produced, it is called Passion: Which efficacy however various, and the effects almost infinite; yet we can, I think, conceive it, in Intellectual agents, to be nothing else but Modes of Thinking, and Willing; in corporal Agents, nothing else but Modifications of Motion. (Locke [1690] 1975, II, xxii, 11)
Thus, a cause is a particular substance putting its power to work. Apparently, Locke conceived causes and effects as particulars, for in his discussion of causation he makes not a single reference to either uniformity or necessary connection. Thus, it may be said, as Gibson (1917), Taube (1936, 20) and Wallace (1974, 29) did, that Locke succeeded in keeping the concepts of power and necessity distinct by linking causation to power, but not to necessity. In Gibson's words: "it will be observed that, throughout Locke's treatment of the subject, there is no suggestion that the idea of causality essentially involves that of uniformity or necessary connection according to law, which is now (ever since Hume) often regarded as its primary implication" (Gibson 1917; quoted in Taube 1936, 20). 'Power' and 'necessary connection' are kept separate in Locke's thought, for although we do perceive powerful or changing objects and thus have the ideas of power and cause, we do not perceive any necessary connections between ideas: "it is impossible we should know, which [ideas] have a necessary union or inconsistency one with another" (IV, iii, l). 16 Thus, by linking causation to power, Locke may be said to have held a singularist approach to causation.17 This view conflicts with the commonly held Humean view that causation involves uniformity or necessary connection according to law. In the next section I will show that Isaac Newton defended a view that, though similar qua basic insights, was even more radical than that of his friend Locke. Where Locke had merely argued that the idea of causation
30
CHAPTER ONE
does not involve the idea of necessary connection, Newton took the far more radical step of separating causation from law-like behavior. 3.2.2
Newton: rejection of the principle of causality
In his masterpiece, Philosophiae Naturalis Principia Mathematica ("The Mathematical Principles of Natural Philosophy"), Newton (1642-1727) set forth the mathematical laws of physics and "the system of the world." The world system consists of material bodies (masses composed of "solid, massy, hard, impenetrable, moveable particles") at rest or in motion and interacting according to his three famous laws of motion, which are stated in implicitly causal terms: (1) Every body perseveres in its state of rest, or of uniform motion in a right line, unless it is compelled to change that state by forces impressed thereon. (2) The alteration of motion is ever proportional to the motive force impressed; and is made in the direction of the right line in which that force is impressed.' 8 (3) To every Action there is always opposed an equal Reaction; or the mutual actions of two bodies upon each other are always equal, and directed to contrary parts. (Newton [1687] 1968,1: 19-20)
In order to understand what Newton means by causation as it is implied in expressions such as "motive forces impressed upon" a body, which "compel" the body to move differently than had they been absent, and how he sees the relationship between the concept of cause, these compelling motive forces, and the laws of motion, one must turn to the Scholium on absolute space and time, in which Newton makes a distinction between "true motion" and "relative motion." True motion refers to some absolute standard (for example: a sailing ship in relation to the immovable earth); relative motion is motion in respect of some relative standard (for example: a man walking on a sailing ship). It is in his attempt to explain the difference between true motion and relative motion that Newton shows what he means by 'cause': The causes by which true and relative motions are distinguished, one from the other, are the forces impressed upon bodies to generate motion. True motion is neither generated
nor altered, but by some force impressed upon the body
moved', but relative motion may be generated or altered without any force impressed upon the body. For it is sufficient only to impress some force on other bodies with which the former is compared, that by their giving way, that relation may be changed, in which the relative rest or motion of this body did
KEY MOMENTS IN THE HISTORY OF 'CAUSATION' consist. Again,
true motion
suffers always
some change from
any
31 force
impressed upon the moving body... (Newton [1687] 1968, 1: 14; italics mine)
Thus, according to Newton, a cause is precisely the above mentioned (in the first two laws of motion) "motive force impressed upon" a body, which "compels" it to move differently. More precisely, causes are forces or constraints that compel moving bodies to behave differently than they would have done without them. Thus 'caused' means constrained or compelled. Newton used the expression "free" motion to refer to unconstrained motions. Thus, the motion of a body that continues in its state of rest, or maintains its uniform rectilinear motion, is uncaused or free. Collingwood rightly concludes that "in Newton there is no law of universal causation; he not only does not assert that every event must have a cause, he explicitly denies it." Any movement that happens according to the first law of motion is an uncaused event. Thus if a body moves freely from A to B to C, the event which is the movement from A to B, is in no way the cause of the event which is the movement from B to C; it is not caused at all. The first law of motion is in fact a law of free or causeless motion (Collingwood [1938] 1991, 159; italics mine). By thus implicitly rejecting the principle of universal causation, Newton moreover defends a fundamental distinction between causation and law-like behavior. For, there are two classes of events in Newton's universe: (a) those that happen according to a law, and (b) those that are the effects of causes. Causation and law-like behavior (or necessary connection according to law) are mutually exclusive notions. In the next section, I will show that Hume neglected (or misrepresented) both Locke's and Newton's basic insights. He simply assumed that the concept of cause involves the concept of necessity, which he identified with the concept of power. 3.2.3
David Hume: causation and regularity
All modern discussions of causation are somewhat infected by Hume's views on the subject, either in an attempt to further develop Hume's basic insights, or by an attack on them. Hume (1711 -76) started from the observation that we think that our concept of causation involves the concept of necessity: events or states of objects follow their causes with some kind of necessity. More specifically, he held that, when we speak of the causal relation, we believe in: (1) a contiguity (in space and time) between cause and effect, (2) temporal priority of a cause to its effect, and (3) a necessary connection between cause and effect. He considered the third factor to be by far both the most
32
CHAPTER ONE
important, because it is the criterion by which we seem to distinguish causal from non-causal relations (Hume [1739] 1978, 77), and the most problematic. The problem is that given the concept of causal necessity, there seems to be no way to rationally justifying it. To Hume such justification could be given only if causal necessity could be shown to be as stringent as logical necessity. But this is impossible. Hence, the necessity that we read into causal relationships is illusory; the illusion is born from our expectations, which are due to habit. Three aspects of Hume's theory have retained the interest of contemporary philosophers: (1) we have no justification for causation, (2) causal relations cannot be known a priori, and (3) causation involves regularity; A is the cause of B entails that things similar to A will cause things similar to B. After having considered these issues successively, I will consider (4) Hume's view of the causal relata, the problematic character of which has been (almost) completely neglected by contemporary philosophers. 3.2.3.1 Causation and necessity According to Hume's empiricism, every idea is based upon a prior impression. Our mind has two kinds of contents: impressions and ideas. The main difference between them concerns their degree of "force and liveliness." Under the name impression Hume comprehended "all our sensations, passions and emotions, as they make their first appearance in the soul." By ideas he meant "the faint images of these [the impressions] in thinking and reasoning." Whereas complex ideas may be formed out of simpler ideas, there is but one way in which simple ideas can enter the mind, namely as "copies of our impressions" ([1739] 1978, 1-3). In accord with the empiricist principle that ideas are derived from impressions, Hume explained that in order to clarify and justify our idea of causation, we must find the impression that has given rise to it. The idea of necessity cannot be derived from our experience of individual cases of causation. For, in a single instance of causation, we can never discover any necessary connection or power. Instead, the idea of necessity arises from our experience of a great many similar instances. The constant conjunction produces an association of ideas - so if we see a flame, by sheer habit an idea of heat will come to mind. But the constant conjunction also produces a feeling of necessary connection in the mind. Thus, there are two roots of our idea of necessity: constant conjunction of the objects, and the feeling of necessary connection in the mind. The habitual transition from impression to idea feels like a necessitation, as if the mind were compelled to go from one to the other. Thus, ultimately, the tie of necessary connection "lies in
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
33
ourselves, and is nothing but the determination of the mind, which is acquir'd by custom, and causes us to make a transition from an object to its usual attendant, and from the impression of one to the lively idea of the other" ([1739] 1978, 266). The necessary connection is not discovered in the world but is projected onto the world by our minds. Hume accordingly gave two different definitions of 'cause.'19 First he defined causes in terms of regular succession: a cause is An object precedent and contiguous to another, and where all the objects resembling the former are plac'd in like relations of precedency and contiguity to those objects, that resemble the latter. (Hume [1739] 1978, 170)
The other definition accords with the experience that a cause always conveys the mind, by a habitual transition, to the idea of the effect. Thus, a cause is An object precedent and contiguous to another, and so united with it, that the idea of the one determines the mind to form the idea of the other, and the impression of the one to form a more lively idea of the other. (Hume [1739] 1978, 170)
Hume granted that these definitions, which were "drawn from objects foreign to the cause" ([1748] 1975, 77), may not be altogether satisfying to us. For they define the causal relation by appeal to objects other than the cause and the effect themselves (to objects resembling the cause and objects resembling the effect, and to ideas, impressions and minds). He insisted, however, that the two definitions were the best that could be given. 3.2.3.2 Causation and the a priori Most contemporary philosophers believe that Hume refuted the rationalists before him (such as Descartes, Spinoza, Leibniz and Hobbes), and the idealist after him (such as McTaggart and Blanshard) who held that there is an element of genuine a priori reasoning in causal inference. According to Hume, causal relations are not logically necessary, and hence they cannot be known a priori. To say that causation is not a logically necessary relation is to say that even if A caused B, it is not logically impossible to suppose that, given A, B might not have occurred. So far as reason and logic are concerned, given a particular event, anything may happen next. This is precisely the reason why causal relations cannot be known a priori; in order to determine whether or not a causal relation holds between A and B we must rely on our experience of similar relations. "There are no objects," wrote Hume, "which by the mere survey, without consulting experience, we can determine to be the causes of any other; and no objects, which we can certainly determine in the same manner not to be the causes" (Hume [1748] 1975, 75).
34
CHAPTER ONE
3.2.3.3 Constant conjunction The third influential element of Hume's theory of causation is the idea that causation involves "constant conjunction." As we have seen, it was Hume's view that causes are objects followed by other objects, such that all objects that are similar (in some respects) to the cause-object are followed by objects that are similar (in some respects) to the effect-object. Thus, if HIV causes AIDS, then all viruses sufficiently similar to HIV will be followed by AIDS. Hume held that constant conjunction or regularity is not only a necessary but also a sufficient requirement for causation. The latter claim, that constant conjunction is a sufficient condition for causation, was easily shown to be false by Thomas Reid (1710-96): there are many examples of constant conjunctions, such as day following night, that are not causal relations (Reid [1846] 1967, 627). However, the idea that regularity is essential to causation is generally accepted. If HIV is the cause of aids, then HIV and aids are constantly conjoined. This view seems to agree with our common sense view. We expect similar causes to have similar effects; we expect that each time that our children shoot a ball with sufficient speed against a window, the window will break. If the window had not broken, we take it for granted that this cause was not sufficiently similar to the causes that did break the window; probably the ball did not have enough speed. 3.2.3.4 The causal relata Though Hume never addressed the problem of the causal relata, and though he often gave examples in which the causal relata were events, his whole discussion (including his definitions of cause) is based on the idea that both causes and effects are 'objects,' conceived as substances having certain qualities. Consider, for example, the following passage by which Hume started his discussion of causation: ... we must consider the idea of causation,
and see from what origin it is
deriv'd. [...] Let us therefore cast our eye on any two objects, which we call cause and effect, and turn them on all sides, in order to find that impression, which produces an idea of such prodigious consequence. At first sight I perceive, that I must not search for it in any of the particular qualities of the objects; since, which-ever of these qualities I pitch on, I find some object, that is not possest of it, and yet falls under the denomination of cause and effect. And indeed there is nothing existent, either externally or internally, which is not to be consider'd either as a cause or an effect; tho' 'tis plain there is no one universal quality, which universally belongs to all beings, and gives them a title to that denomination. The idea, then, of causation must be deriv'd from some relation among objects. (Hume [1739] 1978, 74-75)
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
35
Hume apparently presupposed the causal relata to be substances, and observed that the idea of causation cannot be derived from some particular qualities of the cause-substance, but, instead, must be derived from some relation among things. Similarly, in the following passage he observed that we cannot justify our idea of causation (involving necessary connection) by just observing the essential qualities of certain objects: 'Tis easy to observe, that in tracing this relation [of cause and effect] the inference we draw from cause to effect, is not deriv'd merely from a survey of these particular objects, and from such a penetration into their essences as may discover the dependence of the one upon the other. There is no object which implies the existence of any other if we consider these objects in themselves, and never look beyond the ideas which we form of them. (Hume [1739] 1978, 86-87)
Thus, the idea of necessary connection cannot be derived from certain qualities of an object; by just observing all the qualities of a knife, one can never derive all the qualités of a wound. Hence, eventually, Hume's whole discussion takes place within an Aristotelian-scholastic framework, according to which the world consists in nothing but the set of substances (ousiai) and their qualities. By thus putting the question of causation in terms of an Aristotelian-scholastic conception of the world, Hume was forced to conceive necessitation as a matter of the imagination. Thus, Hume's basic insight concerns the Aristotelian categoreal framework: if the causal relata be substances (rather than events or facts), the idea of necessary connection cannot be justified by regarding individual instances of causation. 3.2.4
Kant: causation as an a priori conception
Hume's discussion of causation played an important role in the development of Kant's critical philosophy (Kant [1783] 1950, 257-60). Kant (1724-1804), much impressed by the obvious success and constant advance of scientific knowledge, and by the triumph of Newtonian physics in particular, could not accept Hume's conclusion that neither causation nor induction can be rationally justified, and that, consequently, we cannot rationally justify scientific knowledge. His basic epistemologica! strategy was to ground the principle of causality in the structure of reason. In order to overcome the epistemologically disastrous consequences of Hume's critique, Kant attempted to justify causality by declaring it an a priori conception. After establishing to his own satisfaction that certain pure concepts or categories, including those of substance and causality, are universally valid with respect to possible experience, Kant argued in the Transcendental
36
CHAPTER ONE
Analytic of the Critique of Pure Reason (1781/87) that he could prove the validity of a set of synthetic a priori principles involving each of his twelve categories. Among the most important of these are the principles (the "Analogies of Experience") that every change has a cause, and that all changes in the phenomenal world are alterations in the properties of enduring substances. In a preliminary section to the Transcendental Analytic, Kant had raised the problem of the transcendental deduction of the concept of cause. According to him, the concept of cause "signifies a special kind of synthesis, whereby upon something, A, there is posited something quite different, B, according to a rule." The problem with this concept of cause is that "it is not manifest a priori why appearances should contain anything of this kind [...]; and it is therefore a priori doubtful whether such a concept be not perhaps altogether empty, and have no object anywhere among appearances" (A 90/b 123). Kant saw only two possible solutions to this problem: either the concept of cause must be completely abandoned as a mere chimera, or it must be grounded completely a priori in the understanding: If we thought to escape these toilsome enquiries by saying that experience continually presents examples of such regularity among experiences and so affords abundant opportunity of abstracting the concept of cause, and at the same time of verifying the objective validity of such a concept, we should be overlooking the fact that the concept of cause can never arise in this manner. It must either be grounded completely a priori in the understanding, or must be entirely given up as a mere phantom of the brain. For this concept makes strict demand that something, A, should be such that something else, B, follows from it necessarily
and in accordance
with an absolutely
universal
rule. Appearances do indeed present cases from which a rule can be obtained according to which something usually happens, but they never prove the sequence to be necessary. To the synthesis of cause and effect there belongs a dignity which cannot be empirically expressed, namely, that the effect not only succeeds upon the cause, but that it is posited through it and arises out of it. This strict universality of the rule is never a characteristic of empirical rules; they can acquire through induction only comparative universality, thai is, extensive applicability. If we were to treat pure concepts of understanding as merely empirical products, we should be making a complete change in [the manner of] their employment. (Kant [1781/87] 1963, A 91-2/B 123-4)
In this passage, Kant follows Hume's account of what it is we believe when we say that A is the cause of B, but, contrary to Hume (who doubts the validity of what we believe), Kant accepts the belief and explores its implications. First, Kant explicitly endorses the belief that effects do not just follow their causes as a matter of fact, but follow them necessarily.
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
37
Moreover, contrary to Hume, who explains our belief in the universality involved in the causal relation in terms of induction, Kant argues that such explanation is mistaken because it does overlook the fact that experience can never provide strict universality ([1787] 1963, B 3-4). Whereas "the concept of cause implies a rule according to which one state follows another necessarily" ([1783], 1950, 315), experience can only show us that one state of things commonly follows another, and therefore neither affords strict universality nor necessity. Thus, an event A is the cause of an event B if, and only if, there is a universal law of the form: events of type A are necessarily followed by events of type B. And, because neither the necessity nor the strict universality involved in the causal relation can be established empirically, they must be grounded a priori. The judgment that A caused B must be grounded in the a priori conditions of objective judgment of possible experience. In his 'Second Analogy of Experience,' Kant's concern is with the principle that "all alterations take place in conformity with the law of the connection of cause and effect" ([1787] 1963, B 232). Kant argued that the objective time order depends on causation; it consists ... in that order of the manifold of appearance according to which, in conformity with a rule, the apprehension of that which happens follows upon the apprehension of that which precedes [...]. In conformity with such a rule there must lie in that which precedes an event the condition of a rule according to which this event invariably and necessarily follows. [...] If the state which precedes is posited, this determinate event follows inevitably and necessarily. (Kant 11781/87] 1963, A193/B238; A194/B239; A198/B244)
Thus, the Second Analogy is the principle that every event has a cause. Whereas Hume had roused Kant from his 'dogmatic slumber' by contending that this principle can never be established by human reason, Kant, on the contrary, maintained that the principle of causality can be shown to be a necessary condition of experience. He argued that we can know that an event has occurred only if we know that the event has a cause. For, the mere successiveness of perceptions is not sufficient to determine whether or not what we are observing is an event. For even when we observe a static object (say, a house) our perceptions occur successively. However, what distinguishes a merely subjective succession of phenomena (the various glimpses of a house) from an objective succession (a ship moving downstream) is that only in the first case can the order of perceptions be reversed. Events always pertain to irreversible processes. Causal regularity is a characteristic of events: "We never in experience attribute to an object the notion of succession [...] and distinguish it from the subjective succession of
38
CHAPTER ONE
apprehension, unless when a rule lies at the foundation." Thus Kant explicitly rejected Hume's view that we first perceive temporal succession between events, and then regard one as cause and the other as effect. The opposite is true: in order to establish an objective order in time, we need cause-effect relationships. This may seem to contradict all that has hitherto been taught in regard to the procedure of our understanding. The accepted view is that only through the perception and comparison of events repeatedly following in a uniform manner upon preceding appearances are we enabled to discover a rule according to which certain events follow always upon certain appearances, and that this is the way in which we are first led to construct for ourselves the concept of cause. Now the concept, if thus formed, would be merely empirical, and the rule which it supplies, that everything which happens has a cause, would be as contingent as the experience upon which it is based. Since the universality and necessity of the rule would not be grounded a priori, but only on induction, they would be merely fictitious and without genuinely universal validity. (Kant [1781/87] 1963, A196; B 241)
Kant noted that a major problem of his formula was that it seemed to limit the principle of causal connection among appearances to their serial succession, while "the great majority of efficient natural causes are simultaneous with their effect" such as a heated stove which is simultaneous with the heat of the room. But he thought he could solve this problem by making a distinction between "the order of time" and "a lapse of time," and between the "immediate effect" and the "complete effect." Whereas the immediate effect is simultaneous with "the causality of the cause" ('die Kausalität der Ursache'), the achievement of the complete effect takes a lapse of time. But, even the relationship between cause and immediate effect takes place in the order of time; cause and effect are distinguished through the time relation of their dynamical connection. "For if I lay the ball on the cushion, a hollow follows upon the previous flat smooth shape; but if [...] there previously exists a hollow in the cushion, a leaden ball does not follow upon it" ([1781/87] 1963, A203; B249). Thus, in the order of time, causes precede their effects. Though in his examples Kant often referred to substances as causal relata (such as stoves, balls, and glasses), he clearly held the causal relata to be events presupposing substances. When he mentioned substances as causes, he often used the expression "the causality of the cause" because he was well aware that a substance in itself does not cause anything. The ball in itself does not make a hole in the cushion, but the ball being laid on the cushion does, which is clearly an event. Thus, "the causality of the cause" refers to an event rather than a substance (this accords with his 'First Analogy of
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
39
Experience,' according to which all changes in the phenomenal world are alterations in the properties of permanent substances). To summarize: Kant claimed to have shown that concepts such as 'cause' (and 'substance,' and 'community'), and the principles drawn from them (for example, that each event "presupposes something upon which it follows according to a rule") "stand a priori before all experience and have their undoubted objective rightness, though admittedly only in respect of experience" ([1783] 1950, 311). The principle of causality is an a priori conception, grounded in the structure of reason. It involves that (a) every event has a cause; (b) the cause of every event is a prior event; (c) the effect follows from the cause necessarily, and (d) in accordance with an absolutely universal rule; (e) this is known to us not from experience but a priori. In the next section I will show that, contrary to Kant's a priorism, John Stuart Mill defended a radical empiricism, thus following in the steps of Locke, Newton, and Hume. 3.2.5
John Stuart Mill: causes and causal circumstances
Mill (1806-73) defined cause as "the antecedent, or the concurrence of antecedents, on which [a given phenomenon] is invariably and unconditionally consequent" (Mill 1874, 245). Mill in effect differs from Hume in one important respect. According to Hume, given two changes A and B, A may be said to be the cause of B and B the effect of A, if and only if A is immediately followed by B and things similar to A are always immediately followed by things similar to B. Mill adds the restriction that the two must unconditionally be conjoined. By so doing, he reintroduces the idea of necessity, which had been repudiated not only by Hume, but paradoxically by Mill himself at the beginning of his essay on the "Law of Causation" where, in speaking of the "supposed necessity" between physical facts, he writes that "[n]o such necessity exists for the purposes of the present inquiry, nor will any such doctrine be found in the following pages" (1874, 236). Apart from his inconsistency, Mill offers some interesting insights regarding causality. Thus he points out that what we usually call the cause of an object or an event is only a partial cause. In ordinary discourse we tend to call the cause the factor to which we wish to call attention, although it is not the only factor. We select one condition out of a whole set of conditions which are together sufficient, and call it the cause. What we call the cause is: (a) the last condition to be fulfilled before the effect takes place, or (b) the condition whose role in the affair is "superficially the most conspicuous" (1874, 238-39). Thus, we say that striking the match caused it to burn, because it was "the one condition which last came into existence" (proximate
CHAPTER ONE
40
event). And we may refer to the gene as the cause of the cancer, because it is the most conspicuous of all the conditions involved. According to Mill, this common sense idea of 'cause' is misleading because there are many more conditions involved that are equally necessary for the effect to occur. Therefore, "philosophically speaking," we have no right to select one of them and give it the name of 'cause.' Accordingly, Mill defined the cause of an event as the set of conditions upon which the event (that is, an event of this type) invariably occurs: The cause, then, philosophically speaking, is the sum total of all the conditions,
positive
and
negative
taken
together,
the
whole
of
the
contingencies of every description, which being realized, the consequent invariably follows. (Mill 1874, 241)
Moreover, the "real cause" of an occurrence is that set of conditions of a certain type which, when upon being realized, is invariably followed by a type of occurrence described as effect. However, the cause involves only those conditions that are needed for the effect to occur; it does not involve "the addition of any superfluous circumstance" (1874, 245). Furthermore, a cause not only involves a set of conditions sufficient for the occurrence of the effect, but it also involves that each condition be necessary, though none is by itself sufficient. Thus, the cause of combustion involves at least the following three conditions: a) oxygen, b) inflammable material, and c) a certain degree of heat. None of these conditions is by itself sufficient, each of them is necessary, and the set of them is sufficient to cause combustion. However, Mill specifies that, in order for a sufficient set of conditions to be a cause, it must not only be the set of conditions upon the occurrence of which the effect invariably occurs, but it must be the set of conditions upon the occurrence of which the effect unconditionally occurs as well. Given the whole constellation of causal circumstances, the effect will occur, whatever all other conditions may be: If there be any meaning which confessedly belongs to the term necessity, it is unconditionalness.
That which is necessary, that which must be, means that
which will be, whatever supposition we may make in regard to all other things. The succession of day and night evidently is not necessary in this sense. It is conditional on the occurrence of other antecedents. That which will be followed by a given consequent when, and only when, some third circumstance also exists, is not the cause, even though no case should ever have occurred in which the phenomenon took place without it. (Mill 1874, 245)
Mill's demand that the cause be unconditional was intended to eliminate non-causal regularities, like day following night. Day is not the cause of
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
41
night, because the occurrence of night is not dependent upon the prior occurrence of day, but on all kinds of circumstances, such as the existence of the sun, the earth rotating on its axis, and the absence of obscuring material between sun and earth. Thus, Mill tacitly reintroduced the idea of a necessary relation between causes and effects. For to say that A and B are "unconditionally" conjoined is to say that they are not only conjoined under all actual circumstances, but also under all imaginable or possible circumstances. By so doing, Mill had abandoned the basic point of empirical analysis, according to which causation must be analyzed in terms of what actually does happen, not in terms of what could possibly happen.20
4.
CONCLUSION: IMPORTANT CHANGES IN THE MEANING OF CAUSE
The development of the history of the concept of cause reveals a remarkable discrepancy between the constancy in the use of terminology and the gradual shift in the meaning of the terms used. This development - which has largely remained unnoticed - requires analysis, if only because most contemporary discussions on the subject, which almost invariably stand in the tradition of Hume, seem to have been victimized by it. For, contrary to what is generally supposed, causation is not a univocal conception (which either can or cannot be further analyzed). It is an ambiguous conception, with at least two (or three) different meanings, each of which requires a critical analysis on its own. Hume's celebrated criticism concerns only one of these senses of cause, which is notably just a derivative sense. In this concluding review of the results of this article, I will point out the conceptual tensions that are inherent to the historical development of the concept of cause. More specifically, I will show that two decisive milestones mark the history of causality: the Aristotelian (-scholastic) Conception (I), and the Scientific Conception (II). It will be shown that these two conceptions of cause are incompatible. (I) Aristotle conceived efficient causes as 'things responsible for a certain state of affairs', An efficient cause is a thing, which by its activity brings about an effect in another thing. Thus, the efficient cause is defined by reference to some substance performing a change: it is the "primary source of the change." That which is produced is either some new substance, such as ashes from wood, or simply a change in some property of a given substance. Furthermore, an efficient cause involves a teleological factor, in as much as each efficient cause acts for the sake of an end. Hard work, for example, is the efficient cause of fitness, which is the end. Thus, according
42
CHAPTER ONE
to the Aristotelian conception, causes are conceived as the active originators of a change that is brought about for the sake of some end. (II) Probably the most radical change in the meaning of cause happened during the seventeenth century, in which there emerged a strong tendency to understand causal relations as instances of deterministic laws. Causes were no longer seen as the active initiators of a change, but as inactive nodes in a law-like implication chain. This change of perspective had its antecedents in Stoicism and medieval philosophy. The Stoics were the first to associate causation with exceptionless regularity and necessity. But, contrary to the modern scientific conception of cause, the necessity involved in the Stoic perception of a causal relation pertained to particular events; it was thought to be necessary that the same particular effect would recur in the same particular circumstances, and that it was not possible that it would be otherwise. Aquinas went further than the Stoics by relating efficient causalityto natural necessity and to law-like behavior; things belonging to the same type act similarly in similar causal circumstances. The dismissal of explanations by final and formal causation by Descartes, Francis Bacon and Galilei brought about the rejection of the Aristotelian-scholastic doctrine of substantial forms as causal factors in natural processes. Their view was that the Aristotelian idea of a substantial form being transmitted from a cause to its effect had no basis whatever in our experience of things. However, paradoxical as it may seem, precisely this concept of formal cause came to play an important role in the development of the new conception of efficient cause, according to which efficient causes were simply instances of general laws, which in turn were general, mathematical principles. For, to a large extent, the concept of law of nature was the inheritor of the concept of formal cause: both concepts were meant to explain the stability of the world. The main difference is that, whereas the formal cause was thought to explain the stability of the world by explaining the structure of things, the laws of nature were thought to explain the stability of the world by explaining the relations between things. An important characteristic of the modern conception of cause is that causation and determinism became virtually equivalent. The crux of the debate between the rationalists and the empiricists pertained to the nature of this determination. (IIa) In the rationalist conception of cause, the relationship between cause and effect is a quasi-logical relation. Necessitation involves implication. Thus, a complete knowledge of the causes is tantamount to knowing the premises from which by reasoning alone the effects can be deduced. Though Descartes, Hobbes, Spinoza, and Leibniz roughly shared this view, they could not avoid some basic ambiguities.
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
43
Thus, Descartes, while regarding the laws of nature themselves (instead of their instances) as efficient causes, also held the view that the relation of a cause to its effect is the relation of ground to consequent: a logical relation. On the other hand, when it came to the minds of human beings, which he thought were free beings, he retreated to the traditional Aristotelian substantialist view of efficient causes as active initiators of a change.21 Hobbes's position was equally ambiguous. For, while he defended Ha (given the cause, "it cannot be conceived but that the effect will follow"), he regarded the concepts of cause and power as complementary notions (Hobbes 1655, 9.3), an idea that is characteristic of the Aristotelian conception of cause (I). Similarly, Spinoza, perhaps the most straightforward defender of the view that necessitation involves implication, held that God is a free cause in the sense of being a real initiator of change, but that the necessary (natural) causes are necessitated by other causes and are therefore just inactive nodes in a chain, each of them logically necessitating its effects and logically necessitated by its effects. Even Leibniz, who held the most original view of causation by rejecting the idea that the ultimate constituents of reality (the monads) have a causal relation to each other, and thus limiting causation to the links of the historical-logical chain constituting each individual substance, could not avoid one major ambiguity. For, whereas, in his view, the necessity involved in the causal relation is as strong as logical necessity (Ha), the innermost significance of causality is that of the active initiation of change (I). For, every monad is "spontaneous," that is to say, its substantial form is the only source of its modifications. Monads are real "centers of activity." Remarkably, Leibniz' originality resulted partly from his 'reactionary' defense of both formal and final causation. Each monad behaves in accordance with its nature or substantial form, which is its original purpose, given by God. Efficient causes are therefore means to ends (I). Thus, it may be argued that the rationalist philosophers all held some hybrid conception of cause, involving a combination of cause IIa (the identification of causes with grounds) and cause I (the originators of a change). (IIb) To David Hume, commonly held to be the main representative of the empiricist approach to causation, our idea of causal necessity is due partly to our observation of the constant conjunction of certain objects, and partly to the feeling of their necessary connection in the mind. The habitual impression of conjunction feels like a necessitation. as if the mind were compelled to go from one to the other. The necessary connection is not discovered in the world but is projected onto the world by our minds.
44
CHAPTERFOLR
However, Hume's view was far from being shared by all empiricist philosophers. Indeed, by suggesting that his fellow empiricists held the belief that necessity is synonymous with power, he seriously misrepresented their views. For, both Locke and Newton explicitly denied that the ideas of causation or power involved the idea of necessary connection according to law. According to Newton, these two notions are even mutually exclusive because complete uniformity or necessary connection would entail a denial of causal efficacy. For Locke, as for Newton, causality is related to the Aristotelian belief that causes are substantial powers that are put to work. Therefore, Hume's famous criticism only concerns the rationalist scientific conception of cause (Ha), which, from a historical perspective, is merely a derivative sense of 'cause.' Kant's concept of cause, by which he tried reconcile the rationalist and empiricist views, is a hybrid of Ha and lib. Because causal relations involve laws (II), Kant in effect says that an event A is the cause of an event B if, and only if, there is a universal law of the form: events of type A are necessarily followed by events of type B. But, while defending the rationalist idea that causality is an a priori conception, which involves strict universality and necessity (Ila), he also holds the empiricist view that causes precede their effects, which from the perspective of Ha (according to which, causes are contemporaneous with their effects) is utterly impossible. Mill too conceived causal relations in terms of law-like generalizations (II). His analysis is about kinds of causes and kinds of effects. The "real cause" of an event is that set of conditions which, upon being realized, is invariably and unconditionally followed by the type described as the effect. All in all, the complex evolution of the concept of cause from the seventeenth century on is marked by the interplay between, at least, two radically different conceptions of cause: the Aristotelian-scholastic conception, according to which causes are the active initiators of a change, and the scientific conception, according to which causes are the inactive nodes in a law-like implication chain. Our common use of causal terms is entirely oblivious of this ambiguity. But, more deplorably, most discussions by modern philosophers have failed to see this basic duality, because all too often the premises of those discussions are infected by it. For instance, the common assumption that causation is inherently law-like behavior masks a basic ambiguity. In short, my analysis of the historical development of the concept of cause shows that each analysis of causation must start from the recognition that causal propositions are ambiguous, and that (at least) two mutually exclusive meanings are often considered as being one and the same. According to I, 'A is the cause of B' means 'A is the initiator of a change in
KEY MOMENTS IN THE HISTORY OF 'CAUSATION'
45
B'; according to II, 'A is the cause of B' means 'Given the occurrence of B, A must necessarily have occurred.' In order to show the relevancy of the Peircean account of causation that I will defend in the course of this book, it is necessary to first briefly explore the major contemporary approaches to the problem. Thus, in the next chapter, I will give a survey of the contemporary philosophical accounts of the concept of cause.
Chapter 2 CONTEMPORARY APPROACHES TO CAUSATION
The attempt to "analyze" causation seems to have reached an impasse; the proposals on hand seem so widely divergent that one wonders whether they are all analyses of one and the same concept. (J. Kim 1995, 112)
This chapter consists of two parts. In the first section, I will give a general outline of the current approaches to causation. In the second, I will discuss a number of pertinent problems the current approaches are afflicted with.
1.
THE CONTEMPORARY DEBATE
The following approaches have been prominent in contemporary discussions of causation: 1 (1) the approach that analyzes causation in terms of necessary and/or sufficient conditions, (2) the counterfactual approach, (3) the instrumental approach, and (4) the probabilistic approach. Moreover, (5) there is the so-called singularist approach to causation, which is a minority view.
1.1
Necessary and/or sufficient conditions
The necessary and/or sufficient conditions approach is by far the most favored line of enquiry into the nature of causation. It may be traced to the philosophy of David Hume and John Stuart Mill. At the heart of their approach, there is the view that causation is a matter of regularity and that
47
48
CHAPTERFOLR
causally related events instantiate a general regularity between like kinds of events. Thus, a is the cause of b, if and only if there are kinds of events, X and Y, such that a is of kind X, b is of kind Y, and events of kind X are regularly followed by events of kind Y. The position, held by most contemporary philosophers, that regularity (or constant conjunction) is a necessary condition of causation, is not without problems. Indeed, the concept of constant conjunction implies a relationship between kinds of causes and kinds of effects. Causation, however, is first and foremost a relationship between an individual cause and an individual effect. Hence, an adequate analysis of causation requires an answer to the question, What is it about this particular A that makes it the cause of this particular effect B? What is it about this particular gene that makes it the cause of this particular malignant tumor in this particular person? But if regularity should be a necessary condition of causation, it certainly is not a sufficient condition. No gene can by itself be the cause of cancer. All kinds of biological conditions must be met before a gene can function at all. And there probably are also many other conditions that must be met before the gene can produce cancer. Therefore, most contemporary philosophers make a distinction between the 'causes' and the 'causal circumstances' under which a given change occurs. Moreover, they distinguish causally relevant from causally irrelevant conditions. 1.1.1
Causes as necessary conditions
Let some of the (positive and negative) conditions of a given fire be: (a) a short circuit, (b) the presence of inflammable material, (c) the absence of a suitably placed sprinkler, (d) the presence of a bird watching the event, and (e) a storm raging violently some hundred miles away. Obviously, a), b), and c) are causally relevant to the fire, while d) and e) are not. But that in itself does not explain why the first three factors are relevant, while the last two are not. For each event there may be an indefinite amount of immensely complex conditions, but only some of them are causally relevant in the sense that, in their absence, the event would not have occurred. In the light of this many recent philosophers (for example, Collingwood 1938; Hart and Honoré 1959; Nagel 1961; Hartshorne 1970), have regarded causally relevant conditions as those that are in some sense necessary for the occurrence of an event. Thus, A is a necessary condition for B i f , under the circumstances, in the absence of A, B would not have occurred. According to the majority view - Ernest Sosa (1980) being a notable exception - the kind of necessity involved here is not logical necessity. There is no logical contradiction between the proposition 'the burning match touches gasoline' and the proposition 'the gasoline touched by the burning match does not burn.'
CONTEMPORARY APPROACHES TO CAUSATION
49
According to the necessary condition analysis, causes may be defined in one of the two following ways: la. A causal condition of an event is any sine qua non condition under which that event occurred. Had the condition in question not obtained, the effect would not have occurred. The cause of the event is the totality of necessary conditions. 2 Ib. A is the cause of B if and only if A and B have obtained, and A was, under the circumstances, necessary for B. Thus, the cause of an event is just one of its necessary conditions. 3
Whereas the first definition is philosophically simpler and more useful for the understanding of causal connections, the second reflects ordinary usage better. According to definition la, an ordinary causal statement such as 'the short circuit caused the fire' would be incomplete because the short circuit was only one of the many necessary conditions, such as the presence of oxygen and inflammable material. According to definition Ib, there is nothing wrong with saying that the short circuit was the cause of the fire. For all the other causally relevant conditions may be regarded as standing conditions, which are covered by the expression 'under the circumstances.' (For reasons of convenience, I will call these other causally relevant conditions from now on the complex condition C.) The short circuit rather than any of the other causally relevant conditions is called the 'cause' of the fire, because it is the one conspicuous factor in the set of standing conditions. As Mill had already emphasized, the difference between la and Ib is more of a practical than of a philosophical concern. From a logical point of view it makes no difference whether the term 'cause' is used to refer to the one conspicuous condition rather than to the whole set of necessary conditions. In deference to ordinary usage most contemporary philosophers prefer the expression 'the cause' for some causal condition that is novel or conspicuous, (or, particularly, one that is within one's control4). Thus, for example, Hart and Honoré concluded that... [t]he notion, that a cause is essentially something which interferes with or intervenes in the course of events which would normally take place, is central to the common-sense concept of a cause, and at least as essential as the notions of invariable or constant sequence so much stressed by Mill and Hume. (Hart and Honoré 1959, 29)
50
CHAPTERFOLR
Thus, if, for reasons of convenience, we restrict ourselves to definition lb, it can be said that A is the cause ofB if and only if A and B have occurred, and there was a condition C such that B necessitated A on condition C. An important problem with the necessary condition analyses was noticed by Michael Scriven (1966, 58). He pointed out that in cases of causal overdetermination, causes are not necessary under the circumstances (or contingently necessary). When a man is killed by a firing squad, he is hit by many bullets each of which may ensure his death. But it would be wrong to infer that each shot is a contingently necessary condition of his death. Thus, in terms of the necessary condition analysis (la and lb), none of the bullets nor the set would be a cause of the man's death.5 1.1.2
Causes as sufficient conditions
The sufficiency thesis, which is closely related to the necessary condition thesis, was held by, for instance, Mill (1879), Braithwaite (1953), Hart and Honoré (1959), Hempel (1965), Popper (1972), and Honderich (1988). Consider, again, definition la, according to which the cause A of an effect B includes every condition necessary for the occurrence of B. According to the defenders of the sufficiency thesis, this entails that the set of A was also sufficient for B, since no other condition was necessary. Thus, according to the sufficiency thesis, the cause of an event may be defined as: IIa. The cause (A) of an event (B) is that set of conditions, among all the conditions that obtained, each of which was necessary and the totality of which was jointly sufficient for the occurrence of B.
However, analogously to definition lb, a (sufficient) cause may also be defined as: lib. A is the cause of B if and only if A and B have obtained and A was, under the circumstances, sufficient for B.
Again, though the first definition may be philosophically simpler, the second one reflects our ordinary use of 'cause' much better. According to definition lib, A is the cause ofB if and only if A and B have occurred and there was a condition C such that A necessitated B on condition C. However, according to the majority view, causal sufficiency does not entail logical necessity. To be sure, there are logically necessary and sufficient conditions, as in the case of 'if this table is round, then it is not square.' The state of affairs expressed in the first part of the sentence (p) is a sufficient condition of the state of affairs expressed in the second part (q), and, correlatively, the state of affairs expressed by q is a necessary condition of the state of affairs expressed by p. But, it cannot be said that the
CONTEMPORARY APPROACHES TO CAUSATION
51
roundness of the table is the cause of its not being square. Causally sufficient conditions are not logically sufficient conditions. Striking this match is, under the circumstances, a causally sufficient condition of the match's lighting, but it is not a logically sufficient condition. In the context of causal expressions, 'sufficient' should not be understood in the logical sense, but in the ordinary sense of "enough," "lacks nothing" (Sanford 1975). If everything necessary for B obtains, the set of conditions is collectively sufficient for B's occurrence, because nothing necessary for B is missing. But this is no argument against the logical possibility that in cases in which all the causally necessary conditions are met, B may still not occur.6 There are at least two important problems with the sufficiency thesis. The first one, which was raised by Richard Taylor (1966), is the problem of undeterminative sufficiency: there are many examples that perfectly satisfy the conditions put forward by the sufficiency thesis, but of which it is obvious that they aren't causes. The roundness of the table is not the cause of its not being square, although sufficient to explain its not being square.7 The second problem (put forward by Anscombe 1971, and many other authors) is that the sufficiency thesis precludes the idea of probabilistic causation. For, if it is true that there are probabilistic causal laws, as some think quantum theory suggests, events may have causes that are not a part of a sufficient condition. However, the issue of probabilistic causal laws is still hotly debated. On the one hand it pertains to the ambiguity of the concept of cause; more specifically, it concerns the question whether the concept of cause entails the idea of determinism. On the other hand, it concerns the questionable empirical evidence for the idea that (some of) the laws of nature are probabilistic rather than deterministic. 1.1.3
Causes as necessary and sufficient conditions
Probably the most intuitive idea of causes is that causes are both necessary and sufficient conditions. According to the necessary and sufficient conditions thesis, the cause A of an event B may be defined as: Ilia. ... that set of conditions, among all the conditions that occurred, each of which was necessary and the totality of which was jointly sufficient for the occurrence of B, or Illb. A is the cause of B if and only if A and B have obtained, and A was, under the circumstances, both necessary and sufficient for B.
Evidently, definition Ilia is identical with definition Ha (according to Ha, the sufficiency thesis involves the necessity thesis).
CHAPTERFOLR
52
Of course, the objections against the sufficiency thesis and the necessary condition thesis also apply to the necessary and sufficiency thesis. Perhaps most nagging, however, are the problem of 'undeterminative sufficiency' and the problem of 'directionality.' The problem of undeterminative sufficiency is the problem that many examples of necessary and sufficient conditions cannot in any way be called causes. The problem of directionality, which was first raised by Michael Scriven (1966), points to the fact that all the accounts of causation in terms of necessary and sufficient conditions fail to distinguish causes from their effects. For, if A and B are events and A is necessary and sufficient for B and therefore the cause of B, then it logically follows that B is necessary and sufficient for A and therefore the cause of A. But this is quite plainly absurd. The ignition of a match is not the cause of its being rubbed, and the warmth of a stone is not the cause of the sunshine. The distinction between causes and effects obviously requires some temporal constraints. In short, the standard accounts of causation in terms of necessary and sufficient conditions are inadequate. In the next section I will consider John Mackie's more sophisticated version (inasmuch as it better reflects our ordinary use of 'cause') of this type of analysis. 1.1.4
Causation and INUS conditions
Mackie suggested that a cause of some particular event is: ... an insufficient
but non-redundant
part of a condition which is itself
unnecessary but sufficient for the result. (Mackie 1974, 62)
Mackie called a condition of this kind an INUS condition, after the initial letters of the main words used in the definition. Thus, the short circuit was in itself Insufficient for the fire; a wider set of conditions was required, such as the presence of oxygen and inflammable material. The short circuit was a Non-redundant part of this complex sufficient condition, for without the short circuit, this complex condition would not have been sufficient to cause the fire. Whereas the whole complex condition, including the short circuit, was Sufficient for the fire to occur, it was Unnecessary for the fire, for the fire could also have come into being under other circumstances. Thus, when experts declare a short circuit to be the cause of the fire, they "... are saying, in effect, that the short-circuit is a condition of this sort, that it occurred, that the other conditions which, conjoined with it, form a sufficient condition were also present, and that no other sufficient condition of the house's catching fire was present on this occasion" (Mackie 1965, 34). Thus, the cause (A) of an event (B) is a part of a complex condition (C) which is sufficient, but not necessary, for the effect (B). (Thus, A is, under the
CONTEMPORARY APPROACHES TO CAUSATION
53
circumstances, not necessary for B; fire could be brought about by different means, for example by lightning.) Mackie defined an INUS condition as follows: A is an INUS condition of a result P if and only if, for some X, (AX or Y) is a necessary and sufficient condition of P, but A is not a sufficient condition of P and X is not a sufficient condition of P. (Mackie [1965] 1993, 35)
Here 'X' stands for the conjunction of terms that together with A constitute a minimal sufficient condition of P; 'Y' stands for the disjunction of other minimal sufficient conditions. Thus, more briefly, we can say that "A is an INUS condition of P when (A ... or ...) is a necessary and sufficient condition of P" ([1965] 1993, 35). The analysis of a singular causal judgment 'A caused P' requires one more notion: a condition A is said to be at least an INUS condition of P if and only if (i) either A is an INUS condition, or (ii) A itself is a minimal sufficient condition, or (iii) A is a component in the only minimal sufficient condition of P, or (iv) A is by itself a necessary and sufficient condition of P. Though Mackie did not pretend to have given a complete analysis of 'A caused P, ' he thought the four following points to be "an important part of the concept of causation." Thus, 'A caused P, ' entails that: (i) A is at least an INUS condition of P. (ii) A is present on the occasion in question. (iii) The factors represented by the 'X,' if any, in the formula for the necessary and sufficient conditions, are present on the occasion in question. (iv) Every disjunct in 'Y' that does not contain A as a conjunct is absent on the occasion in question. (Mackie [1965] 1993, 37) Mackie suggested certain refinements, such as the explanation of the notion of an INUS condition in terms of necessary and sufficient conditions, which in turn he analyzed in terms of counterfactual conditionals.8 These, however, will not concern us here. The INUS condition approach to causation raises a number of problems, most of which it inherits from the accounts of causation in terms of necessary and/or sufficient conditions. First, as is emphasized by Scriven (1966), there is the problem of directionality. Secondly, there is the problem of probabilistic causal laws. In the words of Sosa and Tooley (1993, 9) this entails that if "there can be probabilistic laws, then an event can have a cause that is not a part of any condition which is sufficient, in the circumstances, for the event in question." 9
54
CHAPTERFOLR
Given all the above mentioned problems with the regularity and necessary and/or sufficient conditions approaches, 10 philosophers such as Robert Stalnaker (1968) and David Lewis (1973), thought it was time to give up this type of analysis, and proposed something they thought was more promising: a counterfactual analysis of causation.
1.2
Causes and counterfactual dependency
The idea that causation involves counterfactual dependence goes back to Hume's second definition of cause. Hume defined 'cause' twice over. He wrote, "we may define a cause to be an object followed by another, and where all the objects, similar to the first, are followed by objects similar to the second. Or, in other words, where, if the first object had not been, the second never had existed." (Hume [1748] 1975, 76). According to the first definition, a causal succession is supposed to be a succession that instantiates a regularity. Present-day necessary and/or sufficient condition analyses all descend from this definition" by considering the cause (A) of an event (B) as a necessary part of a set of conditions (C) which are jointly sufficient, given the laws of nature, for the occurrence of the effect (B). But Hume's second definition suggests that causality can also be understood in terms of a counterfactual relationship: "if the first object had not been, the second never had existed." Similarly, according to some modern interpretations, a cause is the necessary condition of its effect, in the sense that the effect could not have occurred but for the occurrence of the cause. Saying that the short circuit was a necessary condition of the fire, is saying that, given the circumstances, this fire could not have occurred without this short circuit. Before continuing our exposition of the counterfactual approach to causation, we must have at least a rough understanding of what is meant by 'counterfactual conditionals.' According to the contemporary philosophy of language, there are two kinds of conditional statements, called 'indicative' and 'subjunctive conditionals.' Conditionals are statements of the form 'if p, then q' (or equivalently 'q if p). A conditional such as 'If Oswald did not kill Kennedy, someone else did,' is called indicative, because both the antecedent and the consequent are in the indicative mood. 12 This appears more clearly by translating the conditional into its disjunctive equivalent: Oswald killed Kennedy or someone else killed Kennedy. Both disjuncts are clearly in the indicative mood. Subjunctive conditionals have the form 'if it were the case that p, then it would be the case that q. ' Counterfactuals, or contrary-to-fact conditionals, belong to the class of subjunctive conditionals: they are called
CONTEMPORARY APPROACHES TO CAUSATION
55
'counterfactual' because they presuppose the falsity of their antecedents. They are often symbolized as 'p []—» q'. An example is: 'If Hitler had won the war, life would have been miserable.' The falsity of the antecedents of counterfactuals is presupposed rather than asserted. The above-mentioned counterfactual statement makes sense only to those who know that Hitler did not win the war. Accepting this counterfactual as true depends on accepting some theory about the counterfactual statement as a whole. According to many contemporary philosophers, causal relations are best defined in terms of relations of counterfactual dependence. According to David Lewis, one of the most prominent defenders of the counterfactual approach to causation, our ordinary use of 'cause' involves the idea of counterfactual dependence: We think of a cause as something that makes a difference, and the difference it makes must be a difference from what would have happened without it. Had it been absent, its effects - some of them, at least, and usually all - would have been absent as well. (Lewis [1973] 1993, 194)
Basic to Lewis's approach are the analysis of causation in terms of the more restricted idea of causal dependence, 13 and the analysis of causal dependence in terms of counterfactual dependence. Suppose A and B to be events that have occurred. To say, then, that 'B is causally dependent upon A' is to say that 'if A had not occurred, then B would not have occurred.' Moreover, to say that A is the cause of B is to say that there is a chain of causally dependent events linking A with B.14 And because causal dependence is understood in terms of counterfactual dependence, Lewis's basic idea may be expressed (roughly) in the following definition of cause: A is the cause of B if and only if there is a chain of counterfactually dependent events linking A and B. Perhaps the greatest difficulty with the counterfactual approach is underdetermination. If taken literally, the counterfactualist says that 'If A had not occurred, B would not have occurred' entails A caused B.' However, as Kim (1973) has pointed out, there are many examples of counterfactual dependency that do not display causal relations. For example: "If my sister had not given birth at t, I would not have become an uncle at t. " Or, "If I had not raised my hand, I would not have signaled" (Kim [1973] 1993, 206). Though signaling depends asymmetrically on raising my hand, raising my hand is not the cause of signaling. The question, then, is whether these noncausal counterfactuals can be distinguished from causal counterfactuals without using causal concepts. According to some (Kim, Horwich), counterfactual dependency is not only too broad to capture causal dependency; it can also be too narrow, as in cases of causal over-determination. Consider Horwich's example of a man
56
CHAPTERFOLR
who was shot in the head simultaneously by two people. The effect is not counterfactually dependent on its causes, for the man would have died even if one of the two had not shot him. Horwich concludes that this example shows that counterfactual dependence is not necessary for causation (Horwich [1987] 1993,210). So far I have discussed a number of attempts to explain causation in terms of conditionals. The difficulties that beset these approaches have led a number of philosophers to abandon the whole strategy of explaining causation in terms of conditionality or lawfulness. Some have tried to explain causation by supplementing conditionality with agency: the agency or means-to-ends approach. Others defend a probabilistic approach, according to which probability rather then conditionality is the basic concept. Finally there are those who have tried to explain causation without in any way banking on conditionality or lawfulness; their approach is called the singularist approach. In the following three sections I will discuss these approaches successively.
1.3
The instrumental approach: causes as means-to-ends
Any adequate analysis of causation should enable one to distinguish analytically between causes and effects. We have seen that the 'necessary and sufficient conditions approaches' fail in that respect. In order to distinguish causes from their effects (in order to be able to say that the warmth of a stone is not the cause of the sunshine) they require temporal constraints. Some philosophers (for instance Horwich 15 ) claim that the counterfactual approach suffers from the problem of directionality too. Due advantage of the explanation of causation in terms of agency, that is to say, in terms of means to ends, is that it can tackle the problem of directionality. Causes differ fundamentally from effects because they are always within the immediate control of agents, and the effects can only be indirectly controlled because of the relation in which they stand to the causes. Thus, the cause is an event or state that we can produce or prevent at will, or otherwise manipulate, in order to produce or prevent a certain other event as an effect. We may make an iron bar glow (end) by means of heating it (cause). This relation is irreversible: we cannot heat the bar by making it glow. Thus, whereas heating may be the cause of the glowing, the glowing cannot be the cause of the heating. The best-known defense of the agency approach to causation was given by R.C. Collingwood (1938). His view, however, was not that causation in general can be analyzed in terms of means-ends relations. Instead, he defended the thesis that causal propositions are ambiguous, and that only one of three different meanings of 'cause' involves agency as an essential
CONTEMPORARY APPROACHES TO CAUSATION
57
element.16 According to this (agency) conception of 'cause,' a cause is a tool, so to speak, by which we can manipulate nature. In Collingwood's own words: A cause is an event or state of things which it is in our power to produce or prevent, and by producing or preventing which we can produce or prevent that whose cause it is said to be. (Collingwood [1938] 1991, 148)
Causes, understood as means toward ends, cannot in themselves produce their effects. They always work in combination with conditiones sine quibus non, which are usually not specified. For example, a short circuit does not cause a fire without other conditions being fulfilled. In the previous chapter, we have seen that Mill refused to make a distinction between causes and conditions, because selecting the cause from the other, equally relevant conditions, would be purely arbitrary. According to Collingwood, however, this selection is by no means arbitrary: "for any given person, the cause of a given thing is that one of its conditions which he is able to produce or prevent" ([1938] 1991, 150). For example, from the driver's point of view, the cause of a car accident may be that he drove too hard, while from the county surveyor's point of view, it may be a defective road surface, and from the motor manufacturer's point of view, it may be a defective design. In this sense of the word cause, only persons that are practically concerned with certain kinds of event can form opinions about their causes. "For a mere spectator there are no causes" (Collingwood [1938] 1991, 151). Collingwood argued that his definition of causes as means concerns general causal statements rather than singular ones. One might put it differently by saying that it concerns universals rather than individuals. Thus, the proposition 'x causes / reads, rightly understood, "any instance of X is a thing whose production or prevention is means respectively of producing or preventing some instance of y" ([1938] 1991, 151). In this sense of the word cause, all causal propositions are general. It would be absurd to look for the cause of an individual event as such. Gaskin (1955) and G.H. von Wright (1971)17 defended the more radical thesis that causation in general can be analyzed in terms of agency. In Gaskin's view, we come close to the meaning of 'A causes B' when we say that A is a "recipe" for producing or preventing B. However, Gaskin acknowledged that there are examples in which some particular event A causes some other particular event B, while we are unable to produce A as a means of producing B. He solved that difficulty by stating that the only thing required is that we be able to produce something like A.
58
CHAPTERFOLR For example, one may say that the rise in mean sea level at a certain geological epoch was due to the melting of the Polar ice cap. But when one can properly say this sort of thing it is always the case that people can produce events of the first sort as a means to producing events of the second sort. For example, one can melt ice in order to raise the level of water in a certain area. We could come rather closer to the meaning of "A causes B" if we said: "Events of the B sort can be produced by means of producing events of the A sort." (Gaskin 1955, 483)
According to Gaskin, the agency approach does not have the unacceptable consequence that an event requires the presence of an agent who either brought it about or could have prevented it. The reign of causation goes far beyond the actual reach of agents. Another important objection against the agency approach is that it begs the question,18 The idea of producing or preventing one thing by producing or preventing another thing appears to presuppose the concept of causation. The concepts of producing, preventing, manipulating, controlling, bringing about, and so on, are all causal concepts. And thus, it does not make us any wiser.
1.4
Probabilistic causation
According to the instrumental view, causes are sufficient conditions. If every event has a sufficient cause, then it is true that all events without exception are entirely necessitated by earlier events. But, according to many scientists and philosophers, modern quantum physics is incompatible with this view. This explains the increasingly popular view that causation is a probabilistic concept. The basic idea of the probabilistic approaches to causation is that a cause is an event A such that it makes the occurrence of another event (B) more likely than if A had not occurred. There are, however, two ways in which this idea has been captured. The first, and most favored, strategy concerns types of events; the occurrence of an instance of the cause type should increase the probability that an instance of the effect type will occur. According to it, an event A may be said to be a probabilistic cause of an event B, i f , given the occurrence of A, the probability of the occurrence of B is higher than the probability of B if A had not occurred. Accordingly, the basic form of a probabilistic causal law is: p(B/A) > p(B/-A)
CONTEMPORARY APPROACHES TO CAUSATION
59
Thus, saying that smoking (A) causes lung cancer (B) is saying that the probability of getting lung cancer is higher for those who smoke than for those who do not. A well-known representative of this strategy is Patrick Suppes, who introduced the notion of a prima-facie cause, which he defined as follows: An event S is a prima facie cause of an event A if and only if (i) B occurs earlier than A, and (ii) the conditional probability of A occurring when B occurs is greater than the unconditional probability of A occurring. (Suppes 1984, 151)
Unfortunately, the suggested strategy fails when it is applied to particular cases. On the whole, smoking in general may increase the probability of lung cancer in general. But when we say that Joe's smoking is the cause of his lung cancer, the translation in terms of probability no longer works. His smoking did not increase the probability of his cancer; it caused it. This explains why, over the last fifteen or twenty years, a number of probabilistic theories of singular causation have been proposed (Mellor 1986; Lewis 1986; Menzies 1989b). The common characteristic of their views is that the idea that a cause makes its effect more likely should not be understood in terms of the conditional and the unconditional probabilities of a given type of event. Instead, it ought to be understood in terms of counterfactuals concerning the objective chances of individual events. The basic idea is, roughly, that A is the cause of B, if A and B are individual events (orfacts19), and the objective, single-case chance ofB's occurrence is raised by the occurrence of A.20 One of the most crucial objections to both probabilistic strategies is that it is probably not true that causes always make their effects more likely. To illustrate this objection, I will cite one version of an example given by Sosa and Tooley: Suppose, then, that there are two types of disease, A and B, satisfying the following conditions. First, it is a law that contracting disease A causes death with probability 0.1, and a law that contracting disease B causes death with probability 0.8. Secondly, it is a law that contracting disease A produces complete immunity to disease B, and a law that contracting disease B produces complete immunity to disease A. Thirdly, it is a causal law that in condition C an individual must contract either disease A or disease B. [...] Fourthly, individual X is in condition C, contracts disease A, and the latter causes his death. Given that these conditions obtain, the question is what would have been the case if X, though being in condition C, had not contracted disease A, and the answer, it would seem, is given by the following counterfactual: if individual X had not contracted disease A, he would have
60
CHAPTERFOLR contracted disease B. But if individual X would have contracted disease B if he had not contracted disease A, then his probability of dying had he not contracted disease A would have been 0.8, and so would have been higher than his probability of dying given that he had contracted disease A, since the latter is only 0.1. (Sosa and Tooley 1993, 21)
Thus, Sosa and Tooley conclude that it is not true that causes always make their effects more likely. In the next section I will consider the one approach that rejects all the attempts to analyze causation in terms of conditionality or laws, whether they be deterministic or probabilistic.
1.5
The singularist approach
The view that causation is merely a relation between concrete individual events, rather than a relation involving some sort of reference to corresponding types of individuals, was held by C.J. Ducasse (1926) and probably also by Elizabeth Anscombe (1971). Since Ducasse was more explicit on this matter than Anscombe, I will first briefly discuss Anscombe's view, before presenting Ducasse's theory. In her inaugural lecture at Cambridge University, 'Causality and Determination' (1971), Anscombe addressed three claims: (1) that causal relations are instances of exceptionless generalizations, (2) that causation is to be identified with necessitation, and (3) that causal relations presuppose some kind of law. According to Anscombe, it is generally assumed that, if one event is taken to be the cause of another, the relationship between the two events instantiates a law. A regularity theory of causation identifies an individual causal relation as an instance of some regularity. Thus, whoever accepts a regularity theory is committed to the view that the identification of a certain individual sequence as causal, involves the possibility of identifying the regularity that makes the individual sequence a causal one. However, it is often perfectly possible to identify a sequence of concrete events as causal without being able to identify any regularity at all. Suppose, for example, that I was in contact with someone suffering from a contagious disease, and I get it myself. Two friends of mine had contact with the same person, but were not infected. Clearly, there is no regularity here. Yet the doctors are pretty sure that I got the disease from the infected person (Anscombe [1971] 1993, 91). Thus, Anscombe calls into question the generally accepted idea that causal relations presuppose underlying laws: Meanwhile in non-experimental philosophy it is clear enough what are the dogmatic slumbers of the day. It is over and over again assumed that any
CONTEMPORARY APPROACHES TO CAUSATION
61
singular causal proposition implies a universal statement running 'Always when this, then that'; often assumed that true singular causal statements are derived from such 'inductively believed' universalities. Examples indeed are recalcitrant, but that does not seem to disturb. Even a philosopher acute enough to be conscious of this, such as Davidson, will say, without offering any reason at all for saying it, that a singular causal statement implies that there is such a true universal proposition - though perhaps we can never have knowledge of it. (Anscombe [1971] 1993, 104)
Though this passage does not explicitly question the idea that causal relations might presuppose underlying probabilistic laws, there is every reason to believe that Anscombe would in fact question it. Indeed, it generally challenges the view that causal relations always fall under some law, whether deterministic or probabilistic. Anscombe's point is that laws are simply not relevant to causation, qua causation: ... causality consists in the derivativeness of an effect from its causes. This is the core, the common feature, of causality in its various kinds. Effects derive from, arise out of, come of, their causes. (Anscombe [1971] 1993, 91-92)
Though analysis in terms of laws of nature may help us to derive knowledge of the effect from knowledge of the cause, it "does not show us the cause as source of the effect:" For example, everyone will grant that physical parenthood is a causal relation. Here the derivation is material, by fission. Now analysis in terms of necessity or universality does not tell us of this derivedness of the effect; rather, it forgets about that. For the necessity will be that of laws of nature; through it we shall be able to derive knowledge of the effect from knowledge of the cause, or vice versa, but that does not show us the cause as source of the effect. Causation, then, is not to be identified with necessitation. (Anscombe [1971] 1993,92)
Thus, analysis in terms of necessitation does not tell anything about the main feature of causality: the derivativeness of an effect from its cause. Though it may inform about some hereditary laws, it does not indicate the parents of this particular child. C.J. Ducasse (1926) was far more explicit in his rejection of analyses of causation in terms of laws. Ducasse's main themes were (a) that the correct definition of the causal relation must be framed in terms of one single case of causal sequence, and that constancy of conjunction is no part of it, and (b) that the causal relation is directly observable (Ducasse [1926] 1991, 133). Ducasse, taking seriously Hume's complaint that his definition of cause "was drawn from circumstances foreign to the cause" (Hume [1748] 1975,
62
CHAPTERFOLR
77), gave an explanation of causation that was independent of anything extraneous to the cause. According to Ducasse, the idea of regularity or recurrence is entirely irrelevant to the meaning of cause. Recurrence is relevant only to the meaning of law, and it only becomes related to causation when there is a law that happens to generalize individual causal facts. General propositions concerning such individual facts may be causal laws, but they are not causal because of their generality. They are laws because they are about classes of resembling facts; and they are causal only because each of these facts is causal in itself. Thus: The causal relation is essentially a relation between concrete individual events; and it is only so far as these events exhibit likeness to others, and can therefore be grouped with them into kinds, that it is possible to pass from individual causal facts to causal laws. (Ducasse [1926] 1991, 137)
Accordingly, Ducasse defined a cause in terms of a relationship between individual events. He thought it was "an admitted fact of language" that the word 'cause' entails that a particular change sufficed to the occurrence of an other change. And he suggested that this 'sufficing' was to be defined as follows: Consider two changes, C and K (which may be either of the same or of different objects), the change C is said to have been sufficient to, i.e. to have caused, the change K, if: 1. The change C occurred during a time and through a space terminating at the instant I at the surface 5. 2. The change K occurred during a time and through a space beginning at the instant I at the surface 5. 3. No change other than C occurred during the time and through the space of C, and no change other than K during the time and through the space of K. More roughly, but in briefer and more easily intuited terms, we may say that the cause of a particular change K was such particular change C as alone occurred in the immediate environment of K immediately before. (Ducasse [1926] 1991, 135) Thus, Ducasse could reinstate the sharp distinction between causes and conditions that Mill denied. For, according to this definition, only events, not states or standing conditions, can be causes. Moreover, the idea of cause requires reference to some definite environment. Thus, rather than involving a dyadic relationship between cause and effect, causation involves a triadic relationship between cause, effect, and environment (Ducasse [1926] 1991, 136).
CONTEMPORARY APPROACHES TO CAUSATION
63
At first sight, Ducasse's definition seems to raise a serious difficulty. For, it demands that the identification of a cause requires a description of the total change observed in the immediate environme it of the effect. Consider the example of a brick that strikes a windowpare, while at the same time a canary sings nearby. We would be inclined to s ly that the impact of the brick was the cause of the breakage of the pane, and lot the impact of the airwaves produced by the canary. Ducasse, however, w )uld be forced to say that the cause of this particular breaking of the windo v was the total change in the contiguous space-time, including the airwaves as well as the brick's impact, How can this be a correct analysis of the mean ng of what we, usually, mean by 'cause'? Ducasse, who provided the exam île, counters by pointing out that our problem is the result of a confusion be tween two different questions (due, in turn, to an ambiguity in the concepì ion of cause): (1) "what did cause, i.e., what did then and there suffice to, t, te occurrence ofthat concrete individual event?, " and (2) "which part ofwha, did suffice would be left if we subtracted from what did suffice such portion of it as were unnecessary to such an effect?" (Ducasse [1926] 1991, 141 Ducasse admitted that it is usually the second question that we have in rr ind when we inquire after the cause of an event, but he hastened to add that, contrary to the first question, the second question does not concern the cau se of one individual concrete event strictly as such: "It is, on the contrary, pn inquiry concerning what is common to it and to the causes of certain other events of the same kind" (Ducasse [1926] 1991, 141). The key to the whole matter is, in his view, that such generalization as is involved in the second question, is not involved in the mere perception of the cause-effect relationship ("we observe [a causal relationship] whenever we perceive that a certain change is the only one to have taken place immediately before, in the immediate environment of another" (Ducasse [1926] 1991, 139)). It is, however, inevitably involved in assigning a name to the cause and the effect. The second question is altogether meaningless with regard to one single event; instead, it is about what was common to the causes of at least two events (Ducasse [ 1926] 1991, 141). Thus, Ducasse's answer to the objection amounts to saying that the cause of the breaking of the window has two different meanings. In the strict sense it means the full concrete event that was sufficient to "all the concrete detail of the breaking of this window." In the derived, but practically more interesting, sense, it means "that which the cause of this breaking of this window has in common with the individual causes of certain other individual events of the same sort" (Ducasse [1926] 1991, 142). There are at least two important problems with Ducasse's view, both put forward by Mackie (1974): first, we can indeed perceive that a certain change C was sufficient, under the circumstances, for K, if we take
64
CHAPTERFOLR
'sufficient under the circumstances' in the weak sense: it was sufficient just because K did occur. However, 'sufficient under the circumstances' can also be taken in the strong, counterfactual sense, that 'if K was not going to occur, C would not occur.' Whether C was also sufficient in this counterfactual sense is not easily decidable. Yet, an answer to this counterfactual question need not be based on any generalization (Mackie 1974, 137). Secondly, and similarly, an answer to Ducasse's second question need not involve any generalization. This second question is equivalent to 'What was necessary under the circumstances for KT According to Mackie, this element of counterfactional conditionality (of the cause being necessary for the effect) is part of our ordinary notion of a singular causal sequence. It does not in any way involve a comparison with other causal sequences (Mackie 1974, 137-140) 21 Now we have seen that the singularist approach is no less problematic than all the previously mentioned approaches, it is time to sum up the main problems that every theory of causation must address. This will be done in the next section.
2.
BASIC ISSUES IN THE CONTEMPORARY APPROACHES TO CAUSATION
As we have seen, the current approaches to causation are afflicted with many problems. In this section the most important of these will be passed in review. First (2.1), I will discuss the five issues mentioned by Sosa and Tooley (1993), which, they maintain, any attempt to formulate an adequate theory of causation must deal with. Next (2.2), I will discuss the problem of the relata of the causal relation. Finally (2.3), I will list a number of other problems that are particularly relevant to the philosophical issues regarding causation.
2.1
Five fundamental requirements
The choice of presenting five major problems as they are stated in Sosa and Tooley's book is based on no other reason than that they seem to mirror rather well the modern attitude regarding the issue of causation. For reasons of clarity, I quote them at length: (1) What relationship is there between causal laws and causal relations? In particular, are causal relations between events logically supervenient upon causal laws together with the totality of non-causal affairs? If not, do
CONTEMPORARY APPROACHES TO CAUSATION
65
causal relations at least presuppose the existence of corresponding, covering laws, or, on the contrary, is a singularist account of causation correct? (2) Are causal states of affairs logically supervenient upon non-causal ones? (3) If not, is an a posteriori reduction of causal states of affairs to noncausal ones possible? Or is a realist approach to causation correct? (4) Is it possible for causal relations to be immediately given, either in perceptual experience, or introspectively? (5) Do causal concepts need to be analyzed, or can they be taken as analytically basic? If they do stand in need of analysis, should the analysis be one that reduces causal states of affairs to non-causal ones, or should it treat causal terms as theoretical, and offer a realist account of those terms? (Sosa andTooley 1993, 5) The following explanation of these five issues is based, for the greater part, on Sosa and Tooley (1993) and Tooley (1987).22 (1) Causal relations and laws One of the crucial issues in any account of causation concerns the relation between, and relative priority of, causal connections and causal laws. In this respect, there are basically three views: the first, and by far most dominant view is that causal laws are more basic than causal connections, and that causal connections between particular events are logically supervenient upon causal laws in combination with the non-causal properties of, and relations between, events. Most contemporary discussions have taken place against the background of this allegedly21 Humean assumption. However, this view has been challenged in two ways: first, by the defenders of a singularist view of causation, who argue that causal relations need not involve causal laws at all; secondly, by those (among whom Dretske 1977, Tooley 1977, 1987, and Armstrong 1983) who, while holding the view that causal relations presuppose (ontologically) real laws of nature, reject the idea of Humean supervenience. (2) Analytical reductionism. The second fundamental issue concerns the question whether all causal relations between events and all causal laws are logically supervenient upon, and therefore reducible to, non-causal facts. The question is: "does the totality of non-causal facts logically determine what causal facts obtain?" (Tooley 1987, 175). The dominant view has been that it does. This view may be characterized as analytical reductionism.
66
CHAPTERFOLR
(3) A posteriori reductionism Analytical reductionism has been challenged in two different ways. First, some philosophers have advanced the idea of a non-analytic or a posteriori reduction of causal to non-causal facts. Thus Bunge (1959), for example, thought that "causation is a mode of energy transfer." Similarly, Fair (1979) identified causation with the transference of energy and/or momentum. One obvious objection against this type of a posteriori reductionism is that the idea of transfer itself presupposes the idea of causation. Armstrong and Heathcote (1991), however, advocate quite a different identity thesis. They claim that singular causal sequences are simply instantiations of strong laws (which are relations between properties). Thus singular causal relations are "instantiations of relations between properties of events or states of affairs" (1991, 71). Their view that causal connections are at bottom nomic connections is the result of an a posteriori inference. They admit that it is conceivable that the world be such that the same sorts of causes do not give rise to the same sort of effects. But, "it is meta-inference from the success of science that the world is not like this" (Armstrong and Heathcote 1991, 69). (4) Causal realism A second alternative to analytical reductionism is the realist approach to causation, which holds that causal relations are not reducible to non-causal properties and relations. There are two sorts of realist approaches: on the one hand there are those who insist that the causal relation is immediately given, either in perceptual experience or by introspection. On the other hand, there is the approach that insists that causation is a theoretical relation that is not directly observable. (5) The analyzability of causal concepts. Most philosophers who hold that causation can be directly experienced, consider it to be analytically basic; they believe that any attempt to analyze the concept of causation is either superfluous or mistaken.24 This view was defended and perhaps best expressed by Richard Taylor. He emphasized that causation is a philosophical category. Thus, "while the concept of causation can perhaps be used to shed light upon other problems or used in the analysis of other relationships, no other concepts can be used to analyze it" (Taylor 1966, 39). A somewhat more moderate version of this view was defended by Michael Scriven, who also rejected the idea that causal concepts can be analyzed in terms of other concepts, but at the same time insisted that they could be explicated by describing the relations between causal terms and non-causal terms:
CONTEMPORARY APPROACHES TO CAUSATION
67
The concept of cause is fundamental to our conception of the world in much the same way as the concept of number: we cannot define it in terms of other notions without conceptual or ostensive circularity. It is probably best to see the notion of cause, like number, as systematically developed from a simple cause which we can exhibit, though not define in non-causal terms. (Scriven [1966] 1993, 56)
However, those who maintain that causation is a theoretical relation, insist that causal concepts stand in need of further analysis, and propose a realist explanation of those terms. The originator of this view is Michael Tooley (1987). He agrees with Hume that causation cannot be directly experienced, and that therefore an analysis of fundamental causal concepts is required. On the other hand, however, he rejects Hume's idea that the analysis must be reductionistic.25
2.2
The relata of the causal relation
Important though these problems may be, they fail to raise a fundamental question, the answer to which may very well determine how the previous questions ought to be explained. This fundamental question regards the relata of a causal relation. Indeed, it is important to distinguish between understanding the relation of causation and understanding what kind of entities causation relates. Suppose that malaria is caused by certain mosquitoes. The first question that is usually asked would be: What is the relation between mosquitoes and malaria? But we must also ask: What kinds of entities must mosquitoes and malaria be so that they can stand in a causal relation? Though both questions are distinct, their answers will undoubtedly affect each other. What we take causation to be will depend on what we take causes and effects to be, and vice versa. In ordinary language, we very often refer to objects or substances as causes. Thus we say that malaria is caused by certain mosquitoes. However, modern discussions tend to treat causality as exclusively or primarily a relation between events, usually understood as changes in the states of things or substances. Contracting malaria, for instance, is a change in one's physical condition and thus an event. It is not caused by a mosquito as such, but by the changes brought about in the body by certain changes in microscopic organisms transmitted by mosquitoes. Together these changes (which in themselves are simple events) compose a complex event. Quite a number of philosophers hold that there is also something like agent causation, which refers to the act of a person in bringing about a change. There is a controversy as to whether agent causation is reducible to event causation. 'Someone's washing the dishes' is an example of agent
68
CHAPTERFOLR
causation. Although this seems to be reducible to event causation - for it may be said that the movement of his arms (which is an event) or his decision to wash (which is also an event) is the cause of the dishes being washed - most philosophers maintain that there cannot be decisions unless there are agents to decide. These agents are conceived as substances. But in ordinary language we also identify causes as facts. This has led some philosophers to suggest that the relata of causation are facts. Whereas event theories give primacy to statements like "The temperature's dropping severely caused the freezing of the pipes" (events), fact theories prefer statements like "The pipes froze because the temperature dropped severely" (facts) (Emmet 1992, 25). I will discuss both approaches successively. 2.2.1
Causation as a relation between events
Events are mostly thought of as anything that happens in a particular place during a particular time (Lombard 1986). According to most accounts, events need not be instantaneous, nor even have a brief duration. The Second World War is as much an event as the movement of my arm. 26 The concept of an event is often associated with the concept of change; an event is a change in an object, for example, the acquisition or loss of a property by the object. However, some philosophers, such as Ducasse (1926), hold that events may include states: "an event can be defined as either a change or an absence of change in the relation of an object to either an intensive or an extensive standard of reference, during a specified time interval" (Ducasse 1926, n.2). For Ducasse, the absence of a change is plainly a state. Thus, the persistent state of an object - say the insistence of malaria - may be causally explained by another persisting state - say the continuing presence of certain organisms in the blood. More recently, the view that events may involve states has been defended by Kim (1973) and Lewis (1986). Though there is some disagreement about whether or not events involve persisting states, there is almost universal agreement about events being somehow related to substances: they are either changes or 'unchanges' of an object. This suggests that an explicit endorsement of an ontology of events is not necessarily incompatible with an ontology of substances. Apparently it is assumed that there can only be events and causal relations between events because there are substances behaving in a certain way. However, a completely different view of events was developed by a number of philosophers. C.D. Broad (1925), A.N. Whitehead (1920; 1929), and Ch. Hartshorne (1970), the latter both philosophers of process, see events neither as changes in substances, nor as presupposing the existence of substances, but as ontologically prior to substances. Every material object,
CONTEMPORARY APPROACHES TO CAUSATION
69
from an electron to a human body, is a persisting pattern in a series of events. This view has far reaching consequences for the theory of causation, and for the conceptions of time and space. Events do not happen in a particular place during a particular time, but they are precisely those relational 'entities' in virtue of which there is time and space. In the contemporary discussions on causation, this challenging view has been almost completely neglected.27 Donald Davidson probably has been the most influential recent representative of the event approach to causation. Like most contemporary philosophers, Davidson takes events to be unrepeatable, dated particulars, which can be truly described in many ways. Thus, according to Davidson, a sharp distinction must be made between causal statements and causal explanations (Davidson 1967). A causal statement, according to Davidson, expresses an extensional relation between individual events. Given that 'a caused b' is true, then it remains true regardless of how we describe a and b. Thus, 'The cause of this match's lighting is that it was struck' is as good a statement of causation as 'The lighting of the match was caused by the match being struck.' Causal explanations, on the other hand, are intensional in the sense that they depend upon how the events are described. Moreover, only causal statements describing the causes so as to provide an answer to a 'why?' question may count as causal explanations. Thus, whereas 'The cause of this match's lighting is that it was struck' is a good statement of causation, it is not a causal explanation. For causal explanations require a reference to laws (or law-like generalizations).28 Laws relate events only insofar as the events are described in a certain way. So an event may instantiate a law under some descriptions but not under others. Thus, an adequate answer to the question, 'Why did the match light?' would be: because it was dry, in adequate oxygen, and it was struck hard enough, and it is a law that under such circumstances matches are lit. Davidson's distinctions may be summarized by saying that most causal statements are elliptical causal explanations, in the same sense that extensional propositions are elliptical intensional propositions. 2.2.2
Causation as a relation between facts
Contrary to the event theory of causation, the fact theory does not make a sharp distinction between causal statements and causal explanations. A causal statement provides an adequate causal explanation whenever it correctly describes the property that is contained in the fact. In order to explain this, we must first see how facts differ from events.
70
CHAPTERFOLR
A fact is, traditionally, the correlate of a true proposition. It is the truthmaker of a true proposition. The fact that John shot Peter contains the objects John and Peter standing in the relation of shooting and being shot. It is the actual obtaining of this state of affairs that makes it true that John shot Peter. Thus a fact is an actual state of affairs. Facts are not particulars. 'John shot Peter' expresses a fact that involves two persons and an action. Whereas these constituents of the fact are particulars, the fact itself is not. While a particular can be given a definite description, the attempt to describe a fact turns it into a statement about an event, substance or person (Emmet 1992, 20). This was clearly expressed by Frank Ramsey: 'That Caesar died' is really an existential proposition, asserting the existence of an event of a certain sort, thus resembling 'Italy has a king' which asserts the existence of a man of a certain sort. The event which is of that sort is called the death of Caesar, and should no more be confused with the fact that Caesar died than the King of Italy should be confused with the fact that Italy has a king. (Ramsey 1927; quoted in Davidson 1980, 135)
Thus, facts do not exist; only the particulars involved in the fact may or may not exist. One of the most influential representatives of the fact approach to causation is D.H. Mellor (1987, 1996). But even he makes a distinction between two basic sorts of causes and effects: facts (understood as actual states of affairs corresponding to true statements), and particulars, which may be things or events. This division is exhaustive: "all singular causes and effects are either facts or particulars" (Mellor 1996, 10). According to Mellor (1996, 11), reports of causation between facts are of the form: (1)'E because C.' Thus, 'Don dies because he falls' represents the cause and the effect as facts. Instances of (1) are called molecular sentences because they contain other sentences 'C ('Don falls') and 'E' ('Don dies'). C and E are linked by the connective 'because,' which is meant to indicate causation (as opposed to the 'because' in sentences such as 'I broke the law because I drove too fast'). On the other hand, reports of causation between particulars are of the form: (2) 'c causes e.' Thus, 'Don's fall causes his death,' represents the cause and the effect as particular events. Instances of (2) are called atomic sentences because they contain no other sentences.
CONTEMPORARY APPROACHES TO CAUSATION
71
Both fact and event theories of causation make a distinction between different sorts of causal claims. The dispute between event and fact theories of the causal relata addresses the question of the primacy of either fact or event. While Davidson takes (2) as more basic, Mellor takes (1) as more basic (Mellor 1987, 201). This question of primacy is indeed very important, and it will be addressed within the discussion of Peirce's concept of causation. In the next subsection, some further issues will be briefly discussed.
2.3
Further issues
Apart from the above-mentioned six basic problems that each theory of causation should take into account, there are a number of other problems that are relevant to the philosophical discussion of causation. Not all of these, however, are equally important, and not all of them will be addressed within the discussion of Peirce's theory of causation. The following problems will be briefly presented: (1) spatio-temporal continuity, (2) probabilistic causal laws, (3) causation and time, (4) the direction of causation, (5) causation and agency, (6) causation and power, (7) teleology, (8) uniformity of causation, (9) universality of causation, (10) reasons and causes, (11) mental and physical causation, and (12) general and singular causal statements. (1) Spatio-temporal continuity. Are causal processes necessarily spatiotemporally continuous? Several philosophers, such as Hume, Ducasse (1926) and Wesley Salmon (1984) hold that all causal processes are characterized by spatio-temporal continuity. This is why they reject the idea of action at a distance (either spatial, temporal, or spatio-temporal) as impossible. But their position is not self-evident. Especially in view of quantum phenomena, which are basically discrete, the issue requires serious consideration. (2) Probabilistic causal laws. If the singularist approach to causation should be abandoned, then the question arises whether causal relations are subject to probabilistic laws or to deterministic laws. (3) Causation and time. Are causal concepts more basic than temporal ones? If so, then it would appear that the analysis of causal concepts precludes the use of temporal concepts. If temporal concepts are thought to be more fundamental, then the question arises whether causes precede their effects, or whether causes can be simultaneous with their effects. (4) Direction of causation. Closely related to the previous point is the question how the direction of causation is to be understood and explained. We have seen that the standard accounts of causation in terms of necessary and sufficient conditions cannot explain what we believe to be the obvious
72
CHAPTERFOLR
fact that, if a state of affairs a causes a state of affairs b, it cannot be that b causes a. The ignition of a match is not the cause of its being rubbed. (5) Causation and agency. Because the ideas of causation and agency are intimately related, the question arises whether one can be analyzed in terms of the other. While most philosophers hold that causation is more basic than agency, some, such as Gaskin (1955) and Von Wright (1971), hold that causation can be analyzed in terms of agency. (6) Causation and power. Some philosophers (for example, Taylor 1966) maintain that causes cannot be properly conceived except as things having the power or efficacy of producing certain changes in other things, and that it is precisely this element of efficacy that distinguishes causes from their effects. According to this view, the conception of power is a metaphysical notion that cannot be analyzed merely in terms of necessary and sufficient conditions. The question is whether the concept of causal power or efficacy is essential or even relevant to the understanding of causal relations. (7) Causation and teleology. Can so-called teleological processes be explained by efficient causation? Though it is usually held that teleological processes can be explained by efficient causation alone, it is far more difficult to say how this should be done. (8) Uniformity of causation. Do similar causes always produce similar effects? The idea of the uniformity of causation, which was common to the majority of past and present philosophers, is rejected by the defenders of the singularist and probabilistic approaches to causation. It will be seen that C.S. Peirce was among the first philosophers to reject this idea of the uniformity of causation. (9) Universality of causation. To say that causation is universal is to say that every event, fact, or substance, has a cause. Throughout the history of philosophy, the universality of causation usually was assumed to be obvious or even self-evident. Today, however, several philosophers consider it quite possible that certain events have no causes at all. Thus, what was once considered self-evident is now controversial. (10) Reasons and causes. Are reasons causes? And if so, in what sense is the concept of cause broader than the concept of reason? (11) Mental and physical causation. Do mental and physical causes belong to different kinds? If so, what is their relationship? (12) General and singular causal statements. We make two sorts of causal claim: singular causal statements, such as 'Charles's smoking caused him getting cancer,' and general causal statements, such as 'smoking causes cancer.' The link between these two kinds of causal statements is subject of controversy.
CONTEMPORARY APPROACHES TO CAUSATION
3.
73
CONCLUSION
The second chapter of this book provides a survey of the contemporary discussions of causation. It was established that the following approaches have been prominent: (1) the necessary and/or sufficient conditions approach, (2) the counterfactual approach, (3) the instrumental or agency approach, (4) the probabilistic approach, and (5) the singularist approach. Probably the most sophisticated version of (1) the necessary and sufficient conditions approach is John Mackie's analysis of causes in terms of so called INUS conditions. Mackie's approach is roughly expressed in the following definition of 'cause:' an event A is the cause of an event B if A is a necessary part of a complex condition C, which, though sufficient, is not necessary for the effect (B). According to (2) the counterfactual approach, a cause is anything that makes a difference that would not have occurred without it. According to (3) the instrumental or agency approach, a cause is an event or state of affairs that we can produce or prevent at will, or otherwise manipulate, in order to produce or prevent a certain other event as an effect. According to (4) the probabilistic approach, an event A may be said to be a cause of an event B, if, given the occurrence of A, the probability of the occurrence of B is higher than the probability of the occurrence of B would have been if A had not occurred. (5) The singularist approach is characterized by the idea that the correct definition of 'cause' must be framed in terms of one single case of causal sequence. Thus, laws are not relevant to causation qua causation. One of the most prominent defenders of this view, C.J. Ducasse, gave the following (rough) definition of a cause: "the cause of a particular change K was such particular change C as alone occurred in the immediate environment of K immediately before" (Ducasse [ 1926] 1991, 135). I discussed a number of issues mentioned by Sosa and Tooley (1993), which, as they maintain, any attempt to formulate a theory of causation must deal with (see 2.1). Some of the most important topics are: (1) the relationship between causal connections and causal laws, (2) the reducibility of causal relations to non-causal facts, and (3) the analyzability of causal concepts. However, though these issues mirror the modern attitude regarding causation fairly well, they fail to raise a fundamental question, the answer to which may be of the utmost importance to the explanation of the issues: the question concerning the status of the causal relata. Are they substances, facts, events, or perhaps something else? Apart from the above-mentioned problems, a number of other problems pertain to the contemporary analyses of causation, such as: (1) spatio-
74
CHAPTERFOLR
temporal continuity, (2) probabilistic causal laws, (3) causation and time, (4) the direction of causation, (5) causation and agency, (6) causation and power, (7) teleology, (8) uniformity of causation, (9) universality of causation, (10) reasons and causes, (11) mental and physical causation, and (12) general and singular causal statements. Most of these problems will be dealt with in the final chapter of this book (chapter 6), in which I present my Peircean approach to causation. In the light of the developments sketched in this chapter and the previous chapter, it will be shown in (the third and) the sixth chapter, in what significant senses the contemporary approaches to causation are inadequate. At the same time, this discussion will provide the background against which Peirce's approach to causation appears as particularly promising. However, before presenting my Peircean approach to causation, I will first discuss some issues that are of paramount importance to the understanding of Peirce's view. Thus, in the next chapter I will discuss Peirce's theory of final causation. It will be shown that on Peirce's account, teleology becomes again a key-concept in the theory of causation.
Chapter 3 PEIRCE ON FINAL CAUSATION
... the non-recognition of final causation ... has been and still is productive of more philosophical error and nonsense than any or every other source of error or nonsense. If there is any goddess of nonsense, this must be her haunt. (C.S. Peirce MS 478; 1903)
1.
INTRODUCTION
The problem of teleology is the question whether all natural processes can be adequately explained in terms of efficient causality. In contemporary philosophy and science there is a strong aversion to explanations by final causation; most approaches consider teleological processes as a special kind of mechanical processes, and try to reduce teleological explanations to explanations based solely on efficient causation. 1 Typical examples of such reductionist strategies are the system theoretical and cybernetic approaches. 2 Furthermore, there are the approaches of certain evolutionary biologists who maintain on the one hand that biology cannot do without teleological language, but on the other hand insist that the explanations of biological processes need to be based on nothing but efficient causation. 3 Although it is currently held that there are no final causes in nature, the proponents of the reductionist strategy do not provide us with a clear theory of causality that shows how teleological processes can be explained by efficient causation alone. At the present moment, no clear theory of teleological processes is available. However, the problem of teleology shows up time and again in all kind of discussions; it is not only prominent in debates on biological evolution and
75
76
CHAPTERFOLR
biological behavior, but also in discussions about other areas of physical science, perhaps most prominently in physical cosmology. 4 The purpose of this chapter is twofold. The first objective is to give an outline of C.S. Peirce's view of final causation. The second objective is to use some of Peirce's insights in order to show that most contemporary debates on teleology are based on dubious presuppositions. More specifically, the second objective is to offer a Peircean critique of the theory of teleology held by Ernst Mayr, one of the most prominent contemporary philosophers of biology.5 My analysis will be confined to Mayr's view, as set forth in his famous 1974 article, "Teleological and Teleonomic, A New Analysis," 6 for two reasons. First, it has had a tremendous influence on later discussions about final causation, even far beyond the scope of biological theory. Secondly, it contains a number of disputable presuppositions shared - and probably borrowed from Mayr - by most contemporary philosophers of biology and of science in general.
2.
PEIRCE'S CONCEPTION OF FINAL CAUSATION
In this section the following themes concerning Peirce's theory of teleology will be discussed: (2.1) final causation, (2.2) the relationship between final causation and efficient causation, (2.3) the difference between mechanical and teleological processes, (2.4) teleology and objective chance, and lastly (2.5), the idea that teleology is creative.
2.1
The nature of final causation
Much of the aversion of contemporary philosophy of science regarding teleology is based on the erroneous view that teleological explanations imply final causes that are concrete future events. Such backward causation is rightly rejected because it is thought to be incompatible with the current views of efficient causation. Indeed, how could future events cause present events at all, if they do not yet exist? Thus, the idea of final causation as backward causation is preposterous. Peirce's critique of this erroneous view of teleology was in this respect in total agreement with Aristotle's view. Moreover, like Aristotle, Peirce endorsed the view that the conception of final causation is explicitly and intentionally anthropomorphic. 7 While warning us not to identify final causes with conscious goals - "a purpose is merely that form of final cause which is most familiar to our experience"
PEIRCE ON FINAL CAUSATION
77
(CP 1.211; 1902) - Peirce used the model of goal-directed experience as point of departure of his analysis. This, of course, raises the problem of anthropomorphism, that is, the problem of the justification of the ascription of human characteristics to nonhuman beings or things. Peirce, however, persisted that anthropomorphism is simply unavoidable. All our ideas in one way or another refer to our human experience (MS 293:1-2; 1906). The same holds for our theoretical concepts and scientific explanations: "every scientific explanation of a natural phenomenon is a hypothesis that there is something in nature to which the human reason is analogous; and that it really is so all the successes of science in its applications to human convenience are witnesses" (CP 1.316; 1903). Ideas of, say, causation, action, force, energy, motion, natural selection, etcetera are all anthropomorphic because they all find their origin in human experience. Consider the idea of causation: "The very conception of causality has its origin in our tendency to seek relations in nature analogous to intellectual relations" (MS 963; c.1893). Or consider the idea of natural selection: it is only by analogy to human acts of selection that this idea makes any sense. Thus, all theoretical ideas in one way or another originate in and refer to human experience. If they did not, they would be meaningless: for, if they are to have any meaning at all, there must be some kind of relationship between them and our daily human experience. Consequently, far from being a problem, anthropomorphism is a sheer necessity. The first thing we notice when considering our own goal-directed behavior is that, contrary to what is usually believed, our goals are neither things nor events. According to Peirce, goals are nothing but 'operative desires,' the objects of which are never concrete, but always general. Something desired is always something of a certain kind. We want a certain kind of apple pie, not one specific individual specimen (CP 1.341; c.1895). Of course, there are all kinds of levels of generality, and one goal may be more specific (less general) than the other: we may want an apple pie made of a special kind of apples and a special kind of dough. But even then, the object remains general. Accordingly, we can see that final causes are general, and not concrete. If final causes are general, they cannot be events either, because events are always individual. Our wishing to eat an apple pie is an event that directs us toward some end. While the wish itself is an event, what it is we wish is of the nature of an idea, or a general type. Consequently, to regard final causes as concrete events is a category mistake (in the Peircean sense of the word).8 Furthermore, our conscious goals do not work from the future toward the present. One may have a purpose and only later be able to realize it, if at all,
78
CHAPTERFOLR
but that does not imply that there is an influence from the future on the present moment. At the present moment, the future is not there yet, nor can it influence anything. Thus, final causes cannot be future events; they are general types, that may be realized in the future. These general types are no actual existences, but general (physical) possibilities for future realization. These preliminary considerations may enable us to better understand Peirce's conception of final causation. Peirce gave the following description of final causation: ... we must understand by final causation that mode of bringing facts about according to which a general description of result is made to come about, quite irrespective of any compulsion for it to come about in this or that particular way; although the means may be adapted to the end. The general result may be brought about at one time in one way, and at another time in another way. Final causation does not determine in what particular way it is to be brought about, but only that the result shall have a general character. (CP 1.211; 1902)
Elsewhere Peirce called a final cause "a tendency to produce some determinate kinds of effect" (MS 682:5-6; c.1895). About such tendencies Peirce made the following observation: By a tendency to an end, I mean that a certain result will be brought about, or approached, and in such a way that if, within limits, its being brought about by one line of mechanical causation be prevented, it will be brought about, or approached, by an independent line of mechanical causation. (NEM IV: 66; 1902)
Thus, the final cause is not an existing thing at all. Indeed, it may be misleading to call it a 'cause,1 for this term - at least in its regular modern sense - suggests that some concrete, existing thing, or event, or fact, has a determinate influence on another thing, or event, or fact. The final cause is not a concrete thing, but it is a type, a mere possibility; it is nothing but an ideal end state that a process tends toward. Peirce called this tendency toward an end state a cause, because he attaches great value to the original meaning of concepts.9 According to its original, Aristotelian, meaning, a cause is some kind of condition without which a thing would not be what it is. Thus Peirce's notion of cause is much more general than the modern notion, which restricts the term to the Aristotelian efficient cause.10 According to Peirce, some reference to a final cause is required in any explanation of a teleological process, because the final cause is a determinant of it.'1
PEIRCE ON FINAL CAUSATION
79
This may be illustrated by the example of thermodynamic processes. These are teleological, because they tend toward an ultimate state of relative stability: Those non-conservative actions which seem to violate the law of energy, and which physics explains away as due to chance action among trillions of molecules, are one and all marked by two characters. The first is that they act in one determinate direction and tend asymptotically toward bringing about an ultimate state of things. If teleological is too strong a word to apply to them, we might invent the word finious, to express their tendency toward a final state. The other character of non-conservative action is that they are irreversible. (CP 7.471 ; c. 1898)
Finious processes are marked by two characters: (a) they tend asymptotically toward an end state, and (b) they are irreversible. Peirce hesitated to call thermodynamic processes teleological in a strict sense, because the end state is only approximated and never completely reached. Although they are teleological in a somewhat weak sense, they are nevertheless teleological, because the convergence cannot be explained by reference to the innumerable separate forces (efficient causes) alone; it can only be explained by statistical laws. Consider for example the diffusion of gasses: whatever the initial state of the gas and the countless small forces which the different molecules exert upon each other may be, these do not suffice to predict the end state of the gas. Such prediction requires knowledge of the relevant statistical laws, which are final causes in the Peircean sense (CP 6.24, 1898; NEM IV: 66, 1902).12 Thus, we may conclude that, according to Peirce, final causes are general types that control the efficient causation; they determine that the effects brought about by efficient causation are of a certain general character. The final cause determines what kind of means is suitable for reaching the general end. Moreover: final causes specify which efficient causes advance the realization of that final cause (CP 2.149; 1902). Whenever someone wants to realize an idea, this idea functions as a principle of selection in the choice of the appropriate means ('lines of mechanical causation') whereby the idea is to be realized. If one wants to build a house, he does not approach his objective by going for a swim, because swimming is not an appropriate means for building a house. The selection of the means may vary, as long as they are appropriate to the building of the house. A house brought about in a different way will no doubt be a (somewhat) different house, but it will still be a house. The fact that the means may be varied and yet may lead to a result of the same general type, can only be explained if we presume that the general type governs the whole process, and this general type is what is meant by 'final cause.'
80
CHAPTERFOLR
We may now define final causes provisionally as follows: final causes are general types that tend to realize themselves by determining processes of mechanical causation. Final causes are not future events, but general (physical) possibilities, which may be realized in the future. The symptoms of final causation are (a) that the end state of a process may be reached in different ways, and (b) that the process is irreversible.
2.2
Final causation and efficient causation
A thorough understanding of Peirce's conception of teleology requires a clear view of how he conceives the relationship between final and efficient causation. Peirce gives the following description of efficient causation: Efficient causation [...] is a compulsion determined by the particular condition of things, and is a compulsion acting to make that situation begin to change in a perfectly determinate way; and what the general character of the result may be in no way concerns the efficient causation. (CP 1.212; 1902)
Thus, efficient causation, as opposed to final causation, is not directed toward an end in any way; it is blind compulsion. To clarify the relationship between final and efficient causation, Peirce gives the example of someone who has the intention to shoot a bird. To hit the bird, he does not shoot directly at it, but a little ahead of it, taking into account the distance that the bird will fly before the bullet reaches it. This activity is end-directed, and thus belongs to final causation. But as soon as the bullet has left the rifle, there is only the stupid, blind efficient causation which in no way is concerned about the results of its activity; the bullet will not follow the bird swooping in another direction. Efficient causation has no regard whatever for results; it simply "obeys orders blindly" (CP 1.212; 1902). While efficient causation, considered apart from its final causational component, is a dyadic (two-term) relation between two concrete individual events or facts, final causation is a triadic (three-term) relation13 between the general final cause, the concrete efficient cause, and its concrete effect. The production of the individual effect (B) by the individual efficient cause (A) is determined, or mediated, by the general final cause (C). The efficient cause functions as a means for the attainment of the end (the motion of the bullet is the means for the shooting of the bird, which is the end). Schematically this may be represented as follows:
PEIRCE ON FINAL CAUSATION
81
C' (final cause)
A
A (cause)
•> C B (effect, realized end) (effect, means)
The relationship between A, B, and C' is triadic. At the time of the causation, the concrete, realized end C does not yet exist. Consequently, the causation of B by A cannot be influenced by C (there is no backward causation). Instead, it is determined by possibility C'. The individual event of the bird's dying does not direct the shooting, but the general purpose of the hunter does. At the moment of the shooting this purpose has not yet been realized; it is a mere idea, that is to say, a physical possibility. Most contemporary philosophers of science think that final causation and efficient causation preclude one another. The example of the shooting of the bird, however, illustrates that final causation and efficient causation are complementary. First, there plainly is no final causation without efficient causation. To shoot a bird it does not suffice to direct one's gun; one also needs the motion of a bullet that blindly obeys the action of the trigger. Conversely, there is no efficient causation without final causation. For, even after the bullet has left the rifle, it conforms to a general law, the causality of which is of the order of final causality:14 "Thus, the relation of a law, as a cause, to the action of force, as its effect, is final causation, not efficient causation" (CP 1.212; 1902). The bullet conforms, among other things, to the law of gravity, which "might without falsity be conceived as a final cause, since it certainly destines things ultimately to approach the center of the earth" (MS 682:7; 1913). Peirce wrapped up this insight regarding the basic complementarity of efficient and final causation in the form of his famous analogy of the court and the sheriff: The court cannot be imagined without a sheriff. Final causality cannot be imagined without efficient causality: but no whit the less on that account are their modes of action polar contraries. The sheriff would still have his fist, even if there were no court: but an efficient cause, detached from a final cause in the form of a law, would not even possess efficiency" (CP 1.213). [...] Final causation without efficient causation is helpless [...]. Efficient causation without final causation, however, is worse than helpless, by far: it is mere chaos; and chaos is not even so much as chaos, without final causation it is blank nothing. (CP 1.220; 1902)
82
CHAPTERFOLR
Hence, final causation and efficient causation are not two different types of causation, each of which would act in different situations. In each act of causation, there is an efficient and a final component. While being caused by previous events, each event is part of a process, which is governed by a final cause. This entails that causal explanations are not more valid than teleological explanations; they just serve a different aim. Whereas causal explanations explain individual events on the basis of previous events and laws, teleological explanations tell us what general principle determines the tendency of a process. To summarize: according to Peirce, every teleological process implies a triadic relationship between an individual efficient cause, a general final cause, and an individual effect. Final causation and efficient causation are complementary inasmuch as each act of causation involves both an efficient component anda final component.
2.3
Teleological and mechanistic processes; Peirce's rejection of dualism
If final causation and efficient causation are complementary, a closer examination of their relationship cannot be avoided. This will be done by focusing upon processes. In this section I will explain the difference between teleological and mechanical processes. It will be seen that the problem regarding the relationship between the mechanical and the teleological aspects of natural processes involves the concepts of continuity and chance. In Peirce's view, mechanical behavior is characterized by the following properties: (1) the end state depends completely upon the situation at the beginning; (2) there is only one way in which the end state can be reached; (3) mechanical behavior is completely reversible in the sense that knowledge of the end state and the relevant laws of nature make it possible to retrodict the initial state (MS 1343:26-7; 1902). But, paradoxically, and contrary to what is sometimes suggested, 'pure' mechanical behavior does not exist in Peirce's universe. For example, in an article about Peirce's conception of final causation, T.L. Short speaks of "completely mechanical" processes, and of entities that "may operate mechanically or by efficient causation [...] [while] others may operate by final causation."15 Expressions such as these are misleading. Peirce rejected the view that there are two kinds of fundamentally different processes for the same reason that he rejected every sort of dualism. The core of his argument is related to his synechism, which is "that tendency of philosophical thought which insists upon the idea of continuity as of prime importance in philosophy and, in particular, upon the necessity of hypotheses involving
PEIRCE ON FINAL CAUSATION
83
true continuity" (CP 6.169; 1902). Peirce tells us how synechism involves the rejection of dualism: ... dualism in its broadest legitimate meaning as the philosophy which performs its analyses with an axe, leaving as the ultimate elements, unrelated chunks of being, this is most hostile to synechism. In particular, the synechist will
not
admit
that
physical
and
psychical
[read:
mechanical
and
teleological 16 ] phenomena are entirely distinct - whether as belonging to different categories of substance, or as entirely separate sides of one shield but will insist that all phenomena
are of one character, though some are more
mental and spontaneous, others more material and regular. Still, all alike present that mixture of freedom and constraint, which allows them to be, nay, makes them to be teleological, or purposive. (CP 7.570; 1892; italics mine)
Clearly, Peirce's synechism, which is a regulative principle of logic and of metaphysics,17 rejects dualistic philosophies because they are all based on inexplicable ultimates. Synechism amounts to the principle that "whatever is supposed to be ultimate is supposed to be inexplicable," and that "inexplicabilities are not to be considered as possible explanations" (CP 6.173; 1902). Thus, synechism is committed to the idea that "all phenomena are of one character." Interestingly, synechism also involves an element of chance. Chance is related to the fact that the laws of mechanics never agree completely with the course of events. What we are inclined to call mechanical processes are processes that approach the laws of mechanics to a high degree, without ever doing so perfectly. Thus, the laws of mechanics are primarily formal laws, and as such they are independent of the matter in which they work. The laws of mechanics are laws that are the same in all possible worlds. As such, they lack an inherent drive toward self-realization. And thus, they fail to represent what is distinctive of real, material laws: "a real and living action in nature."18 But, wherever there is "a real and living action in nature," there is also objective chance. Though objective chance involves novelty, which is not reducible to any law, the action of chance does not yield randomness, for its effects are always subject to certain statistical laws. According to Peirce, this "mixture of freedom and constraint" has "inevitable teleological results" (EP I: 236; 1885). For instance in gambling, the overall end state of a series of games can be predicted on the basis of particular statistical laws, and knowledge of the boundary conditions (the number of players and the amount of money they deposit) and the pertinent statistical laws (which are determined by the rules of the game) enables one to predict the percentage of players who will win a certain amount of the money, the percentage that inevitably will loose, etcetera (EP I: 270-1; c.1887).
84
CHAPTERFOLR
Granted, in "mechanical processes" the degree of deviation from the deterministic laws is minimal, and thus the degree of finality is very low. But even so, the fact remains that in some way, all processes are teleological, even though there is a difference in the degree of finality. Mechanical processes are teleological processes with a negligible degree of finality. Final causation in mechanical processes may be viewed as a degenerate kind of final causation.19 Though I have not explained yet the precise meaning of 'objective chance,' it should be sufficiently clear that, in Peirce's view, a teleological process is not merely an evoluito in its original, Ciceronian sense, according to which evolution consists in the unrolling of a scroll. Instead of being a strictly deterministic process, teleology is decisively creative, albeit in a very specific sense which has yet to be examined.
2.4
Teleology and objective chance
If we are to understand in what sense teleology is connected with creativity, it will be necessary to elucidate Peirce's concept of chance as well as his concept of developmental teleology. In this section, I will first explore the meaning of 'objective chance,' and then the relationship between teleology and chance. In the next section, I will discuss the idea of developmental teleology. Contrary to what is often thought, and contrary to what Peirce himself wrote in his earliest texts on chance, it would appear that 'objective chance' does not refer to a special kind of events that happen only every now and then. Hilary Putnam for example, in his recent introduction to Peirce's Reasoning and the Logic of Things, writes: "such indeterminism as Peirce postulated consists in the very rare occurrence of chance events."20 This formulation smacks of a bifurcation of nature into a realm of chance events and a realm of completely mechanical events, which, as we have seen, Peirce categorically rejected. Putnam seems to view chance events as uncaused events, or in his own words, as "rare spontaneous events." But there is every reason to believe that this cannot possibly have been Peirce's view. Though there is some indication that Peirce believed that there are uncaused events,21 it may be argued that his position is far more subtle in that it affirms that every event is partly uncaused. This view is confirmed by Peirce's explanation of the variety and increasing complexity of the universe:
85
PEIRCE ON FINAL CAUSATION By thus admitting pure spontaneity or life as a character of the universe, acting always and everywhere though restrained within the narrow bounds by law, producing infinitesimal departures from law continually,
and great ones
with infinite infrequency, I account for all the variety and diversity of the universe, in the only sense in which the really sui generis and new can be accounted for. (EP I: 308; 1892; italics mine)
This text was taken from his "Doctrine of Necessity," in which he proposed "to examine the common belief that every single fact in the universe is precisely determined by law" (EP I: 298). Clearly, Peirce not only rejected the view that there are some facts or events that are not precisely determined by law; his position is far more radical: no event is ever completely determined by a law; an aspect of irreducible novelty characterizes each event. Since each event involves an element of objective chance, and since the action of chance has "inevitable teleological results" (which are determined by certain statistical laws), each event has to be part of a causal chain that develops into a definite direction. Hence, objective chance must involve teleology. But, conversely, there can be no teleology without chance. Without chance, all processes would be purely mechanical (deterministic). The variety in the world as well as the possibility of there being different routes leading toward the same end state, can, in Peirce's view, only be explained by chance. Besides, it is chance that accounts for the irreversibility of natural processes, without which there would not be any teleology. Hence, final causation and chance presuppose one another. And since the same holds for final causation and efficient causation, we may conclude that according to Peirce, every event (as part of a natural process) is characterized by an aspect of final causation, efficient causation and chance.22 If this is correct, it entails that all teleological processes are in some sense creative: they presuppose an aspect of irreducible novelty at every stage of the process. This is the reason why final causes cannot specify exact results. By rejecting strict determinism, according to which the definiteness of the world is given throughout all time, Peirce defended the fundamental creativity of the world: reality is always reality in the making.
2.5
Teleology as creative; developmental teleology
Before examining what other forms of creativity may be involved in teleology, we must have a clear idea of what we mean by that term. In an article on cosmic creativity in Peirce, Carl Hausman defines 'creativity' as follows: "a creative act issues in an outcome that is new in kind, which was
86
CHAPTERFOLR
unpredictable, and which has a definite character that is neither reducible to the sum of its elements nor exhaustively traceable to its antecedents" (Hausman 1974, 14). The creativity of a process is seen as equivalent to the irreducibility of the outcome of a process to anything that preceded that outcome. Hartshorne, on the other hand, who was also influenced by Peirce, stresses the positive side of creativity: "it does not mean merely that what happens is not fully specified by the causal conditions and laws; it means that there is more definiteness in reality after a causal situation has produced its effect than before. This increase or growth in richness of determinations is not an absence of something, it is a positive presence" (Hartshorne 1970, 34). Both Hausman's and Hartshorne's approaches may be combined by defining creativity as follows: a process is creative whenever it is incompletely determined by the causal conditions and laws (final causes), and is therefore unpredictable in minute detail. Each stage of the process involves "an addition to the definiteness of reality" (Hartshorne's expression). According to the Peircean interpretation, creativity thus conceived, involves three elements: (1) the events involved within the process, (2) the process itself, and (3) the end state of the process. The first aspect I have considered already in the previous section; each event is creative to the extent that it contains an element of irreducible novelty. Here, I will concentrate on the two other aspects of creativity. (2) It has been explained already in considerable detail, that one of the main characteristics of final causation is that the end state of a process can be reached in different ways. This entails that the ways toward the end state are not determined by the final cause. As appears from Peirce's definitions of final causation (as quoted in the first section), it is precisely this characteristic that distinguishes teleological from mechanical behavior. A limerick by Maurice Evan Hare (1886-?), which was written as a reaction to the idea of determinism, may serve as a funny illustration of this difference: There once was a man who said, 'Damn ! It is borne in upon me I am An Engine that moves In determinate grooves, I'm not even a bus but a tram. The motion of the bus is ideologically constrained by its terminus, without its specific movements being determined by it. A tram, on the other hand, is bound to its tracks. According to determinism, that is to say, according to the view that rejects final causation, man behaves more like a tram than like a bus, because everything he does is completely determined by mechanical
PEIRCE ON FINAL CAUSATION
87
causes. Thus, a teleological process, in general, is creative to the extent that, whenever one way or line of mechanical causation be blocked, it may originate new lines of action.23 (3) So far, it has been assumed that the main purpose remains unchanged in the course of the process. But in fact, it frequently happens that the main purpose develops into another one. The idea of writing a paper on teleology in Peirce, for example, might change into the idea of writing about efficient causation instead. This would be an instance of what Peirce has called "developmental teleology." In Carl Hausman's words this is "the view that there are purposes that may evolve spontaneously" (Hausman 1993, 175). Peirce uses the term "developmental teleology" only once, in his "The Law of Mind" (EP I: 312-33; 1892), which was intended chiefly to explain his synechism. The "law of mind" states, more or less, that all ideas tend to grow to a harmonious whole, and by doing so, lose intensity while gaining generality. Applying this idea to human personality, Peirce concludes that a "personality is some kind of coordination or connection of ideas." One of the most important constituents of a personality is its reference to future ends, which gradually come to be: ... this teleology is more than a mere purposive pursuit of a predetermined end; it is a developmental
teleology. This is personal character. A general idea,
living and conscious now, it is already determinative of acts in the future to an extent which it is not now conscious. This reference to the future is an essential element of personality. Were the ends of a person already explicit, there would be no room for development, for growth, for life; and consequently there would be no personality. The mere carrying out of predetermined purposes is mechanical. (EP I: 331; italics mine)
Thus, growth refers to a process in which the purposes as such may evolve. But, 'developmental teleology' is not restricted to human personality. It is applicable to the idea of teleology in general: learning from the developmental aspect of our own human purposes, we can inductively infer that all final causes in nature are, at least in principle, subject to evolution. Thus, in the process of being realized, the final causes themselves may change. Moreover, final causes differ in degree of generality, and the more general ones can, while being realized, generate less general, or subordinate final causes. In T.L. Short's words this means: "chance thus leads to new ends, but only when the new is a way of fulfilling a more general and already operative end" (Short 1994, 406). The newly developed subordinate final causes must fit within the overall scheme of the more general final cause. For instance, if the more general final cause is the idea of writing a paper on some aspect of Peirce's cosmology, then the change in idea from writing on Peirce's conception of efficient causation to the idea of writing on
88
CHAPTERFOLR
his concept of teleology fits within the overall scheme of the more general final cause. Moreover, the new subordinate purposes, which arise by chance,24 are selected by the more general final cause. This has far reaching consequences, as T.L. Short again so well expresses: "Final causation thus results, not in the dead uniformity of a single plan, but in the unpredictable heterogeneity of enterprises, personalities, and species that fill our world ..." (Short 1994, 406).
3.
A PEIRCEAN CRITIQUE OF ERNST MAYR'S THEORY OF TELEOLOGY25
The problem of teleology is, no doubt, most prominent in philosophical discussions concerning biological evolution and biological behavior. The standard view nowadays is that, although teleologica! language may be indispensable to biology, the explanations of biological processes must be given in terms of efficient causation. Any reference to final causation is rejected. One of the major proponents of this view is Ernst Mayr. In this section I will use some of Peirce's insights to show that Mayr's rejection of final causation is based on a number of dubious presuppositions. Furthermore, I will use Peirce's theory of final causation to argue that Mayr's own solution to the problem of teleology is insufficient. First, I will consider Mayr's rejection of a teleological interpretation of biological evolution, and show that it is based on at least three dubious presuppositions (see 3.1). Next, I will discuss two core ideas of Mayr's own solution: I will examine Mayr's dualism, that is, his distinction between seemingly and genuinely goal-directed processes, and show that it is based on an invalid argument (see 3.2). After that, I will examine Mayr's idea of a 'program as responsible for genuinely goal-directed processes,' and show that, contrary to Mayr's own opinion, it agrees perfectly well with the (Peircean) idea of final causation (see 3.3).
3.1
The goal of evolution
According to Peirce, biological evolution is a perfect example of a finious process because (a) there is a definite tendency toward a state of relative stability, (b) the general end state is independent of whatever the various lines of mechanical causation that lead to it may be, and (c) the process is irreversible. That the best-adapted species (or individuals)26 will survive may easily be predicted, independently of what species are considered, and regardless of the specific mutations that may take place. Thus, explanations
89
PEIRCE ON FINAL CAUSATION
of this tendency do not require any reference to specific, concrete efficient causes, but only to the general principle of "survival of the fittest," which functions as a principle of selection.27 More or less the same point is made by the contemporary evolutionary biologist Ayala: "... the overall process of evolution cannot be said to be teleological in the sense of being directed towards the production of specified DNA codes of information, i.e. organisms. But it is my contention that it can be said to be teleological in the sense of being directed towards the production of DNA codes of information which improve the reproductive fitness of a population in the environments were it lives" (Ayala 1970, 11). Mayr considers Ayala's explanation to be a serious misinterpretation of the term 'teleological:' Natural
selection
is never
goal oriented.
It is misleading
and
quite
inadmissible to designate such broadly generalized concepts as survival or reproductive success as definite and specified goals. (Mayr 1974, 96)
Mayr's rejection of a teleological interpretation of evolution is based on three dubious assumptions. First, as the quotation shows, Mayr has an unsound conception of 'goal'. As I have explained in the first part of this chapter, goals are always general. Though there may be a difference in the degree of generality - some goals are less general than others - there is always an element of generality involved. Survival and reproductive success are, of course, very general concepts, as Ayala boldly admits. Mayr charges that Ayala's referring to "completely generalized processes, rather than to specific goals" has ludicrous consequences. For instance, it would be "the goal of every evolutionary line to become extinct because this is what has happened to 99.9% of all evolutionary lines that have ever existed" (Mayr 1974, 97). Clearly, Mayr here fails to see that teleological explanations - as T.L. Short has made clear in his brilliant, Peirce-inspired essay, "Teleology in Nature" 28 - must explain a general tendency. The extinction of an evolutionary line does not, and cannot, as a final cause should, explain the process that generated that line. Consequently, such extinctions cannot possibly be final causes. Secondly, Mayr identifies teleology with a completely determined, straightforward development toward a specific end, from which every form of creativity is expelled: Natural selection rewards past events, that is the production of successful recombination of genes, but it does not plan for the future. This is, precisely, what gives evolution by natural selection its flexibility. With the environment
90
CHAPTERFOLR changing incessantly, natural selection - in contradiction
to orthogenesis
-
never commits itself to a future goal. (Mayr 1974, 96; italics mine)
According to Mayr, natural selection cannot be teleological because it is highly flexible; contrary to orthogenesis (the theory according to which evolution is nothing but the completely determined development of what was present in the beginning), evolution is not teleological because it does not involve a straightforward progression to a completely predetermined end. Mayr rightly stresses that evolution is a flexible process. There are, however, two problems with his explanation. First, it is not clear that the flexibility Mayr refers to, involves real creativity. Does objective chance play a crucial role in the process of evolution or is it a completely determined process? Only in the former case would biological evolution be unpredictable in principle; thus, even in the fictitious case that we would have knowledge of all the relevant causal circumstances, we would still not be able to predict which species will evolve. The process is not determined in advance. Mayr's insistence on "strictly causal and mechanistic" explanations, however, leads one to suspect that he rejects the idea of objective chance. The very concept of flexibility involves a commitment to a goal. For a strict determinist, the concept of flexibility has no meaning: que sera, sera. Thus, Mayr is faced with the following dilemma: if he acknowledges objective chance, he cannot insist on "strictly causal and mechanistic" explanations. But if he rejects objective chance, he cannot possibly claim that evolution is flexible. The other point is, that there are three reasons why Mayr's identification of teleology with orthogenesis is mistaken: first, teleology does not imply orthogenesis at all, because one of its main characteristics is that it presupposes chance. Next, in teleological processes, the end state can be reached in various ways. Even if one does not agree with this Peircean criterion (but can one imagine other, better criteria?), he has to admit at least that there are processes in which the end state can be reached in different ways. This, however, is exactly what is precluded by orthogenesis.29 Finally, teleology does not presuppose a completely fixed and determinate final end state. As Peirce has argued forcefully (see 2.4), final causes, while being realized, may become modified and developed. Teleology is basically developmental. Hence, contrary to orthogenesis, teleology is creative (in more than one way). Thirdly, Mayr associates final causation with backward causation. In the quotation above, Mayr speaks of 'future goals,' and elsewhere in his article he writes: "The assumption that future goals were the cause of current events seemed in complete conflict with any concept of causality" (Mayr 1974, 93).
91
PEIRCE ON FINAL CAUSATION
Mayr is certainly right in rejecting the idea that future events could have an influence on present events. But he is mistaken in that he concludes that this entails that there cannot be any final causation. He too mistakenly considers final causes as belonging to the same genus as efficient causes. In the first part of this chapter I have used one of Peirce's insights to show that this is a category mistake: final causes do not belong to the category of concrete events, but to the category of general possibilities. Accordingly, final causation has nothing whatever in common with backward causation.
3.2
Mayr's dualism
According to Mayr, nature is split up into two realms of "an entirely different nature:" the realm of genuinely goal-directed processes and the realm of seemingly goal-directed processes. But even the genuinely goaldirected processes can entirely be explained by efficient causation. Two questions are relevant here: (1) is it true that nature is split up into these two realms?, and (2) can all seemingly and all genuinely goal-directed processes be explained by efficient causation alone? In this section, the first question is explored; the second question is addressed partly in the present section, and partly in the next. According to Mayr, the seemingly goal-directed processes which he calls teleomatic processes occur only in inanimate nature. They ... are 'end-directed' only in a passive, automatic way, regulated by external forces or conditions. Since the end state of such inanimate objects is automatically achieved, such changes might be designated as teleomatic.
All
teleomatic processes come to an end when the potential is used up (as in the cooling of a heated piece of iron) or when the process is stopped by encountering an external impediment (as a falling stone hitting the ground). Teleomatic processes simply follow natural laws, i.e. lead to a result consequential to concomitant physical forces, and the reaching of their end state is not controlled by a built-in program. The law of gravity and the second law of thermodynamics are among the natural laws which most frequently govern teleomatic processes. (Mayr 1974, 98)
Genuinely end-directed or teleonomic processes differ in (at least) three respects from seemingly end-directed or teleomatic processes. Teleonomic processes are marked by the following properties: (1) The end state is not reached automatically, and (2) A built-in program controls the process toward the end state, (3) The end state is not determined by (a) external impediments or (b) by the condition that some potential is used up.
92
CHAPTERFOLR
Furthermore, teleonomic processes occur only in living nature. Mayr's so called fundamental distinction between teleonomic and teleomatic processes is based on four concepts: (1) automatism, (2) program, (3a) external impediments, and (3b) potential exhaustion. I will consider these one by one. ( 1 ) The end state of teleomatic processes is reached automatically, that is to say, it is the straightforward effect of natural laws; in contrast, the end state of teleonomic processes is completely determined by a built-in program (e.g. DNA, acquired habits, computer programs30). But obviously, programdirected behavior is just as automatic as 'law-of-nature-directed' processes: "Teleonomic processes are strictly causal and mechanistic" (Mayr 1974, 112; italics mine). Consequently, teleonomic processes are just as mechanical as teleomatic processes. (2) Mayr defines 'program' provisionally as: "coded or prearranged information that controls a process (or behavior) leading it toward a given end" (1974, 102). If I leave out "coded or prearranged information" - the meaning of which is not at all obvious - this conception of program agrees perfectly well with the general concept of a law of nature. For (according to the realistic interpretation), laws of nature induce things to behave in a definite way. I have considered already the law of gravity and the second law of thermodynamics; they are good examples of general principles that "control a process leading it toward a given end." Thus, nothing stands in my way of considering laws of nature also as built-in programs. Consequently, the validity of a real distinction between teleonomic and teleomatic processes on the basis that only the latter are directed by built-in programs, is debatable at best. (3) Finally, according to Mayr, teleomatic processes, as opposed to teleonomic processes, come to an end (a) as the result of some external impediment (as a falling stone hitting the ground), or (b) because some potential is used up (as in the cooling of a piece of iron). But these criteria are both very vague. The first one is certainly ambiguous. For, what Mayr calls an external impediment might be called an internal impediment as well, depending on how the system is defined; and perhaps it even ought to be called an internal impediment. Thermodynamics usually studies isolated systems, and there are of course no isolated systems without boundary conditions. Thus, the boundary conditions are part of the system, and what Mayr calls external impediments may just as well be called internal impediments. The second criterion is, taken by itself, too wide: an athlete who runs a race as fast as he can uses up his potential. According to Mayr's terminology, this would be an example of a teleomatic process, while it is clearly teleonomic.
PEIRCE ON FINAL CAUSATION
93
Hence, the three criteria used by Mayr fail to support the distinction he makes between teleonomic and teleomatic processes. But more importantly, as we have seen in the first part of this chapter, neither teleonomic nor teleomatic processes can be explained on the basis of mechanical causation alone; even our explanations of thermodynamic processes - according to Mayr a plain instance of teleomatics - need to refer to some general type which functions as a final cause. Therefore, there is no reason whatever to regard his distinction as fundamental.
3.3
Mayr's idea of a program as 'causally responsible' for teleological processes
The key word in Mayr's definition of 'teleonomic' is the word 'program.' The process tending to the end state is controlled by a built-in program, which "is causally responsible for the teleonomic nature of a goal-directed process" (1974, 99; italics mine). Apparently, there is no final causation involved in teleonomic processes. Mayr gives two reasons why a program is responsible for teleonomic behavior, and without thereby involving final causation: "a program is [1] something material, and [2] it exists prior to the initiation of a teleonomic process. Hence, it is consistent with a causal explanation" (1974, 102). The argument involves three premises, all of which are dubious: (a) information (a program) is something material, (b) causes are always material, (c) causes always precede their effects. What it means to say that a program is something material is by no means clear. The concept of matter is a philosophical concept with a long history; if one wishes to use it as a key concept, he is obliged to clarify its meaning. But, the concept of matter involves at least something spatio-temporal, i.e. something that can be indicated in space and time. One should know therefore in what sense a 'program' refers to something spatio-temporal. Granted, the information of the program needs a substrate (a 'material' carrier), but this does not imply that it is identical with its substrate. Though the substrate can be indicated in space and time, information itself cannot, because it is general and not concrete. Indeed, the same information content may be inscribed in different 'material' substrates, and the same substrate may carry different information. In Aristotelian terms, one might say that the information refers to form, not to the matter. It is the form that matters, not the matter that (in) forms. The two other presuppositions of Mayr were that (b) causes need to be material, and (c) that they always precede their effects. Are these true? The least one can say is that the debate is not closed; there may be mental causes as well as physical ones, and it would appear that many causes are
CHAPTERFOLR
94
contemporaneous with their effects. Thus, these two presuppositions are by no means obvious. But apart from these considerations, there are other difficulties with Mayr's concept of 'program.' He stipulates that: ... it is in the nature of a teleonomic program that it does not induce a simple unfolding of some completely preformed Gestalt, but that it always controls a more or less complex process which must allow for internal and external disturbances. (Mayr 1974, 99)
Mayr might be very surprised to discover that this description of the functioning of a program agrees perfectly well with Peirce's concept of final causation; the process toward the end state is not "completely preformed" but "flexible," and the end state itself is independent of the various different efficient causes. Final causes - so I have written in section 2 - "are general types that tend to realize themselves by determining processes of mechanical causation." If 'causally responsible' should be interpreted as 'due to final causation,' there would not be any problem. But obviously, this is not what Mayr means, for he holds that "teleonomic explanations are strictly causal and mechanistic" (Mayr 1974, 114). Mayr fails to explain what he means by 'causal.' I can only interpret him as follows: a program is an efficient cause, which in combination with other efficient causes (the internal and external disturbances) completely determines the end state of a process or behavior. If so, then Mayr owes us an explanation of how, under different circumstances, and thus, given different sets of efficient causes, the same program may lead processes to the same general end state. More importantly: how can programs be considered as efficient causes at all? Or put more concisely: is there a theory of efficient causation that meets with the idea of a program? That seems unlikely, for efficient causes are singular events or facts, while programs are not. Programs, however, are, as Peirce would say, of the nature of a habit.31 And habits are always general, and this in a double sense: (a) contrary to concrete efficient causes, habits not only induce one or more lines of mechanical causation at one singular moment, but they start new lines whenever possible. Besides, (b) a habit never completely determines all characters of the end state toward which it leads; for example, the habit of smoking does not (completely) determine how and how often one smokes. Accordingly, I may safely conclude that Mayr's conception of teleonomics is, to borrow his own expression (which he uses to refute teleology), "in complete conflict with any concept of causality" (Mayr 1974, 93). Mayr's critique of teleology is partly based on the mistaken assumption that the conception of (efficient) causality is unproblematic. But this
PEIRCE ON FINAL CAUSATION
95
assumption is a myth, which is, unfortunately, widespread among philosophers and scientists alike. An adequate account of the problem of teleology requires a thorough analysis of the concept of efficient causation. Nothing is gained by unmasking the supposed myth of teleology by banking on yet another myth.
4.
CONCLUSION
In the first part of this chapter an outline was given of Peirce's theory of final causation. In the second part, some of Peirce's insights were used to show that Ernst Mayr's theory of teleology is flawed. I chose to discuss Mayr's view, partly because he has had an enormous influence on contemporary discussions, and partly because it illustrates a number of dubious presuppositions that are to be found in most of the current discussions. According to Peirce, final causes are (a) general types that tend to realize themselves by determining processes of efficient causation. They are (b) not future events, but general (physical) possibilities. The symptoms of final causation are: (I) the end state of a process can be reached in different ways, and (ii) the process is irreversible. Peirce rejected the idea of a bifurcation of nature into two kinds of fundamentally different substances or processes. Hence, he rejects the view that final and efficient causation are two basically different types of causation. On the contrary, they are complementary inasmuch as there is in each act of causation an efficient and a final component. He also rejects the idea that there exists a class of mechanical processes next to teleological processes. All processes are teleological; mechanical processes are simply teleological processes with a low grade of finality. Moreover, teleology presupposes objective chance, and thus, there is an aspect of irreducible novelty at every stage of a process. But teleology also involves novelty in the choice of the different routes that lead to a specified general end state, and in the possible evolution of the final causes themselves ("developmental teleology"). It was shown that Mayr's dismissal of a teleological interpretation of biological evolution is based on the rejection of three properties that he mistakenly attributes to the concept of final causation. Mayr imagines final causes to involve: (a) Individual events, (b) An influence of the future upon the present (backward causation), (c) A straightforward, completely determined course toward the end state (which entails the absence of any form of novelty).
96
CHAPTERFOLR
If this analysis is correct, Mayr's rejection of Final causation is based on a fatal misunderstanding of the nature of teleological processes. Moreover, it was shown that Mayr's distinction between genuinely end-directed or teleonomic processes and seemingly end-directed or teleomatic processes is essentially without foundation. Furthermore, it was shown that, contrary to Mayr's opinion, both teleonomic and teleomatic processes require an explanation by final causation. Finally, I have examined the key concept in Mayr's idea of teleonomics, which is the idea of a program. I have shown that, contrary to Mayr's opinion, programs are good examples of final causes. Mayr's mistakes are largely due to two false premises that are characteristic of nearly all contemporary debates on teleology. The first of these is that he regards final causes as belonging to the same genus as efficient causes. A hundred years ago, Peirce has shown that this is a category mistake: contrary to efficient causes, final causes are not individual events or facts, but general principles. The second false premise is the assumption that we have a clear concept of efficient causation. Harmful though both premises may be, the second of these is by far the most ferocious. The failure to see that causation is a philosophical problem instead of a clear-cut, self-evident idea proves to be fatal to Mayr's analysis. In the next chapter, I will pursue my inquiry into the nature of causation by studying Peirce's conception of natural class. Indeed, Peirce's theory of causation requires an explanation of his conception of natural class, for Peirce held that the condition for causes and effects to be mediated by final causes is that they belong to a natural class.
Chapter 4 FINAL CAUSES AND NATURAL CLASSES
Every classification has reference to a tendency toward an end. If this tendency is the tendency which has determined the class characters of the objects, it is a natural classification. (C.S. Peirce, NEM IV: 65, 1902)
Though Peirce's theory of natural classes is often mentioned in contemporary philosophy of science and metaphysics (Ian Hacking for example gives Peirce a prominent place in the tradition of natural kinds'), it has not as yet been studied thoroughly. Accordingly, the presentation of Peirce's theory is often only partially correct, and sometimes even misleading. Perhaps the main reason for the absence of a thorough study is that Peirce's theory of natural classes is intimately related to his theory of final causation, 2 - a concept which in contemporary philosophy is avoided for being a mystifying idea which neither agrees with the methods nor with the results of modern science. In the previous chapter I have tried to show that this is a biased view, due to a number of false presuppositions that were clearly recognized by Peirce a century ago. In this chapter it will be shown that Peirce's theory of natural classes is intimately linked to his conception of final causation. In the previous chapter it was shown that Peirce held the view that in each act of causation there is an efficient and a final component: final causes are general types that tend to realize themselves by (teleologically) determining processes of efficient causation. They are not future events, but general, physical possibilities. The efficient aspect of causation is that each event or fact is caused by a previous event or fact (the efficient cause); the teleological aspect is that each event or fact is part of a chain of events with a definite tendency. The tendency is determined by the final cause of the process. This entails that each act of causation is mediated by a final cause. 97
98
CHAPTER FOL R
Indeed, Peirce held that the condition for causes and effects to be mediated by final causes is that they belong to a natural class. Thus, Peirce's theory of causation requires an elucidation of his concept of natural class. Conversely, since Peirce defined natural classes in terms of final causation, his concept of natural class must be considered within the perspective of his theory of final causation. Thus, my primary reason for discussing Peirce's conception of natural class is that an adequate explanation of Peirce's conception of causation requires such discussion. There are, however, at least three subsidiary reasons for considering Peirce's theory of natural classes: first, it has (as far as I know) never before been studied within its proper context of teleological causation. Secondly, Peirce's theory of natural kinds is thought by many to be relevant to contemporary discussions. Thirdly, if a Peircean critique of most of the contemporary views of teleology has any merit, and if it is true that there is a close link between the concepts of causation and natural class, then this critique may have serious consequences for contemporary debates on natural classes. An additional reason for discussing Peirce's theory of natural classes is that it provides a good ground for discussing his alleged pluralism, which is a hot topic in contemporary Peirce studies. Accordingly, the first objective of this chapter is to reconstruct Peirce's theory of natural classes. The second objective is to examine Peirce's view of the relationship between natural classes and causation, and to see whether it contains insights that might be relevant to contemporary debates on natural kinds and causation. Thus, the first section will provide a general sketch of some contemporary approaches to the problem of natural kinds. In the second section, some contemporary interpretations of Peircean natural kinds will be considered. These first two sections will raise a number of questions, which will serve as a point of departure for the subsequent sections. The third section will deal with Peirce's discussion of John Stuart Mill's theory of natural kinds. Next, in the fourth section, Peirce's distinction between kinds and classes will be examined. The fifth section will concern the close relationship between the concepts of final causation and natural kind. In the sixth section, the question of demarcation criteria will be dealt with. Whereas the seventh section will provide a provisional summary, the eighth section will deal with Peirce's reasons for believing in natural classes. The ninth section will give an account of Peirce's main examples of natural classes: respectively, (a) examples from the domain of human action, (b) the chemical elements, and (c) the biological species. In the tenth section, I will consider Peirce's (assumed) pluralism. In the eleventh section, the results of this chapter will be summarized, and their relevancy for contemporary philosophical discussions will be considered.
FINAL CAUSES AND NATURAL CLASSES
í.
99
NATURAL KINDS AND CAUSATION IN CONTEMPORARY PHILOSOPHY
That there is a close relationship between causation and natural kinds is not as strange as it may seem at first. On the contrary, in the contemporary philosophy of science and metaphysics it is widely believed that the concepts of causation, explanation, natural law, and natural kind are interrelated. For example, in an influential paper on natural kinds, W.V. Quine (1969, 132) emphasizes that the concept of causation entails the concept of natural kind: "To say that one event caused another is to say that the two events are of kinds between which there is invariable succession." And, for example, D.M. Johnson (1990, 63) defines a natural kind as "a spatiotemporally unrestricted or repeatable category ineliminatively presupposed by at least one true and explanatory law of nature." In their glossary to their anthology "The Philosophy of Science," the editors Richard Boyd, Philip Gasper, and J.D. Trout give a very general definition of natural kind, which, it may be assumed, is supposed to cover most current theories. A natural kind is: A type of property, process, state, event, or object studied by science, mentioned in scientific laws, and assumed to be a causal feature of the world. The primary instances of natural kinds are objects of scientific taxonomy, such as electrons in physics, zinc in chemistry, and species in biology. Natural kinds are contrasted with phenomena that are assigned no such systematic, organizing role, such as an event's occurring after I drop this pen, or an object's being located 34 miles west of the Liberty Bell. (1991, 778-9; italics mine)
According to this view, natural kinds, as opposed to other phenomena, play a systematic role in our explanations of the world; they are supposed to be something like the world's causal joints. The same idea is defended by J. Levinson (1991, 65), according to whom the objects belonging to a natural kind "occupy the same causal role in nature." As it is not al all obvious, however, what it involves to be 'a causal feature of the world' or to 'occupy the same causal role in nature,' it would seem clearly that the concept of natural kind presupposes an elucidation of the concept of causation. Bigelow et al (1992, 373) stress that natural kinds are always associated with essential properties. "If something is of a natural kind, then there will be properties which this thing must have to be a thing of that kind, and which it could not cease to have without ceasing to be a thing of that kind." The idea that things belong to natural kinds seems to involve a commitment to essentialism: what makes a thing a member of a particular natural kind is that it possesses a certain essential property (or a cluster of essential
100
CHAPTER FOL R
properties), a property both necessary and sufficient for a thing to belong to that kind. The essential property is supposed to provide an objective feature that determines to what kind a thing belongs, independently of any context of inquiry. It is also supposed to play an important explanatory role in regard to other properties and relations. The main examples of natural classes used by philosophers are the chemical elements and biological species. Especially the chemical elements are often taken to be paradigm cases of natural kinds. Consider Saul Kripke's famous example of gold:3 the fact that gold is defined by its atomic number, entails that a thing is made of gold precisely when it is composed of atoms that have atomic number 79. It is because "the essence of a natural kind must be necessary, explanatory, and purely qualitative" (Sober 1995, 345; italics mine), that the atomic number 79 is said to provide the essence of the natural kind of gold. Whereas it is an accident that some lump of gold has a specific shape, it is supposed to be a necessary truth that golden things have atomic number 79. Moreover, the atomic number explains many properties of golden things. Finally, specifying the essence of gold does not involve a reference to shape, place or time; the atomic number supplies this general qualitative specification (Sober 1995, 345). The case of biological species is more complicated, for it is by no means clear that biological species have essences. The favored view is that species are individuals. According to this view species are said to be populations that have organisms as parts rather than as members (Hull 1978). Organisms belong to the same species, not by virtue of their similarity, but because of their genealogical relatedness. Despite their common descent, they do not thereby form a natural kind (Sober 1995, 346). Apart from a certain agreement regarding chemical elements, philosophers tend to heartily disagree when it comes to give clear examples of natural kinds. Thus Van Brakel (1992, 243-4) lists a number of different interpretations: while Putnam includes multiple sclerosis, gold, horses, and electricity, Kripke and Quine mention colors, Hacking suggests social kinds, and Churchland does not hesitate to include mass, length, duration, charge, color, energy and momentum. According to Van Brakel, this disagreement is not only due to different opinions regarding the distinction between natural kinds and artificial kinds, but to more fundamentally different views regarding induction, prototypes, universals, scientific realism, meaning and reference. 4 Van Brakel might also have added the problems of causation, explanation and natural law. Apparently there are a great many conflicting theories about natural kinds. To obtain some clarity in the problem, a number of fundamental questions must be answered. Some of the most important are: (a) What are natural kinds?, (b) What argument is there for believing in their existence?.
FINAL CAUSES AND NATURAL CLASSES
101
(c) What use has science for the notion of natural kind?, (d) What are the demarcation criteria by virtue of which one can decide to what natural kind an object belongs?, (e) Is there a uniquely correct grouping of objects into natural kinds, or are there countless legitimate, objectively grounded ways of classifying the objects of the world?, ( f ) What is the precise relationship between causation, natural laws, and natural kinds? But before pursuing Peirce's views on these questions, I will first survey some recent interpretations of Peirce's theory of natural kinds. It will appear that the current interpretations are rather meager, and in many respects contradictory. Their main value for my discussion consists in the questions they raise.
2.
SOME CONTEMPORARY INTERPRETATIONS OF PEIRCEAN NATURAL KINDS
To obtain a first impression of Peirce's view of natural kinds, and of the problems it involves, I shall briefly examine recent interpretations of, successively, Susan Haack, Christopher Hookway, and Sandra Rosenthal.
2.1
Susan Haack's interpretation
A contemporary philosopher who stresses the importance Peirce's concept of natural kind, is Susan Haack. According to Haack in her "Extreme Scholastic Realism: Its Relevance to Philosophy of Science Today" (1992), Peirce's 'extreme scholastic realism' entails that there are real laws of nature and real kinds, which are not dependent upon our characterization of the world. Haack defends Peirce's 'scholastic realism' as a necessary presupposition of the whole scientific enterprise, but she realizes that "a full and detailed defense of this claim would require a better understanding of what makes a class natural" (Haack 1992, 42). As Haack understands it, it is Peirce's view that without real laws and real kinds, no genuine science is possible. Without them there can be neither explanation, nor can there be prediction or induction (Haack 1992, 28; 1993a, 134). According to Haack, Peirce's realism entails that the particular facts and events we observe, are the expression of an underlying pattern of natural kinds and laws. While particular facts and events are concrete, the underlying pattern consists of so called generals. This pattern is real inasmuch as it is independent of how any individual inquirer thinks about it. As science - which is by its very nature co-operative - proceeds, this real pattern will eventually emerge: "Which generals are real is a matter which
102
CHAPTER FOL R
would only be finally settled in a hypothetical completed science" (Haack 1992, 29; also 1993b, 353). If science were to continue long enough, it would yield true classifications and true laws of nature, that is to say, classifications and laws "from which the local and idiosyncratic, the unreal, had been eliminated" (Haack 1992, 32). Haack offers a number of different descriptions of Peircean natural kinds that might help us to obtain a preliminary understanding. Natural kinds are: (1) "clusters of similarities holding together in a lawful manner," (2) generals "that would figure in the laws," and (3) "the kinds of things in the world which really do behave in a law-like way."5 As examples of Peircean natural kinds Haack mentions horses, men and stones. Since there are no laws of nature that are specific to stones,6 the suggestion is that Peirce has a very broad idea of law. Haack observes, almost casually, that genuine Peircean laws are basically habits (1992, 28). Thus, stones as well as fundamental particles are natural kinds because they are the kind of things in the world that behave in a habit-like manner. Moreover, the habitual aspect of natural kinds is illustrated by the fact that, for instance, stones that do not actually fall are nevertheless capable of falling. For laws (habits) sustain subjunctive conditionals; they tell "not just what does happen when ..., but what would happen if ..." (Haack 1992, 28). We expect that if someone would for instance drop the stone he has in his hand, it would fall to the ground.
2.2
Christopher Hookway's interpretation
Hookway's interpretation differs sharply from Haack's position. He interprets Peirce as holding that there are infinitely many, but equally legitimate ways of dividing the world into kinds: It is as if cognitive activity inevitably reads into reality a sort of articulated gesture, a system of classifications, which enables us to bring our experience under control, but which does not correspond to anything real. Indeed Peirce's picture of reality is close to this. He sees it as a continuous spread of reaction and feeling; where we draw the boundaries in thinking of it as containing individual objects, or how we classify the continuous range of possibilities which underlie general laws and characters is up to us. Generality is real, but dividing things into classes reflects our interests and conventional
decisions.
(Hookway 1985, 251; italics mine)
According to this description, Peirce is presented as an "anti-realist" who rejects the possibility of giving an objectively correct system of classifications, which reflects the nature of natural classes (Hookway 1985, 250). While Haack defends the view that our personal considerations do not
FINAL CAUSES AND NATURAL CLASSES
103
interfere in the ultimate (Peircean) scientific classification and that there is but one legitimate way of dividing the world into (Peircean) kinds, Hookway insists that, according to Peirce, classifications are entirely subjective and that there are countless correct ways of classifying the objects of the world.
2.3
Sandra Rosenthal's interpretation
Sandra Rosenthal gives a third interpretation: dividing things into classes partly reflects our interests and conventional decisions, and partly the way things really are. Consequently, there are many, but not infinitely many, equally legitimate ways of dividing the world into natural classes: Knowledge is abstractive and selective. A world, though concrete, is nonetheless selective in the sense that a world, as the concrete content denoted by a system of meanings, is a way in which the concreteness of reality can be delineated or "fixed." A system, once chosen, limits the alternatives possible within it, but alternative systems may be possible. (Rosenthal 1994, 7-8; italics mine)
While our abstractive and selective process of knowledge imposes "cuts" upon the world, the decision regarding where these "cuts" occur is at least partially ours: As Peirce notes, "Truly natural classes may, and undoubtedly often do, merge into one another inextricably" (CP 1.209), and thus boundary lines must be imposed, although the classes are natural. The continuity is there; where the "cut" is imposed is, in part, our decision. (Rosenthal 1994, 8)
Thus, whereas Rosenthal with Hookway grants that there is an arbitrary element in establishing boundary lines between natural classes, she agrees with Haack in insisting that there is an objective ground to our natural classifications. Yet Rosenthal fails to tell us what such arbitrariness or convention entails. Does the absence of clear boundary lines entail that natural classes are not clearly defined? Or does it only mean that there are no clear demarcation criteria by virtue of which it can always be decided to which natural class an object belongs. The three foregoing interpretations raise a fundamental question concerning the nature of Peirce's genuine or assumed pluralism: What, according to Peirce, is the (epistemological and ontological) origin of our dividing the world into classes? Perhaps the best way to introduce Peirce's own views is to consider his critique of John Stuart Mill's (1806-73) definition of natural kind. This will be the subject of the next section.
104
3.
CHAPTER FOL R
PEIRCE VERSUS MILL
In this section I will first (3.1) briefly consider Mill's theory of natural kinds. Next (3.2) we will regard Peirce's earliest thoughts on Mill's theory. Thereupon (3.3) I will discuss one of Peirce's most important definitions of 'kind,' and discuss the critique offered by the mature Peirce of Mill's theory of "Kind" in Baldwin's Dictionary of Philosophy and Psychology (1901). Because this text raises an important problem, section 3.4 will be devoted to an examination of this problem in the light of some of Peirce's later manuscripts.
3.1
Mill's theory of natural kinds
According to Mill, every thing in the world belongs to some natural class or real kind. Mill made a distinction between natural classes and non-natural or artificial classes (Mill did not use the latter term). The main difference is that the things that compose a natural class have innumerous properties in common, whereas the things that belong to an artificial class resemble one another in but a few respects. More precisely: a "real kind" is defined as a class "which is distinguished from all other classes by an indeterminate multitude of properties not derivable from one another..." (Mill 1874, 99). The members of a natural or real class share innumerable properties that are not derivable from its defining character "by some law of causation." Plants and animals, and sulphur and phosphor are examples of real kinds (Mill 1874, 97). In contrast, the members of an artificial class share only a few properties, which "follow, as consequences, under laws of nature, from a small number of primary [characters] which can be precisely determined, and which, as the phrase is, account for all the rest" (Mill 1874, 98-99). The class of white things is an example of an artificial class: White things are not distinguished by any common properties, except whiteness: or if they are, it is only by such as are in some way connected with whiteness. But a hundred generations have not exhausted the common properties of animals or plants, of sulphur or phosphorus; nor do we suppose them to be exhaustible, but proceed to new observations and experiments, in the full confidence of discovering new properties which were by no means implied in those we previously knew. (Mill 1874, 97)
Similarly, the class of flat-nosed animals is an artificial class, because in addition to their flat noses, flat-nosed animals do not have any common properties other than those that are common to all animals (1874, 99).
FINAL CAUSES AND NATURAL CLASSES
105
Mill seems to have been the first philosopher to seriously consider social kinds (Hacking 1991, 118) such as those of Christians, Englishmen, and Mathematicians. He thought them to be artificial kinds, however, because there is no innumerable set of properties "common and peculiar" to Christians or Englishmen or Mathematicians; the limited number of properties shared by the members of the class are logically or causally determined by their definitions: "A Christian, for example, differs from other human beings; but he differs only in the attribute which the word expresses, namely belief in Christianity, and whatever else that implies, either as involved in the fact itself, or connected with it through some law of cause and effect" (Mill 1874, 98).
3.2
Natural kinds and the uniformity of nature; Peirce's earliest discussion of natural kinds
The problem of natural kinds first appears in Peirce's work in "The Fourth Lowell Lecture" (1866). The main objective of this lecture was a critical discussion of Mill's treatment of induction. In the course of the lecture Peirce raised two issues that are directly relevant for our understanding of his mature theory of natural kinds. The first issue concerns the relationship between natural classes and the uniformity of nature; the second issue pertains to the problem of defining a natural class. Mill's justification of the use of induction was based on his belief in the uniformity of nature. Peirce criticized this view by pointing out that it begs the question: the justification of the belief in the uniformity of nature is based on the use of induction (Wl: 414; 1866). Whereas Peirce agreed with Mill that "everything there is belongs to some real kind" (Wl: 416), he questioned Mill's idea that the uniformity of nature consists in the existence of natural classes. First he pointed out that, while uniform relations are one type of relations in nature, there are many more entirely irregular relations: Take any pear. It is sweet and all pears are sweet. There is a uniformity. But it is mine; and all pears are not mine. It is next to a bunch of grapes, and all pears are not next to a bunch of grapes. It is ripe and not all pears are ripe. [...] And so I might go on indefinitely. Indeed when it is remembered that everything in the world is related to every other in countless ways; it is plain that there is no end to the excess of accidental relations over those which present any regularity. (Wl : 417; 1866)
Thus, the number of irregularities in individual objects and in natural classes greatly exceeds the number of regularities. Horses, for example, have all kinds of characters in common, like swiftness, strength, and timidness; they all have backbones and long heads; they are all herbivorous, etcetera. But,
CHAPTER FOL R
106
not all animals with backbones are herbivorous, and not all strong things are swift. "So that even in this respect nature is not very uniform" (Wl: 418). Secondly, the statement that there are exact uniformities is not warranted by experience: Every student of physics knows that a law which is exactly conformed to in nature without interference from other laws, is almost if not quite unknown. Every law that is discovered therefore is found after a few years not to be exact. ( W l : 420; 1866)
Moreover, since there are exceptions to almost any rule (Wl: 419), it is highly improbable that there are uniformities in nature that are exact or without exception.7 Peirce concluded from these observations that the objects that belong to the same natural class, need not have all the characters that seem to belong to the class: "it must be admitted that there are exceptions to almost every rule. Thus many of the characters which seem to belong to a class universally only belong to a part of it" (Wl: 419). Peirce gave the example of a man with two heads. Though having just one head is a character that belongs to the class of Man universally, it is not impossible that one day a man would be borne with two heads. Whereas he would be an exception to the general rule, he would still belong to the natural class of Man. Therefore, if we use Mill's definition of uniformity of nature - "the universe is so constituted, that whatever is true in one case, is true in all cases of a certain description" - then "natural classes cannot constitute a uniformity in nature" (Wl : 420; 1866). The second issue that is relevant to our understanding of the mature Peirce concerns his definition of natural class. According to Peirce, natural classes have at least one characteristic in common apart from their defining character. A natural class "has other properties than those which are implied in its definition" (italics mine), for to define something is just to state the meaning of a word. If we define Man as a rational animal, then the expression "Man is rational" is simply an analytic statement. But each class has other properties than those implied in its definition; what makes a class natural is that there is at least one other universal character over and above the characters that compose its definition: Suppose this black board were dotted all over with chalk; and let these dots represent the individuals in the world. Then let us draw a circle around those which have any common character. Let this circle for example include all the animals and this other all the rational beings. Then what they both include would be rational animals. And this will be represented as a natural class if it be entirely or nearly enclosed by another circle. That will be all that is
FINAL CAUSES ANI : NATURAL CLASSES
107
required to make it a natural class for then it will have a universal character besides the rational and animal which compose its definition. ( W l : 419; 1866)
One year later, Peirce came to the following definition of natural class: ... a natural class is one which can be so defined that something can be predicated of it which cannot be predicated of the genera included in its definition. (W2: 443; 1867)
The other character is, according to Peirce, precisely what makes the difference between a logical definition and a scientific one. Taxonomists cannot do with just a logical definition; they need some reference to empirical facts. Thus, Peirce seemed to suggest that, though the elements of natural classes have a number of common empirical properties in common (Wl : 418), the taxonomist is free to choose those he considers to be relevant. Any empirical fact that belongs to all (or nearly all) the members of the class will do, not just the ones that are specific to the members of the class. Thus, if man be defined as a rational animal, 'man' is the natural class, 'rationality' is the defining character, and 'a humanlike set of teeth' could serve as the empirical character. To summarize: according to the early Peirce, (1) everything in the world belongs to some natural class or real kind; (2) natural classes do not constitute uniformities; (3) a natural class is characterized by (a) a defining character and the characters it implies, and (b) an empirical character which belongs to all or nearly all the members of the class.
3.3
Peirce's Baldwin definition of kind
Having obtained some insight into Peirce's earliest view of natural kinds, I will now consider some of Peirce's later texts, especially his Baldwin definition of "Kind" (1901). In this latter text, Peirce gave, in just a few words, a devastating critique of Mill's theory of natural kinds: first, it is simply not true that artificial kinds like white things have only a few properties in common. When Mill talked of "properties," he must have had in mind, mainly, characters that are interesting to us. For it is obvious that all white things have innumerable common properties. Secondly, Peirce rejected Mill's idea that a class of elements whose common properties are caused by a few primary properties, cannot be a natural kind. Indeed, it is precisely the goal of the man of science to explain the multitude of properties of a kind in terms of a small amount of underlying properties (CP 6.384; 1901). In other manuscripts,8 Peirce offered one simple example to disprove Mill's definition: the class cow is a natural class, while red cow is not. The
108
CHAPTER FOL R
class of red cows has all the features of the class of cows, and moreover, the feature of redness. If Mill would have admitted that the class of cows is a natural class, he also must have admitted that the class of red cows is. For if it is true that cows belong to a natural class because they have a large or even apparently inexhaustible number of properties in common (which are not in any way dependent on its defining character), then the same must hold for red cows, because they even share one more common characteristic. Yet such a class is not a "real kind" (MS 421; 1893-5). Evidently Peirce presupposed that Mill would have accepted the common sense opinion that red cows do not form a natural class. After thus having criticized Mill, Peirce gave the following definition of natural class (or real kind):9 Any class which, in addition to its defining character, has another that is of permanent interest and is common and peculiar to its members, is destined to be conserved in that ultimate conception of the universe at which we aim, and is accordingly to be called 'real.' (CP 6.384; 1901)
Compared to his definition of 1867, Peirce made explicit the idea that a natural class must have, over and above its defining character, another character which must not only be of permanent interest and common, but also peculiar to its members. In the next section, an attempt will be made to further clarify the issue of the non-defining, empirical character.
3.4
The PRE-character
Consider the case of 'man.' If 'man' is a natural class, the elements of that class must share a defining character, for example rationality. But human beings also share other properties. For instance, they all have a 'humanlike set of teeth.' But clearly this property is not directly related to their humanity. The character needed must somehow be relevant to what makes us distinctly human. Such character might for instance be the character of 'having a brain with a certain complex cortical structure.' Such character meets the requirements spelled out by Peirce in his preceding quote: it is not the defining character, it is of permanent interest, and it is common and peculiar to the members of the class. Because this character is genuinely empirical in nature, I shall henceforth refer to the defining character as the D-character, and to the other character as the PRE-character (permanently relevant empirical character). The main difference is that while the D-character is a general principle, the PRE-character is an essential quality embodied in an existing thing. The most salient feature of a PRE-character is the requirement that it be permanently relevant, and thus permanently important.
FINAL CAUSES ANI : NATURAL CLASSES
109
In 1867 Peirce had given a definition of 'an important character', which he still held to be correct in 1902.'" According to his explanation in his 1902 paper "On Classification," a taxonomer considers a character to be important because it involves certain others, be it only "a particular likelihood to taking certain forms." Hence, "importance consists in a character's universally carrying with it certain others, be those others no more than tendencies" (NEM IV: 65; 1902). In the text of 1902, however, Peirce claimed that this 1867-definition needed a fundamental correction: "an important character must not only entrain others, but it must entrain another which has relation to the purpose in view" (NEM IV: 65). This entails that the PRE-character universally carries with it certain other characters, though these may be only tendencies. Moreover, among those other characters there is at least one that is closely related to the D-character, which, as will be shown in section five, is precisely what Peirce meant by "the purpose in view." Thus, human beings, defined as rational animal, do form a natural class. The PRE-character of 'having a brain with a certain complex structure' "entrains," among other things, the tendency to form complex thoughts, which is a necessary condition for being rational. Thus, 'having a brain with a certain complex structure' is a good example of a close relationship between the PRE-character and the D-character. Peirce's considerations about 'importance' allow us to modify his Baldwin definition as follows: a natural class is any class that is characterized by a D-character and a PRE-character. The PRE-character universally carries with it certain tendencies, of which there is at least one that is closely related to the D-character. Given this definition, a number of problems arise. Of these, the first one to be examined concerns the distinction between 'kinds' and 'classes,' which Peirce made near the end of his career (1908).
4.
KINDS AND CLASSES
A kind is an entity that corresponds to a set, the elements of which do not exist; a class is an entity corresponding to a set of which at least one element does exist. Thus, Peirce pointed out for instance that, while at his time, black tulips were non-existent, nevertheless some people (for instance, gardeners) may very well have thought of the possibility of growing black tulips. While the kind 'black tulip' was real, there was no natural class of existing black tulips. For the 'existence' of a natural class requires the existence of at least one specimen of that kind:
110
CHAPTER FOL R For the class is that ens rationis whose existence consists in the actualization of a definite kind. The actualization in an existing singular is one requisite to a class, being requisite to its existence: the character which it is required that every member of the class should have, is a second requisite to the class, being requisite to its entity. The two together make up its ousia, its rational essence. (MS 200:00172; 1908)
This is an interesting definition for a number of reasons. The first thing that draws attention is that a class is called an ens rationis or "being of reason." Peirce applied this term, borrowed from Duns Scotus (and other scholastics), to entities that owe their reality to an operation of the intellect which Peirce called 'hypostatic abstraction.' Contrary to a real being or ens in re extra animant, such as a concrete, individual horse, an ens rationis is a 'thing' that depends for its existence upon reason or thought. Whereas real beings exist independently of thought, beings of reason depend on thought. But there are two kinds of entia rationis: those with a foundation in reality and those without such foundation. Examples of the former are genera and species (for instance, animal and horse); examples of the latter are mythical figures. Thus, according to Duns Scotus, some universals exist only by virtue of the operation of the intellect, but cannot in any sense be said to be mere 'figments' of the mind. We can form universal concepts only because there is an objective correlate of them in the objects themselves: the "common nature." Horseness, for example, is the common nature of all the things called horses. But horseness is neither a universal nor a particular. Horseness is simply horseness. Universals are concepts formed by the mind, but there is an objective basis to them in the "common nature" of the concrete, existing things. Peirce had borrowed Duns Scotus's view to the extent that sometimes our abstractions reflect objectively real general principles: "that wonderful operation of hypostatic abstraction by which we seem to create entia rationis that are, nevertheless, sometimes real..." (CP 4.549; 1906). He also had borrowed Duns Scotus's idea that real generals have the reality oí possibility, not of actuality, albeit with a different twist.11 Peircean real generals are not common natures or forms, but final causes or laws. That generals are possibles entails that, though they may be real, they do not exist. It may be noted that Peirce was somewhat careless when he spoke of the existence of natural classes, for classes cannot strictly be said to exist. The members of a class exist, but the class itself does not. Classes are entia rationis, which are generals, and generals are real but do not exist; they are possibilities. Only individual things exist, that is, only things that occupy a definite space during a certain time. Individuals can be pointed at; generals cannot.
FINAL CAUSESANI: NATURAL CLASSES
111
We now know that a class must meet at least two criteria: it must have at least one existing member, and each member of the class must have both a defining or D-character and an indefinite number of D-related class characters.12 A kind differs from a class on two counts: it does not contain an existing member, and therefore it has only a D-character, which constitutes its essence. Peirce made a distinction between the epistemological essence and the metaphysical essence of a kind: The essence of anything is that thought which renders the thing possible. The epistemological
essence is that thought which renders it possible to conceive
of the things. The metaphysical
essence is that intellectual structure which
renders the being of the thing possible. (MS 200:00145; 1908) 13
In natural classifications, the epistemological essence coincides with the metaphysical essence. According to Peirce, it is usually quite easy to determine the metaphysical essence of an artifact. The metaphysical essence of a lamp, for example, is that it can give light. And that is the purpose that brings lamps about. And the metaphysical essence of a stove is "that it is intended to diffuse warmth" (CP 6.336; c. 1909). But the question regarding the metaphysical essence of a natural object is much tougher. One might wonder whether the distinction between metaphysical essence and epistemological essence makes any sense from a pragmatic perspective. Does it not presuppose a bifurcation between the realm of knowledge and the realm of being? And doesn't the idea of 'metaphysical essence' presuppose what Putnam has called "a God's eye point of view"? I have to admit that Peirce's terminology is confusing, for it suggests that there is an objective reality independent of our knowing processes. But this is not what Peirce had in mind. There is no metaphysical essence independent of our knowing processes; the metaphysical essence is independent of how you and I or anyone else at a specific moment characterizes the world. In the long run, however, the classifications of the scientific society will reflect the metaphysical essence of things. Much of the confusion arises from the fact that, though Peirce tries to break with traditional ways of conceiving the question of universals, essences, or kinds, he does not break with traditional language. Another possible objection is related to the fact that the essence of anything is by nature immutable. But, contrary to Platonic or Aristotelian essences, and to Scotistic "common natures," which are all immutable, static forms, Peircean essences are of the nature of habit; and habits are, at least in principle, subject to evolution. Consequently, one of the most persistent objections against natural classes, namely that they presuppose an immutable essence, does not hold for Peirce's position.
112
CHAPTER FOL R
Yet another difficulty regards the intellectual structure of essences. Peirce, however, did not restrict 'idea' or 'thought' to something that a person has in mind, or to a psychical act of thinking: "by an idea [...] I mean a principle such as may be set before the mind in thought" (MS 1344:11 ; 1902). Thus, the statistical distribution of a large number of things, say the molecules of a gas, expresses a statistical law, which is the 'idea' of the distribution (MS 1344:11 ; NEM IV: 65-66). Moreover, ideas are not only (a) general principles; they are also (b) in a sense purposive or quasi-purposive (end directed). Thus, the statistical law is a general idea, which is the final cause explaining the tendency toward the end state of the gas. Ideas, therefore, have a certain inherent tendency to realize themselves. An idea without efficacy cannot be an idea at all: Imagine such an idea if you can! If it was communicated to you viva voce from another person, it must have had efficiency enough to get the particles of air vibrating. If you read it in a newspaper, it has set a monstrous printing press in motion. If you thought it out yourself, it had caused something to happen in your brain. And again, how do you know that you did have the idea when this discussion began a few lines above, unless it had efficiency to make some record on the brain? (CP 1.231 ; 1902)
We have seen so far that the essence of a natural class is of the nature of an idea, and that ideas are, basically, final causes. The defining idea of a set of objects is its epistemological essence. In natural classes, however, the defining idea or epistemological essence reflects the metaphysical essence. Because the defining idea of a natural class is a final cause, it seems appropriate to further explore the purposive nature of ideas or essences. This will be done in the next section.
5.
CLASSIFICATION ACCORDING TO FINAL CAUSES
In his note "On Classification" of his Carnegie Application (1902),14 Peirce mentioned that he had been a student of Agassiz (in 1861 ), and that his study over the years had convinced him that Agassiz's system of classification was basically correct. Peirce formulated Agassiz's central insight as follows: "every classification whatsoever, be it merely arranging words in alphabetical order, has reference to some purpose, or some tendency to an end" (NEM IV: 65; 1902; italics mine). Thus, classifications are teleological instruments, or a way of handling things for some particular purpose. Now, arranging words in alphabetical order is an example in which it is just our purpose that determines the classification. That is why the classification is
FINAL CAUSES ANI : NATURAL CLASSES
113
artificial. In natural classifications, however, it is not our purpose but the purpose or quasi-purpose of the class itself that is at stake: Every unitary classification has a leading idea or purpose, and is a natural classification in so far as that same purpose is determinative in the production of the objects classified. (NEM IV: 15; 1902)
Similarly, Every classification has reference to a tendency toward an end. If this tendency is the tendency which has determined the class characters of the objects, it is a natural classification. (NEM IV: 65; 1902)
Thus, the defining idea of a natural class ideologically determines the class characters of the objects belonging to the class. To clearly distinguish them from the PRE-character (permanently relevant empirical character), I will call the class characters from now on TDE-characters (ideologically determined empirical characters). Though Peirce sometimes used the term "essential characters" (for example in CP 1.204), for reasons yet to be explained, his term "class characters" is more appropriate. In order to precisely understand the relationship between defining character and TDE-characters, we must consider what is perhaps Peirce's most important text on natural classes, "A Detailed Classification of the Sciences"15 (CP 1.203-283; 1902), where he worked out his view that ideas may be said to be ideologically causal. Properly speaking, the text deals with the problem of finding a classification scheme in which all the sciences find their hierarchical place. But since his anti-nominalistic stance implied that such a scheme is based on natural or real classes, Peirce thought it necessary first to explain what he meant by a natural class. In his attempt to give an exact description of a natural class, he concluded that the final cause is its defining characteristic. Accordingly, a natural or real class is defined as a class "of which all the members owe their existence to a common final cause" (CP 1.204), or as "a class the existence of whose members is due to a common and peculiar final cause" (CP 1.211). The final cause is described in this context as "a common cause by virtue of which those things that have the essential characters of the class are enabled to exist" (CP 1.204). Thus, the defining idea must clearly be understood as causally active in the teleological sense. For instance when Peirce wrote: [e]very class has its definition, which is an idea; but it is not every class where the existence, that is, the occurrence in the universe of its members is due to the active causality of the defining idea of the class. That circumstance makes the epithet natural particularly appropriate to the class... (CP 1.214; 1902),
114
CHAPTER FOL R
the expression 'active causality' must be taken in this teleological sense. In view of this, we must again return to the general characteristics of Peirce's conception of final causation. According to Peirce, final causes are general types that tend to realize themselves by determining processes of mechanical causation. They are not future events, but general (physical) possibilities. The symptoms of final causation are that the end state of a process can be reached in different ways, and that the process is irreversible. Final causes are basically habits: they ('habitually') direct processes toward an end state. Like human habits, habits of nature (laws of nature) are final causes because they display tendencies toward an end state. Final causes stand to laws of nature as genus to species. Moreover, habits are not static entities, for they may evolve in the course of time. Peirce called the possible evolution of final causes "developmental teleology." In view of this, what does it mean to say that a natural class owes its existence to a common defining idea or final cause? Do I mean that the idea calls new matter into existence? Certainly not. That would be pure intellectualism, which denies that blind force is an element of experience distinct from rationality, or logical force. [...] What I mean by the idea's conferring existence upon the individual members of the class is that it confers upon them the power of working out results in this world, that it confers upon them, that is to say, organic existence, or, in a word, life. (CP 1.220; 1902)
Ideas cannot call new matter into existence; they can only work if there is matter to work upon. The action of ideas is typical of final causation; the action of matter is typical of efficient causation. Blind force (efficient causation) and rationality (final causation) are two undeniable elements of our experience; one requires the other. But all this does not as yet explain that "the idea [confers] existence upon the individual members of the class," and that it gives them "organic existence" or "life." The reason must be that, if matter were not governed by ideas or final causes, there would not be any regularity in its behavior, which means that it would not even exist: ... if [matter] were to be deprived of the governance of ideas, and thus were to have no regularity in its action, [...] throughout no fraction of a second could it steadily act in any general way. For matter would thus not only not actually exist, but it would not even have potential existence, since potentiality is an affair of ideas. It would be just downright nothing. (CP 1.218; 1902)
Two examples may illustrate Peirce's intention. The first is taken from the realm of social phenomena: the natural class of socialists. A member of the community of socialists can only be a socialist by virtue of the idea of
FINAL CAUSES ANI : NATURAL CLASSES
115
socialism. In Peirce's view, it is the idea of socialism that creates the socialist, not the other way round. Ideas are not just creations of a particular mind, but on the contrary, they have a capacity, a power, to create or to find their vehicles: "it is the idea which will create its defenders, and render them powerful" (CP 1.217). Of course, the idea of socialism does not create the person who is the socialist. But, given an existent person, the idea of socialism may turn him into a socialist. The idea of socialism confers existence upon the individual members of the natural class of socialists. It gives them "organic existence" or "life" as socialists, that is to say, it makes them behave, at least to a certain extent, as socialists are supposed to do. To exist as a socialist requires a certain amount of regularity in one's behavior; it requires that one's behavior be directed by the idea of socialism. The second example is related to what might be called Peirce's (metaphysical) holism. Final causation is seen as that general principle in virtue of which a whole is more than the sum of its parts. The final cause is the intellectual structure or thought that ties the parts together, and gives them "organic existence" or "life." In Peirce's words: Efficient causation is that kind of causation whereby the parts compose the whole; final causation is that kind of causation whereby the whole calls out its parts. (CP 1.220; 1902)
Thus, it is the final cause that confers "organic existence" or "life" upon the individual members of a natural class. To illustrate this idea, Peirce gave the example of a dissected corpse. No one would consider a man's organs lying separately on a stretcher as a human being. The dissection might give some insight into what parts are required to make the human body work, that is, it would at the most display efficient causation. But it cannot explain the fact why a human body works: "The final causation, which is what characterizes the definitum, it leaves out of account" (CP 1.220). The final cause is that principle whereby a person is something more than just a body; it gives the body "organic existence" or "life." Peirce had a simple and convincing answer to anyone who would object, and who would insist that it is only matter that is essential for the existence of an individual person: whereas the matter we are made of continually changes, our form remains the same. The existence of an individual man is something altogether different from the matter he happens to be composed of, and "which is incessantly passing in and out."16 That which gives continuity to his existence is not the matter but the form, that is, his defining idea. This defining idea or final cause is what traditionally has been called the soul of man. There is every indication that Peirce's conception of the defining idea slowly evolved as he gradually became more committed to a scholastic
116
CHAPTER FOL R
realism.17 Whereas in 1866 the defining idea had a predominantly epistemological status, from 1902 onwards its status was first and foremost ontological. While in 1866 Peirce had defined a natural class as a "class that has other properties than those which are implied in its definition and these other characters would make it a natural class" (Wl: 418; 1866), in 1902, the defining idea is no longer a character which logically implies certain other characters. Instead, it is a final cause to which the members of the class owe their existence: the existence of the objects of the class "is due to the active causality of the defining idea of the class" (CP 1.214; italics mine). Whereas in 1866 the defining idea had supplied the epistemological essence of the class, that is to say, "that thought which renders it possible to conceive of things," in 1902 it displayed the metaphysical essence or "that intellectual structure which renders the being of the thing possible." Whereas until 1901 (in his Baldwin definition) Peirce had emphasized the permanently relevant empirical character (PRE-character), from 1902 onwards he had replaced it by an indefinite number of ideologically determined empirical characters (TDE-characters), which were subservient to the defining idea or D-character. The main difference between PREcharacters and TDE-characters is that, contrary to PRE-characters, which are essential qualities but not teleologically determined by the D-character, the TDE-characters are 'non-essential' qualities (in a sense yet to be explained), which are teleologically determined by the D-character. In the next section I will examine the question whether knowledge of the common final cause is sufficient to determine the class (or classes) to which an object belongs, or whether we need other demarcation criteria.
6.
CRITERIA OF DEMARCATION
Because natural classes must be understood in terms of final causes, it is necessary first to consider some further characteristics of final causes before the question of demarcation criteria can be addressed. Final causes are general. This generality involves both vagueness and longitude. Final causes are general because: (1) they are not spatio-temporal; (2) they determine only some but not all qualities of a class of objects (or of a process). For example, the idea of building a house only determines that the end product will be a house, but not the specific form of the house. This lack of specificity is also called the vagueness of the final cause. Finally (3), final causes are general because they are not exhausted by any finite number of instantiations. Moreover, final causes have a certain longitude. "By this I mean that while a certain ideal end state of things might most perfectly satisfy a desire,
FINAL CAUSES ANI : NATURAL CLASSES
117
yet a situation somewhat different from that will be far better than nothing; and in general, when a state is not too far from the ideal state, the nearer it approaches that state the better" (CP 1.207; 1902). If, for some reason, I do not succeed in realizing my plan to write a book on causation, the second best thing would be to write some articles on the subject. Though there was a definite tendency toward an end state - a book on causation -, external or internal elements kept my purpose from being fully realized. But a partial realization is much better than no realization at all. A third element, next to their longitude and vagueness, is important to the determining cause of a natural class: although a final cause is in itself rather general and simple, it necessarily tends to a greater definiteness and complexity in the course of its realization (MS 1343:15; 1902). Such process usually involves conditions that are specific to every step, as well as 'decisions' regarding the further realization of the general purpose. In the course of building a house, all kinds of decisions must be made about shape, size, material etcetera, and each of these functions as a subsidiary purpose. As a result of (a) the vagueness and (b) the longitude of final causes, and as a result of (c) the action of subsidiary final causes, the class characters of the objects of a natural class (that is, the qualities determined by its final cause) cluster around certain average values. Peirce illustrates this by an example borrowed from human experience: if we are to produce artificial light as economically as we can, we must consider all kinds of additional subsidiary purposes: ... the situation of things most satisfactory to one desire is almost never the situation most satisfactory to another. A brighter lamp than that I use would perhaps be more agreeable to my eyes; but it would be less so to my pocket, to my lungs, and to my sense of heat. Accordingly, a compromise is struck; and since all desires are somewhat vague, the result is that the objects actually will cluster about certain middling qualities, some being removed this way, some that way, and at greater and greater removes fewer and fewer objects will be so determined. Thus, clustering
distributions
will characterize
purposive
classes. (CP 1.207; 1902; italics mine)
This consideration is relevant to the issue of demarcation criteria. Peirce illustrated this with an example taken from archeology: ... Prof. Petrie found in the town of Naucratis some hundred and eighty standard weights. The calculus of probabilities applied to their weight-values proves that they were intended to conform to five different quasi-prototypes; but many of the weights, owing to the imperfection of their manufacture, have intermediate values, so that, as far as their governing intended character goes,
118
CHAPTER FOL R it would be impossible to say to which standard any one such intermediate weight was intended to conform. (MS 1343:14; also CP 1.209-10; both 1902)
This example reveals that closely related classes are not, in general, separated by sharp lines of demarcation. Some forms may just as well belong to one natural class as to another. In such cases, further investigation will usually show that there are other, more or less accidental characters, which may help in directing the forms to their true classes. Such characters, which are not specific to the class, may help us in ascertaining whether a given individual belongs to one class rather than the other: "unless we have some supplementary information we cannot tell which ones had one purpose and which the other" (CP 1.208). In the case of Petrie's example, further information might concern the shapes or the material of the stones, or some other "inessential" character (MS 1343:13-14). The example of the weights also reveals that, though natural classes are characterized by a defining idea, which makes up their metaphysical essence, there are no essential qualities that are both necessary and sufficient for belonging to a specific natural class: [We may want to] enumerate characters which are absolutely decisive as to whether a given individual does or does not belong to the class. But it may be, as our [example of the weights] shows, that this is altogether out of question; and the fact that two classes merge is no proof that they are not truly distinct classes. (CP 1.224; 1902)
Though there are no essential qualities by virtue of which it can unambiguously be ascertained to which natural class the weights with intermediate values belong, they nevertheless were intended to conform to one definite prototype. Each of these weights therefore belongs to one specific natural class. Apparently, in 1902 Peirce had distanced himself from his Baldwin definition (1901), according to which each member of a natural class was characterized by at least one essential quality: its permanently relevant empirical or PRE-character. Thus, things belong to the same natural class, not because of some essential qualities (which are Firsts according Peirce's categoreal system18J, but because of a metaphysical essence, which is an idea or final cause (which is a Third). Moreover: in the last quote, Peirce seems to reintroduce his 1866-idea that the objects that belong to the same natural class, need not have all the qualities that seem to belong to the class: "it must be admitted that there are exceptions to almost every rule. Thus many of the characters which seem to belong to a class universally only belong to a part of it" (Wl: 419; see section 3.1). Class qualities therefore are not essential qualities.
FINAL CAUSES ANI : NATURAL CLASSES
119
In the next section it will be seen what these considerations entail for the definition of a Peircean natural class.
7.
RECAPITULATION: DEFINITION OF PEIRCEAN NATURAL CLASSES
Because the developments in Peirce's conception of natural classes are relatively complex, it may be good to summarize the results obtained so far. This summary will be followed by an attempt to formulate a definition of Peircean natural classes that displays Peirce's mature position as fully as possible. In 1866, Peircean natural classes were characterized by (a) a defining character, (b) characters that are logically implied by its definition, and (c) an empirical character (section 3.2). According to Peirce's Baldwin definition of "Kind" (1901), natural classes were characterized by a defining character (D-character) in combination with one empirical character (PRE-character). Both characters were thought to be "of permanent interest" and were considered to be "common and peculiar" to the members of the class (3.3). I proposed a reformulation of Peirce's Baldwin definition of "Kind," based on Peirce's note "On Classification" of his Carnegie Application (1902). According to this reconstruction, a Peircean natural class is any class that is characterized by a D-character and a PRE-character, both of which are common and peculiar to the members of the class. The PRE-character universally carries with it certain tendencies of which there is at least one that is closely related to the D-character (3.4). In 1908 Peirce made a distinction between natural classes and kinds. Whereas a kind is an entity corresponding to a set the elements of which do not exist, a class is an entity corresponding to a set of which at least one element exists. A class must meet at least two criteria: it must have at least one existing member, and each member of the class must have both a Dcharacter and at least one teleologically determined empirical character (TDE-character). A kind differs from a class on two counts: it does not contain an existing member, and therefore it has only a D-character, which constitutes its essence (4). Peirce adopted Duns Scotus's view of classes as entia rationis, owing their reality to an operation of the intellect. Natural classes are abstractions corresponding to objectively real general principles. They are, however, not pure abstractions (or 'Firsts'), but generals ('Thirds') embodied in concrete existing things ('Seconds'). Because these real generals are possibilities, not
120
CHAPTER FOL R
actualities, natural classes cannot strictly be said to exist, but are nevertheless real (4). In 1902 Peirce related natural classes to their final causes. Accordingly, he defined a natural class as "a class the existence of whose members is due to a common and peculiar final cause" (CP 1.204), by virtue of which the members of the class behave in a regular way which is characteristic for that particular class (5). Peirce's relating natural classes to final causation marked an important shift in the evolution of his conception of natural classes. Whereas in its original definition (1866), the function of the defining idea of a natural class was predominantly epistemological, in his later works he defined natural classes in terms of final causation, and thus the defining character came to display the metaphysical essence of the class (5). Thus, whereas for the early Peirce (1866), the relationship between defining idea and essential qualities was a logical one,19 to the later Peirce (1902), the determination of the class or TDE-characters by the defining idea was (teleologically) causal. This shift toward a causal relationship between the defining idea and its class characters had the advantage that Peirce no longer needed to refer to 'important empirical characters' (PRE-characters), since every class or TDEcharacter is closely related to the defining idea because it is teleologically caused by it. Moreover, TDE-characters are not essential qualities because they need to be neither common nor specific to the members of the class. As a result of (a) the vagueness and (b) the longitude of final causes, and (c) the action of subsidiary final causes, the class qualities of the objects of a natural class (the qualities determined by its final cause) cluster around certain average values. Accordingly, closely related classes are not, in general, separated by sharp lines of demarcation (6). The example of the weights revealed that things do not belong to the same natural class because of some common essential qualities (Firstness), but on account of a similarity in behavior; they conform to the same final cause or law. The locus of universality is final causation, habit, or law (Thirdness) (6). On the basis of this reconstruction I propose to give the following characterization of Peircean natural classes: Things belong to the same natural class, not because of certain essential qualities (Firsts), but on account of a metaphysical essence which is a final cause (or Third). Thus, Peircean natural classes are characterized by (a) a defining character, which is a final cause and (b) a number of class characters or teleologically determined empirical characters (TDE-characters); moreover, (c) the TDEcharacters of the objects of a natural class cluster around certain average values; (d) the TDE-characters are not essential characters because they are neither necessary nor sufficient conditions for making something to be a
FINAL CAUSES ANI : NATURAL CLASSES
121
member of the class; (e) there are no clear boundary lines between closely related natural classes; ( f ) natural classes, though very real, are not existing entities; their reality is of the nature of possibility, not of actuality. In the next section I will (a) consider Peirce's reasons for believing in the reality of natural classes, and (b) see what use science has for the notion of natural class.
8.
WHY BELIEVE IN NATURAL CLASSES?
It is a well-known fact that Peirce was greatly interested in medieval logic, especially in the works of the nominalist William of Ockham and the realist Duns Scotus. He thought that nominalism was the greatest source of the mistakes of modern philosophy (CP 6.348; c. 1909; CP 5.61; 1903). On the other hand, he thought the philosophy of Duns Scotus offered a good basis for a philosophy "which is best to harmonize with physical science" (CP 1.6; c. 1897). Whereas Ockham held that only individuals exist in the real world and that universals are mere names, Scotus insisted that the real world contains real universals or generals. Peirce thought the question regarding the reality of universals not only of great technical philosophical value, but also of great importance for our daily moral concerns: ... the question of realism and nominalism [...] [has] branches [which] reach about our daily life. The question whether the genus homo has any existence except as individuals, is the question whether there is anything of any more dignity,
worth,
and
importance
than
individual
happiness,
individual
aspirations, and individual life. Whether men really have anything in common, so that the community is to be considered as an end in itself [....] is the most fundamental practical question in regard to every public institution ... (EPI: 105, 1871)
Moreover, he thought the nominalistic outlook of most modern philosophers was disastrous for the understanding of science. Nominalistic theories cannot explain that scientific theories are excellent tools for predicting future events. If we say, with Ockham, that all generalizations are subjective because they are based on the mind's capacity to form generalizations on the basis of perceived similarities, then our predictions miss any rational ground. Peirce claimed to have proven the falsity of nominalism by a simple, by now famous, thought experiment (CP 5.93-101; 1903). In a Harvard classroom he held up a stone, and asked his audience whether they could predict that it would fall if he were to drop it. Of course, everyone said he could. Peirce argued that this entails that there are real laws of nature. For if
122
CHAPTER FOL R
laws were merely generalizations of past happenings, there would be no ground for our expectation that the stone would fall to the ground. Hence, he drew the "irrefragable" conclusion that "general principles are really operative in nature" (CP 5.101). Without general principles, which are final causes (laws), prediction, induction and explanation would be impossible (CP 5.100-101; also Haack 1992, 25-29). This view has the immediate implication that science must discover the true laws of nature, and therefore it must establish the kinds of things that are connected by those laws. In other words, science must point out natural classes. Having obtained some understanding of Peircean natural classes and why they are necessary for our understanding of the world, I will now consider some important examples. First I shall examine social classes and artificial objects, then the chemical elements, and finally the biological species.
9.
EXAMPLES OF NATURAL CLASSES
We have seen that a natural classification is one that proceeds according to the purpose or quasi-purpose of the existence of the 'objects' classified. But what does the word 'object' depict in the above given description? Are the objects properties, processes, states, facts, events, or things? To answer this question, it seems advisable first to consider the examples Peirce gave of natural classes. I will begin with the easiest examples, which are taken from the realm of human experience.
9.1
Examples from the realm of human experience: social classes, the sciences, and man-made objects
The examples taken from the realm of human experience are usually easiest to classify, for in this domain it is often easy to discover by what purpose the objects of a class are determined. Social classes are examples of natural classes. Peirce mentions artists, practical men (business men), and scientists (CP 1.43; c. 1896). Each of these groups owes its identity to a specific purpose. One might extend the list to include Christians, Mathematicians, Jews, Muslims and Pagans. While Mill listed these as artificial classes, they are excellent illustrations of Peircean natural classes. On the other hand, Peirce would probably agree with Mill that 'Englishmen' constitute an artificial social class; for there is no specific identity or goal that is common and specific to the English. The sciences provide a second category of examples of Peircean natural classes that are closely related to the realm of human action. Indeed, Peirce
FINAL CAUSES ANI : NATURAL CLASSES
123
developed most of his ideas about natural classes while working out a classification scheme of the sciences. All science is divided into three major branches, each of which has a different purpose: Practical Science, Science of Review, and Science of Discovery or Research. Whereas Practical Science deals with investigations conducted for utilitarian purposes, the Science of Review aims at systematizing available knowledge. Science of Research is not defined in terms of available knowledge, but in terms of the concrete life of those whose "single animating purpose" it is to find out the truth for its own sake (CP 7.54; c.1902). Within each branch, every science is classified according to its specific purpose or object. The classifications are hierarchical; the more general the object, the higher is its place in the hierarchy.20 The artificial objects are the third category of examples from the domain of human culture. A natural classification of artificial objects is a classification according to the purpose for which they were made. Accordingly, it can be said that stoves are different from lamps because they serve a different aim. Often artificial objects may also be classified according to subsidiary purposes. Thus, bicycles may easily be classified into city bikes, mountain bikes, racing bikes, tracking bikes, etcetera, each according to its specific purpose. If we classify bicycles according to their purpose, the classification is natural; a classification according to color would be artificial. To illustrate the precedence of form over matter in natural classifications, Peirce also gave an example from the domain of art: "... who would classify Rafael's paintings according to their predominant tinges instead of according to the nature of the composition, or the stages of Rafael's development?" (NEM IV: 322; c.1906) Only the form or structure of the compositions "renders the composition of the entire classified object rationally intelligible," not their matter. Apparently, knowledge of this structure provides insight into the purposes of the painter. In the next two sections, we will see that there is an important similarity between classifications of works of art and classifications of chemical substances and biological classes; in all of these the final cause is displayed in some kind of structure.
9.2
The chemical elements
According to Peirce, the chemical elements differ in an important respect from all other natural classes: they are grouped not hierarchically, but periodically. Peirce insisted that when forms have developed from other forms, their genetic classification must be hierarchical. However, Mendeleefs classification of the chemical elements is definitely not hierarchical. "It is a cross classification of an exact mathematical type" (MS
CHAPTER FOL R
124
421; 1893-5; italics mine). According to Peirce, this strongly indicates that, contrary to biological forms, the chemical elements have not originated by a development of one from the other.21 Indeed, it was Peirce's view that there are two different kinds of systematic relationships between different natural kinds. Whereas classes are normally grouped according to the Aristotelian hierarchical model, chemistry groups the elements periodically. Dmitri Mendeleef had been the first person to arrange the elements according to their periodic similarities (1869). He found that if the elements are arranged approximately according to their increasing atomic weight, elements with similar physical and chemical properties occur at periodic intervals. His table proved to be a good guide to predicting chemical behavior, because it enabled us to determine what elements should be chemically similar to others. Not only do similar elements act alike, but their compounds may also act alike. For instance, NaCl has properties that are similar to those of both KCl and RbCl, because Na, K, and Rb are chemically alike. Peirce, however, thought that the chemical elements owe their classification first and foremost to their valency. Indeed, natural classification is classification according to structure. But indecomposable chemical elements have no parts, and therefore no internal structure. Thus only their external structure must be taken into account. The external structure of an element was defined by Peirce as "the structure of its possible compounds" (CP 1.289; c.1908). In chemical elements, the basis of all external structure is valency: In classification generally, it may fairly be said to be established, if it ever was doubted, that Form, in the sense of structure, is of far higher significance than Material.
Valency
is the basis
of all external
structure;
and
where
indecomposability precludes internal structure [...] valency ought to be made the first consideration. (MS 292:34; 1906)
The view that elements are indecomposable has been refuted by 20th century physics. But in a way, the idea that elements do have an internal structure which determines their valency and behavior, only confirms the consistency of Peirce's view that (a) natural classification is classification according to the final cause of the objects classified, and that (b) natural classification is classification according to structure. An external structure can hardly be a final cause of the objects classified, because it depends itself upon the existence of those objects. I will try to show that the internal structure can be such a final cause. First, however, I will show that internal structures can never be efficient causes. That the internal structure cannot be an efficient cause appears from three facts: (a) whereas efficient causes are always concrete events or facts,
FINAL CAUSES ANI : NATURAL CLASSES
125
internal structures are always general, for they are displayed in a multitude of events. Moreover, (b) because efficient causation is not directed toward an end in any way, it cannot explain that atomic structures are responsible for the atom's tendency to behave in a regular way. Finally (c), whereas efficient causes only induce one or more lines of mechanical causation at one singular moment, the atomic structure continually induces events to conform to a definite pattern. Because Peirce recognized only two types of causation - efficient causation and final causation - one is forced to conclude that inasmuch as the internal structure has some kind of causal influence, it must necessarily be teleological causation. Indeed, the internal structure has all the characteristics of final causation: (a) it is general, (b) it explains a tendency to behave in a regular way, and (c) it continuously induces processes of causation to conform to a definite pattern. Thus, there is no reason for believing that Peirce would not have agreed with the insights of contemporary physics, according to which the external structure (or valency) of the chemical elements is determined by their internal structure. Therefore, according to my (21st century) interpretation of Peirce, it would be correct to say that the chemical elements are classified according to their internal or atomic structure. Whereas the chemical elements are classified according to their atomic structure, the chemical compounds are classified according to their molecular structure. The classification of compounds is related to the fact that "... all samples of the same molecular structure react chemically in exactly the same way..." (CP 4.530; 1906). The molecular structure is represented by what chemists nowadays call the structural formula, which is to be distinguished from the molecular formula, which merely gives the numbers of the different atoms. This explains that compounds with the same molecular formula do not necessarily belong to the same natural class: "who would for one instant liken ordinary alcohol to methyl ether (which has the same material composition) instead of with the alcoholates?" (NEM IV: 321-22; 1902). Though ordinary alcohol and methyl ether have the same molecular formula (C 2 H 6 0), they still have a different structural formula (respectively: CH 3 -CH 2 -OH and CH 3 -0-CH 3 ). Ordinary alcohol does not belong to the same kind as methyl ether, because it has a different geometrical structure. Similarly, analogous behavior of two compounds may indicate that the molecular structures are similar: "to take a simple example, chlorates KQ0 3 , manganates KMn0 3 , bromates KBr0 3 , rutheniates KRu0 3 , iodates KI0 3 , behave chemically in strikingly analogous ways" (CP 1.223; 1902).22 Similarity of behavior indicates that there is a similarity of molecular
126
CHAPTER FOL R
structure, and a certain degree of similarity of molecular structure is a good reason for believing that I are dealing with the same natural class.23 To summarize: the chemical elements and the chemical compounds are classified, respectively, according to their atomic and their molecular structures. Because Peirce defined natural classifications as those that were made according to the final cause to which the members of the class owe their existence, it may be concluded that he thought the final cause of the atom or the molecule to be expressed in their internal structure. Inasmuch as these structures are expressed in individual entities, they are neither universal nor particular. But qua structures, they are universal. According to Peirce, chemical structures are final causes, because (a) they are general (and therefore possibilities, not actualities), and because (b) they explain the tendencies to behave according to definite patterns. In the next subsection, we will see that in biological species the defining cause is also a chemical structure.
9.3
The biological species
Peirce was a chemist by training, who from his youth onwards had shown a serious interest for the classification of the chemical elements. It may be assumed that his interest for the problem of natural classes arose within that context. This idea is confirmed by the fact that he devoted many more pages to the chemical elements than, for example, to the biological species. This may also explain why Peirce tried to apply his findings about chemical kinds to the biological kinds. Thus he sought the metaphysical essence (final cause) of biological species in their internal structure, which he identified with the chemical constitution of their protoplasm. He felt confident that future research would show that the chemical constitution of the protoplasm is "the sole determining cause of the forms of all animals and plants" (CP 1.262; italics mine). This leads us to believe that, if Peirce had known modern molecular biology, he would not have hesitated to consider the chemical structure of DNA as the metaphysical essence of biological species. DNA is precisely that part of the protoplasm that determines the essential morphological and functional characters of the biological species. Moreover, DNA is related to heredity. Thus, the cause of heredity is the chemical structure of DNA. And thus heredity must be related to final causality: Heredity [...] is not a force but a law, although, like other laws, it doubtless avails itself of forces. But it is essentially that the offspring shall have a general resemblance to the parent, not that this general resemblance happens to result from this or that blind and particular action. No doubt, there is some
FINAL CAUSES ANI : NATURAL CLASSES
127
blind efficient causation, but it is not that which constitutes the heredity, but, on the contrary, the general resemblance. (CP 1.215; 1902)
Thus, whereas classification is always classification according to form, in biological species, the form is the expression of the internal structure of DNA. Because DNA is the final cause of the biological class, it may also be said that classification in biological species is classification according to their final cause. Peirce's approach was broadly Aristotelian inasmuch as natural classification always concerns the form of things (which is that by virtue of which things are what they are) and not their matter. This entails that Peirce borrowed Aristotle's idea that the form is identical to the intrinsic final cause. Therefore it was obvious that natural classification concerns the final causes of the things. From the natural sciences, Peirce had learned that the forms of chemical substances and biological species are the expression of a particular internal structure. He recognized that it was precisely this internal structure that was the final cause by virtue of which the members of the natural class exist. To summarize: whereas natural classes are not defined in terms of essential qualities, but in terms of a final cause (and therefore in terms of possible behavior), the final cause may yet be expressed in some empirical internal structure. The chemical substances, the biological species, and art objects are Peirce's main examples of such natural classes. In these cases, similarity of internal structure indicates that objects belong to the same natural class. In the next section I will discuss Peirce's alleged pluralism regarding natural classes.
10.
WAS PEIRCE A PLURALIST REGARDING NATURAL CLASSES?
I started my investigation by giving a survey of the interpretations of Peircean natural kinds that were given by, respectively, Haack (1992), Hookway (1985), and Rosenthal (1994). One of the most important questions raised by them, was: how pluralistic is Peirce's conception of natural class? Before tackling this question, it may be helpful to distinguish three meanings of pluralism, two of which I borrow from John Dupré. Pluralism (1), as opposed to reductionism or eliminativism, refers to "the insistence on the equal reality and causal efficacy of objects both large and small" (Dupré 1993, 7). This pluralism rejects in principle the reduction of macro-objects to
128
CHAPTER FOL R
subatomic particles. Eliminativism, in its most extreme form, would lead to the conclusion that there is only one natural class: the fundamental particles or processes of physics; micro-reductionism wants us to believe that causes at our normal, common sense level of awareness, are not real. I will call the pluralism that rejects eliminativism and micro-reductionism causal pluralism. Pluralism (2), as opposed to classical essentialism, is "the claim that there are many equally legitimate ways of dividing the world into kinds." Our classifications are partly determined by our interests or purposes, and partly by "the recalcitrance of nature." This pluralism rejects classical essentialism because it denies the idea that things possess essential properties (which are both necessary and sufficient for a thing to belong to a natural class), independently of any context of inquiry. But it maintains that our activity is constrained by events beyond our control. Questions like, To which natural kind does this object belong? are always relative to a context, that is to say, "such questions can be answered only in relation to some specification of the goal underlying the intent to classify the object" (Dupré 1993, 5-6). For this type I borrow Dupré's terms promiscuous realism or radical ontological pluralism (Dupré 1993, 5-18). However, 'pluralism' may also be the doctrine that we are entirely free to classify the world as we would like to. Accordingly, pluralism (3) may be defined as the claim that there are infinitely many equally legitimate ways of dividing the world into kinds; our classifications are not restricted by any 'recalcitrance of nature.' Of course, the use of the prefix 'natural' to 'class' would become disputable, but the fact is that there are philosophers who speak of natural kinds in this way. I will call this type anarchistic pluralism. Whereas pluralism 1 is compatible with pluralism 2 and 3, pluralisms 2 and 3 are incompatible. From my description in section 2, it appears that Haack considers Peirce to be a causal pluralist, Rosenthal sees him as a promiscuous radical ontological pluralist, while Hookway is bound to call him an anarchistic pluralist. I will discuss these interpretations successively. That Peirce was not a pluralist in the promiscuous sense (pluralism 2) and even less so in the anarchistic sense (pluralism 3), appears clearly from the following statement in which he explicitly denies the possibility of more than one system of natural classification: "there are artificial classifications in profusion, but [there is] only one natural classification" (CP 1.275; 1902; italics mine). From my previous discussion it has appeared that natural classes are primarily determined by one and only one defining idea. This defining idea or D-character is the final cause to which the members owe their existence as members of the class. However, as a result of subsidiary purposes, there
FINAL CAUSES ANI : NATURAL CLASSES
129
are different levels of natural classes: "there are several different categories of secondary and subordinate purpose. These categories of purpose must be categories of every system of natural classification" (MS 1343:16-17; 1902). Accordingly, "each class is distinguished by performing its part of the general purpose of the branch [the more general class], or carrying out the general idea in a special way" (MS 1344:15). Thus, a racing car belongs to the natural class of racing cars, as well as to the (more general) natural class of cars, and also to the (even more general) natural class of vehicles. Though natural classes may partially overlap with each other, and each object probably belongs to more than one natural class, there is but one natural system of classification. In this sense Peirce was a monist, not a pluralist. The idea that objects belong to unambiguously discoverable natural classes, is intimately connected with Peirce's specific essentialism: what makes an object belong to a particular natural class is that it be teleologically caused by the D-character. The D-character unambiguously determines to what natural class an object belongs, independently of any context of inquiry. This interpretation is opposed to Rosenthal's, according to which the concept of a Peircean natural class is intimately related to a context of inquiry. Though she does not doubt that it is Peirce's view that within a certain commonly accepted context, investigation can "tend toward an ideal limit of convergence," she insists that it was not Peirce's view that there is also a convergence of contexts. There is not one ultimately correct context which clearly determines to what natural class an object belongs: When a community is operating within a common system of meanings on any one issue, then investigation can tend toward an ideal limit of convergence. However, when different segments of interpreters experience different facts because of different sets of meaning structures for cutting into the indefinitely rich continuity of possibilities of ordering, such convergence cannot occur. The criterion for adequately cutting is workability, but workability can be established only relative to some meaningful network by which experience is "caught." Thus there may be a pluralism of interpretators among varying groups of interpretations on any topic. For each group, identifiable by varying nets or perspective for the catching of experience, is variously structuring some contours of a world. [...] [T]he essential pluralism is often hidden from the view in the misplaced drive toward a common conclusion based on "the evidence." (Rosenthal 1994, 17)
Against Rosenthal I would argue that Peirce's view that in the long run science will discover the final causes of things, entails that there is one ultimate context which is not determined by our purposes, but by the purposes (final causes) of the things themselves. Though there may be countless goals underlying our intent to classify, there is but one natural
130
CHAPTER FOL R
classification. It is not our purpose (final cause) that is at stake, but the final cause of the natural class itself (see section 5). And in that sense, Peirce was not a 'radical onto logical pluralist.' When he writes that "truly natural classes may, and often do, merge into one another inextricably" (CP 1.209), he only meant to say that, though each object definitely belongs to a natural class, there are no sharp demarcation criteria by virtue of which it can always be unambiguously decided to which of two closely related classes an object belongs. Whenever closely related classes are not separated by sharp lines of demarcation, non-TDE (teleologically determined empirical) characters may help us in ascertaining to which of the two classes the object belongs: "unless we have some supplementary information we cannot tell which ones had one purpose and which the other" (CP 1.208). If Peirce's definition of a natural class as "a class of which all the members owe their existence to a common final cause" (CP 1.204) is taken seriously, one cannot but conclude that, according to Peirce, even the members of apparently fuzzy classes belong to specific natural classes, regardless of the difficulty we experience in our attempt at classification. If natural classes are not even partly determined by our interests and conventions, then Hookway's position that "[generality is real, but dividing things into classes reflects our interests and conventional decisions" must also be rejected (Hookway 1985, 251). Hookway concurs with Peirce when he writes that natural classes are real, but he fails to see the implication for natural classes. Natural classes are real, because they are generals. Though dividing things into classes may reflect our interests, natural classes do not. Hookway's interpretation may be due to his mistakenly conceiving natural classes as concrete, existing things. But such conception is the result of a category mistake. But, whereas Peirce was neither a 'radical ontological pluralist' nor an 'anarchistic pluralist,' he most certainly defended a causal pluralism. There are many different levels of natural classes (physical entities, chemical entities, biological entities, sociological entities, artifacts, etcetera), and therefore also many different levels of causation. Each class is characterized by a distinctive final cause, and the objects belonging to a natural class do so by virtue of their ability to exert a type of real causal influence. Thus, it is the task of science to determine exactly what natural classes there are. The idea that Peirce was a causal pluralist agrees with Haack's interpretation of Peircean natural kinds. As we have seen, Haack correctly pointed out that Peircean natural kinds are "the kinds of things in the world that really do behave in a law-like way."24 If so, examples of natural classes are horses, men, and all kinds of other macroscopic objects. 25 She therefore clearly acknowledged that Peirce thoroughly rejected any kind of micro-
FINAL CAUSES ANI : NATURAL CLASSES
131
reductionism. Not only the fundamental processes of physics can have a real causal influence, but so can men, horses, and all kind of other macro-objects. But Haack's paper was not intended to give a detailed account of Peircean natural classes (neither were the relevant parts in Rosenthal's and Hookway's books). Hence we should not be surprised that her picture is in some ways incomplete. For instance, she fails to provide a precise enough definition and precise demarcation criteria, which might help us, distinguishing natural classes from artificial ones. Moreover, by not considering Peirce's theory of natural classes within the context of his theory of final causation, she was not able to explain why the world of physics is not more real than our common sense world, or why there is no ontological conflict between the objects of scientific inquiry and the objects of other endeavors. Scientific objects belong to natural classes, but so do man-made objects, because in both cases the existence of these objects is determined by a final cause, which is common and specific to the members of the class. Finally, she also fails to give a clear insight into the problem of the relationship between natural classes and causation. In the next and last section, I will summarize the obtained results.
11.
CONCLUSION: NATURAL CLASSES AND CAUSATION
The problem of natural classes is important because it is inextricably linked to several philosophical notions, such as induction, universals, scientific realism, explanation, causation, and natural law. The main concern of this chapter has been the relationship between natural classes, causation and natural laws. Natural classes are often seen as the kinds of things that behave in a law-like way; objects belonging to the same natural class are considered to play the same causal role in nature. This somehow involves the notion that lawful causal relations presuppose that there are natural classes. In this chapter, it was established that Peirce's mature discussion of natural classes was intimately related to his theory of causation. According to this theory, while each event is part of a continuous chain of events, each chain of events is characterized by some kind of tendency. Thus, each act of causation has an efficient and a final component. The efficient aspect of causation is that each event is caused by a previous event (the efficient cause); the teleological aspect is that each event is part of a causal chain with a definite tendency. Although this is an original and provocative theory, Peirce unfortunately still clung to the expression 'final cause.' This expression is misleading, because what Peirce meant by a 'final cause' was altogether different from what we nowadays call 'a cause.'
132
CHAPTER FOL R
Peirce's originality in respect of natural classes concerns at least two insights: first, he made clear that all classification, be it natural or artificial, must be related to some purpose. Secondly, natural classifications do not primarily involve our purposes, but the final causes of the classified things themselves. Accordingly, Peirce's view may be summarized as follows: Things belong to the same natural class on account of a metaphysical essence and a number of class characters. The metaphysical essence is a general principle by virtue of which the members of the class have a tendency to behave in a specific way; this is what Peirce meant by final cause. This finality may be expressed in some sort of microstructure. The class characters, which by themselves are neither necessary nor sufficient conditions for membership of a class, are nevertheless concomitant. In the case of a chair, the metaphysical essence is the purpose for which chairs are made, while its having chair-legs is a class character. The fuzziness of boundary lines between natural classes is due to the fuzziness of the class characters. Natural classes, though very real, do not exist entities; their reality is of the nature of possibility, not of actuality. The primary instances of natural classes are the objects of scientific taxonomy, such as elementary particles in physics, gold in chemistry, and species in biology, but also man-made objects and social classes. In respect of the contemporary discussion, Peirce's view involves a rejection of micro-reductionism and eliminativism as viable theories of natural classes; Peirce's theory, which I have labeled a causal pluralism (because it insists on the equal reality and causal efficacy of both micro- and macro-objects), does not reduce our common sense, daily experience in favor of some abstract, physical principles. Though the scientific method may yield knowledge of natural classes, there are many obvious examples that are derived from common, human experience. By denying that final causes are static, unchangeable entities, Peirce avoided the problems attached to classical essentialism. On the other hand, by eliminating arbitrariness, Peirce also avoided pluralistic anarchism. Though Peircean natural classes only come into being as a result of the abstractive and selective activities of the people who classify, they reflect objectively real general principles. Thus, there is not the slightest sense in which they are arbitrary: "there are artificial classifications in profusion, but [there is] only one natural classification" (CP 1.275; 1902). Before presenting my semeiotic approach to causation in chapter 6, I must first discuss the basic conceptions of Peirce's semeiotic. Moreover, I must discuss one of its most serious problems: the role of causal elements within semeiosis. Thus my next chapter will provide a discussion of the socalled problem of semeiotic causation.
Chapter 5 THE RIDDLE OF SEMEIOTIC CAUSATION
In order that a Sign should truly represent that which it undertakes to represent, it must be caused, or, to use a wider term, must be determined by that Object; and then it must determine the mind that it addresses in such a way that that mind is in turn determined mediately by that Object. This is my definition of a Sign and it applies even to mendacious Signs. Of course the objection that would be raised if I used the word "cause" in place of "determine" would be that in that case there could be no Sign of the future ... (C.S. Peirce, CSP. SB 12/8/1909)'
In contemporary Peirce scholarship, there is a consensus that Peirce conceived semeiosis or sign-action as a teleological process, that is to say, as a process directed toward the complete interpretation of the sign.2 If semeiosis is indeed some sort of teleological process, then it must, according to Peirce, involve a combined action of final causation, efficient causation, and chance. 3 However, the secondary literature on Peirce's semeiotic does not provide a clear and unambiguous view of the roles of these causal elements within semeiosis. The objective of this chapter then is to clarify the roles within semeiosis of, respectively, final causation, efficient causation, and chance. For the sake of clarity, I will use the expression 'semeiotic causation' for the role of causal elements within semeiosis. 4 The structure of this chapter is as follows: in the first section I will give a sketch of the basic conceptions of Peirce's semeiotic. In the second and third section, respectively, I will briefly explain the views of two well-known Peirce scholars - T.L. Short and Joseph Ransdell - who have written some important papers that are relevant to our subject. In the fourth section, I will list some problems raised by Short's and Ransdell's interpretations. Two of 133
134
CHAPTERFOLR
the most important of them are: (1) Does the dynamic object have any specific causal role in semeiosis?, and (2) Does the term to determine which is a key concept in Peirce's semeiotic - carry a causal connotation, and if so, in what specific sense(s)? Thus, in section five, I will discuss Ransdell's suggestion that the dynamic object is the final cause of semeiosis. Since it will appear that there is conflicting evidence regarding the role of the dynamic object in semeiosis, the problem will be approached from a different perspective in section six. More in particular, I will inquire whether Peirce's classification of signs into icons, indices, and symbols, may throw some new light on the problem of semeiotic causation. After that, in section seven, I will discuss the meaning of the term to determine within the context of semeiosis. Finally, in section eight, I will summarize the results of my analysis.
1.
SOME FUNDAMENTAL CONDITIONS OF SIGNS AS SUCH
Peirce's writings on semeiotic may be divided roughly into two periods: the period from about 1865 till 1873 (early writings), and the period from 1885 till the end of his life (later writings). From 1874 till 1885 Peirce did not write on semeiotic. Before examining Short's and Ransdell's view about semeiotic causation in the next section, we must have at least some rough idea of the basic conceptions of Peirce's early semeiotic, and of the more complicated views that Peirce would elaborate in his later writings.
1.1
Early basic insights
Peirce used the term semeiotic for the theoretical refinement of our common sense idea of sign. Whereas the young Peirce defined semeiotic as "the science of representations" (Wl: 303; 1865), Peirce later described it as "the scientific study of semeiosis," or the systematic explication of the formal structure of "the general conditions of signs being signs" (1.444; c.1896). 5 From the start, Peirce sees the sign in terms of a triadic relation between sign, object, and interprétant. His conception of a sign as a triad meant a fundamental and indeed a revolutionary break with the traditional analyses of a sign (from Aristotle and the Stoics to de Saussure)6, in which the sign was a member of a dyadic relation between signifier and signified. Obviously, the most innovative part of Peirce's semeiotic was his conception of the interprétant, which he defined as "a mediating representation which represents the relate [ the sign I to he a representation of the same correlate
THE RIDDLE OF SEMEIOTIC CAUSATION
135
[the object] which this mediating representation itself represents" (Wl : 523; 1866). Peirce used the term 'interprétant,' "because it fulfills the office of an interpreter who says that a foreigner says the same thing which he himself says" (ibid.). Thus, Peirce explains, suppose I don't understand the French word homme, and I look it up in a French-English dictionary. I find there the word 'homme' followed by 'man.' Given that, in a dictionary, two nouns printed side by side signify the same thing, I conclude that whatever man represents un homme also represents. But, apart from the dictionary, I know that men are, say, rational male two-legged creatures. Therefore, homme is a sign, of which man is the object, and rational male two-legged creature is the interprétant. Rational male two-legged creature is a mediating sign that represents homme to be a sign of the same object (man) that this mediating sign itself represents (W2: 53; 1867). Peirce developed his early semeiotic exclusively within the context of a theory of thought. Not only did he defend the view that "all thought is in signs," but he also held that the sign qua sign (that is, the sign not qua signvehicle, but qua representative function) and the interprétant are themselves of the nature of thought (W2: 223-25; 1868). According to Peirce, each interprétant is itself a further sign, and as such a translation of an earlier sign. Every sign 'lives' through its translation in a virtually infinite series of interprétants: "a representation is something which produces another representation of the same object and in this second or interpreting representation the 1st representation is represented as representing a certain object. This 2nd representation must itself have an interpreting representation and so on ad infinitum so that, the whole process of representation never reaches a completion" (W3: 63-64; 1873). Thus, "there is no exception [...] to the law that every thought-sign is translated or interpreted in a subsequent one, unless it be that all thought comes to an abrupt and final end in death" (W2: 224; 1868). Peirce distinguished three different kinds of signs: icons (or copies), indices, and symbols. Icons refer by virtue of a similarity of sign and object; indices refer by virtue of a causal relationship between sign and object,7 and symbols refer by virtue of a conventional (or law-like) relation between sign and object. Thus, a map is an icon; a weather vane is an index, and a word is a symbol (Wl : 323; 1865; W3: 65; 1873). In his later writings, Peirce would further develop his distinction, and make clear that icons refer to qualities, indices refer to concrete existents, and symbols refer to general types 8
136
CHAPTERFOLR
1.2
Later developments
1.2.1
The sign
Though his writings on semeiotic show a remarkable evolution in many respects, Peirce always remained loyal to his basic insight that a Sign is a first item so determined by a second item, called its object, as to determine a third item, called its interprétant, to be determined by the same object. This basic insight is expressed by the following later definition: [A sign is] anything which is related to a Second thing, its Object, in respect to a quality in such a way as to bring a Third thing, its Interprétant,
into
relation to the same object... (CP 2.92; 1902)
Thus, a sign involves a three-term relationship between a sign, its object, and its interprétant. This relationship is irreducibly triadic, that is to say, a relationship that is not reducible to a summation of dyadic relations. Notably, Peirce's definition does not refer in any way to an interpreter, who apparently has no place in the formal analysis of a sign. Moreover, it appears that a sign is not a thing as such, but it is something which is what it is in virtue of its relation to an object and an interprétant, thus, anything may be a sign provided it is triadically related to an object and an interprétant. Moreover, every sign involves a virtually infinite series of interprétants. This idea is best expressed in the following definition of sign, which was written by Peirce for Baldwin's Dictionary of Philosophy and Psychology (1901): [A sign is ajnything which determines something else (its interprétant)
to
refer to an object to which itself refers (its object) in the same way, the interprétant becoming in turn a sign, and so on ad infinitum. (CP 2.303; 1901)
However, whereas Peirce has always stuck to his view that every sign must have an interprétant, there are some indications that from 1902 onwards he held the view that an interprétant could be a mere possibility.9 Thus, for example, in 1902 Peirce writes, "It is not necessary that the Interprétant should actually exist. A being in futuro will suffice" (CP 2.92). Thus, in the same year, he comes to a new definition of sign: A Sign, or Representamen called its Object, Interprétant,
is anything which so stands in relation to a second,
as to be capable of determining a third, called its
to be in the same triadic relation to that object in which it stands
itself. That is to say, the interprétant must be itself a sign capable of determining a sign of the same object, and so on, endlessly. (MS 478:43; 1903)
THE RIDDLE OF SEMEIOTIC CAUSATION
137
Though this definition clearly expresses Peirce's later conception of sign, there are indications that he was not quite satisfied with it, for after 1903 he made many attempts to redefine it. 1.2.2
The object of the sign
Peirce never offered a precise definition of object as used in the expression 'object of a sign.' Apart from the formal characteristic of being "that for which the sign stands" (CP 1.339; c.1895), the object of the sign does not have any specific characteristics. Thus anything may be an object: qualities, actual or possible events, facts, things, collections of things, relations, laws, and so forth (CP 2.232; 1910), as long as it is triadically related to both a sign and an interprétant. From 1902 onwards, Peirce made a distinction between the immediate object and the dynamic or real object, which are not two different existents, but two different aspects of one and the same object. Whereas the immediate object is "the idea which the sign is built upon," the real object is "that real thing or circumstance upon which the idea [the immediate object] is founded, as on a bed-rock" (MS 318:00071; 1907). The distinction refers to the difference between the object as it is represented at a specific moment, and the object as it really is, irrespective of the way in which it is represented at a given moment. Put differently: the distinction is between the object as it appears to be at a particular moment, and the object as it is known when our scientific knowledge would be complete. The immediate object, or the object as represented by the sign, is inherent to semeiosis, for "its being is dependent upon the representation of it in the sign" (CP 4.536; 1905). The dynamical object, however, is in some sense transcendent to the sign-action: it is "an object really outside the sign" (MS 634:00025; 1909). In the logical context of disinterested inquiry, the dynamical object is that object to which our thoughts must conform in order to be true (CP 8.183).'° 1.2.3
The interprétant
The concept of the interprétant is the most difficult part of Peirce's semeiotic. Not only did the concept evolve, but Peirce never succeeded in completely clarifying it, a fact that he was well aware of." Peirce's later descriptions of the interprétant varied, depending upon different aspects of the notion that the context seemed to require. Though Peirce mostly described it as the effect produced by a sign upon the mind of some person who understands the sign, the interprétant need not be the effect of a sign on a person. A more general definition of interprétant is "that
138
CHAPTERFOLR
which the sign produces" (CP 4.536; 1906), or "the signification or interpretation" of a sign (CP 8.184),' 2 or "the proper significant effect of a sign" (CP 5.473; 1907). From 1903 onwards, when he reviewed Victoria Welby's What is Meaning?, Peirce distinguished three types of interprétant: the immediate, the dynamical, and the, final interprétant. In a letter of 1909 to Lady Welby he gave the following short characterization: My Immediate Interprétant is implied in the fact that each sign must have its peculiar Interpretability before it gets any Interpreter. My
Dynamical
Interprétant is that which is experienced in each act of Interpretation and is different from that of any other; and the Final interprétant is the one interpretative result to which every Interpreter is destined to come if the Sign is sufficiently considered. The Immediate Interprétant is an abstraction, consisting in a possibility. The Dynamical Interprétant is a single, actual event. The Final interprétant is that toward which the actual tends. (LW: 111 ; 1909)
Thus, the immediate interprétant is, in Ransdell's words, "the range of possible interprétants of a given sign at a given time" (Ransdell 1986, 682). Being a mere possibility, the immediate interprétant is indeed a formal property of the sign, but not a distinct event in the process of semeiosis. The dynamic interprétant is an actualization of the immediate interprétant: the interprétant as it actually occurs. As an actuality it "consists in [the] direct effect actually produced by a Sign upon an Interpreter of it" (LW: 110; 1909). The final interprétant is the ideal interprétant toward which the semeiosis tends under favorable conditions, or "the effect that would be produced on the mind by the sign after sufficient development of thought" (CP 8.343; 1908; italics mine). "The Final Interprétant does not consist in the way in which any mind does act but in the way in which every mind would act. That is, it consists in a truth which might be expressed in a conditional proposition of this type: 'If so and so were to happen to any mind this sign would determine that mind to such and such conduct'" (CP 8.315; 1909). The final interprétant is a possibility in an altogether different sense in which the immediate interprétant is a possibility; it is a would be rather than a mere may be. In the context of scientific inquiry, the final interprétant is "that which would finally be decided to be the true interpretation if consideration of the matter were carried so far as that an ultimate opinion were reached" (CP 8.184; undated). Peirce also distinguished the emotional interprétant, the energetic interprétant, and the logical interprétant. He described the emotional interprétant as a feeling - for example the feelings aroused by a musical
139
THE RIDDLE OF SEMEIOTIC CAUSATION
performance -, the energetic interprétant as an action - for example the muscular effort incited by the command to ground arms -, and the logical interprétant as a habit or habit-change (CP 5.475-6; 1907). Whether or not this distinction is different from the immediate/dynamic/final distinction is a matter of dispute among Peirce scholars.' 3 To summarize: each sign involves a triadic relationship between a sign, its object, and its interprétant. Neither the sign as such nor its object as such has any specific characteristics. Anything may be a sign, and anything may be an object, provided it is triadically related to an interprétant. The interprétant must be understood, primarily, not as an act of interpretation, but as the sign's proper signifícate effect. In the next section I will briefly explain the view of T.L. Short, who has written some important papers that are relevant to the subject of semeiotic causation.
2.
T.L. SHORT
It has been generally agreed upon that, according to Peirce, semeiosis or signification involves a triadic relation of sign, object, and interprétant. In his paper on "Semeiosis and Intentionality" (1981b), T.L. Short argues that Peirce conceived semeiosis as triadic and as teleological. According to Short, however, it is important in this context to distinguish between actually being interpreted, on the one hand, and being interpretable, or significance, on the other. Whereas the former involves a triadic relation between a sign, its object, and an actual interprétant, the latter involves a triadic relation between a sign, its object, and a possible interprétant: ... we must distinguish significance from the relations on which it is grounded, that is the relations that justify the interprétant a sign elicits. And from this it follows that significance need
not be identified with
actually
being
interpreted. For the grounding relation that would justify an interprétant of a certain type suffices to make something a sign. Significance consists in being interpretable. (...) Peirce did not stop saying that every sign must have an interprétant, but from at least 1902 onwards, he identified that interprétant as a mere potentiality (e.g., at CP 2.242, 274, 275, LW: 111). (Short 1996, 490-91)
Given Short's identification of significance (or being interpretable) with the immediate interprétant (1981b, 214), I take this passage to mean that whereas every sign must have an interprétant, it need not have a dynamic interprétant; an immediate interprétant will suffice. Thus, considered from a formal point of view, significance consists in a triadic relation between sign, object, and immediate interprétant.
140
CHAPTERFOLR
Several authors have argued that icons and indices are not genuine signs, because they are not related to their objects in a triadic way. Short, however, explicitly rejects this interpretation; his point is that those who defend the view that only symbols are real signs ignore the distinction between significance and the ground of significance. "Significance is justified interpretability, while the ground is what justifies or determines this interpretability." Though icons and indices are not grounded in triadic relations - the relationship between sign and object is monadic in icons (resemblance), and dyadic in indices (existential)14 - their significance is nevertheless triadic because it involves the reference to a possible interprétant (Short, 1981b, 200-02). Thus, in the example of smoke as an index of fire, the ground of significance is the dyadic (or causal) relation between fire and smoke, whereas the significance or the immediate interprétant of the index is the idea that this smoke refers to that fire. This immediate interprétant, however, need not be actualized in an actual interprétant. According to Short, Peirce conceived semeiosis not only as triadic but also as teleological. In order to explain the relationship between triadicity and teleology in Peirce's semeiotic, Short refers to the only passage in the Collected Papers in which Peirce provides a detailed discussion of the triadic character of semeiosis: the paragraphs 5.472-473 of "A Survey of Pragmaticism" (1907). This text is a relatively small part of the very lengthy, and for the greater part unpublished manuscript 318, which according to Short is "perhaps Peirce's most important manuscript on signs" (Short 1996, 491). In those paragraphs Peirce makes a distinction between "intelligent, or triadic action" and "dynamical or dyadic action." Elsewhere in the same text (CP 5.484) Peirce identifies semeiosis as triadic action, as opposed to dynamical or dyadic action. In dyadic action, An event A, may, by brute force, produce an event, B; and then the event, B, may in its turn produce a third event, C. The fact that the event, C, is about to be produced by B has no influence at all upon the production of B by A [...]. Such is dyadic action, which is so called because each step of it concerns a pair of objects. (CP 5.472; 1907)
In triadic action, however, the fact that event C is about to be produced by B does have an influence on the production of B by A. Thus ... when a microscopist is in doubt whether a motion of an animalcule is guided by intelligence, of however low an order, the test he always used to apply [...] is to ascertain whether event, A, produces a second event, B, as a means to the production of a third event, C, or not. That is, he asks whether B will be produced if it will produce or is likely to produce C in its turn, but will
THE RIDDLE OF SEMEIOTIC CAUSATION
141
not be produced if it will not produce C in its turn nor is likely to do so. (CP 5.473; 1907)
On the basis of these passages, Short carefully explains that only triadic or intelligent action is characterized by the anticipation of a future state. Intelligent action involves the causation of one event as a means to another event as an end. Thus, the action is intelligent if it is teleological, that is to say, if event A produces event B in order to bring about event C. Thus, Short concludes, "it is that which is done as a means to something else that is produced triadically. If B is done as a means to C, it is B that is produced triadically, not C" (Short 1981b, 206). According to Short, there is only one way in which the A, B, C scheme applies to the sign situation. Thus he explicitly rejects Fitzgerald's reading of CP 5.473, as expressed in the following passage: "In saying that the interprétant is triadically produced, Peirce is claiming that the sign, through which the interprétant is determined, must be produced in view of the interprétant. The sign is useful because of its intended effect either on another or on the future self."' 5 Short rejects this interpretation for two reasons: first, it implies that every sign is produced in order to be interpreted, which obviously does not hold for natural signs: "fire does not produce smoke in order to draw attention to itself" (1981b, 206). Moreover, Short points out that Fitzgerald's interpretation is also inconsistent with some other statements that Peirce makes in CP 5.473. For in the same paragraph, Peirce writes not only that "... it seems to me convenient to make the triadic production of the interprétant essential to a 'sign'..." but also: "although this condition [the triadic production of the sign] is most usually fulfilled, it is not essential to the action of a sign." Short therefore concludes "What is essential to all semeiosis, according to Peirce, is not the triadic production of signs, but the triadic production of interprétants." Besides, Short concludes that "As is clear from his description of triadic or intelligent action, in 5.473, as being the production of one thing as a means to another, Peirce regarded all triadic action as teleological or, in other words, as involving the influence of final causes. Semeiosis is, therefore, an example of a teleological process" (1981b: 206-7). Thus, the one way in which the ABC scheme applies to the sign situation is, according to Short, that the dynamic interprétant, B, is produced by the sign, A, as a means toward the realization of the final interprétant, C. According to Short, semeiosis owes its teleological character to the goaldirected activity of some interpreter, be it a human being or some other living creature. In order to explain the teleological character of semeiosis, Short refers to a passage from a letter of Peirce to Lady Welby ( 1909; quoted
CHAPTERFOLR
142
in our previous section) in which Peirce introduced a division of interprétants into the immediate, the dynamic, and the final, and in which he described the final interprétant as "the one Interpretative result to which every Interpreter is destined to come if the sign is sufficiently considered." 16 According to Short, ... the idea of a final interprétant presupposes that of a goal of interpretation. For, apart from such a goal, "consideration" of a sign would not lead the interpreter to any "destined" conclusion: regardless of the amount of consideration made, any conclusion would remain just as good. But if each sign has a unique final interprétant, then each sign is the sign that it is in relation to a ground but also to a goal of interpretation.
It is clear then, that
Peirce conceived of semeiosis as a teleological process and of signs as being what their potential role in semeiosis makes them to be. (Short 1981b, 214; italics mine)
Thus, according to Short, a sign is determined both by its ground and by the goal of interpretation of the interpreter. Consider the example of a fingerprint left behind by a thief. The fact that fingerprints are unique to a person is the ground of the sign; the potential role in semeiosis of a fingerprint is determined by this unique relationship. But, a sign as such functions only in a particular context, which is determined by the goal of the interpreter. In case of the fingerprint the particular context may involve a policeman's wish to solve the theft. Thus, the ideal interprétant toward which the semeiosis tends under favorable conditions "depends on the sign and the goal of interpretation" (1981b, 213). This so-called final interprétant has to be understood as "that which would best fulfill the interpreter's goal" (1996, 507). The teleology involved in semeiosis is therefore due to the goal of interpretation of the interpreter. Hence, it is because of the policeman's wish to solve the theft that the fingerprint functions as a sign of the theft, and that a series of interprétants is generated which will ideally lead toward its complete interpretation, and by virtue of that, to the arrest of the thief. Thus, it would appear that, if our interpretation of Short's view is correct, there are two goals or final causes involved in semeiosis. First, there is the goal of the interpreter: his intention to interpret a particular sign in the light of some purpose. But, because this final cause is not a part of the formal structure of the sign, it is in some sense extra-semeiotic. The second goal is the final interprétant, which, according to Short's interpretation, must be an intra-semeiotic final cause.17 Thus, Short's view would raise at least two problems, which I will consider successively. The first problem concerns the supposed necessity of some interpreting agency: Short's view that the teleology involved in semeiosis presupposes the goal-directed interpretation of some living
THE RIDDLE OF SEMEIOTIC CAUSATION
143
creature (such as for example a Paramecium, the goal of which is nourishment or survival, following a chemical trace that functions as a sign of food) 18 implies that semeiosis presupposes the activity of an interpreting agency.19 This assumption is remarkable because in the last section of his paper on "Semeiosis and Intentionality" Short fully acknowledges that Peirce's conception of semeiosis was so broad as to include irreversible physical processes, which, as Short has shown in another paper,20 are intrinsically teleological. Short concludes that: "Peirce's theory of signs [...] exhibits the continuity of the human mind with the rest of nature" (1981b, 220). Our question then is: if Short's view that Peirce's conception of semeiosis covers all natural processes is correct (as I think it is), how can these processes owe their teleology to some interpreting agency? Our second problem concerns the relationship between Short's conclusion that semeiosis is teleological and his rejection of the view that the word 'determines,' which is a key conception of Peirce's semeiotic, has a causal connotation. The problem is this: if Short is right in thinking that semeiosis is indeed a teleological process, then it must, according to Peirce, involve a combined action of final causation, efficient causation and chance. Thus, it would appear that in the relationship between object, sign, and interprétant, efficient causation (and chance) must somehow play an essential role. While Peirce often said that the object somehow determines the sign, and the sign somehow determines its interprétant, the hypothesis that 'determines' has a causal connotation (in the Peircean sense of causation) is plausible. Thus we would expect that the object somehow causes the sign and the sign somehow causes the interprétant. However, Short explicitly denies that 'determines' has a causal meaning. Referring to CP 8.177 and 6.625, Short concludes that [Peirce's] point is simply that a sign delimits what can be an interprétant of it. Similarly, an object determines or delimits its possible signs (Short 1981b, 221, n.3)
In other words: By 'determines' Peirce means 'delimits the possible' rather than 'causes' (8.177). The nature of an object delimits what can be a sign of it, and a sign delimits the class of its possible interprétants. (Short 1982, 290)
Our question is how Short's claim that semeiosis is a teleological process (that is to say, a causal process directed toward an end state) can be reconciled with his claim that 'determine' has no causal meaning. As far as I can see, there is only one way in which this question can be answered: the causation involved in semeiosis comes - in some way - from the side of the interpreting agent.
144
CHAPTERFOLR
In the next section, we will see that Ransdell does not share Short's view that sign action presupposes an interpreting agency, nor does he share Short's conviction that the term 'determine' does not have a causal connotation.
3.
JOSEPH RANSDELL
Contrary to Short, Ransdell emphasizes that Peircean semeiosis does not need an interpreting agency. Though Ransdell admits that it is possible to speak in terms of interpreting agents, his point is that "if we want to understand the representation relation properly we must avoid any ultimate (i.e. ineliminable) reference to an interpreting mind or to any interpreting agency other than signs themselves. That is, [...] any such reference [...] is legitimate only on the assumption that it is possible in principle to eliminate it by further semeiotical analysis." Moreover, "the term 'semiosis' should always be understood as referring to what the sign does as such, namely, generate interprétants of itself, and never construed as referring to an operation performed on a sign by a person such that the sign ipso facto acquires a meaning or interprétant." It is only because signs have "interpretant-generating powers," that they can be used by some interpreting agency.21 According to Ransdell, signs have three modal aspects: (1) they have a certain appearance; (2) they actually occur; and (3) they have "a power of generating interprétants." This power or disposition is not a rule that stands over and above the sign, but it is an immanent principle. Thus, each sign involves "a law determining the future course of semeiosis in virtue of its specific power of generating an interprétant of itself, affecting thereby the sequence of subsequent interpretation" (Ransdell x, 4). This is the reason why Ransdell characterizes semeiosis processes as 'autonomous' or 'selfgoverning' (x, 3). The idea of the "autonomy of the semeiosis process" implies that reference to interpreters and interpretational acts is always eliminable in principle. Thus, any puristic analysis must be restricted to the conceptions of sign, object, and interprétant (Ransdell 1986, 692). According to Ransdell, semeiosis may be understood either from a causal or from a logical point of view. The causal perspective is the perspective of the sign as sign-action: to act as a sign is to determine an interprétant. The logical perspective is the perspective of the sign as "referring to" or "standing for," or "representing" its object (Ransdell 1986, 675-84). Moreover, whereas expressions such as 'referring to' and 'standing for' carry only a logical sense, the word 'determines' expresses both a causal and a logical sense.
THE RIDDLE OF SEMEIOTIC CAUSATION
145
In its logical sense, [determines] means that whatever the sign refers to must be referred to by its interprétant [...]. In its causal sense, determines means that the sign causes or produces or generates its interprétant, such that the latter in turn causes or produces an interprétant, etc.. and such that, in the course of such a causal chain, there is a real tendency for the object to manifest itself. (Ransdell 1986. 683)
According to this description, semeiosis is a teleological process directed toward a complete manifestation of the object, in which the object seems to function as the final cause of the semeiosis process." Thus, if Ransdell is right, this chapter will cause a series of interprétants, such that in the course of that causal chain, there is a real tendency for our intention to manifest itself. Moreover, Ransdell contends that, though a real tendency always involves a three-term relationship, the production of an interprétant by a sign is in some sense a matter of dyadic, brute force causation. "Thus," he concludes, "semiotic causation is itself telic or final causation, and it presupposes but cannot be explicated in terms of brute force causation" (1986, 684). In order to explain Ransdell's point, we must return to the general characteristics of Peirce's conception of causation. As it was shown in chapter 3, Peirce held the view that each act of causation involves a final component, an efficient component, and a chance component. The teleological aspect is that each event is part of a chain of events with a definite tendency, which is determined by the final cause of the process. The efficient aspect is that each event or fact is caused by a previous event or fact. The chance component is that in each event there is some aspect of irreducible novelty. Thus, final causation involves a triadic relation between a general final cause, a concrete efficient cause, and its concrete effect. The production of the individual effect (B) by the individual efficient cause (A) is determined, or mediated, by the general final cause (C). The effect (B) functions as a means for the attainment of the end (C). Schematically this may be represented as follows: C' (final cause)
A z (cause)
[
* B •> C (effect, realized end) (effect, means)
^ stands for a chain of efficient causation which may or may
CHAPTERFOLR
146
not be mediated by the same general final cause].23 Since Peirce rejected the concept of backward causation, the causation of B by A cannot be influenced by C, but is instead determined by potentiality C'. Each stage of the causal process (level A-B-C in the diagram) involves objective chance. Now, given that Ransdell's interpretation of Peirce's basic trichotomy of interprétants agrees with our description of it in section l, 24 we may represent Ransdell's view schematically as follows: 25 C' (dynamic object)
A ¿ (sign)
> ì B (dynamic (d interprétant)
C (final interprétant)
Thus, the sign (A) is the efficient cause of the dynamic interprétant (B), which itself is a means toward the realization of the final interprétant (C). Thus, there is a chain of efficient causation from A toward C, in which each act of causation is mediated by the object (C), which is the final cause of the semeiosis process. Thus, Ransdell's words as quoted above are a sign of which our schematic representation is the dynamic interprétant, and Ransdell's intention is the dynamic object. The final interprétant would be a complete interpretation of Ransdell's intention. Ransdell's words are the efficient cause of the scheme, but his intention is the final cause. But what to say about objective chance in semeiosis? If our schematic representation of Peirce's conception of (final) causation is correct, and if semeiosis is indeed a teleological process, then each stage of the semeiosis process must involve objective chance. Expressions such as "objective indeterminacy in the sign itself' (Ransdell 1986, 682) and the "spontaneity of the signs themselves" (Ransdell x, 2) suggest that Ransdell indeed subscribes to the view that each stage in the semiosis process involves objective chance. If so, there is objective chance involved in semeiosis inasmuch as signs are never completely determinate, 26 and therefore permit a whole range of different dynamic interprétants. That is to say, every sign is indeterminate to the extent that it does not provide a complete representation of its object (W3: 84; 1873).
THE RIDDLE OF SEMEIOTIC CAUSATION
4.
147
SOME PROBLEMS GENERATED BY SHORT'S AND RANSDELL'S VIEWS
Our previous discussion has made clear that there are at least three important points on which Short and Ransdell seem to disagree: (I) the status of the final cause in semeiosis, (II) the precise meaning of the term 'determine,' and (HI) the supposed necessity of an interpreting agent. I will briefly consider these. (I) Whereas Ransdell points out the dynamic object as the final cause of semeiosis, Short seems to regard the final interprétant as the final cause. This disagreement indicates the need to further explore the relationship between dynamic object and final interprétant. I will do this in the fifth section. Since this exploration raises some further problems concerning the causal role of the dynamic object in semeiosis, Peirce's classification of signs into icons, indices, and symbols, which is based on the relationship of sign and dynamic object, will be addressed in section six. (II) The second problem concerns the meaning of 'determine'. According to Peirce, the object somehow determines the sign, and the sign somehow determines its interprétant. Whereas Short defends the view that 'determines' means 'delimits the possible' rather than 'causes,' Ransdell holds that the word 'determines' expresses both a causal and a logical sense. This problem demands the further exploration of the meaning of 'determines.' I will examine this problem in section seven. (İÜ) According to Short, semeiosis owes its teleological character to the goal-directed activity of some interpreting agency. Ransdell, on the other hand, holds the view that reference to interpreters and interpretational acts is always eliminable in principle: any puristic analysis must be restricted to the conceptions of sign, object and interprétant (Ransdell 1986, 692). But, Ransdell emphasizes that a puristic analysis that applies to all cases, has yet to be developed. He insists that such development is highly desirable, because what is at stake is nothing less than a new conception of mind (and of teleological processes in general) in terms of semeiosis processes: From a philosophical point of view what is at stake is the elimination of implicit reliance on a Cartesian conception of mind in favor of understanding the mental -, or, more generally, the psychical (that which exhibits a telic form) - in terms of semeiosis processes. (Ransdell 1986, 692)
Thus, Ransdell intimates that the activity of a mind would be a semeiosis process itself, and that therefore semeiosis cannot by itself involve the activity of some interpreting mind. Moreover, Ransdell does not only suggest that all semeiotic processes can be analyzed in terms of causal concepts (see our previous section), but also that each teleological process
CHAPTERFOLR
148
can be analyzed in terms of semeiotic concepts. And since final causation and efficient causation go hand in hand, the suggestion is there, though implicit, that causation can be analyzed in terms of Peirce's semeiotic. However, this challenging idea will not be discussed in this chapter; it will be the object of the following chapter.
5.
THE CAUSAL ROLE OF THE DYNAMIC OBJECT
This section consists of two subsections. In the first part I will consider some positive evidence for the hypothesis that the dynamic object is the final cause of semeiosis. In the second part I will consider some negative evidence against this idea.
5.1
Positive evidence
According to Ransdell (1977), the Peircean conception of the semeiotic object is the least understood. Ransdell holds that "what is meant in saying that every sign has an object is that every sign-interpretational process tends toward an end-state, that is, has a final causational form. That end state is the object of the process" (1977, 168). Since Peirce nowhere explicitly defines the semeiotic object in terms of final causes, it is important to see on what evidence Ransdell's claim is based. According to Ransdell, the following passage of the young Peirce is "completely explicit" on the identity of the final cause, the ultimate opinion, and the real object: I suppose that the fundamental proposition from which all metaphysics takes its rise is that opinions tend to an ultimate settlement and that a predestinate one. Upon most subjects at least [,] sufficient experience, discussion, and reasoning will.bring men to an agreement; and another set of men by an independent
investigation
with
sufficient
experience,
discussion,
and
reasoning will be brought to the same agreement as the first set. Hence we infer that there is something which determines opinions and which does not depend upon them. To this we give the name of reality. [...] to say that thought tends to come to a determinate conclusion, is to say that it tends to an end [,] is influenced by a final cause. This final cause, the ultimate opinion, is independent of how you, I, or any number of men think. Let whole generations think as perversely as they will, they can only put off the ultimate opinion but cannot change its character. So the ultimate conclusion is that which determines opinions and does not depend on them and so is the real
THE RIDDLE OF SEMEIOTIC CAUSATION
149
object of cognition. This is idealism since it supposes the real of the nature of thought. (MS: 393.2; c.1870; quoted in Ransdell 1977, 178, n.10.)
However, since the context of the discussion is not explicitly semeiotical,27 but concerns the explanation of the nature of reality, it is not at all obvious that Peirce meant by 'the real object of cognition' the same thing as the 'object' of semeiosis. Thus, the evidence for the view that, according to the young Peirce, the object is the final cause of semeiosis is dubious, let alone "completely explicit." However, this meager evidence is supplemented by some other indications from Peirce's later writings, especially from MS 318 (1907), in which he struggled with the problem of semeiotic causation. Thus, for example, Peirce wrote not only that the "influence" of the object upon the sign "is usually indirect and not of the nature of a force" (MS 318:00339), but also that... It may be asserted, [...] that in every case an influence upon the Sign emanates from its Object, and that this emanating influence then proceeds from the Sign and produces, and is capable of producing and partly at least, in a mental way, an effect that may be called the Interprétant, or interpreting action, which consummates the agency of the Sign. (MS 634:00023; 1909)
Thus, there is an influence from the object upon the sign and from the sign upon the interprétant, such that the sign produces the interprétant (at least partly) in a mental way. This seems to entail that the interprétant is produced by the sign under the influence of the object, which is the final cause; for, as Peirce said a few lines below the quoted passage, "it is strictly true that final causes do act mentally" (MS 634: 00024). Peirce further explained that there is indeed always an influence from the object upon the sign, even if the real object seems to belong to the future of the sign. Peirce gave the example of an advertisement in a newspaper for a theatre performance, which influenced or caused some people to visit the performance. There cannot have been any physical causation from the performance, via the advertisement, upon the people, for the performance was not there yet at the moment of the influence. Peirce concluded that the influence is clearly mental (MS 634:00024-25), which entails that the causation involved is final causation, not efficient causation.28 Because of its mental influence on sign and interprétant, Peirce called the object "the intelligential cause of the sign" (MS 318:00290). I will now consider some evidence that would testify against the idea that the dynamic object is the final cause of semeiosis.
150
5.2
CHAPTERFOLR
Negative evidence
The young Peirce defined a sign or representation as "an object which stands for another so that an experience of the former affords us knowledge of the latter" (W3: 62; 1873). The reason why signs can provide knowledge is that they stand for an object to some interpreting thought in some respect or quality (W2: 223; 1868). According to Peirce, each representation is characterized by three necessary conditions: It must in the first place like any other object have qualities independent of its meaning. It is only through a knowledge of these that we acquire any information concerning the object it represents. Thus, the word 'man' as printed, has three letters; these letters have certain shapes, and are black. I term such characters the material qualities of the representation. In the 2nd place a representation
must have a real causal connection with its object. If a
weathercock indicates the direction of the wind it is because the wind really turns it round. If the portrait of a man of a past generation tells me how he looked it is because his appearance really determined the appearance of the picture by a train of causation,
acting through the mind of the painter. If a
prediction is trustworthy it is because those antecedents of which the predicted event is the necessary consequence had a real effect producing the prediction. In the third place, every representation addresses itself to a mind. It is only in so far as it does this that it is a representation. (W3: 62; 1873; italics mine)
Thus, any representation, be it an icon (portrait), an index (weathercock) or a symbol (prediction), is characterized by a real causal connection with its object. The examples strongly suggest that Peirce had nothing but efficient causation in mind; there is not the slightest evidence that he meant final causation. Thus Peirce writes in a parallel manuscript "there is to be [...] a physical connection between every sign and its object" (W3: 67; 1873). The example of the prediction, however, is somewhat more complicated than those of the portrait and the weathercock: In that case it cannot be said that that which follows after has caused that which precedes it, the prediction, but if the event has been predicted it has been through some knowledge of its cause and this same cause which precedes the event also precedes some cognition of the mind which gave rise to the prediction so that there is a real causal connection between the sign and the thing signified although it does not consist in one's being the effect of the other but in both being the effect of the same cause. (W3: 67; 1873)
Thus, consider the example that we predict rain on the basis of a cloudy sky. Here the cloudy sky is the sign, and the possibility of rain is the object of the sign, while our idea that clouds predict rain is the interprétant. Obviously,
THE RIDDLE OF SEMEIOTIC CAUSATION
151
the object (possible rain) is not the cause of the sign (clouds), and, strictly taken, the sign (clouds) is not the cause of the object (possible rain) either. Both the clouds and the rain are due to a common cause, namely, the condensation of vapor. Thus, Peirce concludes, the relation of sign and object involves a causal relation such that "the causation may be either from the object to the sign, or from the sign to the object, or from some third thing to both; but some causation there must be" (W3: 82; 1873). Again, there is not the slightest evidence that Peirce had final causation in mind. A first appraisal of the positive and negative evidence in the writings of the young Peirce regarding the question whether the dynamic object is the final cause of semeiosis, intimates that the negative evidence outweighs the positive. Certainly, one should take into consideration the context within which Peirce expresses himself on the subject. The context of the passages upon which the view is based that the object is the final cause, is a discussion of "the nature of reality." The context of the passages upon which the view is based that there is an efficient causal relationship between sign and object, is "the nature of signs." And since it is only this latter context that is directly relevant to the problem of semeiotic causation, the young Peirce probably did not regard the object as the final cause of semeiosis, but as the efficient cause of semeiosis, or more precisely, as standing in a relation of efficient causation to the sign. In the later Peirce, there is also some evidence against the idea that he considered the dynamic object as the final cause of semeiosis. For example, in the following passage, taken from MS 318 (1907), he explained that the role of the object of a sign is not just to produce a mental effect, but that it must also convey meaning. Moreover, he explicitly described the object as an efficient cause: ... every sign, in functioning as such, produces a mental effect. How shall we name the entire mental effect which a sign by itself is calculated, in its proper significative function, to produce? The word signification
is somewhat too
narrow, since, as examples will soon show, this mental effect may be of the nature of an emotion or of that of an effort. No existing word is sufficiently appropriate. Permit me to call this total proper effect of the sign taken by itself, the interprétant of the sign. But merely producing a mental effect is not sufficient to constitute an object a sign; for a thunder-clap or avalanche may do that without conveying any meaning at all. In order that a thing may be a true sign its proper significate mental effect must be conveyed from another object which the sign is concerned in indicating and which is by this conveyance the ultimate cause of the mental effect. In order to be the cause of an effect, - or efficient cause, as the old phrase was, - it must either be an existing thing or an actual event. (MS 318:00)33-34; 1907)
152
CHAPTERFOLR
Thus, Peirce does not only call the object the ultimate cause of the interprétant, he also calls it an efficient cause. Moreover, by emphasizing that it must be either an existing thing or an event, he implicitly denies that it can be a potentiality or final cause. (Moreover, Peirce suggests that signs are teleological inasmuch as they are calculated to produce the interprétant. And by defining the interprétant as the "total proper effect of the sign taken by itself," the interprétant is conceived as the intended significate effect of the sign.) In the following passage, Peirce again emphasizes that the real object must be a real fact or thing. Moreover, he claims that "the logical interprétant does not correspond with any kind of object," because there is no object that is, like the logical interprétant, "in a relatively future tense." The object and the interprétant are [...] merely the correlates of the sign; the one being antecedent, the other consequent of the sign. Moreover, the sign being defined in terms of these correlative correlates, it is confidently to be expected that object and interprétant should precisely correspond, each to the other. In point of fact, we do find that the immediate object and emotional interprétant correspond, both being apprehensions, or are "subjective"; both, too, appertain to all signs without exception. The real object and energetic interprétant also correspond, both being real facts or things. But to our surprise, we find that the logical interprétant does not correspond with any kind of object. This defect of correspondence between object and interprétant must be rooted in the essential difference there is between the nature of an object and that of an interprétant; which difference is that the former antecedes, while the latter succeeds the sign. The logical interprétant must, therefore, be in a relatively future tense. (MS 318:00082-83; 1907)
Thus, there is a fundamental difference between the dynamic object and the logical interprétant inasmuch as the former is an actuality, while the latter is "in a relatively future tense," and as such a potentiality or final cause. Being an actuality, the dynamic object cannot have the status of final cause. Apparently, there is conflicting evidence regarding the role of the (dynamic) object in semeiosis. Whereas the young Peirce considered it to be the efficient cause of semeiosis, or more precisely, as standing in a relation of efficient causation to the sign, the later Peirce sometimes considered it as the final cause, and sometimes as the efficient cause. Thus, it would appear that the later Peirce was still struggling with the problem of semeiotic causation, and had no unambiguous conclusion about it. Within the context of his later discussions of semeiotic causation (especially in MS 318), Peirce sometimes regarded the dynamic object as an actual thing or event, and sometimes as a general idea or potentiality. However, there is one further complicating factor: in my general introduc-
THE RIDDLE OF SEMEIOTIC CAUSATION
153
tion to Peirce's semeiotic (1.2.2) I have referred to a passage of 1910 (CP 2.232) in which Peirce also mentions another type of dynamic objects: qualities or possibilities. Moreover, if we consider Peirce's classification of signs into icons, indices, and symbols, it appears that icons refer to qualities (possibilities), indices refer to concrete existents, and symbols refer to general types (potentialities or habits). This fact that the dynamic object may be entities of such different ontological status raises serious doubt regarding the hypothesis that there is one general, encompassing formula (applicable to all kinds of examples) of the role of causal elements within semeiosis. Moreover, though Peirce nowhere explicitly discusses the relationship between the triadic structure of the sign, and his classification of signs into icons, indices, and symbols, (the later Peirce used the symbol as the paradigm example of a sign in terms of which he discussed the triadic sign structure), this classification of signs offers strong evidence against the idea that there is one general formula of semeiotic causation. In fact, there are indications that his icon/index/symbol typology is precisely the expression of three different forms of semeiotic causation. In the next section, I will discuss the relevance of the icon/index/symbol trichotomy for the problem of semeiotic causation.
6.
ICON, INDEX, AND SYMBOL
Before starting this section, it is important to emphasize that icon, index, and symbol need not be different things, but may be different aspects of the same thing. A photograph, for example, is not just an icon, but also an index - as will presently be explained - and it even may have a symbolic function, as for example the photograph of some martyr who died for a political purpose. Indeed, the typology of icons, indices, and symbols concerns the predominant aspects of a sign, considered from a definite perspective. From a causal perspective, the relationship between sign and object is characterized by (1) the absence of a causal relationship in icons, (2) an efficient causal relationship in indices, and (3) a final causational relationship in symbols: A regular progression of one, two, three, may be remarked in the three orders of signs, Icon, Index, Symbol. The icon has no dynamical connection with the object it represents; it simply happens that its qualities resemble those of that object, and excite analogous sensations in the mind for which it is a likeness. But it really stands unconnected with them. The index is physically connected with its object; they make an organic pair, but the interpreting mind has
154
CHAPTERFOLR nothing to do with this connection, except remarking it, after it is established. The symbol is connected with its object by virtue of the idea of the symbolusing mind, without which no such connection would exist. (CP 2.299; C.1895)
Further on I will explain that the idea by virtue of which the symbol is connected with its object, has the connotation of a final cause. Besides, I will explain in some greater detail some of the main characteristics of icons, indices, and symbols in the light of their respective causal roles in semeiosis. According to Peirce, an icon is a sign that refers to its object on account of its similarity with it. Examples of icons are: a photograph of a person, a map of a terrain, and an analogy (CP 2.277). Though each icon has an object, the object of an icon need not exist (CP 2.304; 2.247); the design of a house (sign), for example, usually exists before the house (object) is built. Icons are characterized by the absence of a causal relationship between sign and object. Against those who object and say that the resemblance of photographs is due to a causal relation between object and sign, Peirce argues that photographs are only icons in respect of their similarity to their objects; considered from a causal point of view, however, they belong to the class of indices (CP 2.281; c.1895). An index is a sign that indicates an existing individual object. Thus, a pointing finger may indicate some object, personal pronouns (I, me, you, she, etc.) may indicate persons, demonstrative pronouns (this, that, these, those) may indicate persons or things, and smoke may indicate some fire (CP 2.305). Indices may be distinguished from icons and symbols by three characteristic marks: First, that they have no significant resemblance to their objects; second, that they refer to individuals, single units, single collections of units, or single continua; third, that they direct the attention to their objects by blind compulsion.
(...) Psychologically,
the action of indices depends
upon
association by contiguity, and not upon association by resemblance or upon intellectual operations. (CP 2.306; 1901)
Thus, indices always refer to existing individuals (actualities), never to mere qualities (possibilities) or to generals (potentialities). They don't refer to their objects because of a similarity, but because they are in some way contiguous with their objects. Moreover, they direct our attention to their objects by blind compulsion. For example, a pointing finger more or less blindly compels our attention onto a particular object. The main characteristic of indices is that they are causally or dynamically related to their objects, on the one hand, and to the mind of the interpreter of the sign, on the other hand. An index is:
THE RIDDLE OF SEMEIOTIC CAUSATION
155
A sign, or representation, which refers to its object not so much because of any similarity or analogy with it, nor because it is associated with general characters which that object happens to possess, as because it is in dynamical (including spatial) connection both with the individual object, on the one hand, and with the senses or memory of the person for whom it serves as a sign, on the other hand. (CP 2.305; 1901)
It is easy to give examples of indices in which the object is literally the efficient cause of the sign: a footprint in the sand as a sign of human presence (CP 4.531), a weather-vane as a sign of wind-direction (CP 2.286), the symptoms of a disease (CP 8.335), and a bullet-hole as a sign of a shot (CP 2.304). But, the causal relationship between sign and object is less obvious when, for instance, the index is a pointing finger or a demonstrative adverb such as 'here' or 'there.' According to Peirce, the index in such cases functions as ¿/there were an efficient causal relationship between object and sign. Consider the example that A wants to tell B that there is a fire: If A says to B, "There is a fire," B will ask, "Where?" [...] If A points his finger to the fire, his finger is dynamically connected with the fire, as much as if a self-acting fire-alarm had directly turned it in that direction; while it also forces the eyes of B to turn that way, his attention to be riveted upon it, and his understanding to recognize that his question is answered. If -4's reply is, "Within a thousand yards of here," the word "here" is an index; for it has precisely the same force as if he had pointed energetically to the ground between him and B. (CP 2.305; 1901)
Thus, Peirce holds that indices are characterized by a causal (dynamical) relation between sign and object, which are both actualities. Contrary to an icon, therefore, an index cannot exist as a sign without an existing object; a bullet-hole, for example, cannot exist without a bullet. But contrary to a symbol, an index may lack an explicit interprétant without ceasing to be a sign; the bullet-hole is there, regardless whether there is anybody to recognize it as such (CP 2.304; 1901). A symbol is a sign that represents its object by virtue of some conventional, habitual or law-like relation. A symbol is related to its object by a rule to the effect that it will be interpreted as signifying that object, regardless of any similarity or causal relation that it may have to that object. In Peirce's own words: A Symbol is a sign which refers to the Object that it denotes by virtue of a law, usually an association of general ideas, which operates to cause the symbol to be interpreted as referring to that Object. It is thus itself a general type or law (...). As such it acts through a Replica. Not only is it general itself, but the Object to which it refers is of a general nature. (CP 2.249; c. 1903)
156
CHAPTERFOLR
A symbol denotes its object by virtue of a law-like relation, which causes the symbol to be interpreted as referring to that object. The relationship between the symbol and its object is such that the symbol tends toward being interpreted in a certain way, this tendency being determined by a habitual or lawlike relation. Symbols are of a general nature, and the objects to which they refer are of a general nature too. As a general, a symbol does not exist; only replicas of symbols do exist. Thus, there is a strong tendency to interpret the American flag as referring to the United States of America, because it is a well-known conventional rule that each replica of the American flag stands for the United States of America. Since symbols are not concrete things, but generals, they cannot be efficient causes. As such they cannot exert any real force. Symbols are basically laws: ... a symbol [...] exists only in a replica, contrary to the nature of a real thing; and indeed the symbol only becomes a sign because its interpreter happens to be prepared to represent it as such. Hence, I must and do admit that a symbol cannot exert any real force. Still, I maintain that every sufficiently complete symbol governs things, and that symbols alone do this. I mean that though it is not a force, it is a law. (NEM IV: 250; probably 1904; 29 italics mine)
Symbols cannot be efficient causes because they are not concrete existents. They are laws, which is to say, final causes; for, "the relation of law, as cause, to the action of force, as its effect, is final, or ideal, causation, not efficient causation" (CP1.212; 1902). Indeed, Every sufficiently complete symbol is a final cause of, and 'influences,' real events, in precisely the same sense in which my desire to have the window open, that is, the symbol in my mind of the agreeability of it, influences the physical facts of my rising from my chair, going to the window, and opening it. (NEM IV: 254)
Thus, every (sufficiently complete) symbol is a final cause, that is to say, "a general formula to which real events truly conform" (NEM IV: 251). Moreover, each symbol involves an endless series of interprétants that conform to the general formula: "it is of the essential nature of a symbol that it determines an interprétant, which is itself a symbol. A symbol, therefore, produces an endless series of interprétants" (NEM IV: 261). In this endless series there is a tendency for the object to manifest itself: A symbol is essentially a purpose, that is to say, a representation that seeks to make itself definite, or seeks to produce an interprétant more definite than itself. For its whole signification consists in its determining an interprétant; so that it is from its interprétant that it derives the actuality of its signification.
THE RIDDLE OF SEMEIOTIC CAUSATION
157
[...] The interprétant aims at the object more than at the original replica and may be truer and fuller than the latter. (NEM IV: 261 ; 262)
Thus, symbols are representations with an inherent tendency to develop themselves into more definite representations. The purpose of a symbol is to produce an interprétant that better represents the object than the original replica. Thus, for example, the purpose of this chapter - which is a complex symbol - is to produce a series of interprétants such that, in the course of this series, there is a tendency toward a fuller representation of our intention, which is the dynamic object of this chapter. Thus, it would appear that a symbol owes its teleological character to its dynamic object. This hypothesis is confirmed by Peirce's discussion of the example of the imperative command of an officer of infantry to "ground arms!" The dynamic object of the command is the present will of the officer. The object is not the act of willing, but what it is that the officer wills, that is to say, the type of action that he wills. The effort of obeying the order is the energetic interprétant.30 Short suggests that the final energetic interprétant is the effort that expresses the order in the best possible way (Short 1981b, 216). Schematically this may be represented as follows:
C' (intention officer)
A Z A B (command) (effort soldiers)
C (effort that best expresses C')
Thus, the command A (sign) is the efficient cause of the effort B (dynamic interprétant). This cause-effect relationship is mediated by the intention of the officer, C' (dynamic object). The intention C' is the final cause of the semeiosis process. Because the object of a symbol is always general,31 it is a proper candidate for the role of final cause. In this scheme, the intention of the officer is the object of the sign, and indeed the final cause of the semeiosis process. One might object, however, that the above-mentioned command is a very specific example of a symbol, in which the object of the sign is the intention of the utterer. An intention is, obviously, a final cause, but what to think of those objects of a symbol that are not intentions? Peirce would argue, I suppose, that any object of a symbol is a final cause, for objects are basically general ideas, as we have seen, and general ideas are basically final causes.
158
CHAPTERFOLR
For each general idea is characterized by an inherent tendency to realize itself: "ideas are not mere creations of this or that mind, but on the contrary have a power of finding or creating their vehicles, and having found them, of conferring upon them the ability to transform the face of the earth" (CP 1.217; 1902). An idea abstracted from all efficiency is, according to Peirce, like a court without a sheriff (CP 1.213; 1902).32 Consider once more the example of a replica of the American flag, which represents the United States of America. The United States of America is not something physical, but is a conception or idea. By saying that "it is the idea which will create its defenders" (CP 1.217; 1902), Peirce in fact says that the idea of the United States of America involves a power of creating the means by virtue of which the idea becomes real. In our example, the replicas of the American flag are the means or vehicles by virtue of which the idea of the United States becomes real. Thus, it would appear that in symbolic signs the dynamic object is indeed the final cause of the semeiosis process. Moreover, it would appear that our schematic representation of Ransdell's view (section 3) indeed expresses the role of causal concepts in symbolic signs. In the case of icons and indices, however, the dynamic object cannot be the final cause of the semeiosis, because in those cases, the dynamic object is not of a general nature. Whereas only symbols refer to general types (potentialities), icons refer to qualities (possibilities), and indices refer to concrete existents (actualities). In indices the causal role of the dynamic object is altogether different from its role in symbols. This type of signs is characterized by a causal relationship between object and sign (from the object to the sign, or from the sign to the object, or from a third thing to both). Thus, there is a causal relationship between wind-direction (efficient cause) and the direction in which the weather-vane points (effect). The wind-direction is a concrete event, and cannot therefore be a final cause (which is always a general type). Since icons refer to qualities or possibilities, the objects to which icons refer can be neither final causes nor efficient causes. In icons there is no causal relation of whatever type between sign and object. There is, for example, no causal relation between a color sample and the color of a curtain of which it is a sample. Thus, it would appear that we cannot assign one specific, universally valid, causal role to the dynamic object. We can only say that in symbols the dynamic object seems to function as the final cause, whereas in indices it has the role of efficient cause. In icons, however, it has no causal role at all. But, would there not be an other suitable candidate for the role of final cause so that we can still hold the idea that there is one universally valid scheme of semeiotic causation? It would appear that the final interprétant
THE RIDDLE OF SEMEIOTIC CAUSATION
159
would be that candidate. For semeiosis involves a tendency towards the complete manifestation of the dynamic object in the final interprétant. The final interprétant, so I said in my general introduction, is the ideal interprétant toward which the semeiosis tends under favorable conditions. But, the view that the final interprétant is the final cause of semeiosis, is not unproblematic either. On the contrary. One should remember that Peirce held the view that final causation involves a triadic relation between a general final cause, a concrete efficient cause, and its concrete effect. But, if the final interprétant is indeed the final cause of semeiosis, then its function is the mediation of sign and dynamical interprétant. If this is correct, then the sign is the efficient cause and the dynamic interprétant is the effect. However, if the two basic assumptions of this chapter are true, that (1) semeiosis is indeed a teleological process, and (2) that each teleological process must be, in principle, completely analyzable in terms of final causation, efficient causation, and chance, what then can be the role of the dynamic object in terms of these causal concepts? It appears that there is no such role. For final causation involves a triadic relationship between a final cause, an efficient cause, and its effect. Now, if it is true that the final interprétant is the final cause, the sign is the efficient cause, and the dynamic interprétant is the effect, then the dynamic object cannot be part of the triadic sign-structure. However, since the dynamic object is per definition part of the triadic sign-structure, the final interprétant cannot be the final cause of semeiosis. To conclude: In his later discussion of semeiotic causation (especially in MS 318) Peirce did not distinguish different types of signs. This suggests that he was looking for one general formula that captures the role of causal concepts within semeiosis. However, a number of conflicting statements about the causal role of the dynamic object in semeiosis indicate that Peirce did not solve the problem. Moreover, our discussion of Peirce's typology of signs into icons, indices, and symbols has shown that Peirce's semeiotic does not imply one general formula of semeiotic causation. The suggestion made by Ransdell and others that the dynamic object is the final cause of semeiosis seems to hold only for symbolic signs. Only in this type of signs, can we precisely indicate the role of final causation, efficient causation, and chance. Thus, Ransdell's interpretation of semeiotic causation (which corresponds to our schematic representation of it in section 3) holds for symbols only. In indices and icons the role of causal concepts is highly problematic. Since the objects to which indices and icons refer are, respectively, concrete existents and possibilities, the final cause of these types of signs cannot be their dynamic objects (for final causes are potentialities). But it would appear that the final cause cannot be their final interprétant either, for that
160
CHAPTERFOLR
would introduce a fourth necessary constituent into semeiosis, which is definitely not what Peirce had in mind. Moreover, in the case of icons, (the later) Peirce explicitly denied that the dynamic object plays a causal role. However, this interpretation leaves us with one stubborn fact to be explained: how should we understand Peirce's numerous statements which mention the dynamic object as the immediate cause of the sign, and as the mediate cause of the interprétant (e.g. 6.347; 1909). Moreover, it again raises our question whether the verb 'to determine' carries a causal connotation, and if so, in what specific sense. I will discuss these questions in the next section.
7.
THE MEANING OF 'DETERMINES'
It seems appropriate to start our discussion of the meaning of to determine (or 'determines' or 'determined') by briefly restating Short's and Ransdell's view on the subject. According to Short, when Peirce pointed out that the object determines the sign, and the sign determines the interprétant, he simply meant to say that an object delimits its possible signs, and a sign delimits its possible interprétants (see 2). According to Ransdell, however, to determine has both a logical and a causal sense. In its causal sense it means that, whereas the object is the final cause of the semeiosis process, the sign is the efficient cause of the interprétant (see 3). Before discussing Ransdell's and Short's views, I will first consider Peirce's own explanations of the term determine. There are only a few places in which Peirce explains the term determine. These explanations, which were written in different periods of Peirce's life, show a remarkable similarity. Thus, whereas in 1868 Peirce wrote that he meant by determined or bestimmt "fixed to be this (or thus), in contradistinction to being this, that, or the other (or in some way or other)" (W2: 155-6), in 1903 he pointed out that it meant "specialized or bestimmt" (CP 8.177). Moreover, in 1908 he wrote that "the Object determines the Sign in a particular manner," that is to say, it "renders [the sign] definitely to be such as it will be" (CP 8.344). According to these descriptions, A determines B means that A fixes B to be this rather than that. Thus, it would appear that A determines B means that A is a necessary condition of B. (A is a necessary condition for B if, under the circumstances, in the absence of A, B would not have occurred.) However, though none of the above-mentioned descriptions of 'determine' involves any reference to causation, there is one passage - the one that was also quoted at the beginning of this chapter - in which Peirce explains the difference between 'determine' and 'cause':
THE RIDDLE OF SEMEIOTIC CAUSATION
161
In order that a Sign should truly represent that which it undertakes to represent, it must be caused, or, to use a wider term, must be determined
by
that Object; and then it must determine the mind that it addresses in such a way that that mind is in turn determined mediately by that Object. This is my definition of a Sign and it applies even to mendacious Signs. Of course the objection that would be raised if I used the word "cause" in place of "determine" would be that in that case there could be no Sign of the future ... (CSP. SB 12/8/1909) 33
According to this description, 'determine' is a more general term than 'cause,' and one reason why Peirce prefers the term 'determine' to 'cause,' is that there are signs of which the object does not exist - such as for example the design (sign) of a house (object) that still has to be build. In these cases it would be nonsense to say that the object is the cause of the sign. In most other cases, however, the terms 'determine' and 'cause' may be taken as synonymous. This supposed near identity of 'determine' and 'cause' raises the question, "What does Peirce mean by 'cause' in a context of semeiosis?" It would appear that he does not simply hold one unambiguous conception, for there are some conflicting statements about the role of causation in semeiosis. Thus, whereas Peirce said on the one hand that icons are characterized by the absence of a causal relationship between object and sign, on the other hand, he seems to suggest in the above-quoted passage that even in icons there is a causal relation between object and sign. However, the contradiction is only apparent if we realize that the term 'cause' is used in two different senses. Whereas icons are characterized by the absence of a (Peircean) efficient causal relationship between object and sign, the object is indeed a necessary condition of the icon. It would appear that in the above-quoted passage, Peirce must have understood the term 'cause' first and foremost as a 'necessary condition.' Hence, it would appear that within the context of semeiosis, Peirce used the term 'cause' in a very liberal sense: its meaning is not restricted to Peircean efficient cause or Peircean final cause, but it also involves 'necessary condition.' Thus even in icons, the object may be called the immediate cause of the sign, and the mediate cause of the interprétant, provided that the term 'cause' be understood in the sense of 'necessary condition.' Thus, when T.L. Short writes that the term 'determines' does not have a causal connotation, he is strictly taken right that it does not necessarily involve Peircean final or efficient causation. Peirce himself, however, was in this respect perhaps less Peircean than Short; within a semeiotic context, he often used the term 'cause' to express a necessary condition.
162
CHAPTERFOLR
Taking the conclusions of the previous section into consideration, we may therefore conclude that, within the context of Peirce's semeiotic, the term 'determines' carries three different meanings. Depending on the context, A determines B' may mean: (i) A is a necessary condition of B, or (ii) A is the efficient cause ofB, or (iii) A is the final cause ofB. Furthermore, we may conclude that the term determine is used to express three different types of relations: (1) the determination of the sign by the object (relatively to the interprétant), (2) the determination of the interprétant by the sign (relatively to the object), or (3) the determination of the interprétant by the object (relatively to the sign). This may schematically be represented as follows:
object
1 >
sign
2 >• >
interprétant
3 In the following diagram, these different basic relations are schematically represented for, successively, icons, indices, and symbols:
(I)
object
> nc
icon
3 ec
interprétant
nc (II) object
index ec
interprétant ec
ec symbol
(III) object fc
interprétant ec•
fc Thus: determines 1 means 'necessary condition' in icons, 'efficient cause' in indices, and 'final cause' in symbols; determines 2 means 'efficient cause' in icons, indices, and symbols; determines 3 means 'necessary condition' in icons, 'efficient cause' in indices, and 'final cause' in symbols (final causation involves efficient causation).
THE RIDDLE OF SEMEIOTIC CAUSATION
163
Though Peirce's semeiotic does not imply one general, encompassing formula of semeiotic causation (with unequivocal roles of final causation, efficient causation and chance), we can now prescind from his icon/index/symbol typology, and see what actually can be said from such a general, encompassing perspective. Beforehand it should be clear that whatever A determines B may mean, it implies that A is a necessary condition of B. Thus, if A is the efficient cause of B, then A is also a necessary condition of B. The reverse does not hold. (This implies that, if A is a necessary condition of B, and B is in its turn an efficient cause of C, then A is a necessary condition of C). Thus, comparing the three different meanings of determine 1 for icons, indices, and symbols ('necessary condition,' 'efficient cause,' and 'final cause'), we must conclude that what they have in common, is 'necessary condition.' Thus, it would appear that, within the context of semeiosis in general, determines 1 means that the object is a necessary condition of the sign. According to the same reasoning, determines 2 means that the sign is the efficient cause of the interprétant, and determines 3 means that the object is a necessary condition of the interprétant. Thus, the object determines (1) the sign inasmuch as it delimits (in Short's sense) what can be a sign of it. However, the sign determines (2) not only the class of possible interprétants (in Short's sense) but is also creative in the (Ransdell's) sense that it actually produces a series of dynamic interprétants: No sign can function as such except so far as it is interpreted in another sign [...]. Consequently it is absolutely essential to a sign that it should affect another sign. In using this causal word, 'affect," I do not refer to invariable accompaniment or sequence, merely, or necessarily. What I mean is that when there is a sign there will be an interpretation in another sign. The essence of the relation is in the conditional futurity; but it is not essential that there should be absolutely no exception. [...] If the exceptions are, as they occur, as many or nearly as many as the cases of following the rule, the causality would be in my terminology "very weak." But if there is any would be at all, there is more or less causation; for that is all I mean by causation. I do not pretend that this is an accurate analysis of the ordinary conception, or a parlance to be recommended. It is simply what I mean in this connection. It leaves the whole question of what there may be of a metaphysical character quite open. (MS 427:00025; 1904)
According to this passage, the term 'causation' refers primarily not just to a necessary condition, but to an event or an action, namely the creation of the interprétant by the sign.34 This agrees with Ransdell's interpretation, which
CHAPTERFOLR
164
entails that the production of the interprétant by a sign is in some sense a matter of efficient causation (section 3).35 Peirce makes clear that, within the context of semeiosis, he does not use the word 'cause' in accordance with his own analysis of causation. Instead, he prefers to use the term 'causation' in the sense that, given a sign as a cause, there will be a tendency to create an interprétant. Thus, the sign conceived purely in its capacity of an efficient cause - only determines that (under favorable circumstances) an interprétant will be created, but not what the interprétant will be. What the interprétant will be is determined by the final cause of the semeiosis process (which in symbolic semeiosis is the dynamic object). Thus, considered from a general, encompassing perspective (applicable to all sorts of signs), the object is a necessary condition of the sign, which in its turn is the efficient cause of the interprétant. However, what the status is of the final cause is not clear (except for symbols, in which it is the dynamic object).
8.
CONCLUSION
In contemporary Peirce scholarship, there is a consensus that Peirce conceived semeiosis as a teleological process. If, as Peirce maintained, teleology involves a combined action of final causation, efficient causation, and chance, it should be possible, in principle, to precisely indicate the respective roles of these causal elements in semeiosis. A thorough understanding of Peirce's semeiotic requires a clear understanding of this problem of 'semeiotic causation.' The objective of this chapter was to clarify the roles of, respectively, final causation, efficient causation, and chance within semeiosis. In order to meet this objective, the views of two well-known Peirce scholars - T.L. Short and Joseph Ransdell - have been discussed. It was shown that Short and Ransdell hold not only conflicting interpretations regarding the status of the final cause in semeiosis, but also regarding the precise meaning of the term 'determine,' which is a key concept in Peirce's semeiotic. I have questioned Short's view that semeiosis owes its teleological character to the goal-directed activity of some interpreter, and Ransdell's view that semeiosis owes its teleological character to its dynamic object. It appeared that in Peirce's writings, there is conflicting evidence regarding the causal role of the dynamic object in semeiosis. Whereas the young Peirce considered it as the efficient cause of semeiosis (or more precisely: as standing in a relation of efficient causation to the sign), the later Peirce sometimes considered it as the final cause, and sometimes as the efficient
THE RIDDLE OF SEMEIOTIC CAUSATION
165
cause. Thus, it would appear that the later Peirce was still struggling with the problem of semeiotic causation, and had no unambiguous conclusion about it. However, though Peirce's later writings (especially MS 318; 1907) suggest that he was looking for one general, encompassing formula of semeiotic causation, in his classification of signs into icons, indices, and symbols, I found strong evidence against such idea. Moreover, I came to conclude that his icon/index/symbol typology is precisely the expression of three different forms of semeiotic causation. Regarding the causal role of the dynamic object, it would appear that in symbols the dynamic object functions as the final cause, whereas in indices it has the role of efficient cause. In icons, however, it has no causal role at all. Thus, the suggestion made by Ransdell and others36 that the dynamic object is the final cause of semeiosis, seems to hold only for symbols. Only in this type of signs, can we precisely indicate the role of final causation, efficient causation, and chance. One of the basic disagreements between Short and Ransdell concerns the question whether or not the term determine carries a causal connotation. Whereas Ransdell holds that it does, Short insists that it does not. I argued that the term 'determines' carries three different meanings: depending on the context, 'A determines B' means (i) A is a necessary condition of B, or (ii) A is the efficient cause of B, or (iii) A is the final cause of B. I have indicated the respective roles of these different meanings of 'determine' for, succesively, icons, indices, and symbols. Though I argued that it is in principle impossible to give one general, encompassing formula of semeiotic causation (which clearly indicates the roles of final causation, efficient causation, and chance within semeiosis), I tend to agree with Colapietro that the most one can say about semeiotic causation in general is that, though the function of a sign is to generate a series of interprétants, the function of the dynamic object is "not to generate but to constrain a series of interprétants" (Colapietro 1989, 19). In the words of Carl Hausman this means that, "although objects [...] constrain signs, signs themselves not only constrain but create meaning in determining the interprétant" (Hausman 1993, 69). The basic assumption of this chapter was the commonly accepted view among Peirce scholars that Peirce conceived semeiosis as a teleological process that somehow involves final causation, efficient causation and chance. The attempt to precisely and unambiguously determine the respective roles of these causal elements in semeiosis appeared to be highly problematic. In the next chapter, however, the problem of semeiotic causation will be approached from an altogether different perspective. This
166
CHAPTERFOLR
novel point of view involves a strict distinction between the sign-aspect and the event-aspect of signifying events. It will be argued that, whereas causation is a productive event, semeiosis involves only a dynamic relationship between certain aspects of events. It will be argued, furthermore, that, while semeiosis does not entail causation, causation does entail semeiosis in the sense that it provides the formal structure of causation. My new perspective opens the door, not only toward a new way of looking at the problem of semeiotic causation, but also, and more relevant to my final concern, toward a new, semeiotic approach to causation as such.37 Thus in the next chapter I will try to develop a (partially) new approach to the problem of causation, which is based, first and foremost, both on Peirce's semeiotic and upon his positive insights in respect of causation. Some of the most important insights we gained in the previous chapters will appear to be of paramount importance to my semeiotic approach to causation.
Chapter 6 A SEMEIOTIC ACCOUNT OF CAUSATION
Nature only appears intelligible so far as it appears rational, that is, so far as its processes are seen to be like processes of thought. (Peirce, CP 3.422, 1892)
The objective of this chapter is fivefold: (1) to point out that the current theories of causation are radically inadequate; (2) to show the historical roots of this inadequacy; (3) to formulate some conditions necessary to an adequate theory of causation; (4) to discuss some of C.S. Peirce's insights regarding causation, and (5) to suggest a new approach to the problem of causation, based upon the semeiotic of Peirce. The first two objectives will be discussed in section 1, the other three objectives will be dealt with in sections 2-4.
1.
CRITICISM OF THE RECEIVED VIEW
In this section, some fundamental presuppositions of the received view regarding causation will be discussed. It will be shown that the received view is inadequate in several respects, and that this inadequacy is (partly) due to the fact that it has failed to recognize the historical roots of concepts related to causation. More particularly, it will be shown that the received view is based upon two incompatible categoreal frameworks, which have their origin in, respectively, Aristotle's philosophy and the seventeenth century scientific worldview.
167
CHAPTERFOLR
168
1.1
Contemporary approaches to causation
In chapter 2,1 listed the main contemporary approaches to causation: (1) the approach that analyzes causation in terms of necessary and sufficient conditions, (2) the counterfactual approach, (3) the instrumental approach, (4) the probabilistic approach, and (5) the singularist approach which is a minority view. 1. Cause as ¡NUS condition. The most sophisticated version of the necessary and/or sufficient conditions approach is probably John Mackie's analysis of causes in terms of so-called ¡NUS conditions. Mackie suggested that a cause of some particular event is ... an insufficient
but non-redundant
part of a condition which is itself
unnecessary but sufficient for the result. (Mackie 1974, 62)
Mackie called a condition of this kind an INUS condition, after the initial letters of the main words used in the definition. Thus, when experts declare a short-circuit to be the cause of fire, they "are saying in effect that the shortcircuit is a condition of this sort, that it occurred, that the other conditions which, conjoined with it, form a sufficient condition were also present, and that no other sufficient condition of the house's catching fire was present on this occasion" (Mackie [1965] 1993, 34). Thus, Mackie's view is roughly expressed in the following definition of 'cause:' an event A is the cause of an event B if A is a non-redundant part of a complex condition C, which, though sufficient, is not necessary for the effect (B). 2. Causality as counterfactual dependence. For many other contemporary philosophers, however, causal relations are best defined in terms of counterfactual dependence. To say that ' ß is causally dependent upon A ' is to say that 'if A had not occurred, then B would not have occurred.' Moreover, to say that A is the cause of B is to say that there is a chain of causally dependent events linking A with B (c.f. Lewis [1973] 1993, 200). 3. The instrumental approach to causality. R.C. Collingwood gave the best-known defense of this approach. According to him, A cause is an event or state of things which it is in our power to produce or prevent, and by producing or preventing which we can produce or prevent that whose cause it is said to be. (Collingwood [ 1938| 1991, 148)
The main point of this approach is that there are causes only for persons that are practically concerned with certain kinds of events; "[f]or a mere spectator there are no causes" (Collingwood [1938] 1991, 151). Consequently, what may be the cause of an event in the eyes of one person, need not be the cause from the point of view of another; for example, from the driver's point of view, the cause of a car accident may be that he drove
A SEMEIOTIC ACCOUNT OF CAUSATION
169
too hard, while from the motor manufacturer's point of view, the accident may be due to a defective part. 4. Causality as probable occurrence. The basic idea of the probabilistic approaches to causation is that a cause is an event A, the occurrence of which makes the occurrence of another event (B) more likely than if A had not occurred. The most favored strategy captures this idea in terms of types of events. Thus, saying that smoking (A) causes lung cancer is saying that the probability of getting lung cancer is higher for those who smoke than for those who do not. Accordingly, an event A may be said to be a cause of an event B, i f , given the occurrence of A, the probability of the occurrence of B is higher than the probability of the occurrence of B would have been if A had not occurred. 5. The singularist approach to causality. The singularist approach to causality is characterized by the idea that the correct definition of causality must be framed in terms of one single case of causal sequence. According to this view, laws are not relevant to causation, qua causation. Thus, for C.J. Ducasse, one of the most prominent defenders of this approach, the word 'cause' signifies that a particular change sufficed to the occurrence of another change. Moreover, the causal relation is directly observable: "we observe [a causal relationship] whenever we perceive that a certain change is the only one to have taken place immediately before, in the immediate environment of another" (Ducasse [1926] 1991, 139). According to Ducasse, we are strongly inclined to confuse two questions: (1) "what did cause, i.e. what did then and there suffice to, the occurrence of that concrete individual event?" and (2) "which part of what did suffice would be left if we subtracted from what did suffice such portions of it as were unnecessary to such an effect?" (Ducasse [1926] 1991, 141) Only the first question concerns the cause of an individual event as such; the second question concerns a law-like generalization. There are numerous variations to the five mentioned approaches. Faced with the abundance of such widely divergent approaches, Jaegwon Kim, one of the most prominent philosophers regarding causation, conceded some sense of defeat when he wrote "[t]he attempt to 'analyze' causation seems to have reached an impasse; the proposals on hand seem so widely divergent that one wonders whether they are all analyses of one and the same concept" (Kim 1995, 112). Though I tend to agree with Kim's complaint, I think that, nevertheless, most of these widely divergent approaches share some important common characteristics. From now on, I will use the expression 'the received view' to signify all or part of the set of the following approaches: (1) the necessary and/or sufficient conditions approach, (2) the counterfactual approach, (3) the instrumental approach, and (4) the probabilistic approach. Because the singularist view does not share the idea
170
CHAPTERFOLR
that causation implies law-like behavior, I will not consider it part of the received view. According to the received view, the two main problems regarding causality concern (1) the criteria by virtue of which it can be determined that some item may be called a cause, and (2) the precise nature of the necessity that is presumed to be involved in causation. However, before further examining the received view, I will first briefly discuss the historical roots of a basic ambiguity in our common use of the concept of cause. Possibly, unmasking this ambiguity may prove to be of the utmost importance for clearing the way to a more adequate approach.
1.2
Two mutually incompatible conceptions of cause
The history of the concept of cause reveals a complex evolution, which is marked by two decisive milestones: (I) the Aristotelian (-scholastic) Conception, and (II) the Scientific Conception. It will be shown that these two conceptions are mutually incompatible. According to Aristotle, an efficient cause is a substance, which by its activity brings about an effect in another substance. Thus, an efficient cause was defined as the "primary source of a change" that was brought about for the sake of an end. For example, a carpenter was seen as the efficient cause of a table. Thus, according to the Aristotelian conception, causes are the active originators of a change that is brought about for the sake of an end. Probably the most radical change in the meaning of 'cause' occurred during the seventeenth century, when a strong tendency emerged to understand causal relations as instances of deterministic laws. Causes were no longer seen as the active initiators of a change, but as inactive nodes in a law-like implication chain. Thus, causation and determination became virtually synonymous. The debate between the rationalists and the empiricists was related to the nature of this determination. The rationalists (IIa) conceived the relationship between cause and effect primarily as a logical relationship, in which the causes were the premises from which by reasoning alone the effects could be deduced. David Hume, the main representative of the empiricist approach to causation (lib), initiated the view that our idea of causal necessity is partly due to the constant conjunction of certain objects, and partly to the feeling of necessary connection in the mind. The habitual impression of conjunction feels like a necessitation, as if the mind were compelled to go from one to the other. The necessary connection is not discovered in the world but is projected onto the world by our minds. Contrary to what is often believed, Hume's view was far from common among empiricists. Indeed, he may be said to seriously have misrepresented
A SEMEIOTIC ACCOUNT OF CAUSATION
171
their views. For instance, both Locke and Newton explicitly denied that the ideas of causation or power involved the idea of necessary connection according to law. According to Newton, these two notions are even mutually exclusive; complete uniformity or necessary connection denies causal efficacy. Thus, while Locke and Newton adhered to the Aristotelian view that causality is related to substantial powers that are put to work (I), Hume's famous criticism only concerned the rationalist scientific conception of cause (Ha), which, from an historical perspective, was by far the less common one. Everything considered, it appears that the modern concept of cause as it is usually presented, is the result of the interplay between the Aristotelianscholastic Conception according to which causes are the active initiators of a change, and the Scientific Conception according to which causes are the inactive nodes in a law-like implication chain. And if so, every analysis of causation must begin with the recognition of this basic ambiguity. According to the first meaning of cause (I), 'A is the cause of B' means 'A is the active initiator of a change in B'; according to the second meaning of 'cause' (II), 'A is the cause of B' means 'Given the occurrence of B, A must necessarily have occurred.' Most discussions by modern philosophers have failed to see the basic incongruity of these two ingredients. In the next section, I will continue my examination of the received view by discussing some of its fundamental presuppositions.
1.3
The inadequacy of the received view
Some basic presuppositions of the received view are: (1) causation consists in some sort of relationship; (2) this relationship is external to its relata; (3) the relata of the causal relation are discrete entities; (4) causation is basically a timeless relationship. I will discuss these successively. My discussion is partially inspired by the process approach to reality, as put forward by C.S. Peirce, A.N. Whitehead, and Ch. Hartshorne. 1.3.1
Causation is a relation
Though the Aristotelian Conception has remained an unmistakable aspect of our common sense idea of 'cause,' the received view unashamedly subscribes to the Scientific Conception of cause. According to the latter view, causation means some sort of law-like relation between cause and effect, rather than the production of an effect by its cause. For reasons of economy and clarity, I will henceforth strictly distinguish causation from causality. I will use the term 'causation' exclusively for the production of an effect by a cause, and I will restrict my use of the term 'causality' to the relation between cause and
172
CHAPTERFOLR
effect. If correct, the distinction has far reaching consequences. For, since each of the modern theories concerns causality rather than causation (even the singularist Ducasse regards causes as sufficient conditions),' the contemporary discussions of the concept of 'cause' pertain to causality rather than to causation. By asking, "What are the necessary and sufficient conditions for something to be a cause?" they fail to address the more basic issue of causation. 1.3.2
The causal relation is external to its relata
Contemporary discussions of causation are inherently infected by Hume's views on the subject. One symptom of this 'infection' is the idea that the relationship between causes and effects is a purely external affair. 2 Thus, to say that smoking causes cancer is to say that, although smoking and cancer are completely different sorts of 'things' (which, as such, are not internally related), there is yet a relation such that the occurrence of the one (smoking) increases the probability of the occurrence of the other (cancer). On this account, Hume is right when he writes, "all events seem entirely loose and separate." There certainly is no logical connection between events. To say that causality is not a logically necessary relation is to say that even if A is the cause of B, it is logically possible to suppose that, given A, B might not have occurred. From a logical point of view, one thing does not necessarily yield a determinate other thing; given a particular situation, anything may follow. This is precisely the reason why causal relations cannot be known a priori; in order to determine whether or not a causal relation holds between A and B, we must rely on our experience of similar relations. But Hume also insists that we cannot observe "ties" or "connections" between events: ... there appears not, throughout all nature, any one instance of connection which is conceivable by us. All events seem entirely loose and separate. One event follows from another, but we never can observe any tie between them. They seem conjoined, but never connected. (Hume [1748] 1975, 74)
The question is, however, whether Hume's contention is warranted. In order to answer this question, we must first inquire what evidence Hume provides. Certainly, the evidence is not based upon an analysis of the structure of events, for he never provided such analysis. Moreover, though Hume often spoke of the causal relata as if they were events, and though he sometimes gave examples in which the causal relata were events, his whole discussion is based on the idea that both causes and effects are 'objects,' conceived as substances having certain qualities. Consider, for example, the following passage at the beginning of his discussion of causation:
A SEMEIOTIC ACCOUNT OF CAUSATION
173
We must consider the idea of causation, and see from what origin it is deriv'd. [...] Let us therefore cast our eye on any two objects, which we call cause and effect, and turn them on all sides, in order to find that impression, which produces an idea of such prodigious consequence. At first sight I perceive, that I must not search for it in any of the particular qualities of the objects; since, which-ever of these qualities I pitch on, I find some object, that is not possest of it, and yet falls under the denomination of cause and effect. And indeed there is nothing existent, either externally or internally, which is not to be consider'd either as a cause or an effect; tho' 'tis plain there is no one universal quality, which universally belongs to all beings, and gives them a title to that denomination. The idea, then, of causation must be deriv'd from some relation among objects. (Hume [1739] 1978, 74-75)
Hume apparently supposed the causal relata to be substances, and observed that the idea of causation cannot be derived from some particular qualities of the cause-substance, but, instead, must be derived from some relation between substances. Similarly, he observed that the idea of necessary connection could not be derived from certain qualities of an object ([1739] 1978, 86-87); for instance, we cannot by a careful consideration of the qualities of that knife, derive all the qualities of a wound inflicted by that knife. Thus, remarkably, Hume's entire discussion takes place within an Aristotelian-scholastic framework, according to which the world consists of the set of substances (ousiai) and their properties. By thus implicitly putting the question of causation in terms of an Aristotelian-scholastic conception of the world, he was forced to conceive the relationship between causes and effects as an external affair, which was moreover seen as a matter of the imagination. The Humean idea that causality is a purely external relation was severely criticized by A.N. Whitehead. Though Whitehead agreed with Hume that the idea of the transference of a quality from cause to effect is unintelligible, he thought that the cause is in some sense immanent in its effect: The mere notion of transferring a quality is unintelligible. Suppose two occurrences may in fact be detached so that one of them is comprehensible without reference to the other. Then all notions of causation between them, or of conditioning, become unintelligible. There is [...] no reason why the possession of any quality by one of them should in any way influence the possession of that quality, or any other quality, by the other. [...] The only intelligible doctrine of causation is founded on the doctrine of immanence. Each occasion presupposes the antecedent world as active in its own nature. (Whitehead [1938] 1968, 164-65)
174
CHAPTERFOLR
Causes are not merely external conditions of an effect; there is an 'active immanence' of the causes in the effects. 3 According to Whitehead, we do have concrete experiences of the interconnectedness of events in perception and memory. Thus, the perception of an electric light suddenly switched on makes us blink our eyes. We have the concrete experience of a "compulsion to blink." The flash, though in the immediate past of the blink, is still present ("objectively immortal") in the compulsion to blink (Whitehead [1929] 1978, 175). Similarly, whenever we remember something, the past event is, somehow, actively involved in the present event of our remembering it; there is an "immanence of the past energizing in the present" (Whitehead [1933] 1967, 188). 1.3.3
The causal relata are discrete entities
If Whitehead is right, cause and effect must somehow be continuous. Consider his example of some event arousing some man's anger. How can we explain the fact that the man knows that a quarter of a second ago he was angry? Obviously, the mere reference to 'memory' is putting the cart before the horse. According to Whitehead, [t]he feeling as enjoyed by the past occasion is present in the new occasion as datum felt, with a subjective form conformai to that of the datum. Thus if A be the past occasion, D the datum felt by A with subjective form describable as A angry, then this feeling - namely A feeling D with subjective form of anger - is initially felt by the new occasion B with the same subjective form of anger. The anger is continuous throughout the successive occasions of experience. [...] His anger is the subjective form of his feeling some datum D. A quarter of a second later he is [...] embodying his past as a datum in the present, and maintaining in the present the anger which is a datum from the past. (Whitehead [1933] 1967, 183-184).
Thus, Whitehead makes clear that, without some continuity between a past event as a cause and a present event as its effect, we simply cannot have any recollection of something happening a quarter of a second ago. Though our experiences are in some sense definite, individual events, they are continuous in respect of their subjective forms; there is, in Whitehead's words, an "identity of subjective form inherited conformally from one occasion to the other" (Whitehead [1933] 1967, 186). Similarly, C.S. Peirce concluded on the basis of a detailed analysis of the question, "How can a past idea be present?" that events are not discrete because they do not have a definite beginning and a definite end. According to Peirce, saying that a past idea can be vicariously present explains nothing,
A SEMEIOTIC ACCOUNT OF CAUSATION
175
because it raises once more the question how this vicarious representation is related to the past idea. After thus having rejected the idea of vicarious representation, Peirce presented the only possible alternative: How can a past idea be present? Not vicariously. Then, only by direct perception. In other words, to be present, it must be ipso facto present. That is, it cannot be wholly past; it can only be going, infinitesimally past, less past than any assignable past date. We are thus brought to the conclusion that the present is connected with the past by a series of real infinitesimal steps. (EP I: 314; 1892)
Similarly, Peirce too emphasized that a moment must necessarily cover an infinitesimal interval of time. In this infinitesimally spread-out consciousness "we directly perceive the temporal sequence of its beginning, middle, and end" (EP I: 315; 1892). Thus, the present moment involves an infinitesimal duration; as such it is "half past and half to come" (EP I: 323; 1892). Thus, it would appear that if causes are internally related to their effects, and both cause and effect are two aspects of one continuous process, the distinction between cause and effect is to some extent arbitrary. For, between any event A, designated as the cause of another event B, there is an innumerable series of events affecting B, which might just as well be called the cause of B. There are strong indications that this view was indeed defended by Peirce (see EP I: 323; 1892). Whatever is designated as the cause of an event, will then necessarily depend on some sort of practical considerations such as those suggested by Collingwood (section 1). Paraphrasing Collingwood, one might say that, though "for a mere spectator there are no causes" (and therefore no causality either), causation nevertheless is very much there. In conclusion: the idea that causation consists in an external relationship between discrete events is based upon a mistaken presupposition regarding the nature of the causal relata. Here Hume's implicitly putting the question of causation in terms of an Aristotelian-scholastic framework, together with his failure to make a thorough analysis of the nature of events, proves to be fatal not only to his own conclusions, but to the received view, which uncritically inherited his blind spots. The analyses by both Peirce and Whitehead of the nature of events provide strong evidence in favor of the view that causation involves the production of an event, and that the cause is in some sense immanent in its effect. It is only from the standpoint of the effect that the causal relation may be defined in terms of antecedent entities, which are abstractly separated from the concrete continuous process. Mistaking the nature of the relationship between these abstracted aspects for the core of causation, is committing what Whitehead has called the "fallacy of
176
CHAPTERFOLR
misplaced concreteness," which consists in confusing conceptual abstractions with the concrete reality from which they are derived (Whitehead [1925] 1967, 50-51). 1.3.4
Causation does not involve the passage of time
The most favored contemporary approach to causation defines causes in terms of necessary and sufficient conditions. One of the main problems with this approach - first raised by Scriven (1966) - is that it fails to logically distinguish causes from their effects. For, if A and B are events and A is necessary and sufficient for B and therefore the cause of B, then it logically follows that B is necessary and sufficient for A and therefore must be the cause of A. But clearly, the ignition of a match is not the cause of its being rubbed. Thus, the problem of symmetry is a major problem of the 'necessary and sufficient condition approach;' the distinction between causes and effects obviously requires some temporal constraints. Though this problem of symmetry does not hold for the counterfactual approach, the instrumental approach, and the probabilistic approach, the tendency common to all of these to consider causation as a relation rather than as a process, betrays an ambivalent attitude toward the relationship between causation and time. Though it is said that causes precede their effects in time, it is not stipulated how this precedence affects the relationship. If there is indeed a temporal priority of causes to their effects, then causation must somehow involve the production of the effect by its cause such that the cause precedes its effect. And "if [causation] does involve 'production,' then it is a process rather than a relation" (Bennet 1974, 12).
1.4
Two mutually incompatible categoreal frameworks
If my analysis so far makes any sense, the received view is inadequate both because it conceives causation as a relation instead of a process, and because it falsely assumes the causal relata to be discrete entities between which there is not even a hint of continuity. The most troublesome aspect of the received view, however, is its basic ambiguity regarding the causal relata. Though most contemporary philosophers hold that the causal relata are events, there are also some philosophers who hold that they are facts, and again some who hold that, next to event causality or fact causality, there is also agent causality in which the agents are conceived as substances. This comment requires closer scrutiny.
A SEMEIOTIC ACCOUNT OF CAUSATION
177
In ordinary language, we rarely identify causes as events. Indeed, the grammatical subject-predicate structure of our sentences lures us into thinking causes in terms of things or substances. It was probably that seductive power of language that was the basis of the Aristotelian doctrine that causes are substances. We still do not think there is anything wrong in saying that 'John broke the glass,' or that 'mosquitoes are the cause of malaria.' Only when we are asked to clarify those statements, do we correct ourselves by saying that some action performed by John caused some process occurring within the glass. In other words, we tend to clarify ourselves by interpreting causes in terms of events. I have already pointed out that the continual shift between referring to causes as things and then again to causes as events had characterized Hume's seminal text. But the confusion grows as soon as philosophers try to clarify what they mean by events. For instance, though the concept of event is rightfully associated to the concept of change, for instance the acquisition or loss of a property in a particular object, it has also been linked to states (Kim 1973; Lewis 1986). Thus, it is said that the persistent state of an object - say the insistence of malaria - may be causally explained by another persisting state - say the continuing presence of certain organisms in the blood. But, though there is some disagreement regarding the question whether or not events involve persisting states, there is almost universal agreement about events being somehow related to substances: they are said to be either changes or 'unchanges' in an object. This clearly shows that in many cases, the defenders of event causality who explicitly endorse an ontology of events, are in fact closet defenders of a substance ontology by assuming that relations between events are rooted in a relation between substances. The same holds for those who have insisted that there is a special class of causation which they have called agent causation, and which refers to the act of a person in bringing about a change. There is a controversy whether agent causation is reducible to event causation. For instance, 'someone washing the dishes' would be an example of agent causation. This may be described in terms of event causation, for it may be said that the decision to wash is an event, which is the cause of the dishes being washed. Yet, most philosophers maintain that decisions necessarily require agents who decide. And these agents are again perceived to be substances. In the end, therefore, it would appear that the defenders of event ontology, in spite of frequent explicit rejections of the Aristotelian substance, cannot really extricate themselves from an Aristotelian categoreal framework. According to this framework, 'the basic furniture of the world' consists of substances and their qualities, and every sound explanation of events must therefore be given in terms of substances (or basic entities) and their properties.
178
CHAPTERFOLR
An exception should be made for those philosophers who maintain that the relata of causation are facts. And indeed, this view too is supported by ordinary language, as when, for instance it is said that gravity is the cause of the fall of an apple. Gravity is neither a substance nor an event. It simply is the fact that, given two masses and a certain distance between them, a certain mathematical relation obtains, which allows us to describe the behavior of the apple. A fact is the truth-maker of a true proposition. Thus, the fact that John shot Peter, which contains the objects John and Peter standing in the relation of shooting and being shot, makes it true that John shot Peter. Where event theories favor statements of the form 'C caused E, ' fact theories favor expressions of the form 'E because C' (Mellor 1995, 11). For instance, event theories favor statements like "The temperature's dropping severely caused the freezing of the pipes" (events), while fact theories prefer statements like "The pipes froze because the temperature dropped severely" (facts). 4 The example points out another important difference: facts do not exist; only the particulars involved in the fact do. Clearly, the ambivalence regarding the status of causal relata is a symptom of the circumstance that our modern worldview is strangely caught between Aristotle's substance ontology and the modern scientific fact ontology. The latter view, which became the predominant view during the seventeenth century, was most clearly expressed by Wittgenstein in his Tractatus'. "Die Welt is die Gesammtheit der Tatsachen, nicht der Dinge" (Debrock 1988). This general ambivalence may be particularly relevant to my discussion, for the tension between the two views, neither of which really allows for an event ontology, is reflected in our conception of causality. Consider the following example: our belief in the nutritional value of bread is usually not seen as the belief that my eating of this bread is the cause of my being fed, but as the belief that the nature of bread, being what it is, necessarily involves nutrition in certain organisms. Indeed, Hume's attack on the rational foundation of causal relations assumed that this latter belief was our belief. Since causation is described as a relationship between general types, rather than as a relation between concrete individual events, the causal relata are regarded as facts rather than events. Thus, it would appear that the received view's apparent insistence on events is trapped between Aristotle's substance ontology and the modern scientific fact ontology. This tension may be a heritage of Hume's ambivalence when he described causation as a relation between types of events, thereby implicitly endorsing a fact ontology, while his examples were expressed in the terminology of a substance ontology. The confusion seems hopeless.
A SEMEIOTIC ACCOUNT OF CAUSATION
1.5
179
Conclusion to part 1: criticism of the received view
In the first part of this chapter it was shown that (1) the current theories of causation are radically inadequate, and (2) that this inadequacy has important historical roots. It was seen that, according to the received view, there are two main questions regarding causality, the first of which concerns the demarcation criteria by which causal relations may be identified, while the second regards the precise nature of the necessity involved in the relationship between cause and effect. In their attempts at solving those questions, defenders of the received view inherited Hume's assumption that causes and effects are discrete events. It was argued that this assumption was ambiguous for two reasons, the first of which was that its apparent obviousness was clouded by Hume's examples in which he almost never referred to events, but mostly to existing substantial entities. The second reason for the ambiguity of his assumption is that his theory is not about concrete events at all, but instead about the relationship between general types of entities. It was suggested that A.N. Whitehead gave strong arguments against Hume's idea that "all events seem entirely loose and separate." His extensive analysis of the structure of events gives credit to the idea that causes are somehow 'actively immanent' in their effects. Whitehead's analysis was seen to agree with Peirce's conclusion that we cannot remember past ideas unless there is some continuity between past events and present events. The majority of contemporary philosophers hold that causal relata are events, conceived as changes in a substance. It was argued that this view is inconsistent because it is based upon two incompatible categoreal frameworks: a substance ontology and a fact ontology. Our tendency to implicitly conceive causal relata as substances or substance-like entities is rooted in Aristotle's substance ontology, and contradicts the explicit standard view that causation is some sort of relation, a view that goes back to the fact approach to reality, which became predominant in the seventeenth century. Thus, it would appear that the impasse in the contemporary attempts to analyze causation may be partly due to the fact that one is insufficiently aware of the lack of a consistent categoreal framework, in terms of which every element of our experience can be interpreted. Thus, the history of the concept of cause was seen to be intimately related to the historical evolution from the substance ontology to the fact ontology, which was marked by two decisive milestones: (I) the Aristotelian-scholastic conception, according to which causes are the active initiators of a change, and (II) the (seventeenth century) Scientific Conception, according to which causes are inactive nodes in a law-like implication chain.
180
CHAPTERFOLR
It was argued that the inadequacy of the received view is (partly) due to the fact that it has failed to recognize this basic ambiguity, and that it has taken the Scientific Conception (II) as point of departure, which, from both an historical and an ontological point of view, is a derivative conception of 'cause.'
2.
NECESSARY CONDITIONS FOR A THEORY OF CAUSATION
In the previous section I made a distinction between causation, or the production of an effect by a cause, and causality, or the relation between cause and effect, and I concluded that the received view concerns only causality. But my analysis also offers some suggestions regarding the necessary conditions for a theory of causation. In this section an attempt will be made to make these conditions explicit, and moreover, I will formulate one additional requirement that is not to be found in the above analysis. It is convenient to distinguish between primary requirements and subsidiary requirements, in the sense that the latter will be dependent upon the outcome of the analysis of the former. There are two primary requirements. Obviously, the first requirement for a theory of causation is that it be coherent. Secondly, if it is accepted that causation is a productive event in an ongoing process, then the analysis of causation must involve a coherent theory of events and processes, and of the relationship between events and processes. There are four subsidiary requirements. The first of these is that the theory be able to clarify whether causation involves continuity. The second subsidiary requirement is that, inasmuch as productive events involve the 'passage' of time, the theory of causation must clarify the relationship between time and causation. The third subsidiary requirement is that the theory be able to clarify the status of causes and effects as such, and by the same token, that it be able to clarify the relationship between causes and effects. Finally, if causation is indeed a process, then some sort of direction must characterize it. Thus, since it would appear that causation and directionality go hand in hand, an adequate theory of causation must be able to sort out the problem of teleology: it must be able to decide whether the concept of teleology is meaningful, and if so, it must be able to express what its relationship is in respect of causation. In section 4, I will give a sketch of an alternative approach to causation, which is based upon the semeiotic of C.S. Peirce, and which, I will argue, meets these requirements. However, before developing this alternative
A SEMEIOTIC ACCOUNT OF CAUSATION
181
approach, I must first consider some of Peirce's (more or less explicit) insights into causation as such.
3.
PEIRCE ON CAUSALITY AND CAUSATION
Before developing my semeiotic approach to causation in the next section, I must discuss Peirce's conception of cause. Though Peirce never explicitly formulated a theory of causation, there are a number of places (especially in his so-called cosmological papers) where he severely criticized the principle of causality. Moreover, in 1902, within the general context of a discussion of natural classes, Peirce gave a highly perceptive analysis of the problem of teleology, in which he exposed some remarkable insights regarding causation. In section 3.1, I will discuss Peirce's critique of the principle of causality. It will be seen that Peirce proposed to strictly distinguish the concept of cause from the concept of force (Peirce himself, however, was far from consistent in the application of this distinction). He reserved the former term for the whole of human experience and of nature, and he restricted the latter to the formal laws of physics. Whereas 'cause' pertains to irreversible processes, 'force' deals with reversible processes; whereas the former term deals with concrete reality, the latter deals only with abstractions. In section 3.2, I will briefly discuss Peirce's conception of causation, according to which each act of causation involves a teleological, an efficient and a chance component. It will be shown, furthermore, that Peirce's conception of efficient cause holds a middle way between the Aristotelian Conception and the Scientific Conception of cause. In section 3.3,1 will show that, according to Peirce, the causal relata may be either events or facts, depending on the context of the discussion. While Peirce insisted that the causal relata are facts within an epistemological context (or context of causality), they appear to be events within an ontological context (or context of causation). Because causation is conceived as a productive event in an ongoing process, the relationship between processes and events will be considered in section 3.4. The results of this part of the chapter will be summarized in section 3.5.
3.1
Peirce's critique of the principle of causality
Peirce's sixth and seventh lecture of his Cambridge series on Reasoning and the Logic of Things (1898) are particularly important to the understanding of his critique of the principle of causality. In these papers, Peirce's concern is
CHAPTERFOLR
182
not an elucidation of causation as such, but rather a clarification of the concept of cause within the context of scientific, causal explanation. The sixth lecture - "Causation and Force" - starts with the following important observation: Those who make causality one of the original uralt elements in the universe or one of the fundamental categories of thought, - of whom you will find that I am not one, - have one very awkward fact to explain away. It is that men's conceptions of a cause are in different stages of scientific culture entirely different and inconsistent. The great principle of causation which, we are told, it is absolutely impossible not to believe, has been one proposition at one period in history and an entirely disparate one [at] another and is still a third one for the modern physicist. The only thing about it which has stood [...] is the name of it. (RLT 197, 1898)
Peirce observed that the history of the concept of cause reveals a discrepancy between the constancy in the use of terminology and a gradual increasing ambivalence in the conception itself. At least three different and mutually incompatible meanings may be discerned: (a) the Aristotelian conception (AC), (b) 'the modern physicist's conception' (MPC), and (c) 'the currently accepted view' (CAV) (RLT 197-202, 1898). Peirce approached the problem from three different angles: (1) a logical analysis of the different concepts of cause, (2) an analysis of scientific knowledge of natural processes, and (3) an analysis of mental processes. Each of these will be discussed separately. 3.1.1
Three concepts of cause
The concept of cause was first explicitly formulated by Aristotle. According to Peirce, The original idea of an efficient cause is that of an agent, more or less like a man. It is prior to the effect, in the sense of having come into being before the latter, but it is not transformed into the effect. In this sense, it may happen that an event is a cause of a subsequent event; seldom, however, is it the principal cause. Far less are events the only causes. (CP 6.600, 1893)
A SEMEIOTIC ACCOUNT OF CAUSATION
183
Moreover, It is generally held that the word cause has simply been narrowed to that one of the four Aristotelian causes which was named from the circumstance that it alone produces an effect. But this notion that our conception of cause is that of the Aristotelian efficient cause will hardly bear examination. The efficient cause was in the first place generally a thing not an event, then something which need not do anything; its mere existence might be sufficient. Neither did the effect always necessarily follow. True when it did follow it was said to be compelled. But it was not necessary in our modern sense. That is, it was not invariable. (RLT 198, 1898)
Thus, according to Peirce, Aristotle considered an efficient cause to be (i) an agent who compelled another agent or a thing to behave in a certain way, (ii) prior to its effects, (iii) not transformed into its effects, 5 and (iv) not necessitating its effects (in our modern, Millsian sense of being a concurrence of antecedents to which a given phenomenon is invariably and unconditionally consequent). AC differs in at least two respects from "our conception of cause" (CAV and MPC): (1) Aristotelian efficient causes are usually things or substances rather than events, and (2) there is no necessary (invariable) relation between certain types of causes and certain types of effects. According to CAV, a cause is "an instantaneous state of things perfectly determinative of every subsequent state" (CP 6.600, 1893). More precisely, the commonly accepted view of 'the great principle of causation' - which had its origin in Stoic philosophy (EP I: 299, 1892) - involves three propositions: (1) The state of things at any one instant is completely and exactly determined by the state of things at one other instant. (2) The cause or determining state of things precedes the effect or determined state of things in time. (3) No fact determines a fact preceding it in time in the same sense in which it determines a fact following it in time. (RLT 198-99, 1898) MPC, on the other hand, concerns those phenomena that are governed by the Law of the Conservation of Energy, in which the future determines the past in precisely the same way in which the past determines the future. MPC involves the following three propositions:
184
CHAPTERFOLR
(1) The state of things at any one instant is completely and exactly determined by the state at two other instants.6 (2) Cause and effect are simultaneous. (3) The positions at the two later instants determine the position at the earliest instant in precisely the same way in which the two positions at the two earlier instants determine the position at the latest instant. (RLT 198202;1898) Thus, according to Peirce, there is a flat contradiction between the basic propositions of CAV and those of MPC. CAV and MPC are therefore irreconcilable concepts of cause. Moreover, both are incompatible with AC. 3.1.2
Analysis of natural processes
In his "The Doctrine of Necessity Examined" (1892) Peirce offered a complex, multi-layered argument against the 'doctrine of necessity,' which was commonly supposed to be a postulate of scientific reasoning. According to Peirce, the doctrine of necessity is the idea that "the state of things at any time, together with certain immutable laws, completely determines the state of things at every other time" (EP I: 299, 1892). One of Peirce's main arguments against this postulate was that the variety and increasing complexity of nature cannot be explained solely on the basis of "the rule of mechanical necessity." The operation of mechanical law cannot create diversity where there was no diversity before. Under given circumstances, "mechanical law describes one determinate result" (CP 1.161, 1897). Since "variety can spring only from spontaneity," the laws of nature must be probabilistic rather than deterministic: By thus admitting pure spontaneity or life as a character of the universe, acting always and everywhere though restrained within narrow bounds by law, producing infinitesimal departures from law continually, and great ones with infinite frequency, I account for all the variety and diversity of the universe, in the only sense in which the really sui generis and new can be said to be accounted for. (EP I: 308, 1892)
Peirce insists that the doctrine of necessity is incompatible with the simple fact that wherever we look - be it in geology, astronomy, biological evolution, the history of institutions, languages or ideas - "everywhere the main fact is growth and increasing complexity" (EP I: 308, 1892). "Now, the essential of growth is that it takes place in one determinate direction, which is not reversed. Boys grow into men, but not men into boys. It is thus an immediate corollary from the doctrine of the conservation of energy that growth is not an effect of force alone" (CP 6.555, 1887).
A SEMEIOTIC ACCOUNT OF CAUSATION
185
The fact that each event is characterized by an aspect of irreducible novelty or objective chance 7 makes natural processes irreversible. And because natural processes consist of causally linked events, cause-effect relationships too must be irreversible. Thus Peirce accepts the third proposition of CAV. Further on we will see that (on the basis of an analysis of mental phenomena) Peirce also accepts CAV's second proposition: causes precede their effects in time. On the other hand, Peirce rejects all three propositions of MPC on account of the irreversibility of all natural processes. This irreversibility extends not only to processes that are inexplicable by the Law of the Conservation of Energy (such as birth, growth, life, conduction of heat, combustion) but even to physical processes that appear to be reversible. Though so-called mechanical processes approach the laws of mechanics to a degree, they never do so perfectly. Thus, Peirce held that our scientific study of natural processes forces us to accept the view that (i) cause-effect relationships are irreversible, (ii) causes only partially determine their effects (there is always an aspect of indeterminatene ss or chance involved), (Hi) cause-effect relationships are determined by probabilistic laws. 3.1.3
Analysis of mental processes
According to Peirce, causation, as distinct from the action of force, is not only "a real, and fundamental, and vital element [...] in the outer world," but it is also at work in the inner world (RLT 220, 1898). Indeed, "the very conception of causality has its origin in our tendency to seek relations in nature analogous to intellectual relations" (MS 963, 2, c.1893). Given Peirce's attention to the original meanings of concepts, 8 and given his view that causality is an anthropomorphic conception, he may have held the view that the analysis of mental processes is the very best way to study causality. Whether or not he actually did so, he surely accepted all the conclusions regarding causality of his analysis of mental processes. For Peirce, there is not the slightest evidence that the doctrine of necessity holds in the domain of the mental. That is to say, "from the state of feeling of any instant, there is no reason to suppose the states of feeling at all other instants are thus exactly calculable" (EP I: 309, 1892). Ideas seem to suggest other ideas, rather than to necessitate them: ... no mental action seems to be necessary or invariable in its character. In whatever manner the mind has reacted under a given sensation, in that manner it is the more likely to react again; were this, however, an absolute necessity, habits would become wooden and ineradicable, and no room being left for the
186
CHAPTERFOLR formation of new habits, intellectual life would come to a speedy close. Thus, the uncertainty of the mental law is no mere defect of it, but is on the contrary of its essence. The truth is that mind is not subject to "law," in the same rigid sense that matter is. It only experiences gentle forces which merely render it more likely to act in a given way than it otherwise would be. There always remains a certain amount of arbitrary spontaneity in its action, without which it would be dead. (EP I: 329, 1893)
Thus, Peirce concludes that the attempt to explain away the variety and apparent arbitrariness of mental action in favor of absolute determinism, does not agree with the observed facts (EP I: 329-30). Moreover, none of the three propositions of MPC applies to the realm of our mental experience. Here it is appropriate to quote Peirce at length: ... when from the world of physical force we turn to the psychical world all is entirely different. Here we find no evident trace of any state of mind depending in opposite ways upon two previous states of mind. Every state of mind acting under an overruling association produces another state of mind. Or if different states of mind contribute to producing another, they simply act concurrently, and not in opposite ways, as the two earlier positions of a particle of matter do, in determining a third position. I come down in the morning; and the sight of the newspaper makes me think of the Maine, the breakfast is brought in, and the sight of something I like puts me into a state of cheerful appetite; and so it goes all day long. Moreover, the effect is not simultaneous with the cause. I do not think of the explosion of the Maine simultaneously with seeing the newspaper, but after seeing it, though the interval be but a thirtieth of a second. Furthermore, the relations of the present to the past and to the future, instead of being the same, as in the domain of the Law of Energy, are utterly unlike. I remember the past, but I have absolutely no slightest approach to such knowledge of the future. On the other hand I have considerable power over the future, but nobody except the Parisian mob imagines that they can change the past by much or by little. Thus all three propositions of the law of causation [MPC] are here fully borne out. (RLT 201-02, 1898)
Thus, Peirce's analysis of our mental experience confirms his analysis of our scientific experience of the outer world in all respects: (i) cause-effect relationships are irreversible, (ii) causes only partially determine their effects, and (iii) cause-effect relationships are determined by the laws of association, which are in some sense analogous to the probabilistic laws of nature 9 Moreover, (iv) on the basis of his analysis of mental phenomena Peirce concluded that causes precede their effects in time.
A SEMEIOTIC ACCOUNT OF CAUSATION
187
Though Peirce preferred to use the term 'force' within the realm of pure physics, and to restrict the term 'cause' to pure psychics, he felt that Occam's razor compels us to hope that the one mode of action be somehow reducible to the other (RLT 237-38, 1898). Peirce's solution is that, both from an ontological and from an evolutionary point of view, causal action is primary, not only in the inner world but also in the outer world. Causal action is ontologically more fundamental than force, since all natural processes are irreversible, and therefore causal. Thus so-called purely mechanical processes too must be causal. Moreover, according to Peirce's grand cosmological hypothesis, causal action is also primary from an evolutionary point of view. For cosmic evolution is characterized by a creative advance from pure arbitrariness or chaos toward "an absolutely perfect, rational and symmetrical system" in the infinitely distant future, though at any time in this asymptotic approach toward complete order "an element of pure chance survives" (EP I: 297, 1891). It may seem odd against this background that Peirce refused to reckon cause as a philosophical word. He thought it was too ambiguous. Thus, when he did use it, he tried to reduce its ambiguity by the distinction between the terms cause, force, and explanation. Cause was to be taken in its original, albeit crude, Aristotelian sense, while he reserved force for the context of dynamics, and explanation for a more general, logical context (CP 6.600, 1893). However, despite Peirce's original intention to stick to the Aristotelian conception of efficient cause, some ten years later, in 1902, Peirce developed what might be called a proto-theory of causation, which - though it was Aristotelian in its reliance on the mutual interdependence of final and efficient causation - deviated in some significant respects from Aristotle's conception. Undoubtedly the most important difference is that it was closer to Peirce's categoreal scheme, which involves an event ontology, than to Aristotle's categoreal scheme, which involves a substance ontology. In the next subsection I will briefly explain Peirce's 1902 conception of causation.
3.2
Peirce's conception of causation
In his 1902 paper "On Science and Natural Classes" (EP II, item 9) and in some closely related papers,10 Peirce developed the highly original view that each act of causation involves an efficient component, a final component, and a chance component. The efficient aspect of causation is that each event or fact is produced by a previous event or fact (the efficient cause). The teleological aspect is that each event is part of a chain of events with a definite tendency. The chance component is that each event has some aspect that is determined neither by the efficient nor by the final cause.
188
CHAPTERFOLR
According to Peirce, final causes are general types that tend to realize themselves by (teleologically) determining processes of efficient causation. Final causes are basically habits: they ('habitually') direct processes toward an end state. Like human habits, habits of nature (laws of nature) too are final causes because they display tendencies toward an end state. Moreover, habits are not static 'entities' for they may evolve in the course of time. Peirce called the possible evolution of final causes "developmental teleology." Thus final causes are not future events, but general possibilities. The end state of the process to which the act of causation belongs can be reached in different ways. Moreover, the process involved is irreversible. The idea that efficient causation can only be understood within the context of final causation is absolutely central to Peirce's conception of causation. According to Peirce: Efficient causation [...] is a compulsion determined by the particular condition of things, and is a compulsion acting to make that situation begin to change in a perfectly determinate way; and what the general character of the result may be in no way concerns the efficient causation. (CP 1.212, 1902; italics mine)
Thus, efficient causation, considered apart from its final causational component, is not directed toward an end in any way; it is blind compulsion. We understand better now why, contrary to the mechanical conception according to which causes are the inactive nodes in a law-like implication chain, Peircean efficient causes are the active initiators of a change. Thus, 'A is the cause of B' means that B results partly from some activity or influence originating in A. Efficient causation thus considered is a dyadic relation between two concrete individual events. Final causation, on the other hand, is a triadic relation between the general final cause, the concrete efficient cause, and its concrete effect. The general final cause (C') determines or mediates the production of the individual effect (B) by the individual efficient cause (A). The produced effect (B) is in its turn a novel efficient cause, which functions as a means for the attainment of the end (C). Schematically this may be represented as follows: C' (final cause)
> (cause)
(effect, means)
C (effect, realized end)
A SEMEIOTIC ACCOUNT OF CAUSATION
189
Thus, final causation and efficient causation are just two aspects of one and the same process, which involves a triadic relationship between A, B and C'. At the beginning of the causal process, the concrete realized end C does not yet exist. It therefore cannot in any way influence the causation of B by A. Final causation does not entail backward causation. Thus, the individual event of a batsman hitting a baseball in order to make a home run (C) cannot be fully understood by just referring to the relationship between the movements of his arms (A) and the movement of the ball (B), while leaving aside the intention of the batsman (C'). His wonderful blow wasn't just an accident; he really intended to do so. The intention of the batsman is a general idea or a possibility, which he could have tried to realize in countless different ways. Moreover, according to Peirce, every event (as part of a process) is characterized not only by an aspect of final causation and an aspect of efficient causation, but also by an aspect of objective chance. Each process involves an aspect of irreducible novelty or objective chance at every stage of the process, which cannot be reduced to efficient or final causation. In the above given diagram, each stage of the causal process (level A-B-C) involves an aspect of objective chance. Peirce's conception of causation is therefore characterized by a triple interdependence of final causation, efficient causation, and chance. Keeping in mind that in my brief historical survey we distinguished two mutually incompatible conceptions of cause - the Aristotelian-scholastic Conception, according to which causes are the active initiators of a change, and the Scientific Conception, according to which causes are the inactive nodes in a law-like implication chain - we may conclude now that Peirce's conception of efficient cause forms, in some sense, an ingenious middle way between these two conceptions. On the one hand, Peircean efficient causes are the active initiators of a change (rather than the inactive nodes in a law-like implication chain); 'A is the cause of B' means that B results partly from some activity or influence originating in A. On the other hand, however, Peirce held that "the action of a cause is essentially a case of the operation of a law, and implies a law" (MS 318:00020, 1907). In this context, the term 'law' must be understood in the broad sense of habit, final cause, or general disposition; it involves natural laws as well as personal predispositions to act in a certain way. Finally, Peircean laws are probabilistic rather than deterministic, causeeffect relationships are irreversible, and efficient causes precede their effects. Furthermore, depending on the context, efficient causes are either events or facts. The latter claim, however, needs to be qualified. This will be done in the next subsection.
190
3.3
CHAPTERFOLR
Causality and causation: facts versus events
One of Peirce's main points of critique of the principle of causality concerns the problem of the causal relata. I decided to discuss it separately, not only because the issue is very important for the development of a Peircean approach to causation, but also because Peirce's observations are at first glance ambiguous, and might therefore easily be misunderstood. The later Peirce (from 1893 onwards) insisted that the causal relata are facts rather than events. He went as far as stating that no one will ever understand the problem of causality unless he sees that both the cause and its effect are facts (MS 647:00010, 1910). In a discussion of the nature of the causal relata, Peirce gave the following description of fact: A Fact [...] is so much of the Real Universe as can be represented in a Proposition, and instead of being, like an Occurrence, a slice of the Universe, it is rather to be compared to a chemical principle extracted therefrom by the power of thought; and though it is or may be Real, yet, in its Real Existence it is inseparably combined with an infinite swarm of circumstances, which make no part of the Fact itself. (MS 647: 00010, 1910)
Thus, we must sharply distinguish events from facts, which are only such abstracted parts of an event as are expressible in a proposition (NEM IV: 252, 1904). In the above-quoted passage, an event is described as "a slice of the universe," while in a passage quoted below it is described as "the very objective history of the universe for a short time" (RLT 198, 1898). Thus, it would appear that Peirce meant by event a minimal temporal unit, or cross section, so to speak, of some actual process. In the following passage, Peirce criticizes the idea, defended by Mill, that what is caused, the effect or causatum, is an event. Instead, he insists that it is a fact, which is an abstracted element of an event: Mill's singularity is that he speaks of the cause of a singular event. Everybody else speaks of the cause of a "fact," which is an element of the event. But, with Mill, it is the event in its entirety which is caused. The consequence is that Mill is obliged to define the cause as the totality of all the circumstances attending the event. [...] He thus deprives the word of all utility. As everybody else but Mill and his school more or less clearly understands the word, it is a highly useful one. That which is caused, the causatum, is, not the entire event, but such abstracted element of an event as is expressible in a proposition, or what we call a "fact." The cause is another "fact." (EP II: 315, 1904)
In the following passage, a cause is described as the minor premise of a deductive argument, of which the major is the relevant law of nature and the conclusion the effect. Just like the effect, the cause is a fact:
A SEMEIOTIC ACCOUNT OF CAUSATION
191
So far as the conception of cause has any validity, [...] the cause and its effect are two facts.
Now, Mill seems to have thoughtlessly or nominalistically
assumed that a fact is the very objective history of the universe for a short time, in its objective state of existence in itself. But that is not what a fact is. A fact is an abstracted element of that. A fact is so much of the reality as is represented in a single proposition.
If a proposition is true, that which it repre-
sents is a fact. If according to a true law of nature as major premise it syllogistically follows from the truth of one proposition that another is true, then that abstracted part of the reality which the former proposition represents is the cause of the corresponding element of reality represented by the latter proposition. Thus, the fact that a body is moving over a rough surface is the cause of its coming to rest. It is absurd to say that its color is any part of the cause or of the effect. The color is a part of the reality; but it does not belong to those parts of the reality which constitute the two facts in question. (RLT 198, 1898)
Accordingly, causal propositions never refer to events in their full particularity; they refer only to such abstracted elements of events called facts. For instance, the question "What was the cause of the eruption of that mountain?" means "What is the fact from which [,] according to the principles of physics, necessarily resulted the fact that the mountain suddenly burst?" (MS 478:00155-56, 1903). By thus insisting that the causal relata are facts, Peirce makes clear that the general context of his discussion of the causal relata is epistemological rather than ontological. For, since a fact is defined as the correlate of a true proposition, there are no facts in and by themselves, independent of propositions. The context of facts is inherently epistemological. Contrary to causes and effects, which are facts, abstracted elements of concrete reality, reality itself is purely a matter of events: What is reality? Perhaps there isn't any such thing at all. As I have repeatedly insisted, it is but a retroduction, a working hypothesis [...]. But if there is any reality, then, so far as there is any reality, what that reality consists in is this: that there is in the being of things something which corresponds to the process of reasoning, that the world lives, and moves, and HAS ITS BEING, in [a] logic of events. (RLT 161, 1898)
Thus, while cause and effect are abstracted elements of concrete events, events rather than facts constitute 'the basic furniture of the world.' But contrary to the received view, Peircean events are neither changes in substances, nor do they presuppose the existence of substances. Instead, by saying that existing things are 'laws' unifying series of events, Peirce suggested that events are ontologically prior to substances:
192
CHAPTERFOLR ... individual existence, in the strictest sense, only belongs to a single event which happens when and where it does and has no other being. For though we speak, for example, of Philip of Macedón as an individual, yet 'Philip drunk' and Philip sober' were different. The 'existing' thing is only individual in the sense of being a continuous law regulating and unifying events of a series of instants. (MS 478:47-48, 1903)
Hence, it would appear that Peirce's conception of reality involves an event ontology in the strictest sense of the word. Concrete things and concrete persons owe their genetic identity to some 'law' or final cause, which regulates and unifies a series of events. Peirce in the end dismisses the idea that an event is an adjectival, an abstracted element of something more concrete. The most concrete or most determinate level of reality is not a substance or an enduring individual, but a momentary state or event in an ongoing process. In the words of Charles Hartshorne, this entails that "genetic identity is [just] a special strand of the causal order of the world" (Hartshorne 1970, 185). Whereas events are continuous (inasmuch as they do not have a definite beginning and a definite end (EP I: 314-15, 1892)), only those abstracted aspects of them called facts are trivially discrete. For, while events are temporal particulars, facts are neither temporal nor particular. Only facts, therefore, can be expressed in a proposition, and only facts can be components of a deductive argument. Events in their full particularity never can. Though Peirce nowhere explicitly made the distinction between causality and causation, he clearly did so implicitly. Causality involves a relationship between facts, but causation is purely a matter of events. Talking about the relationship between discrete causal facts implies that one abstracts from a continuous process of causation. Given that, according to Peirce, causation is a productive event that is part of a teleological process, I will now pursue the problem of the relationship between events and processes.
3.4
Events and processes
In the previous subsection I described Peircean events as minimal temporal units, or cross sections of some actual processes. I also said that the most concrete level of reality is a momentary state or event in an ongoing process. This raises the question, 'How are Peircean events related to Peircean processes?' In order to answer this question, I must first clarify what Peirce meant by event and by process.
A SEMEIOTIC ACCOUNT OF CAUSATION
193
(1) Though Peirce would not deny the reality of macro-events, such as witnessing a concert, 'Peircean events,' as I defined them, are micro-events, that is to say, minimal temporal 'slices' of an ongoing process. If my interpretation so far is correct, Peircean events share the following characteristics: (i) temporal extension, (ii) absence of definite temporal limits, and (iii) an infinitesimal duration. Moreover, (iv) they are ontologically prior to substances. (2) Since Peirce never explicitly discussed the concept of process, the best we can do is to consider what he said about natural (that is to say, irreversible, teleological, evolutionary, etc.) processes. In the previous section, it was shown that Peirce held the view that each process consists of a continuous, causal sequence of events, which is regulated by some final cause by virtue of which there is order in the world. This view may be clarified by making use of Dorothy Emmet's definition of process: ... processes are changes with an internal order which distinguishes what is happening within the process from forces acting on it from outside. [...] [They have] a direction, though not necessarily a terminus ad quern. (Emmet 1992, 35)
Indeed, Peircean processes too are characterized by an internal order and by directionality. In fact, the principle of the internal order of a Peircean process is its directionality, which in turn is determined by its final cause. The final cause sets the boundary conditions of the change toward an end state, which itself is, within certain limits, independent of any "forces acting on it from outside."11 Thus, I propose to give the following characterization of a 'Peircean process:' a 'Peircean process' is a continuous sequence of events that derives its unity or internal order (distinguishing it from other processes) from a final cause, which directs the sequence to some end state which itself may evolve. (3) We are ready now to have a closer look at the relationship between processes and events. Because Peircean processes and Peircean events are intimately related inasmuch as each event is part of a process, while each process consists of a continuous sequence of events, processes and events differ mainly in respect of complexity, teleology, and coherence. Consider the example of someone who, while driving a car from A to B, crashes the car before reaching B. While driving a car from A to B is a good example of a process, the several phases of the change involved count as an event. Driving a car is a complex action, while crashing a car is (comparatively) simple. Hence, it would appear that, while processes are characterized by complexity, events are simple.
194
CHAPTERFOLR
Moreover, while driving a car from A to B is a teleological process, the crashing of the car (qua crashing) is bereft of any teleology. Thus, it would appear that, contrary to processes, events (qua events) are not immediately characterized by any teleology. An event just happens. The teleology only comes in if we consider the event as a part of an ongoing process. Thus, processes are characterized by teleology, which events lack. Closely related to the previous point is that processes, such as driving a car from A to B, are characterized by a certain internal order or coherence. Events, however, being simple in themselves, lack this characteristic. Thus, Peircean processes differ from events inasmuch as only the former are characterized by complexity, teleology, and coherence. It is important, however, to realize that the suggested distinction is to some extent arbitrary. It owes its justification primarily to our (physical) inability to observe any teleology during an extremely short time span (events are characterized by an infinitesimal duration). But, as it will be shown further on, though events as such lack any recognizable teleology, as part of a teleological process, they are nevertheless teleologically determined by the final cause of the process. Thus, eventually events are just microprocesses.
3.5
Conclusion to part 3: causality and causation
In the third part of this chapter it was shown that Peirce held the view that (i) cause-effect relationships are irreversible, (ii) causes only partially determine their effects, (iii) cause-effect relationships are mediated by laws, and (iv) causes precede their effects in time. It was shown, furthermore, that causation involves an efficient component, a final component, and a chance component. The efficient component is that each event is brought about by a previous event (the efficient cause). The final or teleological component is that each event is part of a continuous sequence that is marked by a definite tendency. The chance component is that in each event there is some aspect of irreducible novelty. It was argued that, though Peirce nowhere makes an explicit distinction between causation (the production of an effect by its cause) and causality (the relationship between cause and effect), he did so implicitly in his contention that causality involves a relationship between facts, but causation involves the production of events. Finally, it was shown that Peirce held the view that causation is the production of an event, and that such production is part of a teleological process. The main difference between processes and events is that processes share three characteristics that events lack: complexity, teleology, and coherence.
A SEMEIOTIC ACCOUNT OF CAUSATION
195
In the next section, I will propose my new approach to causation, based upon Peirce's semeiotic.
4.
A SEMEIOTIC APPROACH TO CAUSATION
The close relationship between causation and signs is easily shown, for signs play a major part in our acquaintance with causes. Smoke is a sign of fire. In medicine, symptoms are signs that point to causes. Nevertheless, the apparently intimate relationship between causation and signs does not by itself imply that the relationship is transparent. It is all the more surprising that, as far as I know, it has never received any notable attention in contemporary discussions of causation. In this section I will not only pay attention to this relationship, but I will make it the corner stone of my theory of causation. In order to develop my semeiotic approach to causation, I will first (4.1) briefly explain some formal characteristics of semeiosis. Then (4.2) I will examine the hierarchical relation between semeiosis and causation. Next (4.3) I will try to further determine the nature of the relationship between semeiosis and causation. It will be argued that - contrary to the received view among Peirce scholars - semeiosis cannot be analyzed in terms of causal elements, because it allegedly only concerns the formal structure of causation. Finally (4.4), I will list some of the main formal characteristics of causation, and some of the main formal characteristics of signs, and discuss how causation can be analyzed in terms of signs.
4.1
Some formal characteristics of semeiosis
Peirce used the term 'semeiotic' for the scientific study of semeiosis, or the systematic study of the formal structure of "the general conditions of signs being signs" (CP 1.444, c. 1896). The most important formal characteristic of a sign is that it involves a three-term relationship between a sign, its object, and its interprétant. This relationship is irreducibly triadic, that is to say, it is not reducible to a summation of dyadic relations (CP 2.274, c.1902). According to Peirce, a sign is "anything which is related to a Second thing, its Object, in respect to a quality in such a way as to bring a Third thing, its Interprétant, into relation to the same object" (CP 2.92, 1902). Thus, a sign is not a thing as such, but anything may be a sign by virtue of being triadically related to an object and an interprétant. The interprétant is the respect in which sign and object are related; it must be understood, not as an act of interpretation, but as "the sign's proper significate effect" (CP 5.473, 1907). Thus, if clouds are a sign of rain to
196
CHAPTERFOLR
come, the rain to come is the object of the sign, and the clouds perceived as signifying (possible) rain, is the interprétant. Peirce's semeiotic is closely related to his categoreal scheme. The basic idea behind this scheme is that all phenomena are infected by Firstness, Secondness, and Thirdness. Whatever it may be we are describing, be it theories, arguments, or events, there is always an element that does not refer to anything beyond itself (First), an element that is related to this first element but not to anything else (Second), and an element that relates or mediates between the first and the second element (Third) (e.g. CP 8.328, 1904). In other words: Firstness is a monadic relation, Secondness a dyadic relation, and Thirdness a triadic relation. Signs are paramount examples of triadic relations. Thus the interprétant of a sign mediates between the sign and its object. Moreover, the triadic relation involved in a sign is eminently dynamic. Signs do something: All dynamical action, or action of brute force, physical or psychical, either takes place between two subjects (whether they react equally upon each other, or one is the agent and the other patient, entirely or partially) or at any rate is a resultant of such actions between pairs. But by 'semiosis1 I mean, on the contrary, an action, or influence, which is, or involves, a cooperation of three subjects, such as a sign, its object, and its interprétant, this tri-relative influence not being in any way resolvable into actions between pairs. (CP 5.484, 1906)
Though Peirce seemed to define semeiosis in an implicitly causal terminology - the words 'action' and 'influence' being key words in his definition - he explicitly denied that the triadic action constituting semeiosis could be reduced to dyadic cause-effect influences. Semeiosis is irreducibly triadic. Thus, if someone beats his fist on the table (sign) to express his anger (object), the expression of fear in a witness's face (interprétant) is triadically produced, and is therefore a semeiotic effect. The mechanical effects of the blow are not semeiotic effects, at least not from our point of view as interpreters of the sign. This distinction will be of paramount importance for the theory that will be presented. But by thus defining semeiosis in an implicitly causal terminology, Peirce gave the impression that causation is more fundamental than semeiosis inasmuch as semeiosis would presuppose causation, rather than the other way. This suggestion, however, is misleading. In the next two subsections I will argue that, while semeiosis cannot be analyzed in terms of causal elements, causation can be analyzed in terms of semeiotic elements.
A SEMEIOTIC ACCOUNT OF CAUSATION
4.2
197
The problem of semeiotic causation
In the previous chapter, I established that there is a consensus in Peirce scholarship that Peirce conceived semeiosis as a teleological process, that is to say, as a causal process marked by a definite tendency. I concluded, furthermore, that if semeiosis is indeed some sort of causal process, then it must, according to Peirce, involve final causation, efficient causation, and chance. I used the term 'semeiotic causation' for the role of these causal elements in semeiosis. However, it appeared that any attempt to precisely and unambiguously indicate the respective roles of these causal elements in semeiosis, is highly problematic. In this section, I will discuss a suggestion (unpublished) of Guy Debrock that, though the above-given formulation of the problem of semeiotic causation is indeed consistent with a great many of Peirce's texts, it is nevertheless misleading. According to Debrock, the very idea of analyzing semeiosis in causal terms is based upon the failure to distinguish the signaspect from the event-aspect of signifying events: whereas causation is a productive event, semeiosis involves only a (dynamic) relationship between certain aspects of events. While every causal process requires semeiosis, and semeiosis requires a causal process, causal processes and semeiosis do not coincide. (In this respect, the problem of the relationship between semeiosis and causation is analogous to the problem of the relationship between time and change. While time involves change and change involves time, time is not the same as change.) I will argue that this insight of Debrock opens the door, not only toward a new way of looking at 'the problem of semeiotic causation,' but also, and more relevant to our present concern, toward a new, semeiotic approach to causation as such. Debrock's argument may be illustrated by the example of smoke as a sign of fire. Obviously, the fire is the cause of the smoke, which in its turn is causally related to our conclusion that this smoke signifies fire. Moreover, 'fire' is the object of the smoke sign, while 'smoke perceived as signifying danger' is its interprétant. But, this does neither mean that the smoke qua sign is caused by the object 'fire,' nor that the interprétant 'danger' is caused by the smoke qua sign. Since there simply is no sign-object without a sign, and no sign without an interprétant, it is sheer nonsense to assign a causal function to any of these semeiotic elements. Object, sign, and interprétant cannot be causally related one to another, because together they constitute what it is to be a sign. Since semeiosis is irreducibly triadic, its elements cannot, qua semeiosis, be causally related. Thus, Debrock's point is that, contrary to causation, semeiosis as such does not involve the production of events; it only concerns the dynamic relation of certain aspects of events. Consider once more the example of
198
CHAPTERFOLR
someone beating his fist on the table to express his anger. Clearly, beating his fist on the table is an event, which functions as a sign of anger. But this very same event involves, in principle, innumerable signs, because it involves innumerable interprétants. In some exotic culture, for example, beating one's fist on the table might be the expression of feeling happy, or being excited, and would therefore be a completely different sign than the one mentioned above. It is easy to imagine many other different contexts in which the very same act of beating (event) would function as a different sign. Thus, in principle, each event involves innumerable signs, because it potentially involves innumerable interprétants. In the next sub-section I will give two arguments that support Debrock's hypothesis that, while semeiosis does not entail causation, causation does entail semeiosis. More specifically, I will give two - admittedly formal reasons why I think semeiosis provides the formal structure of causation.
4.3
Semeiosis provides the formal structure of causation
In the first place, the hypothesis that semeiosis provides the formal structure of causation is supported by Peirce's hierarchical classification of the sciences (c.1900), which arranges the abstract sciences in a ladder such that each science derives its fundamental principles from the findings of the more abstract science that occupies the rang above. In Peirce's own words: "I would classify the sciences upon the general principle set forth by August Comte, that is, in the order of abstractness of their objects, so that each science may largely12 rest for its principles upon those above it in the scale while drawing its data in part from those below it" (RLT 114, 1898). Thus, according to Peirce's Cambridge lectures of 1898, Reasoning and the Logic of Things,13 there is a hierarchical relationship between Logic (which he defined as Semeiotic) and Metaphysics, such that Logic is one scale higher on the ladder of abstractions. Hence metaphysics must draw its principles from semeiotic: it "must take as the guide of its every step the theory of logic [semeiotic]" (RLT 116; also 123, 1898). The idea that semeiotic is a more general discipline than metaphysics entails that metaphysical concepts presuppose semeiotic principles, while semeiotic concepts do not presuppose metaphysical principles. Thus, the concept of causation (which is a metaphysical concept) presupposes semeiotic principles. My second argument concerns the relationship between semeiosis, Thirdness, and causation. I already established that the most important formal characteristic of semeiosis is its irreducible triadicity. There are, however, some indications that Peirce expanded the view that semeiosis
199
A SEMEIOTIC ACCOUNT OF CAUSATION
involves Thirdness into the much stronger thesis that semeiosis is the genuine form of Thirdness: In its genuine form, Thirdness is the triadic relation existing between a sign, its object, and the interpreting thought, itself a sign, considered as constituting the mode of being a sign. A sign mediates between the interprétant
sign and
its object. (LW 31, 1904; cf. CP 1.537, 1903)
Thus, it would appear that each genuine triadic relation is basically a relation between a sign, an object, and an interprétant. Thus semeiosis is not only irreducibly triadic, but each triadic relation in the strongest sense is also formally semeiotic. Peirce makes a distinction between genuine and (two forms of) degenerate triadic relations.14 Only the former cannot be reduced to a summation of Seconds. Thus "[t]he genuine third category is where there are three objects each having a character which essentially supposes the other two" (MS 304:00041, 1903). Thus, while 'A gives B to C' is an example of genuine Thirdness, 'A putting down B, which is picked up by C' is an example of degenerate Thirdness (LW 29-30, 1904). Now, we know that causation involves a genuine triadic relation between an efficient cause, a final cause, and an effect, for each of these terms "essentially supposes the other two." And if causation involves genuine Thirdness, and if genuine Thirdness is formally semeiotic, then it would appear that causation is formally semeiotic. In other words: semeiosis provides the formal structure of causation. Though these two arguments do not unconditionally prove my hypothesis that causation presupposes semeiosis, rather than the converse, they make clear that my hypothesis fits Peirce's theoretical framework.
4.4
A semeiotic approach to causation
It may be helpful to point out again that the semeiotic approach to causation that will be defended, is Peircean in the sense that it is primarily based upon Peirce's semeiotic and upon his conception of causation. In order to see what new light Peirce's semeiotic may throw on the concept of causation, it is helpful to recall some formal characteristics of Peirce's conception of causation. This will allow us to elaborate an anatomy of (Peircean) events, and finally to discuss some more formal characteristics of signs. 4.4.1
Some formal characteristics of causation
So far it was established (see section 2) that a coherent theory of the relationship between events and processes is one of the primary
200
CHAPTERFOLR
requirements for any theory of causation. Peircean events have beer characterized by (i) temporal extension, (ii) the absence of definite temporal limits, (iii) an infinitesimal duration, and (iv) ontological priority in respect of substances. Moreover, a Peircean process was defined as "a continuous sequence of events that derives its unity or internal order (distinguishing it from other processes) from a final cause, which directs the sequence to some end state, which itself may evolve" (3.4). Peircean processes differ from Peircean events inasmuch as only the former are characterized by complexity, teleology, and coherence. Moreover, each productive event or act of causation was seen to involve a dynamic relation between a general final cause, a concrete efficient cause, and its concrete effect. Thus, each act of causation involves two aspects that must be sharply distinguished: the fact that something is caused, and what it is that is caused. While the former is explained by the efficient cause, the latter is explained by the final cause. Thus, the final cause does not determine that an event occurs; it only determines what type of event it will be, if the event occurs. In other words: it does not determine the action qua action, but it determines the general type or form of the action. In section 3.2, I gave the following schematic representation of Peirce's conception of causation: C' (final cause, habit)
> (cause)
(effect, means)
C (effect, realized end)
The sequence A —» B —» C represents a continuous sequence of events. Thus, between any event A, apparently causing another event B, there is an innumerable series of events affecting B, which may therefore as well be called the cause of B. Describing A as the cause of B abstracts from the aspect of continuity involved in causation; in other words, one changes his perspective from causation toward causality (and thereby from events to facts). From the dynamic point of view, the scheme represents the production of a sequence of events, in which A-B (as mediated by C') must be thought of as one concrete, productive event. This event is part of the continuous sequence A —» B —> C, which derives its unity teleologically from one and the same final cause (C). It is only after the one concrete, productive event A-B has occurred, that abstraction yields A as the cause and B as its effect;
A SEMEIOTIC ACCOUNT OF CAUSATION
201
for it is only after an effect has been produced, that one can speak of 'causes' and 'effects.' In view of this brief survey of Peirce's conception of causation, I may now provide an anatomy of Peircean events. 4.4.2
The anatomy of events
According to Peirce, each event has an aspect of Firstness, Secondness, and Thirdness. The aspect of Firstness concerns the particular qualities - which as such are mere possibles - that are involved in the event. Secondness concerns the aspect of action hie et nunc, independently of any uniformity or law. Thirdness concerns the aspect of habit, or continuity of form; Thirdness "determines the suchness of that which may come into existence, when it does come into existence" (EP II: 269, 1903). Thus, each event owes its definiteness or form to Thirdness, which is of the nature of habit or final cause. Though Firstness, Secondness, and Thirdness seem to reflect the traditional modal categories of, successively, possibility, actuality, and necessity, the necessity of Thirdness is not absolute, but it is conditional necessity of the form: if A, there will be a tendency toward B (e.g. EP II: 271, 1903). Peirce held the view that modal distinctions correspond to temporal ones; thus possibility, actuality, and conditional necessity (or potentiality) are the modes of, respectively, present, past, and future (CP 5.459, 1905). Accordingly, in each Peircean event three distinct temporal phases may be recognized: (1) reception of causal influence from the past, (2) (present) selfdetermining activity, and (3) influence upon subsequent events. Thus, every present event is conditioned by the past and conditions the future. Though the present event conforms to the past, it nevertheless contains an element of irreducible novelty, for causal conditioning always leaves a range of open possibilities, however small it may be sometimes. This insight, which Peirce shared, among others, with Bergson and Whitehead, was formulated by Whitehead in the following words: "An actual entity [that is, an event] is at once the product of the efficient past, and is also, in Spinoza's phrase, causa sui" (Whitehead [1929] 1978, 150). Every present event is self-actualizing; it is an act of decision that selects and actualizes one possibility among various possibilities. Which of the causally possible events occurs is decided in the event itself, rather than beforehand. Though future events are not necessitated in their full particularity, certain abstract, more or less general, features are determined in advance. Hence, given a certain event, it is necessary that a certain kind of event will occur, while it is not necessary that this rather than some other event will
202
CHAPTERFOLR
occur. For, it belongs to the nature of an event, not only that it creates a new event (Secondness), but that it creates a new event of a certain type or natural class (Thirdness). To what type a new event will belong, is determined by a (cluster of) previous event(s), and by the relevant law or final cause. For instance, an egg produced by a turtle will always be a turtle egg, partly because the event emerges from biological processes that occur in the mother turtle, and partly because of the 'law' of the development of a turtle egg. Thus, each event (1) is causa sui, that is, it involves a decision selecting one possibility among various possibilities, (2) is conditioned by a cluster of previous events, and by the relevant final cause, (3) conditions and limits the range of possibilities for subsequent events. This view, which I consider to be totally compatible to Peirce's approach, entails that the relevant final cause is the perspective by virtue of which the efficient cause and its effect are related. This claim needs some clarification. Consider the macro-event of twenty million people watching a live speech by the new president of the United States on some controversial political subject. No doubt, there will be certain recognizable patterns in the reactions of the audience. Democrats will tend to react differently from Republicans. But on a closer view it will appear that the speech will yield twenty million different reactions, for each observer watches the speech event in the light of his own expectations, predispositions, feelings, etc. In other words, it is by virtue of the fact that each spectator observes the event from his own perspective, that the very same event yields such different reactions. Such perspectives are basically habits, which help determine the future reactions of a person to specific events. For instance, if a person's perspective involves an abhorrence of discriminatory rhetoric, each utterance that smacks of discrimination may cause him to be angry. Thus each perspective reflects a conditional necessity, which has the structure of a material implication. The probable or potential reaction of, say, John to an event of the type X may therefore be described as follows: If an event of the sort X occurs, then John will have a strong tendency to react in such and such a way (way Y). Schematically, the occurrence of John's fury may be represented as follows:
A SEMEIOTIC ACCOUNT OF CAUSATION
203
C' (perspective; final cause)
A
A (cause)
B (effect)
While A is John's perception of discriminatory talk, B is John's angry reaction, and C' is John's abhorrence of discrimination, A-B-C' is the one productive event of John's becoming angry. It would appear that each new event assimilates the past event from a definite perspective. The past event is related to the present event as an efficient cause to its effect. This relationship is mediated by the final cause, which is the perspective. Whereas each present event assimilates the past event from one definite perspective, each past event may be assimilated from innumerable different perspectives. For instance, the just-mentioned speech-event of the new president was perceived from twenty million different perspectives. It would seem that we have an important clue here that may be useful to my semeiotic approach to causation. For the very same speech-event functioned as twenty million different signs, because there were twenty million different interprétants. Hence, it would appear that, from a semeiotic point of view, the perspective from which the previous event is assimilated is precisely the interprétant. If the latter hypothesis is correct, and if causation does indeed presuppose semeiosis, then, in order to better understand causation, we must now try to obtain an even more precise idea of what semeiosis involves. We will see that it involves the transmission of forms from causes unto effects. Because the expression "transmission of forms" may reek of Aristotelianism, it is of the utmost importance to spell out what it refers to. 4.4.3
Signs as media for the transmission of forms
As we have seen again and again, Hume's greatest difficulty with the concept of causation was that it was impossible to make a factual observation of the necessity that is implied in the concept of causation. His position, however, was based upon an epistemology and an ontology that had rejected Aristotelianism in toto. Hence, the idea that necessity might be related to the transmission of forms was an option that he could not envisage. According to Aristotle, efficient causation involved a form being transmitted from the efficient cause to the effect. In natural processes the efficient cause was the same in form as the effect, as when "man begets
204
CHAPTERFOLR
man." In artificial productions, the efficient cause was the form in the mind of the artificer. It was his soul, which contained the form, that moved the artificer's hands so as to make an object of the same form (Generation of Animals I: 21-22). Hume certainly was right in rejecting the Aristotelian formulation of the doctrine of forms, which presupposes an ontology of substances. Aristotle never explained how a transmission of forms was to take place, if forms belonged to substances. Indeed, a form was supposed to be precisely that by virtue of which a substance is what it is. Though Hume never even referred to the Aristotelian doctrine, it is clear that he still conceived of the problem of causation in Aristotelian terms. It is precisely because he conceived the problem in Aristotelian terms, that he saw its limitations; and it is because he was aware of those radical limitations, that he felt he must reject the Aristotelian ontology. On the other hand, by recasting the problem in terms of his own epistemology, he was faced with the impossibility of solving the problem. If forms belong to substances, then the transmission of forms must remain a riddle that cannot be solved by a rational account. Causality must be a purely external relation, which could be explained only by invoking the psychology of habit. We have already seen how the Humean idea of causality as a purely external relation was severely criticized by Peirce and Whitehead who insisted that causation cannot be explained except in terms of some active immanence, or by a continuous transmission, of forms. But clearly, Peirce immediately translated the concept of 'form' in terms of his own framework. This view is most clearly expressed in the following passage in which he explains that forms must be seen as general rules: The being of a Form consists in the truth of a conditional proposition. Under given circumstances, something would be true. (EP II: 544, n.22, 1906).
This description of form, though referring to the Aristotelian terminology, cannot be understood without the pragmatic context that is basic to Peirce's thought. Forms are not related to properties that are inherent to things, because our belief in properties is really our belief in the conditional fact that, given certain interactions, such and such effect would produce itself. In other words, our very reference to 'properties' of things already presupposes a possible immanent relation, which is the hallmark of all interaction. But in order to understand the relationship between forms as general rules and causation, two intermediary steps are required. It has been repeatedly pointed out that the logical structure of thought, rules and generals is the same; it is the structure that is expressed in the material implication: if p, then q. This is the logical equivalent of what Peirce called
205
A SEMEIOTIC ACCOUNT OF CAUSATION
Thirdness. If such is the case, and if forms are equivalent to rules, clearly the transmission of forms involves the transmission of Thirdness. Thus, within Peirce's framework the burning question of causation must be understood in terms of the question how Thirdness may be transmitted. And in this respect there is no doubt that the answer must be sought in Peirce's conception of semeiosis. For indeed, the preceding quote regarding forms, which are to be understood as rules, is part of a larger context in which Peirce explains what he meant by semeiosis: ... a Sign may be defined as a Medium for the communication of a Form. [...] That which is communicated from the Object through the Sign to the Interprétant is a Form. It is not a singular thing; for if a singular thing were first in the Object and afterward in the Interprétant outside the Object, it must thereby cease to be in the Object. The Form that is communicated does not necessarily cease to be in one thing when it comes to be in a different thing, because its being is a being of a predicate. The being of a Form consists in the truth of a conditional proposition. Under given circumstances, something would be true. The Form is in the Object, entitatively we may say, meaning that that conditional relation, or following of consequent upon reason, which constitutes the Form, is literally true of the Object. In the Sign the Form may or
may
not
be
embodied
entitatively,
but
it
must
be
embodied
representatively, that is, in respect to the Form communicated, the Sign produces upon the Interprétant an effect similar to that which the Object itself would under favorable circumstances. (EP II: 544, n.22, 1906)
The premise of Peirce's argument is that the object of a sign is always a rule, and inasmuch as a rule has the same logical structure as a thought, the object of the sign may be considered to be a thought. The sign represents this rule or thought, and by doing so, yields an interprétant which the object itself would have yielded, were it not that the relation between the object and the interprétant must be mediated through the sign. When smoke acts as a sign of fire, it represents the rule "if you get close to fire, you will get burnt," and by so representing that rule it yields the interprétant "get away unless you get burnt," which is exactly the rule which fire by itself would yield without the smoke if that were possible. The function of the sign therefore is to mediate in the transmission of a rule from the object to the interprétant. The only problem remaining is to show how the transmission of forms within semeiosis is related to the transmission of forms within causation. At the outset an important point that - though obvious - is often overlooked, is that the question of causation always enters only after a particular effect has struck our attention. The obviousness of this is echoed in the principle that "every event has a cause." Put in another way, this amounts to saying that
206
CHAPTERFOLR
every effect is the 'sign' of some event that is considered to be its cause. Thus causation always implicitly involves semeiosis. According to the received view, causation requires two and only two elements: a cause and its effect. Semeiosis on the other hand, inherently involves three aspects: the sign, its object and its interprétant. It should be clear from the given analysis that the effect is the sign, while the cause is the object. The paramount issue then is, "What sense can be made of the interprétant within the context of causation?" The answer should by now be rather simple and straightforward. An effect can be the sign of a cause only from a certain perspective. It is only because the death of a person is important to us that we look after its cause. However, it is not only the effect that is selected by our perspective: the very same event (effect) may be said to have different causes, depending on one's perspective. Thus, for example, while the cause of the death may be the deprivation of oxygen for a pathologist, it may be the act of a criminal who deliberately deprived the victim of oxygen for a policeman, whereas it may be some traumatic experience of the criminal in the view of a psychiatrist. Speaking of cause and effect without at least implicitly indicating the perspective from which an effect is produced by its cause is as vacuous as speaking of natural classes without specifying the perspective from which certain properties are considered important. This perspective is precisely where interprétant and final causation coincide. Although a reference to final causes has become a virtual taboo in modern philosophy, it should be emphasized again that, within the perspective of Peirce, final causes are not events in the future bringing about events happening now; final causation merely refers to that sort of general rule in virtue of which events that follow upon each other may be thought in relation to one another. Thus, though an apple falling to the earth undeniably involves the mass of the apple, the mass of the earth, and the distance between those masses, there can be causation only inasmuch as the behavior of the earth and the apple in respect of each other happen within the boundaries set forth by the law of gravity; this law states no less and no more than that, given the masses of the apple and the earth, and given the distance between them, the apple and the earth will behave in such determinate way. This conditional relation constitutes on the causal level the final cause, and on the semeiotic level the interprétant. Apparently, Peirce's concept of Form has roughly the same function as the Aristotelian formal cause: it explains the apparent stability of the world. However, it radically differs in at least one respect from the Aristotelian formal cause. Whereas the formal cause was thought to explain the stability of the world by explaining the structure of things, Peircean Forms explain the stability of the world by explaining the dynamic relations between
A SEMEIOTIC ACCOUNT OF CAUSATION
207
events. In other words: the structure of the Forms themselves is relational, rather than something embodied in a substance. If my hypothesis is correct that semeiosis concerns the aspect of the transmission of forms from causes unto effects, then, in order to better understand causation, we must now have a more precise look at what goes on in semeiosis. 4.4.4
Signs conveying the form-aspect of causation
We established that Peirce held the view that signs convey significate forms, which are of the general nature of thought. We established, furthermore, that a productive event assimilates a past event from a definite perspective, which is the final cause, and that, on the semeiotic level, this perspective or final cause is the interprétant. Peirce defined the interprétant as "all that is explicit in the sign itself apart from its context and circumstances of utterance" (CP 5.473, 1907). For instance, the interprétant of the symptoms of measles is the real law-like relationship between measles and its symptoms. The symptoms are there, and so is the interprétant to which they would lead a qualified doctor.15 Thus, the thought is 'there,' even in case there be no thinker of it. Hypothetical thinking may be regarded as that sort of thinking in which we ascribe to a particular event the status of being the sign of an object. And since 'reasoning from effect to cause" is an instance of hypothesis (EP I: 194, 1878), it would appear that causal explanations are abductions or hypotheses in which the effects are conceived as functioning as signs which point to their causes. The relevant 'law ' or habit is the interprétant by virtue of which the effect is related to its cause. For instance, the symptoms of a disease point to its cause. The relevant medical law is the interprétant by virtue of which the symptom (sign) is related to its cause (object). Considered from the perspective of causation as a productive event, the law determines that the symptoms caused by the particular disease belong to a certain type or natural class. Measles, for instance, cause red spots rather than tooth aches. We may now consider the following schematic representation of a productive event in semeiotic terms: C' (interprétant; perspective, final cause)
A (object; cause)
B (sign; effect)
208
CHAPTERFOLR
Thus, while A stands for measles, B stands for red spots, and C' stands for the law that measles cause red spots, A-B-C' represents the structure of the single productive event of the emergence of red spots. Thus the emergence of red spots is a new event, which assimilates the past event - some functional disorder called measles - from a definite perspective. This perspective is the law that measles cause red spots. This law is the Form, which is conveyed in the sign. The Form, which is really embodied in the object, is embodied in the sign only in a representative sense. Thus, the red spots represent the embodied Form of measles. If all this seems fairly abstract, a reference to some insights gained from modern psychological research may illustrate the point. Modern scientific research on memory confirms Whitehead's hypothesis that recollection is a productive event, an activity here and now, rather than some passive effect of a past cause. Moreover, if Whitehead is right that in memory we have concrete examples of the interconnectedness of events, then this modern research on memory confirms my hypothesis that each productive event involves an assimilation of a past event from a definite perspective. For recent research on memory has shown that the way in which people remember past events depends upon their intentions and objectives at the moment they try to remember the event (Schacter 1996, ch.2). In terms of my scheme of productive events, this means that the present recollection event is not only determined by the past, remembered event, but also by the present perspective on the past event. While the past event functions as the efficient cause of our recollection experience, the perspective is the final cause. Thus, a different perspective on the same past event leads to a different type of recollection. Apparently, each recollection is an interpretation of the past event. The above-given examples suggest the paradoxical conclusion that causes are both objective and perspectival; not only are causes part of what objectively goes on, but they are also determined by the perspective on what goes on. They are facts rather than events, because they are abstractly derived aspects from the concrete, objective stream of events. As such they are partially determined by concrete reality and partly by the abstractive power of our thought. (Though in the example of the measles, the objective law was emphasized, it is only by virtue of the modern scientific theoretical perspective that the disease can be explained). We are ready now to understand my guess at the riddle how forms can be transmitted from a cause unto its effect. There is an obvious structural similarity between the assimilation of a past event by a present productive event, and the representation of a past object in a present sign. Consider once more my schematic representation of a productive event:
A SEMEIOTIC ACCOUNT OF CAUSATION
209
C' (perspective; final cause)
A (efficient cause)
B (effect)
Consider now the structural similarity with the semeiotic triangle:
C' (interprétant)
A (object)
B (sign)
The structural similarity of (i) the assimilation of a past event by a present productive event and (ii) the representation of an object by a sign, can only be explained if we assume semeiotic principles to be really operative in each act of causation. The in itself baffling transmission of forms in causation is explained in terms of the transmission of forms in semeiosis, simply because the transmission of forms in causation is basically a semeiotic process. Thus Peirce's contention that "general principles are really operative in nature" (EP II: 183, 1903) is translated into my hypothesis that 'semeiotic principles are really operative in each productive event.,16 The prediction of future events, which is, in some sense, the counterpart of causal explanation, also implicitly refers to semeiosis. In this case, the sign stands in relation to its object as an efficient cause to its effect. Thus, clouds are a sign of rain to come (object) because clouds (condensed vapor) may cause rain. The interprétant - clouds perceived as signifying possible rain - reflects the law-like relationship between clouds and rain under certain conditions. Indeed, the interprétant expresses the law-like relationship between cloud-signs and possible rain (object). Reasoning from an event ('cause') to its possible effect is, just like reasoning from effect to cause, an instance of hypothesis. Thus, it would appear that predictions of future events are hypotheses in which certain events ('causes') are conceived as functioning as signs that point to their would-be effects. The relevant 'law' is the interprétant by virtue of which the cause is related to its effect.
210
CHAPTERFOLR
So far I have explained the role of signs in single productive events, in causal explanations of past events, and in our predictions of future events. There remains, however, one more problem to be discussed: the relationship between signs and causal chains. I established that causes and effects are abstractly derived aspects from a continuous chain of events, and that whatever we call the cause of a certain event depends on a perspective in view of which the continuous sequence is observed. But if my conclusion is correct that effects function as signs pointing to their causes (causal explanations), or that causes function as signs pointing to their would-be effects (predictions), then signs cannot be discrete entities, but must be abstractions of a continuous process of semeiosis. Fortunately, the objection is not serious, for the idea that (Peircean) signs are discrete involves a complete misunderstanding of Peirce's concept of sign for two reasons. First, the objection presumes that signs are a species of entities, while in fact they constitute a certain aspect of events. Furthermore, as Joseph Ransdell has pointed out, atomicity and signs are mutually exclusive concepts: ... the individual 'links' in the chain [of semeiosis] are always potentially analyzable into innumerably many possible sub-links and sub-sub-links [...], since a sign is not a 'logical atom' but only something which it happens to be intellectually profitable to treat as a unit at a certain point in an analysis, though always analyzable in principle into sub-signs if or when the analytic project should require it, and which may itself be a sub-sign within some more exclusive sign. (Ransdell 1986, 677)
Thus it is because both semeiosis and causation involve real continuity, that any of the innumerable events of which a causal chain consists, may function as a sign which points to its cause (causal explanation) or to its would-be effect (prediction). In the next section, we will see whether, and to what extent, my semeiotic approach to causation can handle the main requirements for a theory of causation as put forward in section 2. 4.4.5
My semeiotic approach and the general requirements for a theory of causation
In section 2,1 postulated two primary requirements for a theory of causation: it should be coherent, and it should involve a coherent theory of events and processes. The theory presented meets the first requirement inasmuch as it is based on one, and only one, categoreal framework. Contrary to the received view, which is caught between a substance ontology and a fact ontology, the theory presented is based upon an event ontology in the strictest sense of the
A SEMEIOTIC ACCOUNT OF CAUSATION
211
word (that is to say, one that is not in any way based upon a substance ontology). The theory meets the second requirement by conceiving the relationship between processes and events as expressed in my definition of a (Peircean) process as "a continuous sequence of events that derives its unity or internal order (distinguishing it from other processes) from a final cause, which directs the sequence to some end state which itself may evolve" (3.4). I also listed four subsidiary requirements for a theory of causation: it must explain (1) whether causation involves continuity, (2) the relationship between causation and time, (3) the relationship between causation and teleology, and (4) the status of causes and effects as such, and the relationship between causes and effects. (1) Continuity. It was already established that causation is a productive constituent of a continuous sequence of events, deriving its unity teleologically from one and the same final cause. It was established, furthermore, that between any event A, apparently causing another event B, there is an indefinite series of events affecting B, which are also involved in causing B. Describing A as the cause of B, abstracts from the aspect of continuity involved in causation. Both Peirce and Whitehead made clear that causation must somehow involve a continuous transmission of forms. Given that such transmission cannot be explained on the basis of a substance ontology (as Hume has shown), and given that our most concrete experiences concern events rather than substances, it would appear that the explanation of causation requires an event ontology in the strictest sense of the word. Such ontology is provided by Peirce's categoreal scheme; within the context of this scheme, the urgent question how forms be transmitted from causes unto effects must be understood in terms of the question how Thirdness may be transmitted. And since semeiosis provides the formal structure of genuine Thirdness, the solution to this problem must be thought in terms of Peirce's conception of semeiosis. Peircean forms are general rules, which have the logical structure of a material implication: if p, then q. Thus, unlike Aristotelian formal causes, Peircean forms are relational structures rather than something embodied in a substance. It was shown that the transmission of forms in causation is basically a semeiotic process in which the function of a sign is to mediate in the transmission of a form from the object to the interprétant. Thus, effects may function as signs that point to their causes, and 'causes' may function as signs of future effects, because the transmission of forms from causes to effects is a semeiotic process. (2) Time. According to Peirce, the most distinctive character of time consists is its asymmetry. While earlier members of an event-sequence may affect later members, the converse does not hold. Similarly, simultaneity is
212
CHAPTERFOLR
defined in terms of unaffectibility: two events are contemporaneous if and only if "each is absolutely unaffectible by the other" (EP I: 323, 1892). But since affectability is a causal concept, it would appear that Peirce held the view that time is derivative with respect to causation. If this is correct, then Peircean events do not happen in time, but they are the condition of time, whatever time may be. Whether this is a correct interpretation of Peirce's view or not, I accept it as a valid interpretation of time per se; for, if my hypothesis is correct that the present act of causation (that is, the present productive event) is the most determinate level of reality, and that temporal ordering is generated by relating events to a present productive event, both dimensions of time, the past and the future, are abstractions of an ongoing sequence of productive events. Thus causation, rather than time, is ontologically primary. My analysis of causation conceived as a productive event implies not only that causes and effects are abstractly separated elements of an ongoing causal process, but also that causes precede their effects. For were there no temporal distance between the two, the cause would co-exist with the effect, and the effect of that effect would co-exist with it, and so forth. In such a case, nothing would occur, neither would there be any time. To be sure, it is often suggested that there are clear examples in which causes and effects are simultaneous, such as a ball pressing on a cushion, or a hand raising a pen. However, I think that Paul Weiss's suggestion is correct, that in those examples the simultaneity of cause and effect is only apparent, and that what we are inclined to call the cause, is nothing but an analytic component of the effect (Weiss [1947] 1983, 4). Thus, the movement of my arm is not the cause of the movement of the pen, but (for example) my decision to raise the pen is the cause. Another option would be to accept Michael Tooley's solution that causes and effects cannot be simultaneous because, if relativity theory is correct, "there is a finite speed with which causal processes can be transmitted, so actual examples of simultaneous, causally related events can never involve events that are spatially separated." However, even in cases of physical objects that are in contact, such as a hand raising a pen, there must be some (infinitesimally small) distance separating hand from pen. Hence, causes and effects cannot be simultaneous (Tooley 1987, 208). (3) Teleology. An important part of the problem of teleology is the question whether all natural processes can be adequately explained in terms of efficient causality. Here, Peirce forcefully defends the position that all natural processes are to some extent teleological, and can only be explained on the basis of final causes, which are general principles. The empirical fact that, as far as I know, all (physical, biological, psychological and sociological) processes show a definite tendency toward a general end state -
A SEMEIOTIC ACCOUNT OF CAUSATION
213
whether it be described as chaotic or as involving a higher level complex order - cannot be explained without referring to certain general principles that somehow direct the process. As far as I know, Peirce's theory of teleology is the only theory that explains this fact (see chapter 3). On Peirce's account, teleology becomes again a key-concept in the theory of causation. For there can be causation only inasmuch as there are boundaries set forth by certain general principles. (4) The status of causes and effects as such, and the relationship between causes and effects. It was established that, while causation involves a continuous sequence of events, causality involves a relationship between facts, which are abstractly separated elements of events. Hence, describing A as the cause of B abstracts from the aspect of continuity involved in causation; in other words, one changes his perspective from causation to causality, and thereby from events to facts. Thus, the status of causes is that they are both objective and perspectival; they are facts rather than events, because they are abstractly derived aspects from the concrete, objective stream of events. As such they are not only determined by concrete reality, but also by the abstractive power of our thought. It was established that the 'logic of events' is such that, though future events are not necessitated in their full particularity, certain more or less general features are determined in advance. Thus, given a certain event, it is necessary that a certain type of event will occur, while it is not necessary that this rather than some other event will occur. Given certain causal conditions (that is, a cluster of previous events and the relevant final cause), there is always only a limited range of possible events that can occur. While the common character of all these possible future events (effects) is necessary, the concrete, actual effect is not. The necessity involved in causation (physical necessity) is related to the transmission of Forms in the Peircean sense, which are final causes. They determine to what type or natural class a new event will belong. As such they help explaining the relative stability of the world.
5.
CONCLUSION : A PEIRCEAN APPROACH TO CAUSATION
The objectives of the first part of this chapter were (1) to point out that the current theories of causation are radically inadequate, and (2) to show the historical roots of this inadequacy. One of the main reasons of the inadequacy of the received view of causation is the interpretation of events as changes in a substance. This view is inconsistent because it is based upon two incompatible categoreal
214
CHAPTERFOLR
frameworks: a substance ontology and a fact ontology. The tendency to implicitly conceive the causal relata as substance-like entities is rooted in Aristotle's substance ontology, and contradicts the standard view that causation is some sort of relation; the latter view goes back to the fact approach to reality, which became predominant in the seventeenth century. Thus, the history of the concept of cause was seen to be intimately related to the historical evolution from the substance ontology to the fact ontology. This evolution was marked by two decisive milestones: (I) the Aristotelianscholastic Conception, according to which causes are the active initiators of a change, and (II) the (seventeenth century) Scientific Conception, according to which causes are inactive nodes in a law-like implication chain. The inadequacy of the received view is (partly) due to the fact that it has failed to recognize this basic ambiguity, and that it has taken the Scientific Conception (II) as point of departure, which, from both an historical and an ontological point of view, is a derivative conception of cause. I made a distinction between causation, or the production of an effect by a cause, and causality, or the relation between cause and effect, and I concluded that the received view concerns only causality. I argued that causation is more fundamental than causality, because causes and effects are abstractly separated aspects from a continuous process of causation. Mistaking the nature of the relationship between these abstracted elements for the core of causation is due to a confusion between conceptual abstractions and the concrete reality from which they are derived (Whitehead's 'fallacy of misplaced concreteness'). The apparent impasse that characterizes the contemporary attempts at analyzing causation may partly be due to an insufficient appreciation of the lack of a consistent categoreal framework. It was suggested that Peirce's Categories, which provide such consistent categoreal framework, might offer better opportunities for an adequate analysis of the concept of causation. In the second part of this chapter, I listed a number of requirements for a theory of causation. Not only should it (I) be coherent, and (II) involve a coherent theory of events and processes, but it should also explain the relationship between (i) causation and continuity, (ii) causation and time, and (iii) causation and teleology. Moreover, (iv) it should be able to clarify the status of causes and effects as such, and by the same token, the relationship between causes and effects. The objective of the third part of this chapter was to discuss C.S. Peirce's critique of the principle of causality, and his constructive insights in respect of causation. Peirce's critique of the principle of causality involves the recognition that the concept of cause, as usually presented, is ambiguous. He discerned
A SEMEIOTIC ACCOUNT OF CAUSATION
215
three different and incompatible meanings: (i) the Aristotelian conception, (ii) 'the modern physicist conception,' and (iii) 'the currently accepted view,' which is a hybrid of the two former-mentioned conceptions. Peirce proposed to strictly distinguish the concept of cause from the concept of force. While he reserved the former term for the whole of human experience and of nature, he restricted the latter to the context of the formal laws of physics. Whereas 'cause' concerns irreversible processes, 'force' pertains to reversible processes. 'Cause' deals with concrete reality, 'force' only with abstractions. Peirce held the view that (i) cause-effect relationships are irreversible, (ii) causes only partially determine their effects, (iii) cause-effect relationships are mediated by 'laws' (final causes) and (iv) causes precede their effects in time. According to Peirce, each act of causation involves an efficient component, a final component, and a chance component. While each event is brought about by a previous event (the efficient cause), it is also part of a continuous sequence that is marked by a definite tendency (final cause), and it involves an aspect of irreducible novelty (objective chance). Though Peirce nowhere made an explicit distinction between causation and causality, he did so implicitly in his contention that causality involves a relationship between facts, but causation involves the production of events. Finally, it was shown that Peirce held the view that causation is a productive event as a part of a teleologically determined sequence (process). The main difference between processes and events is that processes share three characteristics which events lack: complexity, teleology, and coherence. The objective of the fourth part of this chapter was to suggest a new approach to the problem of causation, based upon Peirce's semeiotic. I discussed Debrock's hypothesis that, while semeiosis does not necessarily entail causation, causation involves semeiosis. Semeiosis does not involve the production of events; it only concerns the dynamic relation of certain aspects of events. I elaborated an anatomy of (Peircean) events, according to which it belongs to the nature of an event, not only that it creates a new event (Secondness), but also that it creates a new event of a certain type (Thirdness). More particularly, each event (1) involves a decision selecting one possibility among various possibilities, (2) is conditioned by a cluster of previous events, and by the relevant final cause, (3) conditions and limits the range of possibilities for subsequent events.
216
CHAPTERFOLR
It was suggested that (i) each new event assimilates the past event from a definite perspective, (ii) the past event is related to the present event as an efficient cause to its effect, and (iii) this relationship is mediated by the final cause, which is the perspective. It was suggested, furthermore, that each perspective reflects a conditionally necessary relationship, which has the structure of a material implication. Peirce made clear that causation must somehow involve a transmission of forms. However, unlike Aristotle's formal causes, Peircean forms are relational structures rather than something embodied in a substance. They are general rules which have the logical structure of a material implication: if p, then q. Thus, the fundamental difference between Peircean forms and Aristotelian formal causes is that, while the Aristotelian formal causes were thought to explain the stability of the world by explaining the structure of things, Peircean forms were meant to explain the stability of the world by explaining the dynamic relations between events. It was shown that the transmission of forms in causation involves a semeiotic process, which mediates between cause and effect. Effects may function as signs pointing to their causes, and certain events ('causes') may function as signs of future effects. In both cases the interprétant is the mediator in the transmission of forms. Thus, while the symptoms of a disease point to its cause, the relevant medical law is the interprétant by virtue of which the symptom (sign) is related to its cause (object). Similarly, in the example of clouds being a sign of rain to come, the interprétant expresses the law-like relationship between cloud-signs and possible rain (object). Finally, I argued that the theory presented meets the two primary requirements for a theory of causation: not only is it based on one, and only one, categoreal framework, but it also involves a coherent theory of the relationship between events and processes. This relationship was briefly expressed in my definition of a (Peircean) process as "a continuous sequence of events that derives its unity or internal order (distinguishing it from other processes) from a final cause, which directs the sequence to some end state which itself may evolve." The theory presented also meets the four subsidiary requirements for a theory of causation: first, the problem of the explanation of the continuity involved in causation is dealt with by describing causation as a productive constituent of a continuous sequence of events. Moreover, the transmission of forms from causes to effects is a semeiotic process, and it is because semeiosis involves real continuity, that there is a continuous transmission of forms from causes to effects. Next, it was argued that the relationship between causation and time is such that time is derivative in respect of causation. Peircean events do not
A SEMEIOTIC ACCOUNT OF CAUSATION
217
happen in time, but they are the condition of time (whatever time may be). Moreover, my analysis of causation conceived as a productive event implies that causes and effects are abstractly separated aspects of an ongoing process in which causes are said to precede their effects. Thirdly, in respect of the problem of teleology, it was argued that the explanation of natural processes requires reference to final causes, which are general principles. There can be causation only if there are boundaries set forth by general principles. At last, it was established that the status of causes and effects is that they are both objective and perspectival. Since they are aspects abstractly derived from the concrete, objective stream of events, they are not only determined by concrete reality, but also by the abstractive power of our thought. They are, facts rather than events.
NOTES PREFACE 1 This is in fact a description of John Mackie's analysis of causes in terms of INUS conditions. See chapter 2, 1.1.4 ('Causation and INUS Conditions'). CHAPTER 1 1 Aristotle's critique concerns Plato's Phaedo 95e - 107b. See Aristotle, Metaphysics, A 9, 991 b3-9, and De Generatione et Corruptione 335b7-16, 18-24. 2 "We think we know a particular (thing) unqualifiedly [...] whenever we think we know the cause (aitia) through which the thing (pragma) is, because that is its cause, and that it does not happen that it is otherwise." Posterior Analytics, I 2, 71b9-12; Loeb edition, H. Tredennich and E.S. Forster. 3 Translated by R.P. Hardie and R.K. Gaye. In the translated quote we have changed the term 'cause' into 'aitia'. 4 Aristotle gave quite different examples of material causes, such as: the letters of syllables, the premises of a conclusion, and the parts of a whole (Physics II.3, 195a 15-18). These make clear that Aristotle had an altogether different conception of matter than our modern notion of matter. 5 Aristotle's Categories treats the basic kinds of things and their interrelations. Every simple term signifies something in one of ten categories: substance, quantity, quality, relative, place, time, position, having, doing, or being affected. Whereas there are as many proper senses of 'being' as there are categories, there is a fundamental distinction between substance (ousia) and the other categories. There is an asymmetric relation of ontological dependence: the latter, accidental, categories inhere in a substance, and can therefore not exist without a substance. 6
7
See for example, Physics II.9; Analytica Posterior 11.11, 95a3-5. For a thorough discussion, see Sorabji, 1980, 51-56. Cicero, for example, reproduced Carneades argument as follows: "If everything takes place with antecedent causes, all events take place in a closely knit web of natural interconnection; if this is so, all things are caused by necessity; if this is true, nothing is in our power. But something is in our power. Yet if all events take place by fate, there are antecedent causes of all events. Therefore it is not the case that whatever events take place take place by fate" (Cicero, De Fato, 31 ).
8
De Fato 22, 192 22. See Dunphy, 1966, and Lauer, 1974. 10 In fact, there are three levels of causation: (i) the First cause or God, (ii) a cosmic universal causality assigned to the sun or to the first heaven (the sphere of fixed stars), and (iii) the activities of particular causes. In order to explain the order of the world, that is to say, the harmony of particular causes, Aquinas referred to a general causal influence on the cosmic process. This universal causality (which produces the specific essences of things) in direct subordination to God, directs the activities of the particular causes (see: Elders 1974, 1012).
9
219
220
NOTES
11
Other important characteristics of the efficient cause are: (a) secondary efficient causes either precede their effects or are simultaneous with them (SCG II 38.9); (b) the secondary causes are modeled after the primary cause inasmuch as "the agent is distinct from the patient and superior to it" (SCG II 45.4); (c) there is a proportional correspondence of effects to their causes: "we attribute actual effects to actual causes, potential effects to potential causes, and, similarly, particular effects to particular causes and universal effects to universal causes, as Aristotle teaches in Physics II" (SCG II 21.4). 12 It must be noted that, according to Descartes, human beings are in some sense the efficient causes of their actions. Descartes tried to reconcile his idea that "it is certain that all things are pre-ordained by God" (Princ. I: 40) with the "self-evident" idea of freedom of the will (Princ. I: 39). Descartes' solution was that the mind could not change the quantity of motion but that it could change the direction of motion. 13 'Endeavor' is the key concept in Hobbes's attempt to show the compatibility of his philosophy of motion with the explanation of voluntary behavior. 14 This entails that there cannot be an unmoved mover. However, Hobbes does not discuss whether God himself was caused. 15 This should not be confused with the term 'necessary cause,' which Hobbes sometimes used in place of 'entire cause' (DCo 9.5). 16 Also IV, iii, 12. 14, 16. 17 See next chapter, 2.5. 18 Here, the impressed force equals mass times the rate of change of velocity, i.e. acceleration. Hence the formula, F = ma. 19 In his Enquiry Hume gives parallel definitions. Here, however, he held that causation only involved priority and necessary connection: there is no reference to contiguity, which according to the Treatise was the third constituent. According to his parallel regularity definition, a cause is "an object, followed by another, and where all the objects similar to the first are followed by objects similar to the second." On the view of necessity as a connection in the mind, a cause is "an object followed by another, and whose appearance always conveys the thought to that other" (Hume [1748] 1975, 76-77). 20
Eventually, despite his alleged empiricism, Mill appears to be some kind of a Laplacean determinist, according to whom the whole future course of nature is completely determined by antecedent causes: "The state of the whole universe at any instant we believe to be the consequence of its state at the previous moment; insomuch that one who knew all the agents which exist at the present moment, their collocation in space, and all their properties, in other words, the laws of their agency, could predict the whole subsequent history of the universe ..." (Mill 1874, 250). It is obvious that such a conclusion about the future course of the universe cannot be based on empirical data alone.
21
See note 5.
CHAPTER 2 1 There is an enormous amount of recent literature on the problem of causation. I have found Sosa and Tooley (1993), Taylor (1967), and Kim (1995) especially helpful in coming to an overview of the contemporary discussions. 2 Borrowed from Taylor 1966. 3 This is a slightly changed version of the definition given by Sosa en Tooley 1993, 6. 4 This so-called instrumental or agency approach will be discussed in section 1.3 of this chapter.
NOTES
221
It has been questioned whether there are any actual instances of over-determination. For it could be argued that the man's death (effect) would have been different if he had been shot by less bullets. Sosa and Tooley (1993, 6-7) would not accept our interpretation; they defend the idea that causal sufficiency entails logical necessity. Taylor (1966, 35-7) gives several interesting examples of 'undeterminative sufficiency." The counterfactuals in turn are explained on the nomic-inferential model. It is at this point that laws and regularities enter into singular causal judgments. According to Mackie, his analysis "has been given entirely within the limits of what can still be called a regularity theory of causation" (Mackie [1965] 1993, 52). A problem, put forward by Kim (1971; also 1973), that is more specific to the INUS condition approach is that its implicit ontology of events must be carefully worked out. According to Kim, "any discussion of causation must presuppose an ontological framework of entities among which causal relations are to hold, and also an accompanying logical and semantical framework in which these entities can be talked about" (Kim [1971] 1993, 60). Apparently, Mackie often confuses events with their descriptions (Kim [1971] 1993,68). 1
Lewis mentions some problems that were not listed here. According to him, the main problems for the regularity analysis are (a) distinguishing genuine causes from effects, (b) epiphenomenona, and (c) pre-empted potential causes: "regularity analyses tend to confuse causation itself with various other causal relations. If c belongs to a minimal set of conditions sufficient for e, given the laws, then c may well be a genuine cause of e. But c might rather be an effect of e: one which could not, given the laws and some of the actual circumstances, have occurred otherwise than by being caused by e. Or c might be an epiphenomenon of the causal history of E: a more or less inefficacious effect of some genuine cause of e. Or c might be a pre-empted potential cause of e: something that did not cause e, but that would have done so in the absence of whatever really did cause e" (Lewis [1973] 1993, 193-94).
Some important improvements have been made upon this definition of Hume: (a) contemporary analyses distinguish so-called causal regularities from mere accidental regularities, (b) the idea of overall similarity has been replaced by similarity in specific respects, and (c) causes are nowadays mostly considered to be just one part of a situation (not the whole situation) that is followed by the effect in accordance with a regularity. They are often symbolized as 'p —> q', which is true if, and only if, either p is false or q is true (or both). According to Lewis, causal dependence among events implies causation, but the converse does not hold: "If c and e are two actual events such that e would not have occurred without c, then c is the cause of e. But I reject the converse. Causation must always be transitive; causal dependence may not be; so there can be causation without causal dependence. Let c, d and e be three actual events such that d would not have occurred without c and e would not have occurred without d. Then c is a cause of e even if e would still have occurred (otherwise caused) without c" (Lewis [1973] 1993, 200). In Lewis's own words: "one event is a cause of another iff there exists a causal chain leading from the first to the second" (Lewis [1973] 1993, 200). Horwich [1987] 1993,208-11. There has been a historical development from I to II to III. Strictly speaking, sense I is the only proper sense. Whereas I and II originated in ancient Greece, III was a Renaissance
NOTES
invention. According to I, causation entails that a free and responsible agent is influenced by another agent who provided him with a motive for doing something. If we read in the newspaper "Mr. president's speech causes havoc in parliament," this does not mean that the president compelled the members of the parliament to quarrel; it rather means that his speech afforded them a motive for doing so. A cause in sense II is something that can be used to manipulate things in nature. Causes I an II are contingent in a double sense: they are contingent in their existence because their existence depends on human volition. They are contingent in their operation because in their operation they depend for the production of their effects on further conditions. Cause III is a condition or set of conditions that are invariably accompanied by some change. Causes III are necessary both in their existence and in their production. Their existence does not depend on human beings, and the production of their effects does not depend on any further conditions. There can be no relativity of causes, and no diversity of effects depending on other conditions. There is a one-one relation between cause and effect (Collingwood [1938] 1991, 145-54). His approach to the problem of directionality (or causal priority) was: "I now propose the following way of distinguishing between cause and effect by means of the notion of action: p i s a cause relative to q, and q an effect relative to p, if and only if by doing p we could bring about q or by suppressing p we could remove q or prevent it from happening" (Von Wright 1971,70). For a much more detailed criticism of Gaskin's and Von Wright's accounts, see Tooley 1987, 239-242. D.H. Mellor (1986, 1995) takes the relata of the causal relation to be facts rather than events. There is some disagreement about the relationship between causation and probabilistic dependence: whereas, for example, Lewis (1986) takes probabilistic dependence to be sufficient, but not necessary, for causation, Mellor (1986) takes probabilistic dependence to be both necessary and sufficient for causation. Mackie 1974, 134-142 contains an extensive critique of Ducasse's view. More precisely, see Sosa and Tooley 1993, 1-5, 11-14, and Tooley 1987, 173-8, 183-190, 202-204, and 244-50. Until recently, it was generally thought that Hume held the analytical reductionist position. A number of recent philosophers, however, have argued that Hume was a realist rather than a reductionist. For explanation and further references, see Galen Strawson 1989. A.N. Whitehead (1920, 1929) (and the philosophers of process in general) seems to be a remarkable exception to the rule mentioned by Sosa and Tooley; though he insists that we do have an immediate experience of causation, he also insists that it is one of the main tasks of any metaphysics to give an adequate analysis of this experience. Whereas Sosa's and Tooley's interest is primarily epistemological - an analysis of the concept of causation, Whiteheads interest is primarily metaphysical - an analysis of our experience of causation. Though Tooley does not explain (explicitly) what he means by a theoretical relation, he tells us that "to be a realist with regard to theoretical entities is to hold that facts concerning the relevant unobservable entities (or events, etc.) are not logically supervenient upon observable facts" (Tooley 1987, 177). Whereas Tooley holds that causal relations are not logically supervenient upon non-causal facts (1987, 177), he insists that "any knowledge that we have of causal connections must ultimately rest upon knowledge of non-causal states of affairs" (Tooley 1987, 247). Causal terms are analyzable "by means of any method that is applicable to theoretical terms in general." [...] "Moreover, given that causal
NOTES
26
27 28
223
terms are theoretical, there is no need to postulate any special faculty that provides one with direct awareness of causal relations. Causal claims will be epistemologically justified in the same way that theoretical claims in general are justified" (Tooley 1987, 249-50). The term 'event' is ambiguous: it may refer to a particular occurrence, happening only once, with a particular duration and location. These are called 'token events.' There are, however, also 'type events,' such as, for example, the Olympic Games, which happen every four years. Most contemporary philosophers are predominantly concerned with events as particulars. Chisholm (1970) is an exception; he takes events to be universals, capable of recurrence. A notable exception is Emmet 1984, 1992. Davidson's idea that causal explanations need not refer to laws in the strong sense of the word, is clearly expressed by Dorothy Emmet: "When we speak of a singular event as caused or as having effects, this is said to be explanatory when it can be connected with some generalisation - this need not be as elevated as a Law; it can be a commonsensical generalisation as that ships which get great gashes in their holds so that water pours in are likely to sink (the law would be that of the specific gravity of water). So the singular event has a property or properties by which it can be assigned to a class of events, and it is followed by another event whose properties also assign it to a class, and a causal generalisation can be made by stating a relation between properties of events of these two classes. The causal statement about a particular event is being made about it as an instance of a kind, and under the description which assigns it to the kind named by these properties" (Emmet 1984, 20-21).
CHAPTER 3 1 Examples are: Braithwaite 1953; Hempel 1965; Nagel 1961. 2 Important examples: Wiener 1950; Von Bertalanfy 1968. 3 For an overview and critical discussion (in Dutch) of these reductionist strategies, see Soontiëns 1988. 4 See for example the current discussion about "The Anthropic Cosmological Principle" in Barrow and Tippler 1986. 5 In Short 1981, 1983 some of Mayr's ideas are discussed from a Peircean point of view. Though I agree with most of Short's conclusions, I will try to go a little beyond these in my criticism of Mayr. 6 His 1988 book contains the same paper with view minor changes. 7 Probably, the two major differences are that (a) Peirce's theory does not, as Aristotle's does, presuppose a substance ontology, and (b) contrary to Aristotle's conception, Peirce's conception presupposes chance. According to Aristotle, chance entails the absence of a final cause. 8 Considered from the perspective of Peirce's categoreal scheme, efficient causation belongs to the category of Secondness, and final causation to Thirdness. Thus, "it is nonsense and utter confusion to treat [final causes] as forces in the material sense" (CP 1.265, 1902). 9 For Peirce's "ethics of terminology," see Ketner 1981. 10 There are some important differences between Aristotle's notion of efficient cause and the modern notion. Perhaps the most important one is that according to Aristotle, efficient causes are related to things, and according to the modern theory, they are either events or facts.
224
NOTES
11
According to Peirce, the idea of final causation is even more original than the idea of efficient causation. Consider for example the two following quotations: "For the very type, and prototype of what the word cause means is the sense in which, for example, my desire for fresh air may cause me to rise from my chair, cross the room, and open the window" (MS 1343, 1902). There is no doubt that this is an instance of final causation, and not of efficient causation: "... who is acquainted with the fundamentals of dynamics that if physical forces obey the law of the conservation of energy, then a volition cannot be a force. For a volition tends to bring about a result and if circumstances are varied the action will be varied, so far as may seem necessary to bring about that result; while a force acting according to the law of energy [...] does not act in this way" (MS 1343). A last quotation to support our point: "The very conception of causality has its origin in our tendency to seek relations in nature analogous to intellectual relations" (MS 963).
12
According to Short (1981, 372, 375), these statistical laws, and final causes in general, are nothing but tautologies. We think that Short makes a mistake here. Final causes are characterized by an inherent drive toward realization, which can never be said of tautologies. Short seems to consider the statistical laws as "purely mathematical." In a sense he is right: they are mathematical in form. But they always refer to distributions of physical phenomena. Mathematical laws are not in themselves final causes; they do not determine anything but mathematical entities. Final causes are physical possibilities, not just logical or mathematical possibilities. That Short does not refer to the tautological nature of statistical laws in his 1983 article may indicate that he abandoned this view.
13
Characteristic of a triadic relationship is that it cannot be resolved into (two) dyadic relationships. 14 In the next section we will explain that the final causation of mechanistic laws is a degenerate kind of final causation. 15 Short 1981, 374; also 1983, 317. 16 "The distinction between psychical and physical phenomena is the distinction between final and efficient causation" (CP 7.366; 1902). 17 See for example: CP 6.173, 1902; 1.487, 1896; 1.624-25, 1898. 18 See RLT, 218-20, 1898; also W l : 422, 1866. 19 See MS 1343, 26-7, 1902; CP 6.322, c.1909. 20 RLT, 87, 1992; see also note 52, p. 278. 21 The idea that 'objective chance' refers to 'absolutely uncaused events' is held by Peirce from 1880 till at least 1884 (see MS 674 and "Design and Chance," EP I: 217; 1884). During this period, the discussion is not yet placed within the context of his categoreal scheme. It would appear that - according to Peirce's later works - every event involves an aspect of Firstness, Secondness, and Thirdness. This entails that every event involves an element of objective chance, efficient causation, and final causation. 22 23
24
See note 21. A good example that shows that in the choice of routes there is creativity involved is the idea to write a book on a certain theme. It is clear that the way the subject is handled forms a crucial part of the creative process. Different routes may lead to different books on the same subject. During the realization of a final cause, there is always the confrontation with the material world. This is the ground for the evolution of new final causes. New ideas (purposes) do not happen just by chance, but only as a response to an actual problematic situation. We think it is precisely this that Peirce was referring to when he wrote: "The way in which
NOTES
25
26
27 28
29
30
31
225
mind [read: final causation] acts upon matter [read: chains of efficient causation] is by imposing upon it conformity to certain peculiar laws, called purposes; and the manner of the reaction is that the purposes themselves become modified and developed in being thus carried out (MS 478, 18, 1903)." Consequently, developmental teleology requires more than just chance; it presupposes an interrelated activity of chance, efficient causation, and final causation. This agrees with Rosenthal's conclusion that developmental teleology can only be understood in terms of all three of the Peircean categories (Rosenthal 1994, 125). Similar points of critique of Mayr's theory of teleology were made by Soontiëns 1988, which was very helpful in formulating my 'Peircean critique of Ernst Mayr's theory of teleology.' The problem whether natural selection takes place on the level of species or of individuals, will not be discussed here. It is still a hot topic in evolutionary biology. In Peirce's view natural selection concerns types, not individuals (EP I: 272; c.1887). See W 4:46, 1880; CP 6.296,1893; MS 1343, 1902; CP 1.204, 1.269, 2.86: all 1902. We do not claim to be original here: the same point has been made before by Short (1983, esp. 314-5). Short correctly warns against misinterpreting the idea of a teleological interpretation of biological evolution: "a teleological interpretation of biological evolution does not entail that the aim of evolution is a single, "highest" species. On the contrary, if a final cause is a general type, then it might be actualized in any number of different ways. As Darwin emphasized, the principle of the survival of the fittest entails a divergence of species to fill all of the available ecological niches" (Short 1981, 372). Conversely, orthogenesis - if there were such a thing - does not imply teleology either, for it is a completely mechanical process. The word 'program' is taken from the language of information theory. Since the paradigmatic example of a computer program is completely mechanistic (deterministic), the analogous use of the word program to situations that might not be deterministic at all is misleading. Though there may be computer programs that agree perfectly well with Peirce's definitions of final causation, there are serious reasons to believe that Peirce would consider most computer programs as at most quasi-teleological. Usually, the program completely determines not only the results, but also (if there are different ways) which ways will be taken toward the results. Hence, this is a perfect example of a mechanistic process (see MS 1343 and CP 6.322). But there might also be programs, and we bet there are, where there is a randomizer built in, and a principle that selects certain kinds of behavior. In that case there might be real teleology involved. The problem, though very important for our discussion, is too big to handle here. The word habit is used here in a broad sense: Mayr's "completely genetically fixed programs" (Mayr 1974, 102) may also be regarded as genetically fixed habits. Peirce: "habit plays somewhat the same part in the history of the individual that natural selection does in that of the species; namely, it causes actions to be directed towards ends" (W 4:46; 1880).
CHAPTER 4 1 Hacking 1991. 2 Only this can explain that Haack, in an otherwise brilliant paper on Peirce's "Scholastic Realism," could write about the relationship between natural classes and final causation: "I am unable to judge whether Peirce's suggested characterization could be made acceptable" (Haack 1992, 50, note 45). Similarly, Beverly Kent, in her extensive book on Peirce's
226
NOTES
"Logic and the Classification of the Sciences," ignores the importance of final causation for the item of natural classifications because "Peirce was often obscure, if not actually mystical, in some of his writings on final causation within the context of natural classification" (Kent 1987, 229, note 11). A notable exception of someone who does recognize the importance of final causation for natural classifications is Helmut Pape (1989, 1993). Pape's discussion, however, concerns primarily Peirce's idea of a natural classification of the sciences, rather than his idea of natural classes. 3 In Kripke's own words: "Given that gold does have the atomic number 79, could something be gold without having the atomic number 79? ... consider a possible world ... in which, let us say, fool's gold or iron pyrites was actually found in the areas which actually contain gold now ... Would we say ... that in that situation gold would not even have been an element (because pyrites is not an element)? ... One should not say that [this substance] would still be gold in this possible world though gold would then lack the atomic number 79. It would be some other stuff. ... It [is] necessary and not contingent that gold be an element with atomic number 79" (Kripke 1980, 123). 4 Van Brakel 1992,243-4. 5 Successively: Haack 1993a, 134; 1992, 29 and 1992, 25. 6 I don't think that stones are good examples of Peircean natural kinds, for the very reason that there are no laws of nature (final causes) that are specific to stones. See also note 25. 7 Note that, for the young Peirce, the rejection of strict uniformities does not entail the rejection of determinism. It was only from 1880 on that Peirce was going to challenge determinism. 8 For example in Peirce's earliest text about natural kinds, which we have discussed in section 3.2 (W 1:416; 1866); also in his MS 421; c. 1893-5. 9 Peirce would not make the distinction between kind and class until 1908. 10 "An important character is obviously one upon which others depend, that is, one the inclusion of which in a definition renders true general propositions concerning the object defined possible; and the more such propositions a character renders possible, the more important it is" (W2: 443; 1867). 11
The insight that Peirce borrowed his idea that generals are possibles from Duns Scotus, I owe to Mauer 1983, 8. 12 I avoid the expression 'PRE-character' simply because Peirce no longer refers to them beyond the year 1902. This will be further explained in the next section (5). 13 For Peirce's Century Dictionary Definition of'essence,' see W5: 417; 1886. 14 In 1902 Peirce applied for a grant from the Carnegie Application to complete and publish his studies in philosophy. His request was not honored. 15 The most important part of this text was also published in EP II, section 9 (pp. 115-132), entitled "On Science and Natural Classes." 16 Hilary Putnam has made the same observation: "it sounds strange to be told that a human being is not identical with the aggregation of the molecules in his body. Yet on a moment's reflection each of us is aware that he was not that aggregate of molecules a day ago. Seven years ago, precious few of those molecules were in my body. If after my death that exact set of molecules is assembled and placed in a chemical flask, it will be the same aggregation of molecules, but it won't be me." (Putnam 1975, 235) 17 For this evolution from Peirce's early nominalist sympathies toward his mature commitment to scholastic realism, see Fisch 1986.
NOTES
18
19
20
21
22 23
227
For a brief explanation of Peirce's categoreal system, see chapter 6, section 4.1 of this book. Though Peirce does not speak of essential qualities in his 1866 paper, he does speak about 'properties which are implied' in the definition of the natural class (W 1:418). For explanations of Peirce's Natural Classifications of the Sciences, see Kent (1987) and Pape (1989, 1993). Pape explicitly deals with the relationship between final causation and natural classifications. Peirce has a second argument for his idea that chemical elements were not subject to evolution: "The irregularities of biological classification are the traces of the geological vicissitudes through which the earth's surface has passed. There is no trace of anything analogous to this in the chemical classification." (MS 421; 1893-5) Manganates are in fact K 2 Mn0 4 , and rutheniates K 2 Ru0 4 . The question regarding the criteria of 'sufficient similarity' is empirical in nature; it has to be answered by chemistry. In contemporary chemistry, analogous behavior points to analogous molecular structure even when there is no similarity in molecular formulas. In those cases, the chemist will try to rewrite the formulas in an analogous form. For example, the molecular formulas C H 4 0 (methanol) and C 2 H 6 0 (ethanol) are rewritten as the structure formulas CH 3 OH and C 2 H 5 OH.
24
This also agrees with Hacking's interpretation. According to Hacking, that which makes something a Peirce-kind is "its role in a systematic interconnected web of laws of nature" (Hacking, 1990, 120-21). Hacking's formulation contains both a mistake and an important insight. His description is misleading inasmuch as his expression "laws of nature" suggests that Peirce's notion of law is restricted to what we nowadays consider to be the (fundamental) laws of nature. Haack's reference to habit-like behavior is much more appropriate. However, Hacking's description contains an important insight that is lacking in Haack's description. A Peirce-kind is not only characterized by the sum of its habits, but also, and even more so, by a systematic connection of habits. This is provided by the final cause to which the members of the class owe their existence.
25
I think, however, that Haack makes a mistake by mentioning stones as examples of Peircean natural kinds. For, there are no final causes or laws that are specific to stones. The law of gravity is a final cause that makes all things of a certain density approach the center of the earth, not just stones. Peirce's Harvard experiment was not so much about what natural kinds there are; its purpose was only to prove that there are "general principles operating in nature."
CHAPTER 5 1 I owe this reference to The Max Fish Library, which is situated in the Peirce Edition Project (Indiana University). 2 See Colapietro (1989); Hausman (1993); Hookway (1985); Kruse (1986; 1990); Liszka (1996); Pape (1993); Ransdell (1977; 1981; 1986); Rosenthal (1994); Savan (1988); Seager (1988); Short (1981b). 3 See chapter 4. 4 The term 'semeiotic causation' was introduced by Ransdell (1981). I think that, though he does not formulate the problem in terms of the respective roles in semeiosis of final causation, efficient causation, and chance, my formulation of the problem agrees with his view. 5 For a detailed description, see CP 2.227 (c.1897).
228
NOTES
6
There are some earlier thinkers, such as John Poinsot (1589-1644), who conceived the sign triadically. See Deeley 1994. 7 "... the causation may be either from the object to the sign, or from the sign to the object, or from some third thing to both; but some causation there must be" (W3: 82; 1873). 8 See, for example, CP 2.274-3.308 (1895 and later). 9 1 owe this insight to Short (1986, 491), who also gives the following references: CP 2.242, 274, 275, LW: 111. 10 Undated, but probably after 1903. The dynamical object cannot be directly experienced; it cannot be expressed by the sign, but it can only be indicated by the sign (8.314). Moreover, "The sign can only represent the Object and tell about it. It cannot furnish acquaintance with or recognition of that Object; for that is what is meant (...) by the Object of a Sign; namely that which presupposes an acquaintance in order to convey some further information concerning it" (CP 2.231). Thus, for example, we can only discuss 'semeiotic causation' if we have some image or notion of what semeiotic causation is about. This image, which we have gained by "collateral experience," is the immediate object. The sole function of the immediate object is "the identification of the actual or supposed previous experience with which the new meaning, conveyed in the sign, is to be attached" (MS 318:00342-43). The real object is that conception of semeiotic causation "which we should have ultimately in our minds as the result of sufficient information and reflexion" (MS 318:00342). 11
For example, in 1906 Peirce writes about the final interprétant: "I confess that my own conception of this third interprétant is not yet quite free from mist" (CP 4.536). 12 Undated fragment, probably from after 1903. 13 T.L. Short has strong arguments to consider both distinctions as quite different (see Short 1996, esp. 494-6). Short refers to CP 5.475 in which Peirce wrote that in some cases the emotional interprétant is "the only proper significate effect that the sign produces," and where he gives the appreciation of a musical performance as an example. Because this effect is actually produced, Short concludes, "there is no indication (...) that an emotional interprétant is merely potential" (p. 495). The emotional interprétant cannot therefore be identical with the immediate interprétant, which is merely potential. However, in MS 318 Peirce seems to emphasize that emotional meanings (he used the terms 'meaning' and 'interprétant' as synonymous) are yet mere possibilities, which is an argument to treat both trichotomies as identical. According to Peirce: The emotional meaning corresponds to the immediate object, inasmuch as it is involved in the mere presentation of the sign. Only, it is what that presentation brings and not what it finds. It is what is conveyed strictly in the presentation itself without any reflexion, or abstraction, or analysis, or other efficient element. It is not, (to make a very fine point,) even the feeling the sign brings, since that is an actual fact, and so belongs to the existential meaning. This is only the quality of feeling. Because no analysis is involved, it is the total consciousness at the time; and for the same reason there can be no similarity between two emotional meanings. Each is sui generis; and the immediate meaning is, in all ways, strikingly analogous, if it be not identical with, the living personal consciousness, which likewise, be it noted, resembles nothing else. Practically, however, these extreme characteristics are enormously softened by the circumstance that we lack the mental energy which would be
NOTES
229
required to inhibit analysis, etc. So that the purest emotional meaning that we can recognize in the case of an intellectual concept is the plausible feeling of perfectly comprehending the purpose and purport of a sign, in a remarkably clear but as remarkably indistinct an idea of its implications. (MS 318:0034445) 14
For a detailed explanation, see section 6. LW: 111, ftn. 6. Quoted in Short 1981b, 213. 16 LW: 111, ftn. 6. Quoted in Short 1981b, 213. 17 Short apparently does not consider the dynamic object as the final cause, for he gives two examples in which the dynamic object is different from the final cause of semeiosis: (1) a Paramecium following a chemical trace which leads toward food; the trace is a sign of the food, which is the object of the sign. The final cause or end is not the food itself, but nourishment or survival. (2) The flight of a deer elicited by a noise. The deer interprets the noise as being a sign of a predator, which implies danger. The goal of the deer, which is safety, does not coincide with the object of the sign, which is danger (Short 1981b, 20708). 15
18 19
20 21
22
23
Short 1981b, 207. Correspondingly, Short describes semeiotic as "a science both of human thinking and of animal behavior" (Short 1981b, 199). Short 1981a, 371. From: Joseph Ransdell. The Meanings in Things. The Basic Ideas of Charles Peirce's Semiotic; forthcoming. Though Ransdell does not state explicitly in his 1986 paper that the dynamic object is the final cause of the semeiosis process, he did so in two earlier papers. Thus, in his 1977 paper he writes for example that "... what is meant in saying that every sign has an object is that every sign-interpretational process tends toward an end state, that is, has a final causational form. That end state is the object of the process" (Ransdell 1977, 168). In his 1981 paper he writes that the term 'object' should be understood primarily in the sense of "the generic aim of the semiosis process" (Ransdell 1981, 203). There are also some other Peirce scholars who mention the dynamic object as the final cause of semeiosis: Kruse 1990, 214; Pape 1993; Rosenthal 1994, 49.
Inasmuch as the chain of efficient causation is not determined by the general purpose that we had in view, it is still determined by some physical laws, which are final causes too. Even after the bullet has left the rifle, it conforms to a general law, the causality of which is of the order of final causation (CP 1.212). 24 In Ransdell's own words: the immediate interprétant is the "range of possible interprétants of a given sign at a given time;" the dynamic interprétant is "simply the actually occurring interprétant;" the final interprétant is "the range of possible interprétants that would be definitely established with the ultimate cessation of all growth in the powers of the sign as such." The distinction between immediate and final interprétant is purely conceptual; qua content they are indistinguishable (Ransdell 1986, 682). 25 1 suppose that Ransdell would subscribe to my interpretation of Peirce's conception of final causation. 26 About the indeterminacy of signs Peirce writes: "... it is impossible that any sign [...] should be perfectly determinate. If it were possible such sign must remain absolutely unconnected with any other" (CP 4.583; 1906).
230
27 28
29
30 31 32
33 34 35
36
NOTES
The same holds for the following parallel passages: W2: 467-70 (1871); W3: 77-81 (1873). However, there is a complication here: we think it is not the real object that is the final cause, but the immediate object. For the real performance, in its concrete reality, did not exist yet. It is only the idea of the performance as conceived by some advertiser that triadically determines the activity of the visitors. Thus, it was the immediate object that was the final cause of the semeiosis. We owe this information about the year of origin to 'The Peirce Edition Project' (Indianapolis University). MS 318:00037; also CP 5.473; both 1907. "A symbol [...] cannot indicate any particular thing; it denotes a kind of thing" (CP 2.301). Consider also the following passage: "when people say that so and so is a 'mere' word, that adjective 'mere' betrays a grave misconception of the nature of a symbol. A word may be likened to a decree of a court. I has not itself the sheriff s right arm; but it is able to create a sheriff, and to impart to his arm the courage and energy that makes it effective" (MS 478:48; 1903). See note 1. See also CP 2.228. It also agrees with Short's interpretation, which entails that what is essential to all semeiosis, is not the triadic production of signs, but the triadic production of interprétants (see section 2). See note 21.
CHAPTER 6 1 This does not hold for the singularist Anscombe (1971). She holds that causation means the production of an effect by a cause, rather than some sort of law-like relation between a cause and its effect. 2 Aristotle had an altogether different view on the subject. According to him, the complete explanation of the coming to be of a substance (say, a statue) must take into account four explanatory factors: the material, efficient, formal, and final aitiai. These aitiai or 'prerequisite conditions' are primarily internal to the substance. Of these four explanatory factors, only the efficient 'cause' is, in a certain sense, external to the substance. However, the efficient cause is also internal to its effect (substance) inasmuch as efficient causation involves a form being transmitted from the efficient cause to the effect. 3 See also Whitehead [1929] 1978, 237. 4 The examples are borrowed from Emmet 1992, 25. 5 Peirce thinks here of the interpretation of 'the law of the conservation of matter and energy,' according to which causation is transformation of energy and mass. This interpretation, which was rejected by Peirce, was formulated by Paul Carus (a critic of Peirce) as follows: "The law of the conservation of matter and energy rests upon the experience (corroborated by experiments) that causation is transformation. It states that the total amount of matter and the total amount of energy remain constant. There is no creation out of nothing and no conversion of something into nothing" (quoted in Peirce, CP 6.601, 1893). 6 "According to the dominant mechanical philosophy, nothing is real in the physical universe except particles of matter with their masses, their relative positions in space at different instants of time, and the immutable laws of the relations of those three elements of space, time, and matter. Accordingly, at any one instant all that is real is the masses and their positions, together with the laws of their motion. But according to Newton's second law of
NOTES
231
motion the positions of the masses at any one instant is not determined by their positions at any other single instant even with the aid of the laws. On the contrary, that which is determined is an acceleration [the effect is the acceleration]. Now an acceleration is the relation of the position at one instant not to the position at another instant, but to the positions at a second and a third instant" (RLT 199, 1898). For further explanation, see RLT 199-201. 7 See chapter 4, section 2.4. 8 See Peirce's "The Ethics of Terminology" (EP II, item 19). 9 According to Peirce, association is "a habit or disposition of mind in consequence of which an idea of on description is likely to bring into comparative vividness of consciousness an idea of another description" (RLT 232, 1898). 10 For references, see chapters 3 and 4. " Peircean processes are creative in a triple sense: (1) each event involved in the process contains an element of irreducible novelty; (2) the end state of a process can be reached in different ways; whenever one way or line of causation be blocked, it may originate new lines; and (3) the end state toward which a process tends, may evolve spontaneously (see chapter 3, section 2.5). 12 In the explanation of his final scheme of classification (1904), Peirce had changed the word 'largely' by 'exclusively.' "This classification [...] is to be regarded as simply Comte's classification, corrected. That is to say, the endeavor has been so to arrange the scheme that each science ought to make appeal, for its general principles, exclusively to the sciences placed above it, while for instances and special facts, it will find the sciences below it more serviceable." (MS L 107, 1904). 13 Peirce's Classification of the Sciences according to his 1898 Cambridge Lectures is as follows: A. THEORETICAL SCIENCE 1. Mathematics 2. Philosophy a. Logic b. Metaphysics 3. The Special Sciences a. Psychical b. Physical B. PRACTICAL SCIENCE For Peirce's final scheme, see his "A Brief Intellectual Autobiography." MS L 107, 1904. 14 Peirce borrowed the terms 'genuine' and 'degenerate' from the geometry of plane curves, where they refer to the irreducibility or reducibility of a figure to simpler figures (MS 304: 00035, 1903). For a thorough analysis of "Genuineness and Degeneracy in Peirce's Categories," see Kruse 1991. 15 Consider, for example, the following passage: "If a sign has no interpreter, its interprétant is a "would be," i.e., is what it would determine in the interpreter if there were one" (EP II: 409, 1907). 16 This is my interpretation of the meaning of Peirce's "broader conception" of sign (LW 81, 1908).
BIBLIOGRAPHY
Annas, Julia. 1982. "Aristotle on Inefficient Causes." Philosophical Quarterly 32: 311-26. Anscombe, Elizabeth. 1969. "Causality and Extensionality." Journal of Philosophy 66: 152-9. —. 1971/1993. "Causality and Determination." (Cambridge 1971). Reprint in: Sosa and Tooley (eds.). Causation. 88-104. Oxford: Oxford University Press. Aquinas, Thomas. (1973). "The Five ways." Summa Theologica, Part 1. Trans, by the English Dominican Fathers. Reprint in: P. Edwards and A. Pap (eds.). A Modern Introduction to Philosophy. Third edition. 408-10. London/New York: The Free Press. —. (1975). Summa Contra Gentiles. Book Two: Creation. Trans. J.F. Anderson. Notre Dame, London: Notre Dame Press. Aristotle. (1984). The Complete Works of Aristotle, i and ii. Ed. J. Barnes. Princeton, NJ. Armstrong, David M. 1983. What is a Law of Naturel Cambridge: Cambridge University Press. —. and Heathcote, Adrian. 1991. "Causes and Laws." Nous 25: 63-73. Ayala, F.J. 1970. "Teleological Explanations in Evolutionary Biology." Philosophy of Science 37/1. Barrow, John D. and Tipler, Frank J. 1986. The Anthropic Cosmological Principle. Oxford: Clarendon Press. Beauchamp, Tom L. and Rosenberg, Alexander. 1981. Hume and the Problem of Causation. New York and Oxford: Oxford University Press. Bennet, James 0. 1974. "Process and Causality." Tulane Studies in Philosophy. Volume 23. Studies in Process Philosophy I. Den Haag: Martinus Nijhoff. 1-12. Bennett, Jonathan. 1984. "Counterfactuals and Temporal Direction." Philosophical Review 93:57-91. —. 1987/1993. "Event Causation: the Counterfactual Analysis." Philosophical Perspectives 1. Metaphysics. Ed. James E. Tomberlin (1987). Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 217-33. Bigelow, J., B. Ellis and C. Lierse. 1992. "The World as One of a Kind: Natural Necessity and Laws of Nature." The British Journal for the Philosophy of Science 43: 371-388. Blackburn, Simon. 1990. "Hume and Thick Connexions." Philosophy and Phenomenological Research. Suppl. 50: 237-50. Bogdan, R.J. (ed.). 1991. Mind and Common Sense. Cambridge: Cambridge University Press. Boyd, R., Ph. Gasper, and J.D. Trout (eds.). 1991. The Philosophy of Science. Cambridge, Mass., and London: The MIT Press. Braithwaite, Richard B. 1953. Scientific Explanation. Cambridge: Cambridge University Press. Brand, Myles. 1980. "Simultaneous Causation." In: Peter van Inwagen (ed.). Time and Cause. Dordrecht: Reidel. 137-53. Broad, Charlie D. 1925/1951. The Mind and its Place in Nature. London: Routledge and Kegan Paul. Bromberger, Sylvain. 1962. "What Are Effects?" Analytical Philosophy. 1 st. ser. Ed. Ronald J. Butler. Oxford. 15-20. Bunge, Mario. 1959. Causality. Cambridge: Harvard University Press.
233
234
BIBLIOGRAPHY
Capek, Milec. 1951. "The Doctrine of Necessity Re-examined." Review of Metaphysics 5: 4055. —. 1959. "Toward a Widening of the Notion of Causality." Diogenes 28: 63-90. Cartwright, Nancy. 1983. How the Laws of Physics Lie. Oxford: Clarendon Press. Carus, Paul. 1883. Ursache, Grund und Zweck. Dresden. Chisholm, Roderick M. 1970. "Events and Propositions." Nous 4: 15-24. —. 1976. "The Agent as Cause." In: Myles Brand and Douglas Walton (eds.). Action Theory. Dordrecht: Reidei. 199-211. —. and Taylor, Richard. 1960. "Making Things to Have Happened." Analysis 20: 73-82. Cicero. (1942). De Fato. Loeb Classical Library. Cicero in twenty eight volumes. Vol. 4. 189245. Trans. H. Rackham, M.A. Cambridge, Mass. and London: Harvard University Press and William Heineman LTD. Colapietro, Vincent. 1989. Peirce's Approach to the Self: A Semiotic Perspective on Human Subjectivity. Albany: State University of New York Press. Collingwood, Robin G. 1938/1991. "On the So-called Idea of Causation." Proceedings of the Aristotelian Society 38. Reprint in: A.B. Schoedinger (ed.). Introduction to Metaphysics: The Fundamental Questions. 145-162. Buffalo, New York: Prometheus Books. Copleston, F.C. 1955/1973. "Commentary on 'The Five Ways' of Aquinas." F.C. Copleston. Aquinas. Reprint in: P. Edwards and A. Pap (eds.). A Modem Introduction to Philosophy. Third edition. 411-18. London/New York: The Free Press. Davidson, Donald. 1980. Essays on Actions and Events. Oxford: Clarendon Press. Debrock, Guy. 1988. "Aristotle Wittgenstein, Alias Isaac Newton between Fact and Substance." In: P.B. Scheurer and G. Debrock (eds.). Newton's Scientific and Philosophic Legacy. Dordrecht: Kluwer Academic Publishers. — . 1996. "Information and the Metaphysical Status of the Sign." In: Vincent Colapietro an Thomas Olshewski (eds.). Peirce's Doctrine of Signs: Theory, Applications, and Connections. 79-89. Berlin-New York: Mouton de Cruyter. Deely, John. 1994. New Beginnings. Early Modern Philosophy and Postmodern Thought. Toronto: University of Toronto Press. Descartes, René. 1644/1983. Principles of Philosophy. Trans. V.R. Miller and R.P. Miller. Dordrecht: Reidel. Dretske, Fred I. 1977. "Laws of Nature." Philosophy of Science 44: 248-68. Ducasse, Curt J. 1926/1991. "On the Nature and the Observability of the Causal Relation. Journal of Philosophy 23: 57-68. Reprint in: A.B. Schoedinger (ed.). Introduction to Metaphysics: The Fundamental Questions. 133-144. Buffalo, New York: Prometheus Books. Dunphy, William. 1966. "Peter of Auvergne and the Twofold Efficient Cause." Medieval Studies 28: 1-21. Dupré, John. 1993. The Disorder of Things. Metaphysical Foundations of The Disunity of Science. Cambridge, Mass. and London: Harvard University Press. —. and Nancy Cartwright. 1988. "Probability and Causality: Why Hume and Indeterminism Don't Mix." Nous 22: 521-36. Eells, Ellery. 1986. "Probabilistic Causal Interaction." Philosophy of Science 53: 52-64. —. and Elliot Sober. 1983. "Probabilistic Causality and the Question of Transitivity." Philosophy of Science 50: 35-57. Elders, Leo. 1974. "Contineri as a Fundamental Structure of Saint Thomas' Ontology." Aquinas 18: 97-106. Emmet, Dorothy. 1984. The Effectiveness of Causes. London: MacMillan.
BIBLIOGRAPHY
235
—. 1992. The Passage of Nature. Hampshire: MacMillan. Fair, David. 1979. "Causation and the Flow of Energy." Erkenntnis 14: 219-50. Fales, Evan. 1990. Causation and Universals. London and New York: Routledge. Fisch, Max. 1986. "Peirce's Progress from Nominalism toward Realism." In his Peirce, Semeiotic, and Pragmatism. 184-200. Bloomington: Indiana University Press. Ford, Lewis S. 1972. "Reasons, Causes, and Decisions." Southwestern Journal of Philosophy 3: 51-62. Friedman, Michael. 1992. "Causal Laws and the Foundations of Natural Science." P. Guyer (ed.). The Cambridge Companion to Kant. 161-199. Cambridge: Cambridge University Press. Gaskin, Douglas. 1955. "Causation and Recipes." Mind 64: 479-87. Gerwin, M. 1987. "Causality and Agency: a Refutation of Hume." Dialogue 26: 3-17. Gilson, Etienne. 1962. "Notes pour l'Histoire de la Cause Efficiente." Archives d'Histoire Doctrinale et Littéraire du Moyen Age 37: 7-31. Haack, Susan. 1992. '"Extreme Scholastic Realism:' Its Relevance for Philosophy of Science Today." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 28: 19-50. —. 1993 "Pragmatism." In: J. Dancy and E. Sosa (eds.). A Companion to Epistemology. Blackwell Companions to Philosophy. Oxford and Cambridge Mass.: Blackwell. 351-57. Hacking, Ian. 1991. "A Tradition of Natural Kinds." Philosophical Studies 61: 109-126, 149154. Hart, Herbert L.A. and Honoré, Anthony M. Causation in the Law. 1959. Oxford: Oxford University Press. Hartshorne, Charles. 1954. "Causal Necessities: An Alternative to Hume." Philosophical Review 63: 479-499. —. 1958. "The Philosophy of Creative Synthesis." Journal of Philosophy 55: 944-953. —. 1970. Creative Synthesis and Philosophical Method. Lanham: University Press of America. —. 1973. "Creativity and the Deductive Method." Review of Metaphysics 27: 62-74. Hausman, Carl R. 1974. "Eros and Agape in Creative Evolution: A Peircean Insight." Process Studies 4: 11-25. —. 1993. Charles S. Peirce's Evolutionary Philosophy. Cambridge: Cambridge University Press. Hempel, C.G. 1965. Aspects of Scientific Explanation. New York: Free Press. Heymans, Gerard, 1890. Schets Eener Kritische Geschiedenis van het Causaliteitsbegrip in de Nieuwere Wijsbegeerte (Dutch). Leiden: E.J. Brill. Hobbes, Thomas. 1655/1839. "De Corpore (Concerning Body)." The English Works Vol. I. Trans. W. Molesworth. London: Scientia Aalen. —. 1656/1841. "The Questions Concerning Liberty, Necessity, and Chance, clearly stated and debated between Dr. Bramhall, Bishop of Derry, and Thomas Hobbes of Malmesburry. " The English Works. Vol. 5. Honderich, Ted. 1988. A Theory of Determinism. Oxford: Clarendon Press. Hookway, Christopher J. 1985. Peirce. London: Routledge & Kegan Paul. —. 1992. "The Idea of Causation. Some Peircean Themes." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 28:261-288. Horwich, Paul. 1987/1993, "Lewis's Programme." Asymmetries in Time (Cambridge, Mass. 1987). 167-76. Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 208-16. Hull, David. 1978. "A Matter of Individuality." Philosophy of Science 45: 335-60.
236
BIBLIOGRAPHY
Hulswit, Menno. 1996. "Teleology: A Peircean Critique of Ernst Mayr's Theory." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 32: 182-214. — . 1997. "Peirce's Teleological Approach to Natural Classes." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 33: 722-772. — . 1998a. "A Guess at the Riddle of Semeiotic Causation." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 34: 641 -688. — . 1998b. "Semeiotic, Causation, and Semeiotic Causation." In: Jaap van Brakel and Michael van Heerden (eds.). C.S. Peirce: Categories to Constantinople. Proceedings of the International Symposium on Peirce - Leuven 1997. 101-117. Leuven: Leuven University Press. —. 2001a. "C.S. Peirce on Causality and Causation." In: Joäo Queiroz ( e d j . The Digital Encyclopedia of Charles S. Peirce. <www.digitalpeirce.org> —. 2001b. "Semeiotic and the Cement of the Universe: A Peircean Process Approach to Causation." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 37: 339-363. Hume, David. 1739-40/1978. A Treatise of Human Nature. Edited by L. A. Selby-Bigge (1888). Second edition, revised by P.H. Nidditch. Oxford: Oxford University Press. —. 1748/1975. Enquiries Concerning Human Understanding and Concerning the Principles of Morals. Edited by L.A. Selby-Bigge (1777). Third edition, revised by P.H. Nidditch. Oxford: Oxford University Press. Johnson, D.M. 1990. "Can Abstractions be Causes?" Biology and Philosophy 5: 63-77 Kant, Immanuel. 1781-87/1963. Critique of Pure Reason. Trans. N.K. Smith. London: MacMillan. —. 1783/1950. Prolegomena to Any Future Metaphysics. Trans. L.W. Beck. The Library of Liberal Arts. Indianapolis: The Bobbs Merill Company. Kent, Beverly. 1987. Charles S. Peirce. Logic and the Classification of the Sciences. Kingston and Montreal: McGill-Queen's University Press Ketner, Kenneth L. 1981. "Peirce's Ethics of Terminology." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 17: 327-47. Kim, Jaegwon. 1971/1993. "Causes and Events: Mackie on Causation." Journal of Philosophy 68 (1971): 426-41. Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 60-74. —. 1973a. "Causes and Counterfactuals." Journal of Philosophy 70 (1973): 570-2. Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 205-7. —. 1973b. "Causation, Nomic Subsumption and the Concept of Event." Journal of Philosophy 70: 217-36. —.1995. "Causation." In: Robert Audi (ed.). The Cambridge Dictionary of Philosophy. 11012. Cambridge: Cambridge University Press. Kraus, Elizabeth M. 1985. "Existence as Transaction: a "Whiteheadian Study of Causality." International Philosophical Quarterly 81: 349-366. Kripke, Saul. 1980. Naming and Necessity. Oxford: Blackwell. Kruse, Felicia. 1986. "Indexicality and the Abductive Link." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 22: 435-447. —. 1990. "Nature and Semiosis. " Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 26: 211-224. —. 1991. "Genuineness and Degeneracy in Peirce's Categories." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 27: 267-298.
BIBLIOGRAPHY
237
Lauer, Rosemary. 1974. "The Notion of Efficient Cause in the Secunda Via." Thomist 38: 754-67. Ledere, Ivor. 1986. "Leibniz and the Analysis of Matter and Motion." In his The Philosophy of Nature. 75-90. Washington D.C.: The Catholic University of America Press. Leibniz, Gottfried Wilhelm, ca. 1680-84/1976. "First Truths." In: Philosophical Papers and Letters. 267-271. Ed. L.E. Loemker. Dordrecht/Boston/London: Reidel Publishing Company. —. ca. 1696/1976. "Tentamen Anagogicum: An Anagogical Essay in the Investigation of Causes." Philosophical Papers and Letters. 477-85. Ed. L.E.Loemker. Dordrecht/Boston/London: Reidel Publishing Company. —. ca. 1712/1973. "Metaphysical Consequences of the Principle of Reason." Philosophical Writings. Ed. C.H.R. Parkinson. Trans. M. Morris and C.H.R. Parkinson. London: J.M. Dent & Sons. Everyman's Library. 172-78. —. ca. 1714/1976. "Monadology." Philosophical Papers and Letters. 643-653. Ed. L.E. Loemker. Dordrecht/Boston/London: Reidel Publishing Company. Levinson, J. 1991. "Attribute." Handbook of Metaphysics and Ontology (H. Burkhardt and B. Smith, eds.), München: Philosophia. 1: 65-70. Lewis, David. 1973/1993. "Causation." Journal of Philosophy 70: 556-67. Reprint in: Sosa and Tooley (eds.). Causation (Oxford 1993). 193-204. —. 1986. Philosophical Papers II. Oxford: Oxford University Press. Liszka, James J. 1996. A General Introduction to the Semeiotic of Charles Sanders Peirce. Bloomington: Indiana University Press. Locke, John. 1690/1975. An Essay Concerning Human Understanding. Edited by P.N. Nidditch. Oxford: Clarendon Press. Lombard, L.M. 1986. Events: A Metaphysical Study. London: Routledge and Kegan Paul. Long, AA. 1996. Stoic Studies. Cambridge: Cambridge University Press. —. and D.N. Sedley, 1987. The Hellenistic Philosophers. Vol. 1. Translations of the principal sources with philosophical commentary. Cambridge: Cambridge University Press. Mackie, John L. 1965/1993. "Causes and Conditions." American Philosophical Quarterly 2/4 (1965): 245-64. Reprint in: Sosa and Tooley (eds.). Causation (Oxford 1993). 33-55. —. 1974. The Cement of the Universe. A Study of Causation. Oxford: The Clarendon Press. Madden, Edward H., and Harré, Rom. 1975. Causal Powers. A Theory of Natural Necessity. Oxford: Blackwell. Marc-Wogau, Konrad. 1962. "On Historical Explanation." Theoria 28: 213-33. Martin, Raymond. 1972. "The Sufficiency Thesis." Philosophical Studies 23: 205-11. Martinich, A.P. 1995. A Hobbes Dictionary. Oxford: Blackwell. Mauer, A. 1983. "C. S. Peirce and Duns Scotus." Kris is 1: 6-9. Mayr, Ernst. 1974. "Teleological and Teleonomic. A New Analysis." Boston Studies in the Philosophy of Science 14: 91-117. —. 1988. Toward a New Philosophy of Biology. Observations of an Evolutionist. Cambridge, Mass. and London: Harvard University Press. Mellor, D. Hugh. 1980. "Necessities and Universals in Natural Laws." In: Mellor (ed.). Science, Belief and Behaviour. Cambridge: Cambridge University Press. 105-25. —. 1986. "Fixed Past, Unfixed Future." In: Barry Taylor (ed.). Contributions to Philosophy: Michael Dummett (The Hague). 105-25. —. 1995. The Facts of Causation. London and New York: Routledge. Menzies, Peter. 1989a. "A Unified Account of Causal Relata." Australasian Journal of Philosophy 67: 59-83.
238
BIBLIOGRAPHY
— . 1989b. "Probabilistic Causation and Causal Processes: A Critique of Lewis." Philosophy of Science 56: 642-63. Miles, Murray. 1988. "Descartes' Mechanicism and the Medieval Doctrine of Causes, Qualities, and Forms." The Modern Schoolman 65: 97-117. Mill, John Stuart. 1874. A System of Logic. Eighth edition. New York: Harper and Brothers. Nagel, Ernest. 1961. The Structure of Science. London: Routledge and Kegan Paul. Neville, Robert C. 1988. "Motion in Causal Agency." Journal of Speculative Philosophy 2: 175-91. Newton, Isaac. 1687/1968. The Mathematical Principles of Natural Philosophy (Philosophiae Naturalis Principia). Trans. A. Motte (1729), with an introduction by I. Bernard Cohen. London: Dawson. Pape, Helmut. 1989. Erfahrung und Wirklichkeit als Zeichenprozess. Charles S. Peirces Entwurf einer Spekulativen Grammatik des Seins. Frankfurt am Main: Suhrkamp. — . 1993. "Final Causality in Peirce's Semiotics and his Classification of the Sciences." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 29: 581-607. Papineau, David. 1985. "Causal Asymmetry." British Journal for the Philosophy of Science 36: 273-89. Plato. (1971). Timaeus and Critias. Trans. D. Lee. Penguin Classics. Harmonsworth Middlesex: Penguin books LTD. Popper, Karl. 1972. Objective Knowledge: An Evolutionary Approach. Oxford: Clarendon Press. Putnam, Hilary. 1975. "The Meaning of Meaning." In his Philosophical Papers, vol. II. Cambridge: Cambridge University Press. 215-271. Quine, Willard. 1969. "Natural Kinds." In his Ontological Relativity and Other Essays. New York: Columbia University Press. 114-38. Railton, Peter. 1981. "Probability, Explanation, and Information." Synthèse 48: 233-56. Ransdell, Joseph M. 1977. "Some Leading Ideas of Peirce's Semiotic." Semiotica 19: 157-78. — . 1981. "Semiotic Causation: a Partial Explication." Proceedings of the C.S. Peirce Bicentennial International Congress. Ed. Kenneth L. Ketner et al. Texas Tech University. Graduate Studies 23. Lubbock: Texas Tech University Press. 201-06. —. 1986. "Peirce, Charles Sanders." Encyclopedic Dictionary of Semiotics 1: 673-95. Berlin: Mouton de Gruyter. —. X. (unpublished paper). "Teleology and the Autonomy of the Semiosis Process." 10 pp. —. Forthcoming. The Meanings in Things. The Basic Ideas of Charles Peirce's Semiotic. Reichenbach, Hans. 1956/1971. The Direction of Time. Berkeley: University of California Press. Reid, Thomas. 1846/1967. Philosophical Works. Vol i and ii. With notes and supplementary dissertations by W. Hamilton. Hildesheim: Georg Olms. Rosen, Deborah A. 1978. "In Defense of a Probabilistic Theory of Causality." Philosophy of Science 45: 368-86. —. 1980. "A Probabilistic Theory of Causal Necessity." Southern Journal of Philosophy 18: 71-86. Rosenberg, Alex. 1991. "How is Eliminative Materialism Possible?" In: Bogdan (1991). 12143. — . 1992. "Causation, Probability and the Monarchy." American Philosophical Quarterly 29: 305-18. Rosenthal, Sandra. 1994. Charles Peirce's Pragmatic Pluralism. Albany: State University of New York Press.
BIBLIOGRAPHY
239
Russell, Bertrand. 1912/1991. "On the Notion of Cause." Proceedings of the Aristotelian Society 13 (1912): 1-26. Reprint in: A.B. Schoedinger (ed.). Introduction to Metaphysics: The Fundamental Questions. 119-132. Buffalo, New York: Prometheus Books. Sachsse, Hans. 1979. Kausalität, Gesetzlichkeit, Wahrscheinlichkeit: die Geschichte von Grundcategoriën zur Auseinandersetzung des Menschen mit der Welt. Darmstadt: Wissenschaftliche Buchgesellschaft. Salmon, Wesley C. 1977. "An At-At' Theory of Causal Influence." Philosophy of Science 44: 215-24. —. 1978. "Why Ask 'Why?'" Proceedings and Addresses of the American Philosophical Association 51/6: 683 -705. —. 1980a/1993a. "Probabilistic Causality." Pacific Philosophical Quarterly 61 (1980): 50-74. Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 137-53. —. 1980b/1993b. "Causality: Production and Propagation." Proceceedings of the 1980 Biennial Meeting of the Philosophy of Science Association. Vol. 2 (ed. Peter D. Asquith and Ronald N. Giere): 49-69. Reprint in Sosa, Ernest andTooley, Michael (eds.). Causation (Oxford 1993). 154-71. —. 1984. Scientific Explanation and the Causal Structure of the World. Princeton, New Jersey: Princeton University Press. Sanford, David H. 1975. "The Direction of Causation and the Direction of Conditionship." Journal of Philosophy 73: 193-207. Savan, David. 1988. An Introduction to C.S. Peirce's Full System of Semeiotic. Monograph of the Toronto Semiotic Circle. Number 1. Schacter, Daniel L. 1996. Searching for Memory: the Brain, the Mind, and the Past. New York: BasicBooks. Scriven, Michael. 1956-57. "Randomness and the Causal Order." Analysis 17: 5-9. — 1 9 6 6 / 1 9 9 3 . "Defects of The Necessary Condition Analysis of Causation." Philosophical Analysis and History. Ed. W. Dray (1966). Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 56-59. —. 1971. "The Logic of Cause." Theory and Decision 2: 49-66. —. 1975. "Causation as Explanation." Nous 9: 3-16. Seager, William E. 1988. "Peirce's Teleological Signs." Semiotica 69: 303-14. Shoemaker, Sydney. 1980. "Causality and Properties." In: Peter van Inwagen (ed.). Time and Cause (Dordrecht 1980). 109-35. Short, Thomas L. 1981a. "Peirce's Concept of Final Causation." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 17: 369-382. —. 1981b. "Semeiosis and Intentionality." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 17: 197-233. —. 1982. "Life Among the Legisigns. " Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 18: 285-310. —. 1983 "Teleology in Nature." American Philosophical Quarterly 20: 311-20. —. 1986. "What they said in Amsterdam: Peirce's Semeiotic Today." Semiotica 60: 103 -128. —. 1994. "Review of Carl Hausman's 'Charles S. Peirce's Evolutionary Philosophy'." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 30: 401-13. —. 1996. "Interpreting Peirces Interprétant: A Response to Lalor, Liszka, and Meyers." Transactions of the Charles S. Peirce Society. A Quarterly Journal in American Philosophy 32: 488-541. Skyrms, Brian. 1980. Causal Necessity. New Haven, Conn., and London: Yale University Press.
240
BIBLIOGRAPHY
Sober, Elliot. 1995. "Natural Kinds." In: J. Kim and E. Sosa (eds.). A Companion to Metaphysics. Blackwell Companions to Philosophy. Oxford and Cambridge, Mass.: Blackwell. 345-346. Soontiëns, Frans. 1988. Evolutie en Finaliteit (Dutch). Dissertation. Nijmegen: University of Nijmegen Press. — . 1991. "Evolution: Teleology or Chance?" Journal for General Philosophy of Sciences 22: 133-141. Sorabji, Richard. 1980. Necessity, Cause and Blame. Perspectives on Aristotle's Theory. Ithaca, New York: Cornell University Press. Sosa, Ernest. 1980/1993a. "Varieties of Causation." Grazer Philosophische Studien 11 (1980): 93-103. Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 234-42. — . and Tooley, Michael (eds.). 1993. Causation. Oxford Readings in Philosophy. Oxford: Oxford University Press. Spinoza, Baruch. 1677/1949. Ethics. Trans. J. Gutman. New York: Hafner Publishing Company. Sprigge, Timothy L.S. 1971. "Final Causes." Proceedings of the Aristotelian Society. Supplementary 45: 149-70. Stalnaker, Robert C. 1968. "A Theory of Conditionals." In: N. Rescher (ed.). Studies in Logical Theory (Oxford 1986). 98-112. Strawson, Galen. 1989. The Secret Connexion: Causation, Realism, and David Hume. Oxford: Clarendon Press. Suppes, Patrick. 1970. A Probabilistic Theory of Causality. Amsterdam: North Holland Publishing Company. — . 1984. Probabilistic Metaphysics. Oxford: Blackwell. Swoyer, Chris. 1982. "The Nature of Natural Laws." Australasian Journal of Philosophy 60/3: 203-23 Taube, Mortimer. 1936. Causation, Freedom and Determinism. London: George Allen and Unwin LTD. Taylor, Richard. 1962-3. "Causation." The Monist 47: 287-313. —. 1966. Action and Purpose. Englewood Cliffs, NJ: Prentice Hall. —. 1967. "Causation." In: P. Edwards (ed.). Encyclopedia of Philosophy 2 (New York and London). 56-66. Tooley, Michael. 1977. "The Nature of Laws." Canadian Journal of Philosophy 7/4: 667-98. —. 1987. Causation: A Realist Approach. Oxford: Clarendon Press. — . 1990a. "The Nature of Causation: A Singularist Account." In: David Copp (ed.). Canadian Philosophers: Celebrating Twenty Years of the CJP. Canadian Journal of Philosophy. Suppl. 16: 271-322. — . 1990b. "Causation: Reductionism Versus Realism." Philosophy and Phenomenological Research 50. Suppl.: 215-36. Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 172-92 Turley, Peter T. 1977. Peirce's Cosmology. New York: Philosophical Library, van Brakel, Jaap. 1992. "Natural Kinds and Manifest Forms of Life." Dialéctica vol. 46: 243261. van Fraassen, Bas C. 1989. Laws and Symmetry. Oxford: Clarendon Press. van Inwagen, Peter (ed.). 1980. Time and Cause: Essays Presented to Richard Dordrecht: Reidel Publishing Company von Bertalanffy, L. 1968. General System Theory. New York: Braziller.
Taylor.
BIBLIOGRAPHY
241
von Wright, Georg Henrik. 1971. Explanation and Understanding. London: Routledge and Kegan Paul. —. 1973/1993. "On the Logic and Epistemology of the Causal Relation." Logic, Methodology and Philosophy of Science IV. ed. P. Suppes et al. (Amsterdam 1973). Reprint in: Sosa, Ernest and Tooley, Michael (eds.). Causation (Oxford 1993). 105-24 Wallace, William. 1974. Causality and Scientific Explanation. Vol. 1. Medieval and Classical Science. Vol. 2. Classical and Contemporary Science. Ann Arbor: The University of Michigan Press. Weiss, Paul. 1947/1983. Nature and Man. Lanham/New York/London: University Press of America. Welten, Willibrordus. 1961. Causaliteit in de Quantummechanica (Dutch). Dissertation. Groningen: Erven P. Noordhoff N.V. Whitehead, Alfred N. 1919. The Principles of Natural Knowledge. Cambridge: Cambridge University Press. —. 1920/1964. The Concept of Nature. Cambridge: Cambridge University Press. —. 1925/1967. Science and the Modern World. New York: The Free Press. —. 1927/1958. Symbolism, its Meaning and Effect. Cambridge: Cambridge University Press. —. 1929/1978. Process and Reality. An Essay in Cosmology. New York: The Free Press. —. 1933/1967. Adventures of Ideas. New York: The Free Press. —. 1938/1968. Modes of Thought. New York: The Free Press. Wiener, N. 1950. The Human Use of Human Beings. New York: Avon. Wolf-Gazo, Ernest. 1985. "Whitehead and Locke's Concept of 'Power'." Process Studies 14: 237-51.
INDEX
Behavior goal-directed behavior, 10, 77 law-like behavior, 15, 30, 31,42, 44, 170 mechanical behavior, 82, 86 program-directed behavior, 92 teleonomic behavior, 93 Bennet, J.O., 176 Bigelow, J., 99 Biological evolution, 75, 88, 90, 95, 184 Biological species, 88, 90, 98-100, 110, 122, 126, 127, 132 Blanshard, B„ 33 Blind force, 114 Boundary condition, 5, 83, 92, 193 Boyd, R„ 99 Boyle, R., 27 Braithwaite, R.B., 50 Broad, C.D., 68 Bunge, M., 66
A Detailed Classification of the Sciences, 113 Action dyadic action, 140 dynamical action, 140, 196 intelligent action, 140, 141 mental action, 185, 186 triadic action, 140, 141, 196 Actuality, 11, 110, 120, 121, 126, 132, 138, 152, 154-156, 158,201 Agassiz, L., 112 Agency, xiii, xvi, 7, 11-16, 19-21,29, 56-58, 67,68,71-74, 142-144, 147, 149, 176, 177, 182, 183, 196 Aitia Aristotle's theory of the four aitiai, 3-5,219 Annas, J., 5 Anscombe, E., xiii, 51, 60, 61 Anthropomorphism, 24, 76, 77, 185 Aquinas, 2, 8-15, 18, 2 3 , 4 2 , 2 1 9 Aristotle, 2-5,7-9, 11-13, 15,41,76, 127, 134, 167, 170, 178, 179, 182, 183, 187,203, 2 0 4 , 2 1 4 , 2 1 6 , 2 1 9 Armstrong, D.M., 65, 66 Art objects, 127 Artificial
Categoreal framework Aristotelian categoreal framework, 35, 177 Aristotelian-scholastic framework. 35, 173, 175 incompatible categoreal frameworks, 167, 176, 179,214 Categories, 1, 4, 35, 36, 83, 129, 182, 201,214,219 Causal
artificial classifications, 113, 123, 128, 132 artificial kinds, 100, 104, 105, 107, 122, 131 artificial objects, 122, 123 Ayal a, F.J., 89
243
244
INDEX
causal circumstances, 13-15, 39, 40, 42, 48, 90 causal concepts, analyzability of, 66, 73 causal condition, 6, 14, 49, 86, 213 causal explanations, 69, 82, 93, 182, 207, 209,210 causal necessity, 32, 43, 170 causal over-determination, 50, 55 causal pluralism, 128, 130, 132 causal realism, 66 causal relata, xix, 1, 4, 19, 32, 34, 35, 3 8 , 7 1 , 7 3 , 172-176, 178, 179, 181, 190, 191,214 causal relation, xiii, xvi, 5, 7, 19, 21,31-34, 37, 42-44, 55, 60-62, 64-68,71-73, 131, 151, 154, 155, 158, 161, 168-172, 175, 178, 179 causal statements, 49, 57, 61, 69, 71,72,74 causally relevant conditions, 48, 49 causally sufficient conditions, 51 probabilistic causal laws, 51, 53, 58,71,74 Causality causation and causality, xiv, 171, 215 contemporary analyses of causality, xix definition of causality, 169 efficient causality, 11-13, 15, 18, 26, 42, 75, 81, 94, 212, See also efficient causation final causality, 12, 13, 18, 26, 27, 81,126, See also final causation intrasubstantial causality, 24 law of causality, xiii Peirce on causality and causation, 181
Peirce's approach to causality, xviii principle of causality, xix, 2, 6, 30, 35, 37, 39, 181, 190,214 Causation backward causation, xvii, 76, 81, 90, 91, 95, 146, 189 causation and causality, xiv, 171, 215 causation and necessity, 11, 32 causation and power, 27, 71, 72, 74 causation and regularity, 31 causation and teleology, xxi, 72, 211,214 causation and time, xvi, xxi, 71, 74, 176,211,214,216 causation as a philosophical category, 66 causation as a relation between events, 68, 178 causation as a relation between facts, 69 causation as a theoretical relation, 66, 67 causation as an a priori conception, 35 counterfactual approach to causation, xv, 47, 54-56, 73, 168, 169, 176 current theories of causation, 167, 179,213 direction of causation, 71, 74 efficient causation, xvii, xviii, xx, 5, 11, 15,26, 72, 75,76, 79-82, 85, 87, 88,91,94-97, 1 14, 115, 125, 127, 133, 143, 145, 146, 148-152, 156, 159, 161-165, 187-189, 197, 203, See also efficient causality final causation, xvi-xviii, xx, 10, 13, 15, 19,21,24, 26, 43,74,
245
INDEX 75, 76, 78, 80-82, 84-86, 88, 90, 91,93-98, 114, 115, 120, 125, 131, 133, 143, 145, 148-151, 159, 162-165, 188, 189, 197, 206, See also final causality final causation in mechanical processes, 84 formal causation, 13, 15, 18, 19, 42, 43 formal characteristics of causation, 195, 199 formal structure of causation, 166, 195, 198, 199 form-aspect of causation, 207 immediate causation, 20 instrumental approach to causation, xv, 47, 56, 168, 169, 176 law of universal causation, 31 means-to-ends approach to causation, 56 mechanical causation, 78-80, 87, 88, 93,94, 114, 125 mediate causation, 20 natural classes and causation, 98, 131 Peirce on causality and causation, 181
Peirce's conception of causation, xiv, xv, xvii, 74,98, 145, 188, 200, 201 Peirce's theory of final causation, xvi, 74, 76, 88, 95 physical causation, 71, 72, 74, 149 primary requirements for a theory of causation, 210, 216 principle of causation, 1, 17, 182, 183 probabilistic approach to causation, xvi probabilistic causation, 47, 51, 56, 58,72, 73, 168, 169, 176
probabilistic theories of singular causation, 59 received view regarding causation, 167 semeiotic approach to causation, xviii, XX, 166, 181, 195 semeiotic causation, xviii, xix,
132-134, 139, 149, 151-153, 158, 159, 163-166, 181, 195, 197, 199, 203,210 singularist approach to causation,
29, 47,71 singularistic approach to causation, xvi subsidiary requirements for a
theory of causation, 211,216 teleological causation, 26, 98, 125 uniformity and causation, xvi uniformity of causation, 71, 72, 74 universality of causation, xvi, 71, 72, 74 Cause ambiguity in the conception of cause, 63 Aristotelian conception of cause, xv, 3 5 , 4 3 , 4 4 , 173, 179 cause and effect, xiv causes and conditions, 57, 62 causes and counterfactual dependency, 54 causes and effects, xiii, 23, 27-29, 3 1 , 3 4 - 3 8 , 4 1 , 4 2 , 52,56, 62, 67,70, 96, 98, 105, 170-176, 179, 180, 191, 194, 206,210214,216,217 causes as active initiators of a change, 43 causes as facts, 68 causes as inactive nodes in a lawlike implication chain, 42, 44, 170, 171, 179, 188, 189,214 causes as means-to-ends, 56
246
INDEX
causes as necessary conditions, 48, 51 causes as substantial powers, 29 causes as sufficient conditions, 50, 51, 172 concept of cause, xiii, 2, 8, 15, 19, 3 0 , 3 1 , 3 6 - 3 8 , 4 1 , 4 4 , 45,51, 67,72, 170, 171, 179, 181, 182, 214,215 definition of cause, 34, 54, 55, 61 entire cause, 20 external cause, 6, 7, 22, 23 external final cause, 12 final cause of semeiosis, 147-149, 151, 159, 165 first cause, 8, 10, 12, 219 free cause, 23, 43 general cause, 18 general final cause, 80, 82, 87, 88, 145, 146, 159, 188,200 genuine cause, 16, 23 history of the concept of cause, xv, xix internal cause, 6, 7 internal final cause, 12 intrinsic final cause, 127 longitude of the final cause, 117,
120 loose causes, 14 mechanical cause, 17, 87 mutually incompatible conceptions of cause, 170, 189 necessary causes, 21, 23 partial cause, 39 particular causes, 7, 11, 18,219 Peirce's three concepts of cause,
182 primary causes (causae primae), 18, 19 primary efficient cause, 11 real cause, 24, 40, 44 reasons and causes, 71,72, 74
scientific conception of cause, xv, 42, 44, 171, 181 secondary causes (causae secundae), 1, 2, 8, 10, 12, 15, 18, 19, 23 secondary efficient cause, 11,15 signs and causes, xiv status of causes and effects as such, 180,211,213,214 subordinate final causes, 87 subsidiary final causes, 117, 120 substances as causes, 38, 67 tight causes, 14 transcendental cause, 10 uncaused events, 6, 31, 84 universal cause, 11, 26 vagueness of the final cause, 116, 117,120 Chance action of chance, 79, 83, 85 objective chance, xvii, xix, 59, 76, 83-85, 90, 95, 146, 185, 189, 215 Chemical chemical compounds, 124, 125,
126 chemical elements, 98, 100, 122-
126 chemical substances, 123, 127 Chrysippus, 7 Cicero, 6, 7, 219 Class artificial classifications, xvii class qualities, 118, 120 concept of natural classes, xvii definition of natural class, 106, 107, 108 different levels of natural classes, 129, 130 essence of a natural class, 112 examples of natural classes, 98, 100, 122, 130
INDEX kinds and classes, 98, 109 natural class of socialists, 114, 115 natural classes, xviii, 97, 98, 102107, 110-113, 116, 118-123, 126-132, 181,206 natural classes and causation, 98, 131 natural classes and the uniformity of nature, 105 natural classifications, xvii Peirce's theory of natural classes, xvii Peirce's definition of a natural class, 130 purposive classes, 117 real classes, 104, 113 social classes, xviii, 122, 132 Classical essentialism, 128, 132 Classification of the sciences (Peirce's), 113, 198 Collingwood, R.G., 14, 31, 48, 56, 57, 168, 175 Condition causal condition, 6, 14,49, 86, 213 logically sufficient conditions, 51 necessary conditions, xv, 13, 21, 26, 37,48-52, 54, 109, 150, 160-165, 180 sufficient conditions, 34, 47, 48, 50-54, 56, 58, 71-73, 120, 132, 168, 169, 172, 176 Conditions sufficient conditions, xv, xvi, 51, 53, 168, 169, 172 Constant conjunction, 32, 34, 43, 48, 61, 170 Contiguity, 19, 31, 33, 63, 154 Contingency, 21, 23, 24, 38, 40, 50 Continuity, xvi, xxi, 68, 71, 74, 82, 83, 103, 115, 129, 143, 174, 176, 177, 179, 180, 200,201,210,211, 213,214,216
247 Counterfactual conditionals, 53-55, 64 Counterfactual dependence, 54-56, 168 Creativity, 11, 27, 76, 84-87, 89, 90, 163, 187 Davidson, D„ 61, 69, 70, 71 D-character (defining character), 104, 106-109, 111, 113, 116, 119, 120, 128, 129 Debrock, G„ 178, 197, 198, 215 Demarcation criteria, 98, 101, 103, 116-118, 120, 130, 131, 179 Descartes, R„ 2, 16-19, 21, 22, 24, 27, 33, 42, 43 Determine, 133, 134, 143-145, 147, 160-163, 165 Determinism, 2, 7, 15, 16, 19, 27, 42, 51, 85, 86, 186 Directionality, 52, 53, 56, 180, 193 Disposition, 5, 14, 144, 189 DNA, 89, 92, 126, 127 Dualism, 82, 83, 88,91 Ducasse, C.J., 60-64, 68, 71, 73, 169, 172 Duns Scotus, 110, 119, 121 Dupré, J., 127, 128 Eliminativism, 127, 128, 132 Emmet, D., 68, 70, 193 Empedocles, 2 Ens rationis (being of reason), 110, 119 Essence epistemological essence, 111, 112, 116 essence and existence, 8, 9 essence of a natural class, 112 metaphysical essence, 111, 112, 116, 118, 120, 126, 132 Peircean essences, 111
248 Essential qualities, 35, 108, 116, 1 18, 120, 127 Essentialism, 99, 128, 129, 132 Event anatomy of (Peircean) events, xx, 199, 201,215 concept of an event, 68 events and processes, 180, 192, 199,210,214,216 future events, 21, 76, 78, 80, 91, 95,97, 114, 121, 188, 201,206, 209,210,213 logic of events, 191, 213 nature of an event, 175, 202, 215 productive events, 166, 180, 181, 192, 197, 200,203,207-210, 212,215,217 uncaused events, 6, 31, 84 Evolution, 1, 2, 15, 44, 75, 84, 87-90, 95, 111, 114, 120, 136, 170, 179, 184, 187, 188, 214 Exceptionless regularity, 5, 6, 42 Existence, 8-10, 14, 21-23, 26, 28, 3 5 , 3 9 , 4 1 , 6 5 , 6 8 , 7 0 , 78, 100, 105, 109, 110, 113-116, 120-122, 124, 126, 128, 130, 131, 183, 190192, 201 Explanation, Aristotle's four types of, 2 Extension, 17, 18, 22, 25, 193, 200
INDEX Force, 18, 26, 30-32, 77, 79, 81, 91, 114, 126, 140, 145, 149, 155, 156, 181, 184-187, 193, 196,215 Form classification according to form, 127 forms as general rules, 204 Peirce's concept of form, 206 Peircean forms, xx, 206, 211,216 substantial form, 9, 17, 26, 27, 42, 43 transmission of forms, xx, 203205,207, 209,211,216 transmission of forms within causation, 205, 209, 211, 216 transmission of forms within semeiosis, 205, 209 Free will, 14 Future, 21, 25, 76-78, 80, 87, 89-91, 95, 97, 114, 121, 126, 133, 141, 144, 149, 152, 161, 183, 186-188, 201,202, 206, 209-213,216 Galileo, G„ 27 Gaskin, D„ 57, 58, 72 Gasper, P., 99 General (physical) possibilities, 78, 80, 91, 95, 97, 114, 188 General types, 77-80, 93-95, 97, 114, 135, 153, 155, 158, 178, 179, 188,
200 Fact ontology, xix, 178, 179, 210, 214 Fales, E., xiii Fallacy of misplaced concreteness (A.N. Whitehead), 176, 214 Fate, 6, 7, 219 Finality, 12, 84, 95, 132 Finious, 79, 88 Firstness, 120, 196, 201 Fitzgerald, 141
Generality, 62, 77, 87, 89, 102, 116 Generals, 101, 102, 110, 119, 121, 130, 154, 156, 204 Gilson, E„ 8 Goal of interpretation, 142 Goal, conception of, 89 God, 2, 8-12, 15, 16, 18, 19,21-24, 2 6 , 2 7 , 4 3 , 111,219 Gravity, 13, 8 1 , 9 1 , 9 2 , 178,206
INDEX
Haack, S., 101-103, 122, 127, 128, 130, 131 Habit, 32, 92, 94, 102, 111,114, 120, 139, 153, 185, 186, 188, 189, 201, 202, 204, 207 Hacking, I., 97, 100, 105 Hare, M.E., 86 Hart, H.L.A., 48-50 Hartshorne, Ch., 48, 68, 86, 171, 192 Heathcote, A., 66 Hempel, C.G., 50 Heredity, 126, 127 Hobbes, T., 2, 16, 19-21, 24, 27, 33, 42, 43 Honderich, T., 50 Honoré, A.M., 48-50 Hookway, C., 101-103, 127, 128, 130, 131 Hull, D., 100 Hume, D., 2, 17, 29, 31-39, 41, 43, 44, 47, 49, 54, 61, 67, 71, 170-173, 175, 177-179, 203,204,211 Hypostatic abstraction, 110 Hypothesis, 5, 77, 143, 148, 153, 157, 187, 191, 198, 199, 203, 207209,212,215 Icon, 134, 135, 140, 147, 150, 153155, 158-163, 165 Indeterminacy, 146 Index, 134, 135, 140, 147, 150, 153155, 158, 159, 162, 163, 165 Induction, 35-38, 100, 101, 105, 122, 131 Interprétant dynamical interprétant, 138, 139, 141, 146, 157, 159, 163 emotional interprétant, 138, 152 energetic interprétant, 138, 139, 152, 157
249 final interprétant, 138, 141, 142, 146, 147, 158, 159 immediate interprétant, 138-140 logical interprétant, 138, 139, 152 INUS condition, 52, 53, 73, 168 Johnson, D.M., 99 Kant, I., 2, 17, 35-39,44 Kim, J., 47, 55, 68, 169, 177 Kind artificial kinds, 100, 105, 107 natural kinds, 97-105, 107, 124, 128, 130 Peirce's Baldwin definition of kind, 107 social kinds, 100, 105 Kripke, S., 100 Law law-like behavior, 15, 30, 31, 42, 44, 170 laws of nature, 18, 42, 43, 51, 54, 6 1 , 6 5 , 8 2 , 9 1 , 9 2 , 99-102, 104, 114, 121, 122, 131, 184, 186, 188-191 Ledere, I., 25 Leibniz, G.W., 2, 16, 24-27, 33, 42, 43 Levinson, J., 99 Lewis, D„ 54, 55, 59, 68, 168, 177 Liber de Causis, 8 Locomotion, 15, 18, 24 Long, A.A., 6, 7 Mackie, J.L., 52, 53, 63, 64, 73, 168 Mathematics, 24 Matter as extended substance, 25 Mayr, E„ xvii, 76, 88-96 McTaggart, J., 33 Means to ends, 12, 15, 43, 56 Mechanistic worldview, 17
250 Mellor, D.H., 59, 70,71, 178 Memory, 155, 174, 208 Mendeleef, D„ 123, 124 Menzies, P., 59 Micro-reductionism, 128, 131, 132 Mill, J.S.,2, 17, 39, 4 0 , 4 1 , 4 4 , 47, 49, 50, 57, 62, 98, 103-108, 122, 190, 191 Mind, 4, 7, 11, 22, 28, 29, 32, 33, 43, 63, 87, 107, 110-112, 115, 121, 133, 137, 138, 143, 144, 147, 150, 151, 153, 154, 156, 158, 160, 161, 170, 185, 186, 189, 204 Modern science, 15, 27, 97 Monad, 25-27, 43 Natural classification, 97, 103, 111, 113, 122-124, 126-130, 132 Natural kinds Haack's interpretation of Peircean natural kinds, 101, 127, 130 Hookway's interpretation of Peircean natural kinds, 102 Mill's theory of natural kinds, 104 Mill's definition of natural kinds, 103 Mill's theory of natural kinds, 98, 104, 107 natural kinds and the uniformity of nature, 105 Peirce's earliest discussion of natural kinds, 105 Rosenthal's interpretation of Peircean natural kinds, 103 Natural selection, 77, 89, 90 Necessity absolute necessity, 13, 14, 185 causal necessity, 32, 43, 170 conditional necessity, 13, 201, 202 doctrine of necessity, 85, 184, 185 logical necessity, 23, 27, 32, 43, 48, 50
INDEX natural necessity, 13-15, 42 necessary condition, 13, 21, 26, 37,48-52, 54, 109, 150, 160165, 180 necessary connection, 14, 20, 293 3 , 3 5 , 4 3 , 4 4 , 170, 171, 173 necessary relation, 23, 33, 41, 172, 216 relative necessity, 13 Newton, I., 2, 17, 27, 29-31, 39, 44, 171 Nominalism, 121 Novelty, xvii, xix, 83, 85, 86, 95, 145, 185, 189, 194, 201,215 Ockham, W. of, 121 Orthogenesis, 90 Plato, 2 , 2 1 9 Pluralism anarchistic pluralism, 128 causal pluralism, 128, 130, 132 radical ontological pluralism, 128 Popper, K„ 50 Possibility, 6, 21, 51, 60, 78, 80, 81, 8 5 , 9 1 , 9 5 , 9 7 , 102, 109, 110, 114, 119, 121, 126, 128, 129, 132, 136, 138, 150, 153, 154, 158, 159, 188, 189, 201,202,215 Power active power, 28, 29 passive power, 28 PRE-character (permanently relevant empirical character), 108, 109, 113, 116, 118-120 Pre-established harmony, 26, 27 Process Emmet's definition of process, 193 final causation in mechanical processes, 84 finious processes, 79, 88
INDEX genuinely goal-directed processes, 88,91 irreversible processes, 37, 181, 215 mechanical and teleological processes, difference between, 76 mechanical processes, 75, 82-84, 95, 185, 187 mental processes, 182, 185 natural processes, 4, 17, 42, 75, 82, 85, 143, 182, 184, 185, 187, 203,212,217 Peircean processes, 192-194, 200 reversible processes, 181, 215 seemingly goal-directed processes, 91 semeiotic processes, 138, 147, 210 teleological processes, xvii, xviii, xx, 72, 75, 76, 78, 82, 84, 85, 87,90, 9 3 , 9 5 , 9 6 , 133, 141143, 145-147, 159, 164, 165, 192, 194, 197 teleomatic processes, xvii, 91-93, 96 teleonomic and teleomatic processes, distinction between, 92, 93 teleonomic processes, xvii, 91-93, 96 thermodynamic processes, 79, 93 Program built-in program, 91-93 program as responsible for teleonomic behavior, 93 program-directed behavior, 92 Promiscuous realism, 128 Purpose, 3, 16, 24, 27, 39,43, 76, 77, 8 1 , 8 7 , 8 8 , 109, 111-113, 117, 118, 122, 123, 128-130, 132, 142, 153, 156, 157 Putnam, H„ 84, 100, 111
251
Quasi-purpose, 113, 122 Quine, W.V.O., 99, 100 Ramsey, F., 70 Ransdell, J.M., 133, 134, 138, 144149, 158-160, 163-165,210 Rationality, 107, 108, 114 Real generals, 110, 119 Realism, 66, 100, 101, 116, 121, 128, 131 Reality, 9, 43, 85, 86, 102, 103, 110, 111, 119, 121, 127, 132, 148, 149, 151, 171, 176, 179, 181, 191-193, 208,212-215,217 Reasoning and the Logic of Things (C.S. Pierce), 84, 181, 198 Reductionism a posteriori reductionism, 66 analytical reductionism, 65, 66 Reid, T., 34 Relata, 1,4, 19, 32, 34, 35, 38,64, 6 7 , 6 8 , 7 1 , 7 3 , 171-176, 178, 179, 181, 190, 191, 214 Representamen, 136 Rosenthal, S„ 101, 103, 127-129, 131 Russell, B., xiii Sachsse, H„ 19 Salmon, W.C.,71 Saussure, F. de, 134 Scholastic realism, 101, 116 Scriven, M., xiii, 50, 52, 53, 66, 67, 176 Secondness, 196, 201, 202, 215 Semeiosis, xviii, xix, xx, 132-134, 137-149, 151-154, 157-161, 163166, 195-199, 203, 205-207, 209211,215,216 Semeiotic, xiv, xviii-xx, 132-137, 139, 140, 142, 143, 147-149, 151153, 158, 159, 161-167, 180, 181,
252 195-199, 203, 206, 207, 209-211, 215,216 Short, T.L., 82, 87, 88, 89, 133, 134, 139-144, 147, 157, 160, 161, 163165 Sign definition of sign, 136 dynamic object (of a sign), 134, 137, 146-149, 151-153, 157160, 164, 165 immediate object (of a sign), 137, 152 object of a sign, 137, 151,205 real object (of a sign), 137, 148, 149, 152 sign and object, 135, 140, 151, 153-155, 158, 195 sign-action, 133, 137, 144 signs as media for the transmission of forms, 203 triadic sign-structure, 159 Significance, ground of, 140 Sober, E„ 100 Sosa, E., 48, 53, 59, 60, 64, 65, 73 Spatio-temporal continuity, 71, 74 Spinoza, B., 16, 21-24, 27, 33, 42, 43,201 Spontaneity, 26, 27, 85, 146, 184, 186 Stalnaker, R.C., 54 Stoics, 2, 5, 6, 42, 134 Subjunctive conditionals, 54, 102 Substance Cartesian concept of matter as extended substance, 25 dismissal of substantial forms, 17 substance ontology, xix, 4, 5, 15, 17, 177-179, 187,210,211,214 Sufficiency principle of sufficient reason, 24, 27 sufficiency thesis, 50-52
INDEX sufficient conditions, 34, 47, 48, 50-54, 56,58,71-73, 120, 132, 168, 169, 172, 176 undeterminative sufficiency, 51, 52 Sufficient reason, principle of, 24, 27 Summa Theologiae, 10 Suppes, P., 59 Symbol, 135, 150, 153-157, 163, 165 Symbol, purpose of a, 157 Symmetry, 176 Synechism, 82, 83, 87 Taube, M., 20, 29 Taylor, R., xiii, 51, 66, 72 TDE-characters (teleologically determined empirical characters), 113, 116, 119, 120 Teleology Mayr's theory of teleology, xvii Mayr's theory of teleology, 88, 95 Peirce's conception of teleology, 2, 76, 80, 213 problem of teleology, 1, 75, 88, 95, 180, 181,212,217 teleological explanations, 75, 76, 82, 89 teleological language, 75, 88 teleological principle, 26 teleology as creative, 85 Teleonomic explanations, 94 Telos, 3 Tendency, 24, 42, 77-79, 82, 88, 89, 97, 109, 112, 113, 117, 125, 131, 132, 145, 156-159, 164, 170, 176, 179, 185, 187, 194, 197,201,202, 212,214,215 Thirdness, 120, 196, 198, 199, 201, 202, 205,211,215 Time, objective order in, 38 Tooley, M., 53, 59, 60, 64, 65, 67, 73, 212
253
INDEX Trout, J.D., 99
Von Wright, G.H., 57, 72
Uniformity of nature, 105, 106 Universals, 57, 100, 110, 111, 121, 131
Wallace, W., 17, 29 Welby, V., 138, 141 Whitehead, A.N., 68, 171, 173-176, 179,201,204, 208,211,214 Wittgenstein, L.J.J., 178 Would-be, 209,210
Valency, 124, 125 Van Brakel, J., 100