Purdue University
Purdue e-Pubs
Open Access Dissertations
Theses and Dissertations
8-2016
Leibniz on intra-substantial causation and change
Davis White Kuykendall Jr.
Purdue University
Follow this and additional works at: https://docs.lib.purdue.edu/open_access_dissertations
Part of the Philosophy Commons
Recommended Citation
Kuykendall, Davis White Jr., "Leibniz on intra-substantial causation and change" (2016). Open Access Dissertations. 790.
https://docs.lib.purdue.edu/open_access_dissertations/790
This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact epubs@purdue.edu for
additional information.
Graduate School Form
30 Updated 12/26/2015
PURDUE UNIVERSITY
GRADUATE SCHOOL
Thesis/Dissertation Acceptance
This is to certify that the thesis/dissertation prepared
By Davis White Kuykendall
Entitled
Leibniz on Intra-Substantial Causation and Change
For the degree of
Doctor of Philosophy
Is approved by the final examining committee:
Jan Cover
Chair
Jeffrey Brower
Michael Jacovides
Dan Frank
To the best of my knowledge and as understood by the student in the Thesis/Dissertation
Agreement, Publication Delay, and Certification Disclaimer (Graduate School Form 32),
this thesis/dissertation adheres to the provisions of Purdue University’s “Policy of
Integrity in Research” and the use of copyright material.
Approved by Major Professor(s):
Approved by:
Jan Cover
Matthias Steup
Head of the Departmental Graduate Program
28 June 2016
Date
i
LEIBNIZ ON INTRA-SUBSTANTIAL CAUSATION AND CHANGE
A Dissertation
Submitted to the Faculty
of
Purdue University
by
Davis White Kuykendall Jr.
In Partial Fulfillment of the
Requirements for the Degree
of
Doctor of Philosophy
August 2016
Purdue University
West Lafayette, Indiana
ii
For Lauren.
iii
ACKNOWLEDGEMENTS
I would first of all like to thank my dissertation director Jan Cover. This dissertation
would be non-actual were it not for the many dozens of hours in the summer of 2012 he
spent patiently brainstorming with me in his office. I always left his office with a stack
of books to read, a word of encouragement, and a spot in his calendar to meet again. I
also would like to thank committee members Jeff Brower and Mike Jacovides for their
guidance. During many seminars they taught, works of theirs I read, and one-onguidance they gave me during my graduate studies, I learned a lot about how to approach
the history of philosophy. I would also like to thank my fourth reader, Dan Frank, for his
encouragement and feedback during the last stages of my project.
I would like to thank my parents for their encouragement at every stage of my
graduate studies.
Lastly, I would like to thank my wife, Lauren Kuykendall. I met her at Purdue in
my first year of graduate studies and I don’t want to imagine what the remaining five
would have been like without her.
iv
TABLE OF CONTENTS
Page
ABBREVIATIONS ........................................................................................................... vi
ABSTRACT ...................................................................................................................... ix
CHAPTER ONE: INTRODUCTION................................................................................. 1
1. Efficient Causation and Change.............................................................................. 2
2. Leibniz on Efficient Causation and Change ........................................................... 6
3. Dissertation Roadmap............................................................................................. 9
4. The Idealistic Interpretation of Leibniz ................................................................ 18
5. Methodological Approaches ................................................................................. 23
CHAPTER TWO: LEIBNIZ’S TRANSFERENCE ARGUMENT AGAINST
CREATURELY TRANSEUNT CAUSATION, PART 1: THE TRANSFERENCE
CONDITION .................................................................................................................... 27
1. The Transference Condition ................................................................................. 29
2. Leibniz’s Defense of the Transference Condition ................................................ 35
2.1 Philosophers who denied the Transference Condition.................................... 36
2.2 Was Leibniz arguing against a straw man?..................................................... 38
2.3 Divine transeunt causation and transference................................................... 41
2.4 The argument for the Transference Condition................................................ 42
2.5 The argument for the Transference Condition is a Leibnizian Argument ...... 57
3. A Weakened Transference Condition................................................................... 65
3.1 The act/potency Distinction and the transference condition........................... 66
3.2 The transference condition and modifications................................................ 69
CHAPTER THREE: LEIBNIZ’S TRANSFERENCE ARGUMENT, PART 2: AGAINST
TRANSFERENCE............................................................................................................ 82
1. Leibniz’s Substance-Accident Ontology .............................................................. 83
2. Why nothing in Leibniz’s Ontology can be Transferred ...................................... 86
2.1 Why accidents cannot be transferred .............................................................. 86
2.2 Why substances cannot be transferred............................................................ 96
CHAPTER FOUR: LEIBNIZIAN SIMPLE SUBSTANCES AND THE REALITY OF
ACCIDENTS .................................................................................................................. 101
1. Leibniz’s “De Realitate Accidentium” .............................................................. 103
2. Leibniz’s Changing Stance on the Merely-Modal-Reality thesis ....................... 112
2.1 Different Types of Realities.......................................................................... 114
2.2 The Arguments of DRA, Absolute-Realities and Limited-Realities ............ 116
2.3 Leibniz’s Mereology and Accidents ............................................................. 118
2.3.1 Shapes not Homogenous with what they Shape ............................... 131
2.3.2 Limitations are not Homogenous with what they Limit ................... 135
v
Page
2.3.3 Absolute-Realities and Homogeneity ............................................... 137
CHAPTER FIVE: DETERMINISTIC AND INTELLIGIBLE LEIBNIZIAN
SUBSTANCE CAUSATION ......................................................................................... 141
1. Leibniz’s Prima Facie Inconsistent Views on the Efficient Cause of Accidents 144
1.1 Creaturely Spontaneity.................................................................................. 144
1.2 The Efficacious-Substance Interpretation..................................................... 145
1.3 The Efficacious-Accident Interpretation....................................................... 146
1.4 Why not both Efficacious-substance and Efficacious-Accidents ................. 149
1.5 Against the Efficacious-Perception Interpretation........................................ 152
2. The Argument for the Efficacious-Appetition Interpretation ............................. 154
2.1 Rutherford on the Efficacious-substance Interpretation, Determinism and
Intelligibility ....................................................................................................... 155
2.2 Why the Law-of-the-Series does not help the Efficacious-Substance
Interpretation....................................................................................................... 157
2.3 Rutherford on the Efficacious-Appetitions, Determinism, and Intelligibility
............................................................................................................................. 159
3. Leibnizian Substance Causation ....................................................................... 162
3.1 Suarez on the Efficient Principle Cause Quod and Quo ............................... 163
3.2 Why the distinction applies to Leibniz ......................................................... 165
3.2.1 Appetitions are powers .............................................................................. 165
3.2.2 Substances and Appetitions are Principles of Change............................... 166
3.2.3 Substances are Principle Quod Causes and the Appetitions are Principle Quo
Causes ........................................................................................................ 168
3.3 The Efficacious-Substance Interpretation and Determinism, Intelligibility,
and PSR........................................................................................................ 172
3.3.1 Scholastic Substance Causation and Determinism ................................... 173
3.3.2 Deterministic Leibnizian Substance Causation ........................................ 178
3.3.3 The necessity involved in Leibnizian Substance Causation ..................... 181
APPENDICES
Appendix A. Leibniz’s Missing Overdetermination Premise........................................ 184
Appendix B. Leibniz on Transubstantiation .................................................................. 201
Appendix C. Translation of “De Realitate Accidentium” ............................................ 209
BIBLIOGRAPHY........................................................................................................... 213
VITA ............................................................................................................................... 222
vi
ABBREVIATIONS
Leibniz
A
Sämtliche Schriften und Briefe. Darmstadt and Berlin: Berlin Academy, 1923-1.
Cited by series, volume, and page.
AG
G.W. Leibniz: Philosophical Essays. Translated and edited by Roger Ariew
& Daniel Garber. Indianapolis: Hackett Publishing Co., 1989. Cited by
page.
C
Opuscules et fragments inédits de Leibniz. Edited by Louis Couturat. Paris: Felix
Alcan, 1903. Reprint, Hildesheim: Georg Olms, 1966. Cited by page.
DSR
De Summa Rerum: Metaphysical Papers, 1675-1676. Translated by G.H.R.
Parkinson. New Haven: Yale University Press, 1992. Cited by page.
G
Die philosophischen Schriften von Gottfried Wilhelm Leibniz. Edited by C.I.
Gerhardt. Berlin: Weidman, 1875-90. Reprint, Hildesheim: Georg Olms, 1965.
Cited by volume and page.
Grua
Textes inédits. Edited by Gaston Grua. Paris: Presses Universitaires de France,
1948. Cited by page.
L
Gottfried Wilhelm Leibniz: Philosophical Papers and Letters, 2nd ed.
Translated and edited by Leroy E. Loemker. Dordrecht-Holland: D. Reidel
Publishing Co., 1969. Cited by page.
NE
New Essays on Human Understanding. Translated and edited by Peter Remnant
and Jonathan Bennett. Cambridge: Cambridge University Press, 1996. Cited by
page.
PW
Philosophical Writings. Edited and translated by Mary Morris & G.H.R.
Parkinson. London: Dent (Everyman’s Library), 1995. Cited by page.
T
Theodicy. Translated by E.M. Huggard. Yale University Press, 1952.
Reprint, LaSalle, Il: Open Court, 1985. Cited by page.
vii
WF
G.W. Leibniz: Philosophical Texts. Translated and edited by F.S. Woolhouse and
Richard Francks. Oxford: Oxford University Press, 1998. Cited by page.
Other Figures
CSM Rene Descartes. The Philosophical Writings of Descartes, vols. I & II.
Translated and edited by John Cottingham, Robert Stoothoff, and Dugald
Murdoch. Cambridge: Cambridge University Press, 1985. Cited by page
number.
CSMK ____________. The Philosophical Writings of Descartes, Volume III. Translated
and edited by John Cottingham, Robert Stoothoff, Dugald Murdoch, and Anthony
Kenny. Cambridge: Cambridge University Press, 1991. Cited by page number.
DM
Francisco Suarez. On Efficient Causality: Metaphysical Disputations 17, 18, and
19. Translated and edited by Alfred J. Freddoso. New Haven: Yale University
Press, 1994; __________. On Creation, Conservation, and Concurrence:
Metaphysical Disputations 20-22. Translated and edited by Alfred J. Freddoso.
South Bend: St. Augustine’s Press, 2009; ____________. On Various Kinds of
Distinctions. Translated by Cyril Vollert, SJ. Marquette: Marquette University
Press, 1947, Cited by Disputation, section, and page number; and____________.
Disputationes metaphysicae. Vols. 25 and 26 of Opera Omnia. Edited by Charles
Berton. Paris: Viv`es, 1856–61. Reprinted Hildesheim: Georg Olms, 1965.
E
Baruch Spinoza. Ethics. Translated by Shirley Samuel Indianapolis: Hackett
Publishing Company, 1992. Cited by Part.
LO
Nicolas Malebranche. The Search after Truth: With Elucidations of The Search
after Truth. Edited by Thomas M. Lennon and Paul J. Oscamp. Cambridge:
Cambridge University Press, 1997. Cited by page number.
SCG
Thomas Aquinas. Summa Contra Gentiles. Translated by Anton Pegis. Notre
Dame: Notre Dame University Press, 1975. Cited by book, chapter, and section.
DPN ___________. Aquinas on Matter and Form and the Elements: A Translation of
the De Principiis Naturae and the De Mixtione Elementorum of St. Thomas
Aquinas. Translated by Joseph Bobik. Notre Dame: Notre Dame University
Press, 1998. Cited by chapter and section.
ST
___________. Basic Writings of Saint Thomas Aquinas: Volume One. Translated
and edited by Anton Pegis. Indianapolis: Hackett Publishing Company, Inc.,
1945. Cited by Book, question, and article.
QDP ___________. On the Power of God. Translated by the English Dominican
Fathers. Westminster: The Newman Press, 1952. Cited by question and article.
viii
QDV ___________. Truth. Translated by James V. McGlynn, Robert W. Mulligan, &
Robert W. Schmidt. Chicago: Henry Regnery Company, 1954. Cited by book,
question, and article.
RB
Roger Bacon. Roger Bacon’s Philosophy of Nature. Edited and translated by D.
Lindberg. Oxford: Oxford University Press, 1983. Cited by page number.
ix
ABSTRACT
Kuykendall, Davis W. Ph.D. Purdue University, August 2016. Leibniz on Intrasubstantial Causation and Change. Major Professor: Jan Cover.
Leibniz argued that in natural world, only intra-substantial or immanent causation is
possible— the causation that takes place within an individual, when an individual brings
about a change in itself. In this dissertation, I address issues arising from Leibniz’s
arguments against the rival view that posits a world of causally interacting substances and
issues pertaining to Leibniz’s own positive metaphysics of immanent causation and
change.
Chapter 1 is devoted to stage setting for the remainder of the dissertation. I first
offer a historically informed overview of efficient causation and change before
introducing Leibniz’s novel views, including his criticisms of competing accounts and his
own positive account. After presenting a detailed roadmap of my project, I articulate the
idealistic interpretation of Leibniz assumed in this dissertation, where the only genuine
substances are simple monads. Finally, I articulate the methodological approaches I
employ.
In Chapters 2 and 3, I reconstruct and assess Leibniz’s most frequent argument
against transeunt causation (the causation that occurs when a substance produces a
property in a numerically distinct substance), what I call the “Transference Argument.”
Leibniz argued that a created substance’s causing an accident in a numerically distinct
x
substance is possible only if the agent (the cause) transfers the accident from itself to the
patient (the recipient of the effect), where upon transference the agent no longer
possesses the accident it transfered. Call the transeunt causal requirement that the agent
transfer the accident produced from itself to the patient the “Transference Condition.”
Chapter 2 is devoted to two problems with Leibniz’s transference condition.
First, Leibniz stated the transference condition throughout his career, but offered little
argument for it. Second, God is a transeunt cause in Leibniz’s metaphysics yet God’s
causation does not consist in transference. Thus, Leibniz needs a principled way to
require transference for creaturely causation while denying that divine transeunt
causation consists in transference. I shall argue that Leibniz thought that if an agent
transeuntly caused an accident without transferring the accident, the agent created the
accident. For the recipient substance contributed no reality to the accident and the agent
lost no reality in causing the accident. However, only an omnipotent being—God—can
create. Therefore, only God can transeuntly cause without transferring what is caused.
Finally, I close off chapter 2 by drawing attention to an important weakness with the
transference condition that has not yet been recognized by Leibniz scholars. Based on
arguments Leibniz develops against occasionalism in his Theodicy concerning the
production of modifications, I shall argue that Leibniz ultimately only had reasons to
require transference for the transeunt production of non-modal accidents, such as real
qualities.
In Chapter 3, I argue that there is nothing in Leibniz’s ontology that could be
transferred from the cause to the recipient of the effect. I first argue that
xi
Leibniz’s ontology consists of simple non-corporeal substances and their modifications.
Second, I present and articulate a number of important theses Leibniz affirmed about
substances and their modifications, which entail that neither could be transferred. I also
show that most of these theses were not unique to Leibniz, but were in fact widely
endorsed by his predecessors who defended the possibility of creaturely transeunt
causation.
In chapter 4, I continue the study of the nature of Leibnizian accidents, shifting
the focus from their role in Leibniz’s critique of creaturely transeunt causation to their
positive role in change and as causal relata, where such accidents are the effects of
immanent causation. Specifically, I reconstruct and assess Leibniz’s reasons for holding
that accidents are modifications or limitations. Drawing from Leibniz’s 1688 essay “De
Realitate Accidentium” and his later mereological writings, I shall argue that Leibniz’s
thesis that accidents are modifications or limitations allowed him to posit mereologically
simple substance that have a multitude of accidents at a time and change accidents over
time.
In Chapter 5, I address an issue that has divided Leibniz scholars concerning the
precise relata in Leibnizian immanent efficient causation. In many passages, Leibniz
writes as if it is the substance or individual itself that efficiently causes its later properties
or accidents. Call this the “Efficacious-substance” account. The efficacious-substance
account is difficult to reconcile with Leibniz’s requirements that change be intelligible
and deterministic. In plenty of other passages, he writes as if it is the substances earlier
properties or accidents that cause its later accidents. Call this the “Efficacious-accident”
account. The efficacious-accident account explains how change is intelligible and
xii
deterministic but it faces a “plurality of agents” objection. If a substance’s accidents are
the efficient causes of later accidents, then prima facie there is a plurality of efficient
causal agents in a substance. This view is incompatible with Leibniz’s requirement that
substances be simple, unified entities.
Drawing upon a Scholastic distinction made between two kinds of efficient
causes— principle quod efficient causes (efficient causal agents) and principle quo
efficient causes (powers by which agents cause), I shall argue that for Leibniz, substances
are principle quod efficient causes and their appetitions (desire-like accidents that are a
subset of a substance’s accidents) are principle quo efficient causes. This interpretation
combines the strengths of the Efficacious-substance and Efficacious-accident accounts
while overcoming their weaknesses. There is just one causal agent, the substance, but
change is both intelligible and deterministic because as what an agent produces is
explained by its appetitions.
1
CHAPTER 1. INTRODUCTION
In the 17th century, causation took center stage as one of the most debated topics by
philosophers who were increasingly forced to rethink natural philosophy given the
challenges the scientific revolution posed to the Aristotelian-Scholasticism, which still
dominated university curricula.1 One of the most important disputants was the German
philosopher Gottfried Wilhelm Leibniz (1646-1716), whose views on causation and
change are the subject of my dissertation.2 Leibniz argued that at the fundamental level
of the natural world, only intra-substantial or immanent causation occurred— the
causation that takes place within an individual, when an individual brings about a change
in itself.3 Leibniz argued that his counter-intuitive theory overcame serious defects that
plagued the other dominant causal theories of his time, while also providing a
1
For helpful overviews of the causation debate in the late middle ages and early modern era, see Kenneth
Clatterbaugh, The Causation Debate in Modern Philosophy (New York: Routledge, 1999), Steven Nadler
ed. Causation in Early Modern Philosophy (University Park: The Pennsylvania State University Press,
1993), and Walter Ott, Causation and Laws of Nature in Early Modern Philosophy (Oxford: Oxford
University Press, 2009).
2
For general overviews of Leibniz’s life and thought, see Nicholas Jolley, ed., The Cambridge Companion
to Leibniz, (Cambridge: Cambridge University Press, 1995); Brandon C. Look, "Gottfried Wilhelm
Leibniz", The Stanford Encyclopedia of Philosophy (Spring 2014 Edition), Edward N. Zalta (ed.), URL =
http://plato.stanford.edu/archives/spr2014/entries/leibniz/, The Continuum Companion to Leibniz, (London:
Continuum, 2011); and Benson Mates, The Philosophy of Leibniz: Metaphysics and Language, (New York:
Oxford University Press, 1986).
3
A classic statement of this thesis is found in Leibniz’s Monadology, where he writes, “It follows from
what we have just said that the monad’s [individual’s] natural changes come from an internal principle,
since no external cause can influence it internally.” See (G VI.608: AG 214). See also AG 33, (G IV.439:
AG 47), (G VI.351-52: T 396), and T 400 (G V.353-54).
2
philosophical underpinning for the increasingly successful enlightenment physical
theories.
In the first half of this dissertation, I address issues arising from Leibniz’s
criticisms of what I’ll call the Traditional Causal View of his time— the view that created
substances genuinely causally interact. In the second half, I address issues pertaining to
his positive account of change and creaturely immanent efficient causation. In this
introductory chapter, I present a historically informed overview of efficient causation and
change in §1. In §2, I segue to an overview of Leibniz’s own distinctive views on
causation and change before presenting a detailed roadmap of my project in §3. In §4, I
articulate the idealistic interpretation of Leibniz’s metaphysics that I assume in this
dissertation. Finally, I discuss the methodological approaches I employ in this
dissertation in §5.
§1 Efficient Causation and Change
In section 1, I draw attention to some important but mostly non-controversial features of
what I mean by efficient causation and change, which will come from a brief historical
overview. Giving such an account might strike the reader as unnecessary, as efficient
causation is the type of causation that is the most familiar to present day philosophers. In
fact, since the early modern era, it has largely been the only type of causation considered.
However, as will become apparent in this project, the nature of efficient causation has
historically been one of the most controversial topics in metaphysics. I note that my aim
in this overview is not to precisely define efficient causation. Instead, I highlight some
3
important features found in some of the most influential accounts of efficient causation
prior to Leibniz, and which are also found in Leibniz’s own account.
While efficient causation—of some sort—played a role in metaphysics prior to
Aristotle, it is appropriate to start with Aristotle’s account, as his influence will loom
large in what follows.4 According to Aristotle, all causes are principles of change. As
principles of change, causes explain change.5 Thus, the efficient cause of some change is
also a principle and therefore explainer of that change. Specifically, in some change, the
efficient cause is the origin or source of the change.6 Aristotle’s famous example is the
coming-to-be of a statue.7 Take some clay that has been molded into a statue with the
shape of Socrates. The clay is the statue’s material cause, the shape is the statue’s formal
cause, and to-be-admired could be the statue’s final cause. The efficient cause of the
statue is the sculptor, who molds the clay into Socrates shape. In this scenario, the
sculptor efficiently causes the statue by giving or creating a new form in the clay— the
shape of Socrates. The clay acquires a new property or more appropriate to the
metaphysics of Leibniz’s time, an accident— the accident of a particular shape. What’s
key here is that the effect produced by the efficient cause is, in some sense, a new being
or entity— such as a new shape in the clay.8
4
For a helpful and in depth overview of Aristotle’s account of efficient causation, see Thomas M. Tuozzo,
“Aristotle and the Discovery of Efficient Causation,” in Tad M. Schmaltz, ed., Efficient Causation: A
History (Oxford: Oxford University Press, 2014).
5
See Aristotle, Physics, II.3, 194b16-24. For Aristotle’s definition of a principle, see Metaphysics, V.1,
1012b33-1013a23.
6
See Physics, II.3, 194b30-32 and Metaphysics, I.3, 984a27.
7
See Physics II.3 and Metaphysics V.2.
8
In just what sense it is a new being or entity which is produced was a matter of great controversy which I
address in significant portions of chapters 2-4.
4
We can build on Aristotle’s account by turning to Aquinas, as Aquinas’s
influential account is heavily informed by and influenced by Aristotle’s. According to
Aquinas, the efficient cause—such as the sculptor—is the principle that acts.9 Another
way Aquinas puts it is that the efficient cause is a cause insofar as it acts.10 This,
according to Aquinas, distinguishes the efficient cause from the material, final, and
formal cause. The efficient cause, as the cause that acts, is the causal agent.11
Since the efficient cause is one of the principles and explainers of change, it’s also
worth briefly discussing what I mean by change in this section. As with efficient
causation, my aim is not to offer a precise definition or metaphysics of change. Instead
it’s to draw attention to some important features of it that can serve as a launch pad for
this project. By change, I mean an individual’s acquiring and losing properties. For
example, the clay changes when it acquires the property of Socrates’ shape and loses its
previous shape.12
So the efficient cause is to be understood as a principle and explainer of change,
specifically the cause from which the change or effect originates, as the efficient cause is
the agent that acts and by acting produces a new form or property or accident in an
individual, which the individual acquires. Once one probes further into these concepts
and inquires into just what the new beings are that are produced, what it is precisely that
does the producing, and how the new beings produced are related to the individual they
9
See Aquinas, On the Principles of Nature 3.15, DPN 1.4, In Meta I.12.199,, V.2.775, and QDV q28, a8c.
Leibniz is in agreement with Aquinas, arguing that the efficient cause is the “active cause” or the “cause
through action.” See C 472 and A.VI.ii.490.
11
See On the Principles of Nature 3.15. See also Aristotle, Metaphysics, VII.7, 1032a13-14.
12
I explore Leibniz’s precise definition of change in much greater depth in chapter 5.
10
5
are beings in, one enters into more difficult terrain— terrain that I’ll be in throughout this
project.13
Philosophers before and during Leibniz’s era further distinguished between two
types of efficient causes— transeunt and immanent causes.14 An individual is a transeunt
cause when it brings about a change or produces a property in a numerically distinct
individual. For example, when a particle p1 collides with a different particle p2, p1 is the
transeunt cause of p2’s change in velocity. In contrast, an individual is an immanent cause
when it produces a property in itself. While less discussed in studies of causation, we are
more intimately acquainted with this second kind, as many examples come from human
action. When a person moves her hand, she is the immanent cause of her hand moving.
When a person imagines a cloud, she is the immanent cause of the mental image formed.
It is with this distinction between transeunt and immanent causation that Leibniz’s
views on causation and change merit attention. The central early modern philosophical
debate about causation was whether (i) both immanent and transeunt creaturely causation
are possible; (ii) only one of the two is possible; or (iii) neither are possible. AristotelianScholastics, such as the late medieval/renaissance philosopher Francisco Suarez and
notable early modern philosophers such as Rene Descartes argued for (i), what I’ll call
13
While the focus of my study is creaturely or secondary efficient causation, I note that the case is different
with divine efficient causation as the effect is not always simply a form or accident in some pre-existent
substrata, such as prime matter or a substance, but instead the whole substance in cases of creation, a type
of efficient causality exclusive to God alone, according to most classical theists.
14
The distinction traces back to Aristotle but finds more detailed expression in Medieval philosophers. For
example, see Aquinas, SCG II.1 and ST 1a, q. 18, a. 3 ad 1.
6
the “Traditional” view.15 Second-generation Cartesians, such as Malebranche, La Forge,
and Cordemoy, challenged (i) and argued for (iii), the view known as Occasionalism.16
Occasionalists argued that God is the only real cause of change and any creaturely
causation so-called was merely apparent. First defended centuries earlier by the medieval
Muslim philosopher Al Ghazali, the early modern occasionalists revived the theory with
novel and powerful arguments.17
§2 Leibniz’s on Efficient Causation and Change
Leibniz entered into this debate by defending a unique and strikingly counterintuitive
option, which he rigorously defended throughout his career. On the one hand, with the
Aristotelians and contra the Occasionalists, Leibniz argued that the very essence of
substances consists in their being causally efficacious.18 With the Aristotelians and
Descartes, Leibniz also affirmed the fundamentality of immanent causation, again, contra
the Occasionalists.
On the other hand, Leibniz rejected the Traditional view that posited a world
populated by causally interactive created substances. Instead, Leibniz defended a world
of spontaneous substances. A substance is spontaneous when it is solely causally
15
See Suarez, DM 18.1; Aquinas, SCG III.I.69, 28. Descartes’ views are more complicated, as scholars
debate whether he thought one body could transeuntly cause a change in a different body. However,
Descartes did argue that mental substance could transeunt cause changes in extended substance. The best
resource for Descartes’ more complicated causal views is Tad M. Schmaltz, Descartes on Causation,
(Oxford: Oxford University Press, 2008).
16
For a helpful overview of occasionalism, see Sukjae Lee, "Occasionalism", The Stanford Encyclopedia of
Philosophy (Spring 2014 Edition), Edward N. Zalta (ed.), URL =
<http://plato.stanford.edu/archives/spr2014/entries/occasionalism/>.
17
In this first part of my study where I examine Leibniz’s criticisms of competing causal theories, I am
primarily concerned with Leibniz’s response to the Traditional view.
18
See (G VI.608: AG 214), AG 33, (G IV.439: AG 47), (G VI.351-52: T 396), and (G V.353-54: T 400).
7
responsible for all of its natural changes and properties.19 In this world, immanent
efficient causation is fundamental while transeunt causation is merely phenomenal and
reducible in some sense to the immanent causal activity of individuals. 20 So Leibniz
defended position (ii) against (i) and (iii). The spontaneity of substances and the
fundamentality of immanent rather than transeunt causation are foundational theses of
Leibniz’s metaphysics, which Leibniz also utilized in his philosophical theology. He also
argued that these counterintuitive causal views provided a powerful philosophical
underpinning for the increasingly successful enlightenment physics.
Leibniz’s views are challenging in at least three ways, which I take up in my
dissertation. First, Leibniz rigorously defends some aspects of his criticisms of competing
causal theories, while he passes over other crucial premises without pausing to develop
them. One explanation for this is that the reasoning would have been obvious to his 17th
19
I use the term “natural” because Leibniz allows for the possibility of miracles, such as a case where God
is solely responsible for some of a created substance’s accidents. Additionally, Leibniz was a concurrentist
who, with other concurrentists, argued that God’s causal input is required for the production of even nonmiraculuous accidents. For a treatment of Leibniz’s theory of divine concurrence, see See Timothy Allan
Hillman, “Leibniz on Monadic Action & Divine Concurrence,” (PhD Diss., Purdue University, 2008).
20
The reduction of transeunt causal activity to immanent causal activity actually occurs in multiple levels
in Leibniz’s metaphysics. First, it is a crucial feature of Leibniz’s theory of pre-established harmony. On
the theory of pre-established harmony, all non-initial properties of created individuals are immanently
caused by such individuals. However, God created the world in such a way that all the immanently caused
properties are coordinated or harmonious. Second, corporeal individuals—such as organisms and the
particles studied by impact mechanics—are reducible to individual simple non-extended individuals or
substances (what Leibniz calls “monads”) and their accidents. Thus, any transeunt causal activity between
two bodies is reducible to the immanent causal activity of monads. Third, immanent causal activity is
prominent when focusing solely on impact mechanics and Leibniz’s science of dynamics. Leibniz argued
that when two bodies b1 and b2 collide, rather than b1 causing b2’s change in velocity, b2’s change in
velocity is caused by the elastic nature of the particles composing b2. The focus on my dissertation, I note,
is not the reduction of transeunt causal activity to immanent causal activity. Instead, it is the metaphysics
of immanent causation as such— what happens when an individual causes a property in itself. A coherent
account of immanent causation as such is a necessary condition of Leibniz’s reduction of transeunt
causation to immanent causation.
8
century peers, even though the reasoning is lost on us. I dredge out the missing support
for some of the key premises that have baffled scholars.21
Second, Leibniz never stops to rigorously and systematically develop at length his
own positive account of what happens when an individual produces a change in itself.22
This absence leaves many unanswered questions. A task confronting scholars—as yet
undone—is to reconstruct his positive account from both his many scattered criticisms of
alternative causal theories and his equally scattered remarks of the positive elements of
causation.23 In the second half of my dissertation, I contribute to such an eventual
positive account of a substance’s producing a change in itself by carefully examining two
issues: how simple substances can have a multitude of accidents at a time and over time
21
I describe in greater detail the missing premises and how one might fairly go about supplying them in
Chapters 2-5.
22
One explanation for why Leibniz never produces a lengthy, rigorous, systematic treatise on his positive
account of immanent causation is that most of his writings were letters to various philosophers, scientists,
and theologians he dialogued with. Hence, his remarks on causation are scattered, written in response to
specific concerns raised by disputants, usually using the technical vocabulary of his interlocutors instead of
using Leibniz’s own carefully worked out terminology.
23
While very little work addresses creaturely immanent efficient causation, specifically the metaphysics of
what happens when a created individual causes a property in itself, there has been flurry of work on (i)
Leibniz’s account of divine efficient causation, some notable work on (ii) creaturely final causation and (iii)
Leibniz’s theory of pre-established harmony— Leibniz’s account of how God coordinates the immanent
causal activity of all created individuals. For examples of (i), see: J. Von Bodelschwingh, “Leibniz on
Concurrence, Spontaneity, and Authorship,” Modern Schoolman, 88(2011): 267–297; Marc Bobro.
“Leibniz on Concurrence and Efficient Causation,” Southern Journal of Philosophy, 46 (2008): 317–38;
Nicholas Jolley, “Causality and Creation in Leibniz,” The Monist, 81(1998): 591–611; Sukjae Lee,
“Leibniz on Divine Concurrence,” Philosophical Review, 113(2004): 203–48. For examples of (ii), see
Lawrence Carlin, “Leibniz on Final Causes,” Journal of the History of Philosophy, 44(2006): 217–33 and
Jeffrey K. McDonough, “Leibniz’s Two Realms Revisited,” Nous 42(2008): 673-696. For examples of
(iii), see H. Ishiguro, “Pre-established harmony versus constant conjunction,” Proceedings of the British
Academy, 63(1977): 239–63. Mark Kulstad, “Causation and Preestablished Harmony in the Early
Development of Leibniz's Philosophy,” Causation in Early Modern Philosophy, Steven Nadler (ed.),
(University Park: Pennsylvania State Press, 1993), 93–118.
9
and how substances can be efficient causal agents that deterministically and immanently
efficiently cause their accidents.
Finally, Leibniz’s causal views are relevant to contemporary theories of causation
in at least three respects. First, Leibniz offers novel reasons both for and against
persistence theories of causation.24 Second, and related to the first, his views also point
to some consequences of persistence theories of causation, which, while not entailing
their truth or falsity, are important.25 Mainly, persistence theories of causation are
inconsistent with causal overdetermination. Third, Leibniz, I mentioned above and shall
argue in greater depth, has principled reasons to wed substance causation with
determinism—two metaphysical views typically viewed as at odds with each other.26
§3 Dissertation Roadmap
With the broad overview of efficient causation, change, and Leibniz’s distinctive views
on both, I now turn to the specific issues that arise his metaphysics of change and
causation, which will be the subject of my dissertation. In Chapter 2, I address Leibniz’s
criticisms of the dominant version of the Traditional view of his time that defended the
24
A persistence theory of causation holds that causation consists in the persistence of an entity from the
cause to the effect. Persistence theories are a family, with each theory distinguished by what it holds to
persist, e.g., energy of momentum, mass-energy, tropes, or properties. See Hector-Neri Castaneda,
“Causes, Causity, and Energy,” Midwest Studies in Philosophy 9 (1984): 17-27; Douglas Ehring, Causation
and Persistence: A Theory of Causation (Oxford: Oxford University Press, 1997); David Fair, “Causation
and the Flow of Energy,” Erkenntnis 14 (1979): 210-250; S.D. Rieber, “Causation as Property
Acquisition,” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition 109,
No. 1 (2002): 53-74; and Wesley Salmon, “Causality Without Counterfactuals,” Philosophy of Science 61
(1994): 297-312.
25
In Appendix A, I argue that one sort of persistence theory of causation which Leibniz attributes to the
Traditional view of efficient causation is inconsistent with causal overdetermination.
26
I address this in Chapter 5 and give a more detailed summary of this issue in the dissertation roadmap
below.
10
fundamentality of natural or creaturely transeunt causation— Physical Influx. Physical
influx was endorsed by figures of wide-ranging views, such as the renaissance
Aristotelian-Scholastic philosopher Francisco Suarez and enlightenment philosophers
such as Rene Descartes.27 Physical influx’s central tenant was that transeunt causation
consisted in the communication of the effect from the cause to the recipient of the
effect.28 Physical Influx was utilized to explain a wide variety of change, including
perception and impact mechanics. In perception, the perceived object communicates a
likeness or representation—what Scholastics called a “species”—of itself to the
percipient. In impact mechanics, one particle changes the velocity of another by
communicating its motion. On Leibniz’s understanding of Physical Influx, which is
strikingly similar to many contemporary persistence theories of causation, the agent
substance’s (cause) communication of the effect consisted in the accident caused by the
agent first detaching itself from the agent and being sent to the recipient of effect (the
patient).29 Call this literal detachment of the accident from the agent and its being sent to
the patient ‘Transference’, where upon transference, the agent no longer possesses the
accident it causes.
Leibniz’s criticisms of the fundamentality of creaturely transeunt causation then
consisted of two claims:
27
For an overview of Leibniz’s understanding of Physical Influx, see See Eileen O’Neill,
“Influxus Physicus,” in Steven Nadler (ed.) Causation in Early Modern Philosophy (University Park: Penn
State Press, 1993), 27-56.
28
See for example Suarez, DM 17.1.6. For an overview of Aristotelian-Scholastic theories of physical
influx and transeunt efficient causation, with special attention given to Suarez’s theory, who in turn heavily
influenced Descartes, see A.J. Freddoso’s introduction to Francisco Suarez, S.J., On Creation,
Conservation, & Concurrence: Metaphysical Disputations 20-22 (South Bend: St. Augustine’s Press,
2002), xliii-lix.
29
See for example AG 33, (G II.251: AG 176), (G VI.607-8: AG 213-14), and G IV.498f.
11
C1.
C2.
Creaturely transeunt causation is fundamental only if the transference of
accidents is possible.
The transference of accidents is not possible.
C1 is confronted with two problems—interpretive and philosophical—that I address in
Chapter 2. First, Leibniz stated C1 throughout his career, but offered little argument for
it. The lack of defense is startling as most who affirmed the fundamentality of creaturely
transeunt causation also surprisingly but nevertheless vehemently denied C1. In fact,
Thomas Aquinas called the transference condition “laughable.”30 Second, God is a
transeunt cause in Leibniz’s philosophy. Yet God’s transeunt causation does not consist
in transference. Thus, Leibniz needs a principled way to affirm C1 and deny that divine
transeunt causation consists in transference.31
I argue that Leibniz had a solution to both puzzles. While Leibniz never
explicitly states it, I make the case that Leibniz thought that if an agent transeuntly caused
an accident without transferring the accident, the agent created the accident. For the
recipient substance contributed no reality to the accident and the agent lost no reality in
causing the accident. However, only an omnipotent being—God—can create. Therefore,
only God can transeuntly cause without transferring what is caused. I also argue in
Chapter 2 that the argument for C1 is consistent with creaturely immanent causation.
Finally, I close off chapter 2 by drawing attention to an important weakness with C1 that
has not yet been recognized by Leibniz scholars. Based on some important arguments
Leibniz develops against occasionalism in his Theodicy concerning the causation of
modifications, I argue that Leibniz ultimately only has reasons to hold that the
30
31
See Aquinas, SCG, Bk. III, Pt. 1, Ch. 69, 28.
Freddoso raises this problem, which to date has not been addressed. See Freddoso, Ibid., xlix.
12
transference condition is a condition of the production of non-modal accidents, such as
real qualities.
In Chapter 3, I address Leibniz’s support for C2. Unlike C1, Leibniz provides
ample support for C2 throughout his career, making C2 much easier to defend than C1.
Additionally, unlike C1, C2 enjoys much support throughout the history of philosophy,
especially by defenders of creaturely transeunt causation who denied C1. I argue, in two
stages, that there is nothing in Leibniz’s ontology that could be transferred from the cause
to the recipient of the effect. First, I articulate the thesis that Leibniz’s ontology consists
of simple non-corporeal substances and their accidents.32
Second, I present and articulate a number of important theses Leibniz affirmed
about substances and their accidents, which entail that neither could be transferred. I also
show that most of these theses were not unique to Leibniz, but in fact were widely
endorsed by his predecessors who affirmed the fundamentality of creaturely transeunt
causation.33 The majority of the second half of Chapter 3 is devoted to Leibniz’s theses
on accidents, specifically his claims that accidents are modifications or limitations, as
these features of accidents are most relevant to why they could not be transferred. These
theses also provide material which will be utilized in the next chapter when I investigate
32
A classic statement of Leibniz’s ontology can be found in a 1715 letter to Des Bosses, where Leibniz
writes, “Whatever is not a modification can be called a substance.” (G II.503-4: L 614). Leibniz’s most
extensive defense and articulation of his metaphysics of fundamental, simple and non-extended individuals
is his Monadology. See (G VI.607-23: AG 213-25)
33
The two most important theses concern properties. First, for any property P, P exists only if P is some
individual s such that P is a property of s. Second, for any property P and any individual s1, if P is a
property of s1 then there is not some individual s2 such that P is a property of s2. The first thesis entails that
properties must exist in an individual— they cannot float free. The second thesis entails that a property
cannot exist in more than one individual (i) at the same time; (ii) at different times; (iii) or in different
possible worlds. The second thesis was affirmed by nearly every medieval and early modern philosopher
and has recently been revived by contemporary philosophers. Contemporary metaphysicians call it the
thesis of “non-transferability”.
13
the nature of accidents insofar as they are the effects of immanent causation and the role
they play in change.
In chapter 4, I continue my study of the nature of Leibnizian accidents, shifting
the focus from their role in Leibniz’s critique of the Traditional view to their positive role
in change and as causal relata, where such accidents are the effects of immanent
causation. Specifically, I reconstruct and assess Leibniz’s argument that accidents are
modifications or limitations34, a thesis which was widely held by early modern
philosophers and which set them apart from their medieval predecessors and also which
played an crucially important role in Leibniz’s positive account of change and immanent
causation.35 In a key passage, in a 1703 letter to De Volder, Leibniz clarifies what it is
for an accident to be a modification or limitation:
. . . a modification is a varying limitation, and modes merely limit things but do
not increase them and hence cannot contain any absolute perfection which is not
in the thing itself which they modify. Otherwise, in fact, these accidents must be
thought of in the manner of substances, namely, something which stands per se.36
I shall argue that holding that accidents are modifications or limitations allowed Leibniz
to posit that substances are mereologically simple while synchronically and
diachronically complex— they have a multitude of accidents at a time and change
accidents over time.
34
See (G VI.598: AG 207), (G VI.590: AG 265), (G II.458: L 606), and (G II.503-4: L 614).
For a helpful paper on why it was significant and controversial that early modern philosophers only
posited one type of accident—modifications—in their ontology, see Stephen Menn, “The Greatest
Stumbling Block: Descartes’ Denial of Real Qualities,” in Roger Ariew and Marjorie Grene, ed.,
Descartes and His Contemporaries: Meditations, Objections, and Replies (Chicago: The University of
Chicago Press, 1995), 182-207.
36
(G II.257: L 532).
35
14
To do so, I first reconstruct Leibniz’s argument in his not yet translated 1688 “De
Realitate Accidentium,” (DRA) which I also provide a translation of in Appendix B.37
DRA is one of the few texts in which Leibniz goes into any length in arguing that all
accidents are modifications. In this lesser known essay, Leibniz draws from premises the
he utilized throughout his career.
After I reconstruct the argument in DRA, I address an issue arising from his
arguments in DRA. The issue is that while Leibniz’s arguments in DRA prima facie
support the conclusion that accidents are modifications, he gives a surprisingly noncommittal or agnostic conclusion as to whether there are any accidents at all. I argue that
Leibniz hesitates to posit even modifications because he worried that the problems he
raised with non-modal accidents apply to modifications as well.38 Specifically, Leibniz’s
arguments against non-modal accidents stem from such accidents being parts of
substances, which is inconsistent with substantial simplicity. The arguments of DRA
prima facie also entail that not only are non-modal accidents parts, but modal accidents
are parts as well. Not much later, Leibniz changed his mind and posited modal accidents
in his ontology, without, however, ever explicitly addressing how such accidents could be
in a substance without being a part of it. Drawing from Leibniz’s later mereological and
geometrical writings and his understanding of modifications as limitations, I argue that
Leibniz had the resources to posit simple substances that have a multitude of
modifications at a time and change such modifications over time. By doing so, I fill one
scholarly void in this chapter by applying Leibniz’s developed mereological theses to not
37
38
A.VI.4A.994-996.
By “non-modal accidents,” I mean accidents that are not modifications.
15
only showing how substances can have a multitude of modes at a time and over time, but
also showing why simple substances cannot have non-modal accidents (such as real
qualities).
In Chapter 5, I address an issue that has divided Leibniz scholars concerning the
precise relata in Leibnizian immanent efficient causation. In many passages, Leibniz
writes as if it is the substance or individual itself that efficiently causes its later properties
or accidents.39 Yet in plenty of other passages, he writes as if it is the substances earlier
properties or accidents that cause its later accidents.40 Call the former view the
Efficacious-substance interpretation, which Bobro, Clatterbaugh, and Jorati defend.41
Most recent scholars such as Rutherford, Carlin, Kulstad, and Bolton, who defend the
latter view, argue that it isn’t just any accident in Leibniz’s ontology but rather
appetitions, which strictly speaking, produce a substance’s later accidents.42 Call this the
Efficacious-appetition interpretation.43
In this chapter, I present and defend a novel version of the Efficacious-substance
interpretation which incorporates the strengths of the Efficacious-appetition
interpretation. I focus primarily on Donald Rutherford’s arguments for the efficacious-
39
See (G V.194: NE 210), (G VI.295-6: T 300), and (G IV.509: AG 160).
See (G IV.439: AG 47), (G II.91-2: AG 82), (G VI.356-7: T 403), and G IV, 532-3.
41
See Marc Bobro and Kenneth Clatterbaugh, “Unpacking the Monad, Leibniz's Theory of Causality,” The
Monist, (1996) 79: 409–26 and Julia Jorati, “Leibniz on Causation –Part 1,” Philosophy Compass (2015)
10: 389-397.
42
See Martha Brand Bolton, “Change in the Monad,” in Eric Watkins, ed. The Divine Order, the Human
Order, and the Order of Nature: Historical Perspectives (Oxford: Oxford University Press, 2013), 178;
Laurence Carlin, “Leibniz on Final Causes,” Journal of the History of Philosophy (2006) 44: 231; and
Donald Rutherford, “Laws and Powers in Leibniz,” The Divine Order, the Human Order, and the Order of
Nature: Historical Perspectives. Ed. Eric Watkins. (Oxford: Oxford University Press, 2013), 167 and
“Leibniz on Spontaneity.” Leibniz: Nature and Freedom. Eds. Donald Rutherford, and Jan A. Cover,
(Oxford: Oxford University Press, 2005), 167.
43
Textual support can be found in passages where Leibniz argues that appetitions are principles of change.
See (G VI.598: AG 207) and (G VI.609: AG 215).
40
16
appetition interpretation, as he has presented the lengthiest and strongest case for it.
Rutherford argues that the Efficacious-substance interpretation is incompatible with
Leibniz’s determinism and his requirement that change be intelligible— the change must
explained by the substance’s own nature.44 Instead, Rutherford argues that if appetitions
are what produce the later accidents of a substance, then monadic change is both
deterministic and intelligible. A substance s is determined to change from state N to N+1
because the appetitions of s that partially constitute N are both appetitions for the
accidents of state N+1 and what produce the accidents which make up N+1. The change
is intelligible because it is explained by s’s nature— specifically s’s nature as modified
by its appetitions. What is key in Rutherford’s argument is that what does the explaining
is what does the producing— the efficient causal agent of the change.
However, the Efficacious-appetition interpretation succumbs to a serious
objection originally raised by Locke and endorsed by Leibniz himself in his New Essays
on Human Understanding, what I’ll call the Multiplication of Agents objection.45 If
appetitions, rather than substances, are efficient causes of a substance’s later accidents,
then there is a plurality of distinct efficient causal agents in a substance, a view that
Leibniz explicitly rejects, and, moreover that, runs afoul of the simplicity and unity of
created substances. The Efficacious-substance interpretation avoids this objection as
there is just one efficient causal agent— the substance.
44
Leibniz writes, “Whenever we find some quality in a subject, we ought to believe that if we understood
the nature of both the subject and the quality we would conceive how the quality could arise from it. So
within the order of nature (miracles apart) it is not at God’s arbitrary discretion to attach this or that quality
haphazardly to substances. He will never give them any that are not natural to them, that is, that cannot
arise from their nature as explicable modifications.” See A.VI.6.66.
45
See John Locke, An Essay Concerning Human Understanding (Oxford: Oxford University Press, 1975),
2.21.20/243-44 and (G V.159: NE 174).
17
I further argue that the efficacious-substance account can be reconciled with
Leibniz’s determinism and strictures on explanation. Leibniz utilized a distinction found
as early as Aquinas and developed at length by Suarez between two different kinds of
efficient causes— the principle quod efficient cause or efficient causal agent and the
principle quo efficient cause or power by which the agent acts.46 Scholastics such as
Suarez held that in most cases, substances are efficient causal agents yet they also had
principled accounts of how substances could be efficient causal agents and yet act
deterministically or of necessity in some sense.47 I argue that Leibniz had similar reasons
to consistently hold that substances are efficient causal agents—and so avoid the
Multiplication of Agents objection—but also hold that such substances deterministically
produce their effects in a way that satisfies Leibniz’s strictures on explanation. Mainly,
appetitions are powers by which a substance efficiently causes its later accidents.
In the Appendix A, I argue that there are resources within Leibniz’s metaphysics
of causation to provide the support for a premise that scholars have argued is missing but
needed in a different argument Leibniz made against creaturely causal interaction. Early
in his career, Leibniz denied creaturely causal interaction because of his thesis of
substantial spontaneity, specifically a variant of spontaneity in which all of a substance’s
states follow from its complete concept or notion.48 Scholars have rightly pointed out
that Leibniz cannot deny causal interaction from spontaneity alone. Instead, Leibniz also
needs a premise ruling out causal overetermination. Otherwise, there is no inconsistency
46
See, for example, Aquinas, ST 1a q36 a1 and Suarez, DM 22.1.19.
DM 19.1
48
See AG 33 and (G IV.439: AG 47).
47
18
in affirming both that substances are spontaneous and that some states of a substance are
caused by distinct substances.
Leibniz never offers an argument against overdetermination, nor does he even
deny it. However, in Appendix A, I argue that given Leibniz’s understanding of
creaturely transeunt efficient causation— specifically his Transference condition, Leibniz
has the resources to deny causal overdetermination. That is, if creaturely causal
interaction consisted in transference, as Leibniz insists it must if it occurred—with a
literal detaching of the accident caused from the agent and its being transferred to the
patient, then overdetermination is not possible. If Leibniz can rule out overdetermination
from his understanding of creaturely transeunt causation, then he can deny that an
accident transeuntly caused is also immanently and spontaneously caused. However, if
Leibniz can deny spontaneity from creaturely transeunt causation, then he can deny
creaturely transeunt causation from spontaneity via contraposition.
In the Appendix B, I address early and late Leibniz’s views on Transubstantiation.
While at various times in Leibniz’s career, he offered accounts of the metaphysics of
transubstantiation, I argue that the mature Leibniz, ultimately, did not affirm
Transubstantiation. In Appendix C, I offer a translation of Leibniz’s “De Realitate
Accidentium.”
§4 The Idealistic Interpretation of Leibniz
As dissertation is a study of Leibniz’s metaphysics of intra-substantial causation and
change, it’s important to get clear on what count as substances in Leibniz’s metaphysics.
In this dissertation, I assume the idealistic interpretation of the mature Leibniz’s
19
metaphysics—the period starting around the publication of his Discourse on Metaphysics
in 1686 up until his death in 171649—in which the only entities that are substances,
strictly speaking, are simple substances or monads.50 We find Leibniz expressing this
thesis in his “Against Barbaric Physics,” written sometime between 1710 and 1716,
where he claims that “only monads (among which the best are souls, and among souls,
the best are minds) are substances.”51 Writing to his friend Nicolas Remond, in the last
year of his life, Leibniz again claims that “Absolute reality rests only in monads and their
perceptions.”52
While I will draw upon texts that both clarify and support the idealistic
interpretation in this section, my aim is not to offer a full-scale defense of the idealistic
reading of Leibniz. Adequately defending the idealistic or non-idealistic interpretation of
Leibniz is a task demanding its own dissertation. Space and time constraints wouldn’t
allow me to do justice in addressing the numerous issues that must be dealt with in
defending one interpretation over the other. I do note, however, that the idealist
interpretation has been the dominant interpretation, both historically and presently.
To clarify the idealist interpretation in which, strictly speaking, the only entities
that are substances are simple substances, it’s important to get clear on the reasons
49
I will draw upon earlier texts at various stages in this project, when relevant, but will indicate when I do
so.
50
For scholars who defend the idealist interpretation of Leibniz, see J.A. Cover and John Hawthorne,
Substance and Individuation in Leibniz (Cambridge: Cambridge University Press, 1999); Christia Mercer,
Leibniz’s Metaphysics: It’s Origins and Development (Cambridge: Cambridge University Press, 2001);
Robert Merrihew Adams, Leibniz: Determinist, Theist, Idealist (Oxford: Oxford University Press, 1994);
Donald Rutherford, Leibniz and the Rational Order of Nature (Cambridge: Cambridge University Press,
1995); and R.C. Sleigh Jr., Leibniz and Arnauld: A Commentary on their Correspondence (New Haven:
Yale University Press, 1990).
51
AG 319.
52
(G III.636: L 659).
20
Leibniz’s offers in favor of that thesis. To do so, I start by looking at some features of
Leibnizian substances. According to Leibniz, a substance is a true unity or an unum per
se. In a letter to Arnauld, Leibniz writes, “that what is not truly one being is not truly one
being either.”53 In order to be a true unity, however, a substance must be indivisible. In
another letter to Arnauld, Leibniz writes, “A substantial unity requires a thoroughly
indivisible and naturally indestructible being, since its notion includes everything that
will happen to it, something which can be found neither in shape nor in motion (both of
which involve something imaginary, as I could demonstrate) but which can be found in a
soul or substantial form, on the model of what is called me.”54 Thus, according to
Leibniz the only genuine substances, which have true unity, are simple substances. We
find Leibniz expressing this thesis several decades later in a 1704 or 1705 letter to De
Volder, writing that simple substances “alone have unity and absolute reality.”55
Bodies, however, are divisible and therefore cannot be substances. Bodies, being
shaped, are extended and therefore have parts.56 Instead, of counting as substances,
bodies are divisible aggregates of substances, as Leibniz argues in his “Comments on
Michel Angelo Fardella”.57 Given that bodies are beings by aggregation, they “have their
unity in our mind only,” according to Leibniz.58 Bodies are “unities”, Leibniz argues, in
53
(G II.97: AG 86).
(G II.76: AG 79).
55
AG 181.
56
AG 207.
57
AG 103.
58
AG 86.
54
21
the sense that circle of men holding hands are a unity.59 Since bodies are aggregates, they
are not substances. In a 1703 letter to De Volder, Leibinz writes, “Since only simple
things are true things, what remains are only entities by aggregation.”60
As aggregates, such as bodies, are not substances, they are instead phenomena.61
Specifically, bodies are well-founded phenomena that result from simple substances.62
Bodies, then, are not eliminated from Leibniz’s metaphysics. Instead they are reduced, in
some sense, to simple substances and their accidents. Support for this understanding of
bodies is found, for example, in a fictional dialogue between Philarete and Ariste written
in 1712, when Philarete (speaking for Leibniz) claims, “My friend [Leibniz], whose
opinion I have just related, gives enough evidence that he leans in this direction, since he
reduces everything to monads, or to simple substances and their modifications. . .”.63
Earlier, in a 1704 letter to De Volder, Leibniz writes: “Considering the matter carefully,
we must say that there is nothing in things but simple substances, and in them, perception
and appetition.”64 Bodies, reduced in some sense to simple substances and their
accidents, have a phenomenal existence. Leibniz continues, “Moreover, matter and
motion are not substances or things as much as they are the phenomena of perceivers, the
59
Ibid. Leibniz also writes to Arnauld: “We can therefore say of these composites and similar things what
Democritus said so well of them, namely, they depend for their being on opinion or custom. And Plato
held the same opinion about everything which is purely material. Our mind notices or conceives some true
substances which have certain modes; these modes involve relations to other substances, so the mind takes
the occasion to join them together in thought and to make one name account for all these things together.
This is useful for reasoning, but we must not allow ourselves to be misled into making substances or true
beings of them.” See AG 89.
60
AG 177.
61
Ibid.
62
Ibid.
63
(G VI.590: AG 265).
64
(G II.270: AG 181).
22
reality of which is situated in the harmony of the perceivers with themselves (at different
times) and with other perceivers).65
This is the idealistic interpretation I shall assume in this dissertation. The only
substances, strictly speaking, are simple substances. Bodies are reducible in some sense
to simple substances and their accidents. While bodies are not substances, but wellfounded phenomena that result from, in some sense, simple substances, there is a
complication. There is a distinction between plain old composite bodies (e.g., a pebble)
and organisms or animals. In a letter to De Volder, Leibniz writes:
I distinguish: (1) the primitive entelechy or soul; (2) the matter, namely, the
primary matter or primitive passive power; (3) the monad made up of these two
things; (4) the mass or secondary matter, or the organic machine in which
innumerable subordinate monads come together; and (5) the animal, that is, the
corporeal substance, which the dominating monad makes into one machine.66
An animal, according to Leibniz, is a corporeal substance made up of a very large
number of monads or simple substances. Prima facie, an animal is a composite—made
up of a large number of monads—but also a substance. Specifically, an animal consists
in a dominant monad and a plurality of subordinate monads. In the Monadology, Leibniz
writes, “Thus we see that each living body has a dominant entelechy, which in the animal
is the soul; but the limbs of this living body are full of other living beings, plants,
animals, each of which also has its entelechy, or its dominant soul.”67
There are two interpretations one could take on Leibniz’s writings on organisms.
First, these animals—which consist of a plurality of monads—are genuine corporeal
substances, which while composite are nonetheless active true unities. The downside of
65
Ibid.
(G II.252: AG 177).
67
(G VI.619: AG 222).
66
23
this interpretation is that it conflicts with Leibniz’s claims about substantial simplicity.
The second interpretation is that these corporeal substances are not genuine substances.
Instead, they are aggregates of monads, like pebbles, only with additional special
relations between the dominating monad and its subordinate monads. On this latter
interpretation, composite corporeal substances or animals are reducible as well, in some
sense, to simple substances and their accidents. I will assume this latter interpretation in
this dissertation. However, there is still much of value in this dissertation to scholars who
subscribe the non-idealist reading of Leibniz. Scholars who believe that the mature
Leibniz posited corporeal substances in addition to simple substances should understand
my project not as a study of intra-substantial causation and change but as a study on
Leibniz’s metaphysics of intra-monadic causation and change.
§5 Methodological Approaches
As this dissertation is a work in the history of philosophy, I close off this introductory
chapter by explaining the methodological approaches I’ll be utilizing in the remaining
chapters. There is more than one worthwhile approach to the history of philosophy, and
most worthwhile approaches have exemplars in Leibniz scholarship. Using a distinction
recognized by Leibniz scholars such as Mates, Sleigh, Cover and O’Leary-Hawthorne,
these approaches can broadly be construed as “Exegetical” history of philosophy and
“Philosophical” history of philosophy.68
68
See J.A. Cover and John O’Leary-Hawthorne, Substance and Individuation in Leibniz (Cambridge:
Cambridge University Press, 1998), 5-9; Benson Mates, “Individuals and Modality in the Philosophy of
Leibniz,” Studia Leibnitiana 1972: 83-84; and R.C. Sleigh, Leibniz & Arnauld: A Commentary on their
Correspondence (New Haven: Yale University Press, 1990), 2-6.
24
I’ll start with exegetical or what can also be called “historical” history of
philosophy. Mates characterizes exegetical history philosophy as “an attempt to discover
and set forth, as accurately, objectively, and completely as possible the philosophical
views of various historical figures”.69 Building on Mates’s description, Sleigh
distinguishes two components of exegetical history of philosophy— a fact-finding
component and an explanatory component. According to Sleigh, the fact-finding
component is not simply collecting various statements by a philosophy. Instead, it
involves the careful task of formulating a philosopher’s central views on a topic using
sentences whose meaning is obvious in that we know what propositions the sentences
express.70 While the outcome of mere fact-finding is—in and of itself—rarely if ever a
significant contribution to the history of philosophy, it is at the very least a necessary
component of work in the history of philosophy. My own project is no exception-- there
will be such fact-finding when appropriate in this project, which will be used in the
service of what Sleigh calls the explanatory component.
Sleigh describes the explanatory component as not simply determining what a
philosopher said but explaining why the philosopher said it.71 By explanation, Sleigh
means the rational basis, as opposed to the psychological motivations for a philosopher’s
views.72 The explanatory component is more difficult. As Sleigh writes, “Usually, many
aspects of the intellectual setting that bear on philosophical theses ultimately accepted by
69
Mates, Ibid.
Sleigh, 4.
71
Ibid., 5-6.
72
Ibid.
70
25
our philosopher go unstated.”73 This is especially true with some of the reasons for some
of Leibniz’s most important and daring metaphysical theses, some of which are the
subject of my project. I highlight this in order to note that the aims of my project are
primarily explanatory. I aim to explain why Leibniz held certain theses, especially in
Chapters 2-4.
Sleigh also notes that the explanatory component can benefit from what he calls
philosophical history of philosophy as well, which Jonathan Bennett (whom Sleigh cites)
describes as discussing some philosophical topic “in the company” of a historical
figure.74 The style of philosophical history of philosophy I adopt in parts of this project
can be described as follows. Leibniz may argue that some proposition P1 is true because
of a different proposition P2. The reasoning for why P2 entails P1 may be lost on us.
Operating on a principle of charity that is more than appropriate when studying a
philosopher of Leibniz’s caliber, one should assume—initially, at least—that Leibniz had
good reasons for arguing that P1 follows from P2. The work of figuring out how P2
entails P1 requires finding additional premises, usually premises that are not explicitly
stated by Leibniz as linking P1 and P2. Finding these premises can involve: (i) searching
throughout Leibniz’s thought for theses he did explicitly defend which can also link P1
and P2, even though Leibniz does not show that or even state that such theses link P1 and
P2; (ii) looking broader at theses widely held by historical figures whom Leibniz was
familiar with which could link P1 and P2 and which Leibniz would have no reason to
reject; or (iii) engaging in plain old metaphysical reasoning oneself to rationally
73
74
Ibid., 6.
Ibid., 3.
26
reconstruct a link between P1 and P2. Oftentimes, it involves a combination of (i)
through (iii), with a high preference for (i) and (ii). With respect to (ii) and (iii), good
philosophical history should start and end with the figure being studied.75 In parts of this
dissertation, I also engage in such “philosophical” history of philosophy, with the
intention of utilizing it to serve explanatory history of philosophy. In the main body of
this dissertation, I try to stick with methods (i) and (ii) but I will go into the territory of
(iii) in parts of the main chapters and in the entirety of Appendix A.
75
As Cover and O’Leary-Hawthorne note and have done in their own work on Leibniz’s metaphysics of
substance. See Cover and O’Leary-Hawthorne, Ibid., 8-9.
27
CHAPTER 2. LEIBNIZ’S TRANSFERENCE ARGUMENT CREATURELY
TRANSEUNT CAUSATION, PART 1: THE TRANSFERENCE CONDITION
In this chapter and the next, I assess Leibniz’s criticisms of the traditional view of
creaturely causation of his time— what Leibniz called “physical influx”, which defended
the fundamentality of creaturely transeunt causation and posited a world of causally
interacting created substances.76 A central tenant of the traditional view was that
transeunt causation consisted in the giving or communication of the effect from the
cause—the agent—to the recipient of the effect— the patient.77 On Leibniz’s
understanding of physical influx, which is strikingly similar to many contemporary
persistence theories of causation, the agent substance’s communication of the effect
consisted in the accident caused by the agent first detaching itself from the agent and
being sent to the patient.78 Call this literal detachment of the accident from the agent and
its being sent to the patient “transference”, where upon transference, the agent no longer
possesses the accident it caused. Leibniz’s criticisms of the possibility of creaturely
transeunt causation then consisted of two claims. First, creaturely transeunt causation is
76
For an overview of Leibniz’s understanding of Physical Influx, see See Eileen O’Neill, “Influxus
Physicus,” in Steven Nadler (ed.) Causation in Early Modern Philosophy (University Park: Penn State
Press, 1993), 27-56.
77
See for example Suarez, DM 17.1.6. For an overview of Aristotelian-Scholastic theories of physical
influx and transeunt efficient causation, with special attention given to Suarez’s theory, which in turn was a
major influence on Descartes, see A.J. Freddoso’s introduction to Francisco Suarez, S.J., On Creation,
Conservation, & Concurrence: Metaphysical Disputations 20-22 (South Bend: St. Augustine’s Press,
2002), xliii-lix.
78
In the appendix to this chapter, I explore in greater depth the similarities between Leibniz’s
understanding of physical influx and contemporary persistence theories of causation.
28
possible only if the transference of accidents is possible. Second, the transference of
accidents is not possible.
In this chapter, I address two problems—interpretative and philosophical—that
confront the first claim, what I’ll call the “transference condition”. First, Leibniz stated
the transference condition throughout his career but offered little argument for it. This
lack of any sustained defense is striking because most who affirmed the fundamentality
of creaturely transeunt causation vehemently denied the transference condition. In fact, as
we’ll see soon, Thomas Aquinas, a proponent of creaturely transeunt causation called the
transference condition “laughable.”79 Second, God is a transeunt cause in Leibniz’s
philosophy and yet God’s transeunt causation does not consist in transference.80 Thus,
Leibniz needs a principled way to hold that the transference condition is a requisite of
creaturely transeunt causation while denying that divine transeunt causation consists in
transference.
I argue that Leibniz had a solution to both problems. While Leibniz never
explicitly states it, I argue that Leibniz thought that if an agent transeuntly caused an
accident without transferring the accident, the agent created the accident. For the
recipient substance contributed no reality to the accident and the agent lost no reality in
causing the accident. However, only an omnipotent being—God—can create. Therefore,
only God can transeuntly cause without transferring what is caused.
In §1, I articulate and defend a specific interpretation of Leibniz’s transference
condition, where creaturely transeunt causation consists in the effect being transferred
79
Aquinas, SCG, Bk. III, Pt. 1, Ch. 69, 28.
According to Leibniz, God’s transeunt causal activity consists in creation, conservation, and concurrence.
See for example (G VI.118-19: T 27) and (G IV.457-58: AG 63).
80
29
from the agent to the patient, where upon transference, the agent no longer possesses
what it produces in the patient. In §2, I address the two puzzles that scholars have raised
against Leibniz’s transference condition. After showing that Leibniz has principled
reasons for holding that creaturely transeunt causation must consist in transference while
divine causation does not, I draw attention in §3 to a weakness that has not yet been
addressed with the transference condition. Based on some important arguments Leibniz
develops against occasionalism in his Theodicy concerning the causation of
modifications, I argue that Leibniz ultimately has reasons only to hold that the
transference condition is a condition of the production of non-modal accidents, such as
real qualities.81
§1 The Transference Condition
Take a non-initial accident A inhering in some created substance s. What could be the
efficient cause of A? Adherents of the traditional view (nearly everyone except Leibniz
and the occasionalists) would say that either s itself or God or a different created
substance s’ could have caused A. For example, suppose Davis’s face turns red. Davis
could have produced the redness in his face by holding his breath. On the assumption of
Theism, God could have produced the redness in Davis’s face. But it also seems trivially
true that some other created substance, such as Davis’s grandmother, could have
produced the redness in Davis’s face (perhaps by slapping him). I’ll call cases where a
81
Non-modal accidents are accidents that are not modifications.
30
substance produces or efficiently causes an accident in a different created substance cases
of transeunt causation.82
In this chapter and the next, I address Leibniz’s (in)famous denial of the
possibility of creaturely transeunt causation, specifically his requirement that some of the
being of the agent substance be transferred to the patient substance in such causal
interaction, which Leibniz in turn claimed is impossible. I will call Leibniz’s argument
the “Transference argument”. The argument is worth focusing on for two reasons. First,
some of the key metaphysical theses and constraints used in Leibniz’s transference
argument also play important roles in Leibniz’s positive views on efficient causation and
change, which are the subjects of chapters 4 and 5. Second, the transference argument is
the argument Leibniz most frequently invokes against the possibility of creaturely causal
interaction throughout his career. For example, in his 1686 “Primary Truths”, Leibniz
writes:
Strictly speaking, one can say that no created substances exerts a metaphysical
action or influx on any thing. For not to mention the fact that one cannot explain
how something can pass from one thing into the substance of another, we have
already shown that from the notion of each and every thing follows all of its
future states.83
In a later 1703 letter to de Volder, expressing his more mature metaphysics, Leibniz
writes, “Properly speaking, I don’t admit the action of substances on one another, since
there appears to be no way for one monad to flow into another.”84 Around the same time,
in his Monadology, Leibniz continues to deny the possibility of creaturely causal
82
In this chapter, I’m primarily focused on Leibniz’s criticisms of creaturely transeunt causation, so
transeunt causation should be understood as creaturely transeunt causation. I’ll call cases where God is a
transeunt cause, “divine transeunt causation”.
83
AG 33.
84
(G II.251: AG 176).
31
interaction, writing, “There is also no way of explaining how a monad can be altered or
changed internally by some other creature.” Leibniz’s reason again is that “Accidents
cannot be detached, nor can they go about outside of substances, as the sensible species
of the Scholastics once did. Thus, neither substance nor accident can enter a monad from
without.”85
All of these passages express roughly the same argument, which has the following
structure: Creaturely causal interaction (or transeunt causation) is not possible because x
is not possible. The variable x is a placeholder for statements in the passages above about
some entity passing or flowing or detaching from the cause to the effect— or better put,
from the agent to the patient. Call this “passing” or “flowing” or “detaching” of some
entity from the agent to the patient “Transference”. Leibniz’s argument then is that
creaturely transeunt causation is not possible because transference is not possible. Given
that Leibniz argues that creaturely causation is not possible because transference is not
possible, it is evident that Leibniz takes such transference to be a necessary condition of
creaturely transeunt causation.
It is important to get a grip on how Leibniz understands transference. I argue that
Leibniz thinks such transference consists in the effect moving from the agent to the
patient, such that upon transference, the agent no longer possesses what was transferred
to the patient. That is, when an agent produces an effect in a patient, the agent transfers
some of its own being to the patient. Key passages support this understanding of
transference. For example, in his 1695 A New System of Nature, Leibniz writes:
85
(G VI.607-8: AG 213-14).
32
Further, the action of one substance upon another is not an emission or a
transplanting of some entity, as is commonly supposed; and it can be understood
reasonably only in the way just shown. It is true that we can easily conceive of
both the emission and the reception of the parts in matter and can in this way
reasonably explain all the phenomena of physics mechanically. But since
material mass is not a substance, it is clear that the action of substance itself can
be only what I have just described.86
Similar support is found a year later in his 1696 “Second Explanation of the New
System,” where he writes:
The way of influence is that of the common philosophy. But since it is impossible
to conceive of material particles or species or immaterial qualities which can pass
from one of these substances into the other, this view must be rejected.87
Further support is found a decade later in Leibniz’s New Essays on Human
Understanding:
I am not surprised that you encounter insurmountable problems when you seem to
be entertaining something as inconceivable as an accident’s passing from one
subject to another; but I see no reason why we have to suppose such a thing. It is
almost as strange as the Scholastics’ notion of accidents which are not in any
subject; though they are careful to attribute theirs solely to the miraculous
workings of divine omnipotence.88
In these passages and the previous set of passages we looked at, Leibniz is criticizing the
possibility of creaturely causal interaction—specifically “the way of influence,” which he
describes as some entity being detached (detacher) or emitted (emission) from the
substance and being passed (passer) or transplanted (transplantation) from the agent to
the patient. This detaching of the entity implies that the entity was previously attached to
the subject. That the entity emitted and detached by the substance goes outside of the
substance implies that the entity was previously inside. Hence, whatever it was that was
86
(G IV.486: L 459).
(G IV.498-99: L 460).
88
(G V.208: NE 224).
87
33
passed or transplanted or detached or emitted to the patient first belonged to the agent
until its reception in the patient. Leibniz’s statement that “It is true that we can easily
conceive of both the emission and the reception of the parts in matter” is especially
strong evidence for this understanding of transference. For when a material part is passed
from one body to another, the material part surely does not belong to the originating body
upon its reception in the receiving body but was a part of the originating body prior to
transference. Similar reasoning applies when the entity transferred is a material particle.
Leibniz entertains several different kinds of entities as candidates for transference,
such as atoms, sensible species, accidents and monads. For three reasons, however, I’ll
focus primarily on accidents in this chapter and the next. First, Leibniz has other reasons
for denying the possibility of one created substance causing the existence of a different
created substance. Mainly, he holds that all created substances begin to exist by divine
creation and can only cease to exist via divine annihilation.89 Second, Leibniz is an
idealist who—in his strict ontology—denies the existence of physical atoms. His
idealism also leaves no place for sensible species.90 This leaves accidents. Exploring
why accidents could not be transeuntly caused or transferred, in addition to the first two
reasons, also sheds much light on his own positive account of the created substances
immanently causing and changing their own accidents.
89
Leibniz denies the possibility of created substances efficiently causing other created substance to come
into existence because he argues that created substances are mereologically simple. Given that they are
simple, they can only begin to exist and cease to exist via divine creation and annihilation. See (G VI.607:
AG 213).
90
I articulated and to a limited extent defend the idealistic interpretation of Leibniz’s metaphysics in
Chapter 1.
34
So in this chapter, I focus on Leibniz’s reasons for his claim that one substance
can transeuntly cause an accident in a different substance only if the agent transfers the
accident from itself to the patient, with such transference resulting in a loss of some of
the being of the agent— the being of the accident gained by the patient. This is the
transference condition, which I formally express thus:
The Transference Condition: For any created substance s1 and any created
substance s2 and any accident A, s1 efficiently causes A to inhere in s2 only if s1
transfers A from s1 to s2.91
s1’s transferring of A from itself to s2, as I argue Leibniz understands it, entails that s1
loses A or loses the being of A that it transfers to s2. The accident A or the being of A first
belonged to s1 and then is gained by s2, upon which A no longer belongs to s1. That is
how I argue that Leibniz understands it and the version of transference I shall work with
throughout this chapter. While not a definition, I can at least offer some constraints to
further clarify what transference consists in:
For any created substance s1 and any created substance s2 and any accident A, s1
transfers A to s2 only if:
(i)
(ii)
(iii)
91
s1 is not identical to s2;
A first belongs to s1;92
A second belongs to s2.
An alternative formulation, which will make sense in the third section of this chapter when I articulate
the causal adequacy principle, can be expressed thus: For any created substance s1 and any created
substance s2 and any accident A, s1 efficiently causes A to inhere in s1 only if s1 transfers the being of A
from s1 to s2. For now, I stick with the original formulation.
92
Prima facie, the use of ordinals in conditions (iv) and (v) suggest a temporal ordering, where A is a part
of s1 at a time earlier than when A is a part of s2. Invoking time in elucidating transference is prima facie
controversial then, as Leibniz defines time in terms of causality rather than vice versa. However, the use of
ordinals in (iv) and (v) doesn’t necessarily mean that transference is defined partly by temporal concepts.
Ordinals in a statement do not always entail a temporal ordering. For example, “B” is the second letter of
the alphabet. That doesn’t mean that “B” occurs temporally later than “A”. For works addressing the
relation between time and causality in Leibniz, see J.A. Cover, “Non-Basic Time and Reductive Strategies:
Leibniz’s Theory of Time,” Studies in History and Philosophy of Science 28 (1997): 289-318 and Michael
Futch, Leibniz’s Metaphysics of Time and Space (New York: Springer, 2008).
35
(iv)
(v)
If A belongs to s1 then A does not belong to s2;
If A belongs to s2 then A does not belong to s1;
I note two things about these constraints on transference. First, I am not claiming that (i)
– (v) are sufficient for transeunt causation, as it could be the case that God moved A from
s1 to s2, in which case we would say that God, rather than s1, caused s2 to have A.93
Second, the term “belong” is undefined. In the case of s2, A’s belonging to s2 just is A’s
inhering in s2. However, I’ll assume, for now, that “belonging” is broader in extension
than just inherence, leaving open the possibility that A’s belonging to s1 does not entail
that A inheres in s1. This will be relevant in §2 when I address the causal adequacy
principle.
With transference clarified, Leibniz’s transference argument against creaturely
transeunt causation can be expressed thus:
(P1)
(P2)
(C)
Creaturely transeunt causation is possible only if the transference of
accidents is possible.
The transference of accidents is not possible.
So, creaturely transeunt causation is not possible.
I address each premise in turn, focusing on Leibniz’s reasons for (P1) in this chapter and
(P2) in chapter 3.
§2 Leibniz’s Defense of the Transference Condition
(P1), a restatement of the transference Condition, is the most controversial premise in
Leibniz’s transference argument. Appropriately, I’ll devote the rest of this chapter to
93
This is in contrast to the recent persistence theory of causation defended by S.D. Rieber who claims
“causation is nothing more than a property moving from one object to another.” See S.D. Reiber,
“Causation as Property Acquisition,” Philosophical Studies: An International Journal for Philosophy in the
Analytic Tradition 109, no. 1 (2002): 53-74.
36
addressing what reasons Leibniz could and does give in support of it. The controversy
concerning the transference condition is three-fold. First, most philosophers who
defended the possibility of creaturely transeunt causation denied the transference
condition. Second, Leibniz offers very little in defense of the transference condition
himself. Instead, he oftentimes just asserts the transference condition, as noted in the
passages above. Third, God is a transeunt cause in Leibniz’s metaphysics, and yet divine
transeunt causation (unlike creaturely transeunt causation) does not consist in
transference. Thus, Leibniz needs principled reasons for holding that creaturely transeunt
causation must consist in transference while divine causation does not.
2.1 Philosophers who denied the Transference Condition
Many philosophers who not only devoted significant space to defending the possibility of
creaturely transeunt causation but also developed sophisticated accounts of it vehemently
denied the transference condition and therefore the truth of (P1). For example, in an
often-quoted passage, where Aquinas is arguing against a medieval version of
occasionalism that assumed something like the transference condition, Aquinas writes:
Again, it is laughable to say that a body does not act because an accident does not
pass from subject to subject. For a hot body is not said to give off heat in this
sense, that numerically the same heat which is in the heating body passes over
into the heated body.94
Aquinas denies the very same thing we saw Leibniz deny above— that an accident which
belonged to one subject could be detached and sent to a different subject. Instead,
Aquinas argues that such creaturely transeunt causation—such as a hot body heating a
different body—occurs through what Scholastics called “eduction”:
94
Aquinas, SCG, Bk. III, Pt. 1, Ch. 69, 28.
37
Rather, by the power of the heat which is in the heating body, a numerically
different heat is made actual in the heated body, a heat which was previously in it
in potency. For a natural agent does not hand over its own form to another
subject, but it reduces (reducens) the passive subject from potency to act.95
Roger Bacon, who defended a similar account of transeunt causation known as the
“Multiplication of Species” account, also denied that creaturely transeunt causation
consists in transference. Bacon writes, “Acting does not destroy and corrupt an agent,
but perfects it, since . . . a thing is perfect when it is able to produce a like thing.”96
Bacon’s denial of the acting of an agent destroying or corrupting it just seems to be a
denial that in acting, an agent loses any of its own being. Focusing on species, when an
agent produces a species in a recipient, the agent does not transfer the species, but rather,
by a “bringing forth out of the active potentiality in the recipient matter”, the species is
generated.97 Bacon’s denial is noteworthy because the language Leibniz often uses in
describing transference—such as the transmission of species—is heavily drawn from
Bacon’s account of transeunt causation.
Finally, Suarez—from whom Leibniz gets the term “influx”—also denied that
creaturely transeunt causation consists in transference. Suarez writes:
The efficient cause . . . causes by means of a proper action that flows from it.
And in this it is also included that the efficient cause does not give its own proper
and formal esse to the effect, but instead gives another esse that emanates from it
by means of an action . . . The efficient cause . . . is an extrinsic cause, that is, a
cause that does not communicate its own proper and (as I will put it) individual
esse to the effect but instead communicates to it a different esse, which really
flows forth and emanates from such a cause by means of an action.98
95
Ibid.
RB 45.
97
RB 53.
98
DM 17.1.6. See also DM 12.2.7.
96
38
Here Suarez is explicit—the being (esse) that the cause gives to the effect is not any of
the efficient causes own being. Thus, the efficient cause does not lose any of its own
being in producing an effect.
2.2 Was Leibniz arguing against a strawman?
The similarities between the language Leibniz uses in his transference argument and the
terminology of Aquinas’s, Bacon’s, and Suarez’s accounts of creaturely transeunt
causation are not at all surprising. Aquinas’s, Bacon’s, and Suarez’s accounts were
heavily influential and well known in Leibniz’s time. Prima facie, Leibniz was arguing
against a straw man. He argues that creaturely transeunt causation is not possible
because transference is not possible, and he often describes such transference as the
transmission of species, or accidents detaching, or influx. This leads to a dilemma:
Either Leibniz did not understand the dominant accounts of transeunt causation or he did
understand but nonetheless argued against a caricatured account.
A passage in one of Leibniz’s earlier writings—his 1670 “Preface to an Edition of
Nizolius”—calls into doubt the first horn of the dilemma. Criticizing Suarez’s account of
causation, Leibniz writes:
. . . so far as we have shown that technical terms are to be avoided as far as
possible. Now we must note that whether terms are popular or technical, they
ought to involve either no figures of speech or few and apt ones. Of this, the
Scholastics have taken little notice, for strange though this sounds, their speech
abounds with figures. What else are such terms as to ‘depend’, to ‘inhere’, to
‘emanate’, and to ‘inflow’ (influere)? On the invention of this last word Suarez
prides himself not a little. The Scholastics before him had been exerting
themselves to find a general concept of cause, but fitting words had not occurred
to them. Suarez was not cleverer than they, but bolder, and introducing
ingeniously the word influx (influxus), he defined cause as what flows being into
something else, a most barbarous and obscure expression. Even the construction
is inept, since influere is transformed from an intransitive into a transitive verb;
and this influx is metaphorical and more obscure than what it defines. I should
39
think it an easier task to define the term ‘cause’ than this term influx, used in such
an unnatural sense.99
In this passage, the young Leibniz indicates that he is aware that terms such as ‘influx’
and ‘inflow’ have a metaphorical meaning when used in defining efficient causation.
Leibniz’s criticism of the usage of such terms is that he argues that one should not use a
metaphorical term at the ground level of one’s metaphysics. This is especially the case in
defining a term such as ‘cause’, for using a metaphorical term obscures, rather than
clarifies causation. With the above passage as evidence, I herewith suggest that Leibniz
was aware that scholastics such as Suarez do not literally mean that the agent gives a part
of its being to the patient. However, if Leibniz understands that causation does not
consist in transference in sophisticated and influential accounts of transeunt causation
(such as Suarez’s), then Leibniz is in a worse position, leading to the second horn of the
dilemma: Leibniz knowingly argued against a caricatured understanding of transeunt
causation.
In response to this second horn, Eileen O’Neill—who in her own work has
contributed much towards understanding Leibniz’s transference condition—draws
attention to two points in Leibniz’s favor indicating that he was not arguing against a
straw man but rather thought that transeunt causation was committed to transference.100
First, in spite of all the talk about eduction and the strongly worded denials of
transference from scholastics such as Bacon, Aquinas and Suarez, they use the language
99
L 126.
See Eileen O’Neill, “Influxus Physicus,” in Nadler, Steven ed. Causation in Early Modern Philosophy
(University Park: Penn State Press, 1993), 27-56.
100
40
of transference when describing body-to-body causation. Take the following quote by
Suarez on sensible species:
It is known by many experiences that species shoot forth from an object. The first
is, because we see ourselves in another’s pupil, which cannot be understood to
happen otherwise than by some little form which represents me having been
impressed on the other’s pupil.101
Second, when describing causation between entities of a different nature, such as bodies
and minds, the scholastics often use language suggesting occasionalism or preestablished harmony— further implying a denial of transeunt causation in these cases
because transference between entities of a different nature is difficult to conceive. Again,
Suarez writes:
The phantasm and also the intellect of man are rooted in one and the same soul.
For, here it turns out that they have a wonderful order and agreement in their
operation, whence . . . for the same reason that the intellect operates, the
imagination also senses. Therefore, in this way, I think . . . there is spiritual force
in a rational soul for bringing about, in the possible intellect, species of these
things . . . , while sensible cognition itself does not at all concur efficiently to that
action.102
According to O’Neill, Leibniz takes the scholastics’ reverting to the language of
transference when discussing interaction between entities of similar natures or Preestablished Harmony/occasionalism when discussing interaction between entities of
different natures as evidence that Leibniz thought they ultimately have no intelligible
notion of creaturely transeunt causation. Instead, transference at the phenomenal level of
description—bodies—is intelligible and the only conceivable model we have. At the
101
Francisco Suarez, Opera Omnia, edited by L. Vives (Paris, 1856-78), V 164, quoted in O’Neill,
“Influxus Physicus,” 49.
102
Suarez, Opera Omnia, V 719, quoted in O’Neill, “Influxus Physicus,” 50.
41
metaphysical level of description of simple substances, there is no conceivable model of
creaturely transeunt causation.103
2.3 Divine transeunt causation and transference
Suppose Leibniz’s argument is as O’Neill suggests it is— that creaturely transeunt is
unintelligible if no model can be given for it and there is no way of describing it without
using language suggestive of either transference (between entities of similar natures) or
pre-established harmony/occasionalism (between entities of different natures). Suppose
further that such unintelligibility is evidence of impossibility. Alfred Fredosso has a
response: “. . . it is better to have mysteries emerge at the end of one’s investigation into
an obvious starting point than to deny the obvious starting point itself – in this case, the
reality of action as an observable, basic primitive.”104 If Fredosso is right, then creaturely
transeunt causation ultimately bottoms out in a mystery or a primitive. While
dissatisfying, especially to one like Leibniz who places a high premium on explanation,
such mystery does not entail impossibility.105 Figures such as Suarez might respond that
the cost of rejecting something as obvious as creaturely transeunt causation is too high of
a price to pay to avoid a mystery or primitive at some level of one’s explanation.
In this chapter, I do not aim to resolve this particular issue, as doing so requires a
detailed examination of Leibniz’s strictures on metaphysical explanation and its
connection to possibility.106 Instead, I turn here to what I argue is a more pressing
problem for Leibniz’s Transference argument: Leibniz is prima facie guilty of having a
103
Ibid., 52.
Fredosso, MD 20-22., xlix.
105
At least on the strictures of explanation endorsed by figures such as Aquinas, Bacon, Suarez, and others.
106
I address in greater depth Leibniz’s strictures on explanation in chapter 5.
104
42
very similar mystery emerge in his own metaphysics of causation, specifically in cases of
divine transeunt causation. Continuing his critique of Leibniz and others who denied
creaturely transeunt causation, Fredosso argues that philosophers such as Leibniz who
deny creaturely transeunt causation while affirming divine transeunt causation must
answer the question they themselves pose to defenders of creaturely causation: “What
does God’s transeunt action consist in?”107 This question is important for Leibniz, as
divine transeunt causation is not just a mere possibility but a crucial element of his
systematic metaphysics. There would not even be created substances without God’s
creation and conservation, and these are kinds of transeunt causation. Such divine
transeunt causation does not, because it could not, consist in transference. That would
entail God losing some of His being— an impossibility on the classical conception of a
simple God. Leibniz then needs a principled way to argue that creaturely transeunt
causation must consist in transference while divine transeunt causation does not. In what
follows, I develop a way that Leibniz could take up this challenge.
2.4 The Argument for the Transference Condition
Leibniz needs reasons to defend the transference condition or (P1) of his transference
argument:
(P1)
Creaturely transeunt causation is possible only if the transference of
accidents is possible.
But the reasons for (P1) must not entail that divine transeunt causation consists in
transference or that the transference condition is a necessary condition of divine transeunt
causation. The reasons for the transference condition instead must be consistent with
107
Ibid., xlviii.
43
denying that divine causation consists in transference. Further, rather than ad hoc
reasons, Leibniz needs principled reasons for affirming (P1) while denying that divine
causation consists in transference. I argue that Leibniz has such reasons.
My argumentative strategy is as follows. I first articulate three key theses or
constraints on creaturely transeunt causation that were endorsed and utilized by notable
medieval and early figures in their accounts of creaturely transeunt causation. These
theses are connected to one another and follow upon one another in a logical progression
of sorts. From these three theses, I develop an argument for (P1) that I contend was
lurking behind Leibniz’s transference condition. The argument for (P1), as I shall show,
is also consistent with denying that divine transeunt causation consists in transference. In
fact, the argument is an argument for why specifically creaturely transeunt causation
must consist in transference. After developing the argument, I argue in 2.5 that it was the
argument behind Leibniz’s repeated statements of the transference condition.
I note that the argument for (P1) I develop and defend as Leibniz’s is drawn from
what is at most a very thin skeleton in Leibniz’s writings. In this section, then, I trudge
into the terrain of philosophical history of philosophy. I do so with the aim, however, of
serving explanatory exegetical history of philosophy. My aim is to explain why Leibniz
repeatedly endorsed the transference condition. I note as well that I do not conclude that
Leibniz ultimately had a successful argument for (P1). In fact, I draw attention to a
significant weakness with it in §3. Instead, I aim to make sense of why Leibniz thought
(P1) was true while denying that divine causation required transference.
The first thesis is what I’ll call the “Causal Adequacy Principle”. The causal
adequacy principle was not merely endorsed by both medieval and early modern
44
philosophers but was central to some of their most important metaphysical views.
Expressed in various ways, the causal adequacy principle required that effects pre-exist,
somehow, in their causes or that causes must, somehow, contain their effects. The
principle traces back at least to Aristotle, who argued that something that is potentially F
can only be made actually F by something that is or contains the actuality of F.108 Many
centuries later, we find Descartes—known for defending a metaphysics which is largely
anti-Aristotelian—putting the principle to use as a key premise in his causal argument for
God’s existence in his third meditation. Descartes writes, “there must be at least as much
reality in the efficient and total cause as in the effect of that cause.”109
In between Aristotle and early modern figures such as Descartes, we find the
principle explicitly defended by some medieval philosophers. For example, Aquinas
endorsed and utilized the principle both in his own metaphysics of causation and in other
important areas of his metaphysics, including his account of the divine nature. A classic
statement of the principle is found in Part I of the Summa Theologiae: “For effects
proceed from the agent that causes them, in so far as they pre-exist in the agent; since
every agent produces its like.”110 Centuries later, Suarez endorses the very same
principle, writing, “It is proved that nothing of perfection is in the effect that it does not
108
Aristotle writes, “So far as the things formed by nature or by human art are concerned, the formation of
that which is potentially is brought about by that which is in actuality; so that the Form, or conformation, of
B would have to be contained in A.” On Generation and Corruption, 734a30-32.
109
CSM II, 28.
110
ST 1, q19, a4. Commenting on Aquinas’s usage of the principle, John Wippel writes, “This [the causal
adequacy principle] is to say simply that the agent has the power to produce the effect.” Metaphysical
Thought of Aquinas, 490
45
have from its causes”111 and “The effect can have no perfection which does not pre-exist
in some of its causes.”112
One might suppose that the causal adequacy principle is all Leibniz needs to
argue against the possibility of creaturely transeunt causation instead of the more
complicated transference argument. Consider the familiar but philosophically knotty
cases of transeunt causation between substances of vastly different natures, such as mindbody causation in the context of Cartesian metaphysics.113 Suppose, for example, some
mind efficiently caused a body to change its shape, such as when—within the
metaphysics of Cartesian substance dualism—Socrates decides to rise from sitting and
changes his shape. In this case, Socrates’ mind produced a new accident in his body— a
new shape. Given the causal adequacy principle, Socrates’ mind can only produce a new
accident in his body—a new shape—if the accident pre-exists in his mind. However, a
shape is an accident appropriate for an extended thing while the accidents of a mental
substance—such as Socrates’ mind—would be accidents appropriate to non-extended
mental things, such as beliefs and desires. The burden for defenders of transeunt
causation between entities of very different natures—such as Descartes—was to explain
how a mental substance could cause a change in a bodily substance (or vice versa) when
the mental substance lacked the accidents it caused in the body. The challenge, that is,
was to explain how Socrates’ mental substance—which willed the standing—produced
111
DM XXVI.1, quoted in Tad Schmaltz, Descartes on Causation, 52.
DM 4 XXVI 1.2
113
According to Descartes, there are two fundamentally different kinds of created substances: bodies and
mental substances. The essence of bodily substances is extension and all changes in bodily substances are
grounded in changes in their extension. The essence of mental substances, on the other hand, is thought.
Descartes’ most rigorous articulation of his metaphysics is found in his Principles of Philosophy. See CSM
I, 177-292.
112
46
the change in shape in his body when his mental substance isn’t shaped or capable of
being shaped.
However, in addition to the fact that it takes us no further towards understanding
Leibniz’s reasons for (P1), there are at least two reasons why the causal adequacy
principle is not sufficient to argue even against the possibility of transeunt causation.
Clearly, the argument against transeunt causation from the causal adequacy principle, as
presented above, gains purchase only only in cases of causation between entities of
different natures, such as minds and bodies. It would not apply to entities of the same
nature, such as cases of mind/mind or body/body causation. An extended substance
would be a suitable candidate to produce an accident in a different extended substance,
unlike a non-extended substance, because the agent extended substance itself possesses
the types of accidents it causes in the patient. Leibniz, however, is adamant that transeunt
causation is not possible between any created substances.
Second, the few brief passages and argument given above present an overly
simplistic and grossly misrepresentative version of the causal adequacy principle. As
initially presented, the causal adequacy principle requires that a cause have an accident of
the exact same type it causes in the effect, whether it be numerically the same (as Leibniz
would argue) or a different token of the same type. This is plausible in causes of what
the scholastics call univocal causation, such as fire heating a pan.114 But it scarcely
applies to cases such as fire’s heat hardening clay, where the cause does not have an
114
See DM 17.2 for Suarez’s discussion of univocal causes.
47
accident as the same type as the effect. A more pressing burden for many theists would
be to explain how a non-extended God could create extended corporeal substances.
Instead, the simplistic version of the causal adequacy principle that entails the
impossibility of causal interaction between substances of different natures is not the
version of the principle that figures such as Aquinas, Suarez, and Descartes endorsed,
even though they sometimes wrote as if it were. Instead, they endorsed a more refined
version of the principle that is consistent with causal interaction between substances of
different natures. Thus, Aquinas writes:
I answer that, all created perfections are in God. Hence He is spoken of as
universally perfect, because He lacks not (says the Commentator, Metaph. v) any
excellence which may be found in any genus. This may be seen from two
considerations. First, because whatever perfection exists in an effect must be
found in the effective cause: either in the same formality, if it is a univocal
agent—as when man reproduces man; or in a more eminent degree, if it is an
equivocal agent—thus in the sun is the likeness of whatever is generated by the
sun's power.115
From Suarez:
An effect cannot exceed in perfection all of its causes taken together. It is proved
that nothing of perfection is in the effect that it does not have from its cause;
therefore the effect can have nothing of perfection that does not pre-exist in any of
its causes, either formally or eminently, because causes cannot give what they in
no way contain.116
And Descartes:
A stone, for example, which previously did not exist, cannot begin to exist unless
it is produced by something which contains, either formally or eminently,
everything to be found in the stone [i.e. it will contain in itself the same things as
are in the stone or other more excellent things; similarly, heat cannot begin to
exist unless it is produced in an object which was not previously hot, except by
115
116
ST 1, q4, a2, emphasis added.
DM XXVI.1, quoted in Tad Schmaltz, Descartes on Causation, 52. Emphasis added.
48
something of at least the same order (degree or kind) of perfection as heat, and so
on.117
Aquinas, Suarez, and Descartes all argue that the cause must formally or eminently
contain the effect. The requirement of causes formally containing the effect applies to
cases of univocal causation and is what was assumed in the initial argument against
transeunt causation from the causal adequacy principle. That requirement drove figures
such as Descartes’ pupil Princess Elizabeth to question the possibility of transeunt
causation between substances of different natures. In 1643, she writes to Descartes, “I
admit that it would be easier for me to concede matter and extension to the mind than it
would be for me to concede to an immaterial thing the capacity to move the body and be
moved by one.”118 Elizabeth assumed a version of the causal adequacy principle such as
the following:
For any substance s1 and any substance s2 and any accident A, s1 efficiently causes
A to inhere in s2 only if s1 formally contains A.
But as the texts above make clear, Aquinas, Suarez, Descartes, and many others in fact
affirmed a more generous version:
For any substance s1 and any substance s2 and any accident A, s1 efficiently causes
A to inhere in s2 only if s1 formally contains A or s1 eminently contains A.
Formal containment is fairly easy to understand in contrast to eminent containment. Fire
can heat a metal pan because the fire itself is hot.119 A precise formulation of eminent
containment is more difficult to offer. There is indeed much disagreement on how a
117
CSM II, 28. Emphasis Added.
AT iii, 661
119
Formal containment can be expressed thus: For any substance s and any accident A, s formally contains
A if and only if there is an accident F such that F is the same type of accident as A and F inheres in s. In
this definition, A and F may be numerically distinct or identical. Many texts suggest that Leibniz thought
A=F.
118
49
cause can eminently contain an effect.120 Rather than shoulder that burden, it will suffice,
for the purposes of this section, to briefly draw attention to a feature of eminent
containment that most philosophers familiar to Leibniz would have recognized. When an
effect is eminently contained in its cause, the effect exists in the cause, in some “higher
way” than in the effect.121 How, though, can an effect exist in a higher way in its cause?
While various answers have been posed and defended, all answers agree that it is because
the cause is more perfect or the cause has more reality than the effect. For example,
Suarez writes, “Nevertheless, it should be briefly said that to contain eminently is to have
such a perfection of a superior ratio, which virtually contains whatever is in the lower
perfection.”122
Notably, at least two philosophers influenced by Descartes—Malebranche123 and
Spinoza124—rejected eminent containment because of difficulties in giving a precise
120
Most of the discussion has centered around Descartes’ usage of the notion of eminent containment. See,
for example, Kenneth Clatterbaugh, The Causation Debate in Modern Philosophy and 17-46, Tad M.
Schmaltz, Descartes on Causation, 49-86.
121
See, for example, Aquinas, SCG I, 30 and ST 1, q13, a4. See also Descartes, CSM II, 30. Suarez gives
the lengthiest defense and articulation of eminent containment that I have found. See DM XXX.1.
122
DM XXX 1.10.
123
Consider the following passage from Malebranche (where a univocal cause formally contains the effect
and an equivocal cause eminently containst the effect) in LO 277: “And if one objects against their false
and incomprehensible suppositions that fire must be composed of very highly agitated particles because it
produces such violent motion, and that a thing cannot communicate what it does not have (which is
certainly a very clear and well founded objection), they never fail to confuse everything by some frivolous
and imaginary distinction (such as that between equivocal and univocal causes) in order to appear to say
something when in effect they have said nothing. For at bottom it is a common notion among attentive
minds that there is no such thing in nature as a true equivocal cause (in the sense they understand it) and
that only the ignorance of men has invented them.”
50
account of it. Spinoza’s rejection of eminent containment, and thus likely affirmation of
the stricter version of the causal adequacy principle, would suffice to secure the falsity of
transeunt causation between substances of different natures (but not suffice alone to
secure the falsity of causal interaction between substances of the same nature). Leibniz,
however, could not have appealed to that option in arguing against creaturely transeunt
causation. In addition to that version of the causal adequacy principle still not entailing
the falsity of causal interaction between substances of the same nature, Leibniz himself
assumed eminent containment, at least in cases of divine causation. In the Discourse on
Metaphysics, Leibniz wrote, “This simple primitive substance [God] must eminently
(eminemment) include the perfections contained in the derivative substances which are its
effects.”125
The causal adequacy principle alone then does not provide Leibniz with sufficient
reason to deny the possibility of creaturely transeunt causation. The principle will,
however, provide Leibniz with support for the first premise of his transference argument
and it does so in a way consistent with denying that divine causation consists in
transference. To see the role the principle plays in the argument, we must see why
philosophers endorsed the causal adequacy principle.
124
Spinoza denies transeunt causation between substances of different natures in his Ethics, writing “When
things have nothing in common, one cannot be the cause of the other.” See Ethics 1, Proposition 3.
Spinoza seems to give epistemological reasons for Proposition 3 in the Ethics, writing, “If things have
nothing in common, then (Ax. 5) they cannot be understood through one another, and so (Ax. 4) one cannot
be the cause of the other. (Ethics I, prop. 3.) However, Francesca di Poppa has recently argued that the
proposition in the Ethics is ultimately due to problems Spinoza found with eminent containment, leading to
a metaphysics where all modes are formally contained in God, who immanently causes them. See
Francesca di Poppa, “God Acts from the Laws of His Nature Alone: From the Nihil Ex Nihilo Axiom to
Causation as Expression in Spinoza’s Metaphysics,” (PhD Diss., University of Pittsburgh, 2006).
125
See (G VI.602: AG 210). See also (G VI.613: AG 218).
51
Recall the passage from Suarez where he writes, “. . . the effect can have nothing
of the perfection that does not pre-exist in any of its causes, either formally or eminently,
because causes cannot give what they in no way contain.” The causal adequacy principle
is the constraint expressed before the word ‘because’. Granted, Suarez expresses the
principle somewhat differently that I have, requiring that the perfection of the effect preexist or be contained in the cause, while I my formulation of the principle requires the
accident to pre-exist in the cause. I’ll continue to use the term accident. I do so,
however, while noting that scholastics such as Aquinas and Suarez tended to use the term
‘perfection’ (perfectiones) in expressing the causal adequacy principle while Descartes
tended to use the terms ‘reality’ and ‘perfection’ interchangeably.126 However, my usage
of the term ‘accident’ will not impact the argument, thus I’ll assume the following biconditional:
For any substance s and any accident A, s formally contains A or s eminently
contains A if and only if s contains the perfection of A.
If s contains A, surely s contains the perfection of A. For similar reasons, I’ll also assume
the bi-conditional:
For any substance s and any accident A, s formally contains A or s eminently
contains A if and only if s contains the reality of A.
Notice then that in the passage from Suarez, whatever is written after the word ‘because’
is a reason for the causal adequacy principle. The phrase Suarez uses is “causes cannot
give (dare) what they in no way contain”, expressing what I’ll call the “Causing-as-
126
The term “reality” has a nuanced technical meaning in Leibniz’s metaphysics, which I soon address in
§3 in this chapter and in much greater depth in chapter 4.
52
Giving” principle, the second thesis.127 The causing-as-giving principle is a reason for
the causal adequacy principle and the causal adequacy principle is a necessary condition
of the causing-as-giving principle. This relation between the two principles is expressed
in other ways, most notably as “nihil dat quod non habet.” If some pan heats some
water— i.e., the pan gives heat to the water, then the pan must itself contain heat. To put
it slightly differently: what reason might one offer for the causal adequacy principle?
The answer is the causing-as-giving principle. If x gives F to y, x must possess F. I can’t
give my sister fifty dollars for her birthday unless I have fifty dollars. The same
restriction applies to causal transactions.128
Leibniz can agree with Suarez and others that the causing-as-giving principle is a
reason for the causal adequacy principle and that the causal adequacy principle is a
necessary condition of the causing-as-giving principle. That is, Leibniz can agree that:
For any substance s1 and any substance s2 and any accident A, s1 gives the reality
of A to s2 only if s1 formally contains A or s1 eminently contains A.
But Leibniz holds further that once you give what you have, you longer possess it
yourself. While Leibniz doesn’t explicitly make the following claim, there are plausible
reasons to suppose that it is lurking behind his repeated assertions of the transference
condition:
For any created substance s1 and any created substance s2 and any accident A, s1
gives the reality of A to s2 only if s1 transfers A from s1 to s2.
127
Aquinas writes, “In De divinis nominibus Dionysii 4.5: “For these three things seem to belong to the
notion of an efficient cause: to give being, to move, and to conserve.” Quoted in Michael Rota,
“Causation,” in The Oxford Handbook of Aquinas, edited by Brian Davies & Eleonore Stump (Oxford:
Oxford University Press, 2012), 104-114.
128
In DM 12.2.3, Suarez argues that a cause’s “pouring being into another” is synonymous with the cause
“giving or communicating being to another thing.” Quoted in Stephan Schmidt, “Efficient Causality: The
Metaphysics of Production,” 85.
53
An objection immediately comes to mind with respect to arguing that the causing-asgiving principle entails the transference condition. It’s not necessarily the case that if you
give something, you no longer possesses what you gave. For example, Paul’s giving
Timothy advice does not entail that Paul no longer possesses that advice. Yet Paul can’t
give Timothy advice unless Paul possesses, in some sense, the advice he gives. Thus,
Paul’s giving advice must satisfy a constraint similar to the causal adequacy principle
while at the same time not satisfying a constraint such as the transference condition. So
giving what one has doesn’t entail that one loses what they give.
Leibniz would have to respond that there is an equivocal meaning to ‘give’
between ‘giving’ advice and ‘giving’ money, where the giving in causation is like the
latter rather than the former. An immediate objection is that this move is ad hoc. Leibniz
could respond, however, that it is not ad hoc, as the crucial difference between giving
advice and the notion of giving in the causing-as-giving principle lies crucially in what is
given. In cases of causing-as-giving, what is given are accidents— genuine beings that
the patient substance gains.129
One might further object that God can give an accident to a patient without God
losing the reality of that accident. Moreover, God’s giving an accident is not necessarily
equivocal in meaning to a created substance’s giving an accident, unlike Paul’s giving
Timothy advice. This returns us to Fredosso’s objection: If God can give the reality of
an accident to a patient substance without God losing what he gives, why can’t a created
substance do the same?
129
I address in greater depth how to understand that accidents are beings in the third part of this chapter and
throughout chapters 3 and 4.
54
There are reasons available to Leibniz—which I shall argue were his reasons in
2.5—that address both objections: The reason causing-as-giving in transeunt causation is
more like the giving of money when the giver is a created substance while God’s giving
is more like the giving of advice is that only God can create. In support of this claim, I
turn to the third thesis—in addition to the causal adequacy principle and the causing-asgiving principle—that applies to transeunt causation, what I’ll call the “No-Reality-fromPatient” principle. The No-reality-from-patient principle is the common sense principle
that the patient does not contribute reality to the effect in transeunt causation, where the
reality of the effect is the perfection of the effect or more simply put, the accident
produced. While this principle needs little defense, it’s helpful to look at least one
example of a philosopher who affirmed it. In Aquinas’s account of an agent substance’s
educing an effect in a patient, the agent contributes actuality to the effect, while the
patient does not. Instead, the patient supplies the potency for the effect, which is reduced
to act by the agent. A pan must actually be hot if it is to produce heat in water that is
only hot in potency. In the following passage, Aquinas defends the no-reality-frompatient principle:
Now it is plain that the effect pre-exists virtually in the efficient cause: and
although to pre-exist in the potentially of a material cause is to pre-exist in a more
imperfect way, since matter as such is imperfect, and an agent as such is perfect;
still to pre-exist virtually in the efficient cause is to pre-exist not in a more
imperfect, but in a more perfect way.130
While Aquinas is contrasting the perfection contributed by the efficient cause with the
perfection contributed by the material cause, and while he does not use the term
130
ST 1, q4, a2
55
“patient”, his claims supports the no-reality-from-patient principle. The efficient cause is
the agent and the agent contributes a form that inheres in matter— in this case either in
prime matter (if the agent contributes a substantial form) or in the patient substance (if
the agent contributes an accidental form). If the effect exists in a more imperfect way in
the patient or material cause, the patient or material cause surely cannot be what
contributes the perfection—or in other words the reality—of the accident. I note further
that while Descartes does not avail himself of the act/potency distinction, his description
of the causal adequacy principle conveys basically the same idea. That is, in transeunt
causation, the substance acted on is not the source of the reality of the effect.
Further support for the no-reality-from-patient principle can be found in how the
causal-adequacy principle is often expressed. For example, Suarez writes, “An effect
cannot exceed in perfection all of its causes together. It is proved that nothing of
perfection is in the effect that it does not have from its cause.”131 Suarez’s statement is
consistent with the effect exceeding in perfection the patient substance of a transeuntly
caused accident. The reason is that the patient does not contribute perfection to the
effect. If perfection is co-extensive with reality, then the patient does not contribute
reality to the effect. The no-reality-from-patient principle can then be stated thus:
For any substance s1 and any substance s2 and any accident A, if s1 efficiently
causes A to inhere in s2 and it is not the case that s2 efficiently causes A to inhere
in s2, then it is not the case that s2 gives the reality of A to s2.
It is from these three principles that Leibniz has an argument for (P1) that is consistent
with denying that divine transeunt causation must consist in transference. Leibniz would
131
DM XXVI.1
56
argue that those who defend the possibility of creaturely transeunt causation could not
endorse the causal adequacy principle, causing-as-giving principle, and no-reality-frompatient principle while at the same time rejecting the transference condition. While he
never supplies it, his reasoning could be as follows: If the patient contributes no reality
to the effect and the agent does not lose the reality it contains when it gives the reality of
the effect to the patient, then the reality of the accident caused to inhere in the patient
must have been created.132 However, only God can create, as only an omnipotent being
can create and only God is omnipotent. So one of these must be rejected by defenders of
creaturely transeunt causation: the causal adequacy principle, the causing-as-giving
principle, the no-reality-from-patient principle, or the denial of the transference
condition. If you give up the denial of the transference condition and instead affirm (P1),
then the production of an accident in a patient is not a case of creation. So the denial of
(P1)—the denial of the transference condition for created substances—is the culprit.
Affirm, rather than deny, the transference condition and you no longer have created
substances themselves creating. Leibniz thus has an argument for the transference
condition and (P1) of the transference argument that is not only consistent with holding
that divine causation does not consist in transference but also makes sense of it.
132
I note that Thomas Pinkston, in his dissertation on Suarez’s account of efficient causation, claims that
efficient causation for Suarez—even when the cause is a created substance—is a type of creation. Pinkston
writes: “Moreover, inasmuch as the efficient cause makes to be something that was not, and, inasmuch as
the esse of the cause is strictly separated from that of its effect, one might reasonably speak of a cause
“creating” its effect. Obviously, such a use of ‘create’ excludes any connotation of supernatural ex nihilo
creation. Nevertheless, use of the term, suitably qualified, underscores the radical nature of efficient
causality as understood by most scholastics, including Suarez.” As will be obvious in the following pages,
the argument I give on behalf of Leibniz can be understood to press Pinkston (or Pinkston’s reading of
Suarez) on just how the transeunt causation of accidents can satisfy the causal adequacy principle, causingas-giving, and the no-reality-from-patient principle while also denying the transference condition, and not
being the type of creation traditionally understood to require omnipotence. See Pinkston, 56.
57
2.5 The argument for the Transference Condition is a Leibnizian Argument
While it cannot be stated with certainty, two broad reasons point in favor of the argument
developed in 2.4 being Leibniz’s reasons for (P1). First, the argument for (P1) makes
sense of why Leibniz thinks creaturely immanent causation is possible even though
creaturely transeunt causation is not, something which to date has not been dealt with in
the scholarly literature. Specifically, the argument makes sense of how the causal
adequacy principle can be a constraint on immanent causation without immanent
causation requiring the transference or creation of accidents. Take a created substance s
immanently producing an accident A, so that A inheres in s. On the causal adequacy
principle, s must have contained or possessed the reality of A prior to causing A to inhere
in itself. The reality of A did not move from s to another substance when s produced A.
The reality of A also didn’t just pop into being from another substance, nor did it just pop
into being it itself. Instead, the reality of A was already in s. So the production of A
resulted in no net gain of reality in s.
Using a crude but appropriate metaphor given our discussion of the causal
adequacy principle, s is a container, and the reality of A never left the container, nor did
new reality appear in the container when A was produced, in addition to the reality
already present in s. Instead, the reality was “moved around” but within the same
substance.133 Were s to efficiently cause A to inhere in a different substance s’, and were
133
In the Theodicy, Leibniz claims as much, writing, “. . . substances produce accidents by the changes of
their limits.” The nature of accidental reality, how accidents are connected to their substances, and how
substances immanently produce their accidents will occupy a significant portion of this dissertation,
including the next section, Chapter 3, and Chapter 4. See (G VI.351: T 395). For similar claims by
Leibniz, see (G II.257: L 532) and (G II.270: AG 180).
58
this case of transeunt causation to satisfy the causal adequacy principle, causing-as-giving
principle, no-reality-from-patient principle, and the denial of transference, then Leibniz
would call that creation. There would be a net gain in reality in the patient substance—
the reality of the accident produced by the agent—with no loss of the agent’s being.
Hence, Leibniz can say that creaturely immanent causation is possible because it requires
neither creation nor transmission, unlike transeunt causation.134
Second, the argument for the transference condition has historical precedent in
philosophers not merely familiar to Leibniz but whom he claimed to have influenced
important areas of his thought. Notably, the argument for the transference condition
makes sense of two important areas of agreement between Leibniz and the occasionalists:
(i) creaturely transeunt causation is impossible while (ii) divine transeunt causation is
possible. With respect to (i), Malebranche—the most well known occasionalist and one
who corresponded with Leibniz—also argued that one influential account of creaturely
transeunt causation was inconceivable without transference. Arguing against the
134
In conversation, Jeffrey Brower has raised an important objection to this account of immanent causation.
Brower argues that Leibniz faces a dilemma with respect to the causal adequacy principle, the possibility of
immanent causation, and the defense of the transference condition. Take again a created substance s
immanently causing an accident A to inhere in itself. If s contains A, as the causal adequacy principle
requires, then s either formally contains A or eminently contains A. If s formally contains A, then s didn’t
really change, as what it is for a substance to formally contain an accident just is for the accident (or an
accident of the same type) to inhere in and be an accident of that substance. If s eminently contains A, on
the other hand, then s created A.
I concede the first horn: If s formally contained A, then in immanently producing A, s didn’t really
change. The next section of this chapter and the next two chapters do deal with topics relevant to the
second horn of the dilemma raised by Brower, including what kind of reality accidents have, what kind of
reality substances have, how they are related, and how a substance changes its accidents. In the next
section of this chapter, I draw attention to a similar problem: Leibniz’s argument for the transference
condition ultimately only applies to non-modal accidents, such as real qualities. It does not, I argue, apply
to modifications—the only kind of accidents Leibniz posits in his ontology, as I argue in Chapter 3. Were
Leibniz to hold that monads produce non-modal accidents, such as real qualities, then the dilemma applies:
Either the accident would be created or the monad would not change. Modifications, Leibniz argues, have
no reality of their own but instead are limitations on the positive reality of a monad, thus the immanent
causation of a modification by a monad would not result in a net gain in reality.
59
multiplication of species model of causation developed at length by Roger Bacon,
Malebranche wrote in The Search After Truth III.2.2, “Finally, it is inconceivable how a
body that does not sensibly diminish could continually emit species in all directions, or
how it could continually fill the vast spaces around it with them—and all this with
inconceivable speed.”135 Malebranche finds it puzzling that a body (the causal agent)
could emit species—the first effect of a natural cause in Bacon’s account—without the
body diminishing. In other words, Malebranche finds it inconceivable how a body could
produce a species distinct from it without losing some of its own being. So Malebranche
argued that a major account of transeunt causation was inconceivable without
transference. We find here Malebranche endorsing a premise very similar to (P1) of
Leibniz’s transference argument. Malebranche, of course, also agreed with the
conclusion of the Transference argument—that creaturely transeunt causation is
impossible.136
For a different but related line of evidence that the argument for (P1) expresses
Leibniz’s reasons for the transference condition, and one again based on Leibniz’s
agreement with Malebranche, I turn to a key passage in Leibniz’s 1695 “A New System
of the Nature and Communication of Substances, and of the Union of the Soul and
Body.” This work expresses some of Leibniz’s most mature metaphysics. In it, Leibniz
not only agrees with the occasionalists on (i) and (ii), but also explicitly states that he
135
LO 220-21.
I leave open the question of whether or not Malebranche would have agreed with (P2) of the
transference argument— substances can’t diminish in being in the way required for transference.
Malebranche in fact develops different reasons for the conclusion of the transference argument, which I
soon present.
136
60
agrees —to some extent—with the their reasons for (i) and (ii). I present the passage in
full:
For I could not find any way of explaining how the body makes anything happen
in the soul, or vice versa, or how one substance can communicate with another
created substance. Descartes had given up the game at this point, as far as we can
determine from his writings. But his disciples, seeing that the common opinion is
inconceivable, judged that we sense the qualities of bodies because God causes
thoughts to arise in the soul on the occasion of motions of matter, and that when
our soul, in turn, wishes to move the body, it is God who moves the body for it.
And since the communications of motions also seemed inconceivable to them,
they believed that God imparts motion to a body on the occasion of the motion of
another body. That is what they call the system of occasional causes, which has
been made very fashionable by the beautiful reflections of the author
[Malebranche] of The Search after Truth.
I must admit that they have penetrated the difficulty by articulating what
could not possibly be the case, but their explanation of what actually happens
does not appear to eliminate the difficulty. It is quite true that, speaking with
metaphysical rigor, there is no real influence of one created substance on another,
and that all things, with all their reality, are continually produced by the power of
God. . . .
Therefore, since I was forced to agree that it is not possible for the soul or
any other true substance to receive something from without, except by divine
omnipotence, I was led, little by little, to a view that surprised me, but which
seems inevitable, and which, in fact, has very great advantages and rather
considerable beauty. . .”137
Leibniz agrees with the occasionalists’ denial of the possibility of creaturely transeunt
causation. Further, Leibniz admits that the occasionalists “penetrated the difficulty”,
suggesting that Leibniz agrees to some extent with their reasoning against the possibility
of creaturely transeunt causation. Finally, Leibniz agrees with the occasionalists that
divine transeunt causation is the only possible kind of transeunt causation, and the
reasons have something to do with God’s omnipotence. According to both Leibniz and
the occasionalists, only an omnipotent being can produce an accident in a different
137
(G IV.482-83: AG 142-43).
61
substance. So the reason Leibniz and the occasionalists deny creaturely transeunt
causation but affirm divine transeunt causation is because they are convinced that only
God is powerful enough or has the right kind of power to produce accidents in other
substances, while created substances do not.
It is important to note a reason why Leibniz did not agree with the occasionalists
that only an omnipotent being can be a transeunt cause. Malebranche famously argued
that one reason that only an omnipotent being can produce an accident in another
substance is because there must be a necessary connection between a cause and an effect,
something that obtains only when the cause is God’s will.138 Malebranche’s reasoning
can be expressed thus:
(M1) Transeunt causation is possible only if there is a necessary connection
between the cause and the effect.139
(M2) There is a necessary connection between a cause and an effect only when
the cause is an omnipotent being.140
(M3) So, Transeunt causastion is possible only when the cause is an omnipotent
being.141
138
Malebranche writes, “A true cause as I understand it is one such that the mind perceives a necessary
connection between it and its effect.” See LO 450.
139
Actually, Malebranche defends an even stronger conditional: (M1’) Efficient causation is possible only
if there is a necessary connection between the cause and the effect, where even immanent efficient causes
require a necessary connection between cause and effect. Thus, Malebranche also affirms the stronger
conclusion (M3’) So, efficient causation is possible only when the cause is an omnipotent being. For the
purposes of this section, the weaker M1 and M3 suffice.
140
Malebranche continues, “Now the mind perceives a necessary connection only between the will of an
infinitely perfect being and its effects. Therefore, it is only God who is the true cause and who truly has the
power to move bodies. I say further that it is inconceivable that God could communicate his power to
move bodies to men or angels, and that those who claim that our power to move our arms is a true power
should admit that God can also give to minds the power to create, annihilate, and do all possible things; in
short, that He can render them omnipotent, as I shall show.” See Ibid.
141
Malebranche’s argument was originally given centuries earlier by Al-Ghazālī' in The Incoherence of the
Philosophers. See Al-Ghazālī', The Incoherence of the Philosophers, translated by Michael E. Marmura
(Provo: Brigham Young University Press, 1997), 170-181 and Malebranche, The Search After Truth, VI.2.3
in LO 446-452.
62
Leibniz denied that a necessary connection is the link between transeunt causation and
omnipotence.142 But he agreed with the conclusion: something related to power is the
reason why divine transeunt causation is possible while creaturely transeunt causation is
not. The reason, I propose, is that creaturely transeunt causation without transference is
creation. But once again, only an omnipotent being can create—a conclusion that theists
of all stripes have generally agreed upon. They just have not agreed that creaturely
transeunt causation involves creation, a point I take Leibniz and the occasionalists to be
pressing them on. Recall Fredosso’s challenge: If divine transeunt causation doesn’t
consist in transference, why must creaturely transeunt causation consist in it? Leibniz
and the occasionalists would both say that it is because created substances are not
powerful enough. With respect to Leibniz, the reason is that if you deny the transference
condition while affirming the familiar causal adequacy principle, the causing-as-giving
principle, and the no-reality-from-patient principle, then the accident is created—
something that only God can do.
The argument for the transference condition also bears a strong resemblance to an
argument Aquinas attributed to Avicenna concerning the production of substantial forms
in Aquinas’s Quaestiones Disputatae de Potentia Dei.143 Aquinas characterizes
Avicenna’s argument thus:
142
Leibniz writes, “Malebranche’s strongest argument for why God alone acts reduces to this in this end—
a true cause is that which the effect follows from necessarily, but an effect follows necessarily from the will
of God alone. However, it should be noted that if the state of any entity is known perfectly, then the state
of any other entity can be inferred infallibly (although not, I grant, necessarily, i.e., not in such a way that it
could ever be demonstrated that the contrary implies a contradiction, since analysis goes on ad infinitum.”
Quoted in Robert C. Sleigh, Jr., “Leibniz on Malebranche on Causality,” in J.A. Cover and Mark Kulstad,
ed., Central Themes in Early Modern Philosophy: Essays Presented to Jonathan Bennett (Indianapolis:
Hackett, 1990), 171.
143
Kara Richardson has recently argued that the argument Aquinas attributes to Avicenna is partly
63
That which has no matter as a constituent part cannot be made of matter. Now
forms have no matter as a constituent part: because form is contradistinguished
both from matter and from composite things. Since, then, forms are made since
they have a beginning of existence, it would seem that they are not made out of
matter; and consequently are made out of nothing and therefore are created.144
The argument Aquinas attributes to Avicenna specifically concerns the production of
substantial forms. However, Aquinas later argues that if the argument applies to
substantial forms, it also applies to accidental forms.145 In keeping with the focus on the
production of accidents in this chapter, the argument can be reconstructed to apply to
both substantial and accidental forms:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Forms have a beginning of existence.
So, forms are made.
Whatever is made is made out of something or made out of nothing.
Something can be made out of matter only if that something has matter as a
constituent part.
Forms do not have matter as a constituent part.
So, forms cannot be made out of matter.
So, forms are made out of nothing.
Whatever is made out of nothing is created.
So, forms are created.
With the emphasis on accidental forms, let’s take (1) to mean that an accident caused to
inhere in a patient substance has a beginning of existence simpliciter. This is certainly
true if the accident was not transferred from the agent in the way transference is
characterized, where once the agent transfers an accident, the agent no longer possesses
misleading. Her reasons are complicated but can be summarized thus: First, Aquinas’s interpretation of
Avicenna renders Avicenna an Occasionalist, which Avicenna is not. Second, Aquinas’s reading of
Avicenna conceals many important areas of agreement he has with Avicenna, especially ideas Aquinas
adopts for his eduction model of generation. See Kara Richardson, “Avicenna and Aquinas on Form and
Generation,” in Dag Nikolaus Hasse and Amos Bertolacci, ed., The Arabic, Hebrew and Latin Reception of
Avicenna's Metaphysics (Boston: Walter de Gruyter GmbH & Co., 2011), 251-74.
144
Aquinas, QDP, q3, a8, objection 6.
145
Aquinas writes, “Moreover, just as matter is not a part of the substantial form so neither is it a part of the
accidental form. Hence, if the reason why substantial forms must be produced by creation is because they
have no matter, the same argument will apply to accidental forms.” See QDP, q3, a8, sed contra.
64
it. While the text makes no mention of transference, it is also fair to say that the
argument assumes the denial of transference. Were the accident transferred, it would not
have a beginning of existence simpliciter in that causal transaction. Instead, it would
only begin to exist in the patient. The denial that an accident can be made out of matter
can be taken to be a variant of the no-reality-from-patient principle. The accident gets no
reality from the matter, i.e., the accident produced in the patient gets no reality from the
patient. Hence, Avicenna is arguing (on Aquinas’s gloss) that because the accident does
not come from the agent in the sense that the agent loses the accident (the denial of the
transference) and because the patient doesn’t contribute the accident (the affirmation of
the no-reality-from-patient thesis), the accident is created.
Aquinas rejects Avicenna’s conclusion in (9) in part because while Aquinas
agrees that only God can create, he also argues that created substances can produce
substantial and accidental forms.146 Therefore, the created substances that produce the
forms do not create them. Leibniz would agree with Aquinas that only God could create.
But Leibniz would also agree with Avicenna that the denial of transference and the
affirmation of no-reality-from-patient thesis commit to an accident’s being created when
produced.
Of course, the argument differs in important details from the argument I’m
attributing to Leibniz. For example, Leibniz did not utilize a form/matter distinction in
the way many Aristotelian scholastics did. However, for the purposes of this chapter, I
need not develop the argument further. I draw attention to it only to note that the reasons
146
Aquinas writes, “Moreover, God alone can create. Hence, if forms are created, they will be the work of
God alone, so that all nature’s work, the purposes of which is the form, will be useless.” Ibid.
65
I argue are lurking behind Leibniz’s repeated statements of the transference condition
have a notable historical precedent in an influential medieval Aristotelian. Further, there
is textual evidence that Leibniz was aware of Avicenna’s argument. In a section of the
Theodicy where he criticizes eduction, Leibniz writes, “Some have thought that forms
were sent from heaven, and even created expressly, when bodies were produced.”147
While Leibniz doesn’t specify who thought forms were created as opposed to educed, the
most likely candidate is Avicenna, as the view is distinctive to Avicenna as an alternative
to eduction with respect to the production of substantial forms.148
§3 A Weakened Transference Condition
Having developed an argument for (P1) in §2 that is consistent with denying that divine
transeunt causation requires transference, and having argued as well why the argument is
plausibly given to Leibniz, I now draw attention to two weaknesses with the argument.
The first problem is that the argument for (P1) does not take into account the Aristotelian
act/potency distinction operative in most scholastic accounts of causation, such as the
eduction account and Bacon’s multiplication of species account. Such scholastics would
argue that Leibniz’s transference condition is the absurd implication of any metaphysic
that did not take into account the act/potency distinction. A key passage in the Theodicy
will show that Leibniz was in fact aware of the role the act/potency distinction played in
scholastic accounts of causation, and sheds light on why he rejected the possibility of
creaturely transeunt causation.
147
148
(G VI.150-51: T 88).
I address in more depth Leibniz’s criticisms of eduction in §3.
66
However, the first problem leads to a second. In several passages in the Theodicy
where Leibniz criticized both the act/potency distinction central to scholastic accounts of
causation, and in a passage criticizing one of Bayle’s arguments for occasionalism,
Leibniz himself gives reasons that in turn weaken the transference condition. For reasons
that will soon be apparent, I conclude in this final part of the chapter that Leibniz only
has reasons for the transference condition to be a condition of creaturely transeunt
causation when the accident produced is a non-modal accident.
3.1 The act/potency Distinction and the transference condition
The argument for the transference condition would not have persuaded Aristotelian
scholastics who utilized the Aristotelian distinction between act and potency. Such
philosophers argued that efficient causal agents transeuntly produce accidents in patients
not by handing over their own being or actuality, nor by creating the accident. Rather,
the agent produces the accident in the patient by reducing the potency for the accident in
the patient to act or by educing the accident from the potentiality of the patient. For
example, in a passage we’ve already seen, Aquinas writes:
Rather, by the power of the heat which is in the heating body, a numerically
different heat is made actual in the heated body, a heat which was previously in it
in potency. For a natural agent does not hand over its own form to another
subject, but it reduces the passive subject from potency to act.149
In On the Multiplication of Species, Bacon addresses the question of how species—the
first effects of natural agents in his metaphysics—are produced. He immediately rejects a
transference account (in which the agent emits species) on the grounds that it would
149
SCG IIIa, 69
67
entail the corruption of the agent.150 He also rejects an account where the species are
created ex nihilo. Instead:
Since the generation of a species occurs in none of the aforementioned ways, it is
apparent that it must occur in a fifth way, namely, by true alteration and bringing
forth out of the active potentiality of the recipient matter.151
Leibniz’s transference condition prima facie does not take into account the act/potency
distinction and its role in causation, where the agent is said to draw out the effect
produced from the potentiality of the matter. Instead, rather than by being educed from
the patient’s potentiality, the underlying assumptions of the transference condition entail
that the accident produced was already actual and its being produced in the patient
consisted simply in its being transferred from the agent to the patient.
In the Theodicy, Leibniz shows that he is aware but critical of the act/potency
distinction’s role in scholastic causal accounts, especially the eduction model. In a
section discussing the causal origin of substantial forms, Leibniz writes, “Now
philosophers have troubled themselves exceedingly on the question of the origin of
substantial forms. For to say that the compound of form and matter is produced and that
the form is only comproduced means nothing.”152 Aquinas, amongst many other
scholastics, is a likely target in this passage as Aquinas responded to Averroes’s
argument against the possibility of creatures producing substantial forms with a very
similar line of reasoning:
150
Bacon writes, “Acting does not destroy and corrupt an agent, but perfects it, since . . . a thing is perfect
when it is able to produce a like thing.” See RB 45.
151
Bacon also writes, “But a species is the effect of a natural agent and is naturally produced; therefore, the
species must be generated out of the potentiality of matter.” See RB 47.
152
(G VI.150-51: T 88).
68
Now that which is made is said to become according to the way in which it is
because its being is the term of its making: so that properly speaking it is the
composite that is made per se. Whereas the form properly speaking is not made
but is that whereby a thing is made, that is to say it is by acquiring the form that a
thing is said to be made.153
Aquinas’s claim that substantial forms are not made (or as Leibniz writes, produced) is a
crucial component of his eduction model of causation. Aware of this, Leibniz has equally
harsh words for eduction, writing, “The common opinion was that forms were derived
from the potency of matter, this being called Eduction. That also meant in fact nothing . .
.”.154 Instead of an explanation of causation, Leibniz claims that transeunt causal
accounts that rely on some notion of the effect being derived from potency is a pseudoexplanation. However, Leibniz finds one exception:
. . . [B]ut it was explained in a sense by a comparison with shapes: for that of a
statue is produced only by removal of the superfluous marble. This comparison
might be valid if form consisted in a mere limitation, as in the case of shape.155
The only way Leibniz can make sense of a form being educed from the potentiality of
matter is in the way a statue is chipped away from a block of marble. However, this is
restricted to a very narrow class of effects and a far cry from the scope of what
scholastics thought were produced by creatures. Scholastics such as Aquinas argued that
substantial forms of living beings are educed from the potentiality of matter as well as
many different kinds of accidents such as real qualities. Further, the ‘eduction’ of the
shape of a statue from a block of marble is in one sense a removal of the being of the
153
QDP III, 8, respondeo. Aquinas also writes, “. . . that which is made is not the form but the composite,
which is made from matter and not out of nothing. See Ibid.
154
(G VI.150-51: T 88).
155
Ibid.
69
excess marble rather than an addition of being found in many scholastic accounts, such as
when the substantial form of a living organism is educed in some secondary matter.
3.2 The transference condition and modifications
Although Leibniz’s scathing remarks would not have persuaded scholastics such as
Aquinas and Bacon, they at least shed light on what he thought about the role of the
act/potency distinction in explaining creaturely transeunt causation. In short, Leibniz
thought that the act/potency distinction did not explain such causation. This highlights a
deep tension between Leibniz’s approach to metaphysics, on the one hand, and the
scholastics’, on the other.
In this final section of the chapter, however, I draw attention to a more troubling
tension that lies not between Leibniz’s account and others but within his own account.
This problem, I shall argue, weakens the transference condition, ultimately entailing that
it only applies to non-modal accidents. That is, Leibniz himself gives reasons that count
against the original transference condition. The original transference condition applied to
the transeunt production of any kind of accident, which was formulated thus:
For any created substance s1 and any created substance s2 and any accident A, s1
efficiently causes A to inhere in s2 only if s1 transfers A from s1 to s2.
I shall argue that Leibniz only has the resources to defend a more restrictive version of
the transference condition, one that applies only to the production of non-modal accidents
(e.g., real qualities):
For any created substance s1 and any created substance s2 and any non-modal
accident A, s1 efficiently causes A to inhere in s2 only if s1 transfers A from s1 to
s2 .
70
In the same passage where Leibniz argued that the act/potency distinction ultimately
explained nothing in accounts of creaturely causation, Leibniz writes: “Eduction is not
inexplicable with accidental forms, since they are only modifications of the substance,
and their origin may be explained by eduction, that is, by variation of limitations, in the
same way as the origin of shapes.”156 While Leibniz dismisses the eduction account of
substantial forms (as explaining nothing), he allows that the production of accidents may
be explained by eduction. However, he does so with two important caveats. First, the
accidents are modifications. Second, the eduction of those modifications is really just the
variation of limitations, where the varying of limitations is importantly analogous to a
change of shape (such as the shape of a statue carved out of some marble).
In a later chapter in the Theodicy, Leibniz makes a similar claim that seems to
challenge the very conclusion of the transference argument. Responding to Bayle, who
argued that if created substances could produce accidents, they would create them,
Leibniz writes:
As for the so-called creation of the accidents, who does not see that one needs no
creative power in order to change place or shape, to form a square or a column, or
some other parade-ground figure, by the movement of the soldiers who are
drilling; or again to fashion a statue by removing a few pieces from a block of
marble; or to make some figure in relief, by changing, decreasing or increasing a
piece of wax? The production of modifications has never been called creation,
and it is an abuse of terms to scare the world thus. God produces substances from
nothing, and the substances produce accidents by the changes of their limits.157
Prima facie, this passage challenges the argument for the transference condition: Leibniz
seemingly denies that the production of accidents in other substances—such as blocks of
156
157
(G VI.151-52: T 89).
(G VI.351: T 395).
71
marble, wax and statues—is creation. Of course, the argument for the transference
condition is not only consistent with but further clarifies why Leibniz thought the
immanent causation of accidents does not require transference or the power to create.
But all the examples in the passage are examples of transeunt causation. So Leibniz
himself challenges the argument for his transference condition: he denies that the
transeunt causation of accidents is creation.
However, there is reason to deny that the passage challenges the argument for the
transference condition. Leibniz’s Theodicy was a book written for a popular audience.
Depending on his audience, Leibniz often concealed or at least omitted aspects of his
deeper metaphysical views.158 His deeper metaphysics, expressed more in works such as
the Monadology, held that bodies are only phenomenally real and also denied the
possibility of creaturely transeunt causation. What we would call a corporeal substance,
such as a block of marble, would only be phenomenally real in the idealistic
interpretation of Leibniz’s metaphysics, being reducible in some sense to an aggregates
of a monads and their perceptions.159 When a body b1 produces an accident in a
158
John Whipple points to two important distinctions Leibniz draws concerning the communication of his
philosophical views: exoteric versus esoteric form and exoteric versus esoteric content. Roughly, doing
philosophy in an esoteric form is, according to Leibniz, the proper way to do strict philosophy and follows
the geometric manner of demonstration. This is in contrast to an exoteric form, which is philosophy written
in a less rigorous and formal style but more accessible and popular level instead. Philosophical writings
with esoteric content express Leibniz’s deep metaphysical views, which may conflict strongly with either
the general views of the population of the particular philosophical persuasions of Leibniz’s interlocutors.
Leibniz’s writings that contain exoteric content are written either with a popular audience in mind; hence
concealing in some manner Leibniz’s deeper views or the writings are tailored to the philosophical
persuasions of Leibniz’s interlocutor. For example, when dialoguing with Cartesians, Leibniz uses
Cartesian technical terms even though at root, Leibniz’s philosophy is not Cartesian. John Whipple,
"Leibniz's Exoteric Philosophy", The Stanford Encyclopedia of Philosophy (Summer 2013 Edition),
Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/sum2013/entries/leibniz-exoteric/>.
72
numerically distinct body b2, strictly speaking, immaterial monads are immanently
causing changes in their perceptions. For example, at t1 a monad s1 produces perceptions
of a block of marble and at t2 the same monad produces perceptions of the marble with a
chip of the top. 160
Given that the examples in the Theodicy are examples of the transeunt causation
of accidents in bodies and facts about bodies are reducible in some sense to facts about
immaterial monads and their perceptions, at least on the idealistic interpretation held in
this dissertation, this passage does not challenge the argument for the transference
condition. For the passage, ultimately, does not establish that one created substance can
produce an accident in a different created substance without requiring transference or
creative power. Instead, when understood against the background of Leibniz’s
monadological metaphysics, the passage can be taken to establish that the immanent
production of accidents does not require creative power. That is consistent with the
argument for (P1).
While the above argument would not persuade scholars who endorse the nonidealistic interpretation of Leibniz’s metaphysics, there is a different problem arising
from that passage and others that plagues both the non-idealistic and idealistic
interpretations. This problem will lead me ultimately to conclude that Leibniz had
159
Of course, I’m taking a stance on an important debate in the recent scholarly literature on the ontological
status of bodies in Leibniz’s metaphysics. Specifically, I’m assuming the idealistic interpretation, in which
there are no corporeal or bodily substances. For those who disagree with the idealistic reading of Leibniz,
my argument can be understood as the conditional with the antecedent “If Leibniz is an idealist” and the
consequent being the argument I’ve just given. For an overview of the arguments pro and con for whether
or not Leibniz was an idealist, see Brandon C. Look, “Leibniz’s Metaphysics and Metametaphysics:
Idealism, Realism, and the Nature of Substance,” Philosophy Compass 5, No. 11 (2010): 871-79.
160
Thus, Leibniz’s argument in the passage from the Theodicy has what John Whipple calls exoteric
content.
73
reasons to hold only that the creaturely transeunt causation of accidents requires either
transference or creative power only when the accidents are not modifications but instead
are non-modal accidents such as real qualities.
Before denying that the production of accidents requires creative power, Leibniz
quotes Bayle in full:
'One of the absurdities', says M. Bayle (p. 779), 'that arise from the so-called
distinction which is alleged to exist between substances and their accidents is that
creatures, if they produce the accidents, would possess a power of creation and
annihilation. Accordingly one could not perform the slightest action without
creating an innumerable number of real beings (d’estres reels), and without
reducing to nothingness an endless multitude of them. Merely by moving the
tongue to cry out or to eat, one creates as many accidents as there are movements
of the parts of the tongue, and one destroys as many accidents as there are parts of
that which one eats, which lose their form, which become chyle, blood, etc.'
Bayle argues that the production of accidents requires the power to create and annihilate,
a power only available to an omnipotent being— God. Leibniz, on the other hand, denies
that the production of accidents requires omnipotence. Note that Bayle writes,
“Accordingly one could not perform the slightest action without creating an innumerable
number of real beings”. Here, Bayle assumes that accidents are real beings (d’estres
reels), an assumption that is important, as I soon show. Bayle’s argument, focusing on
creation and bracketing annihilation, can be reconstructed as follows:
(1) Accidents are real beings.
(2) The production of real beings requires creative power.
(3) So, the production of accidents requires creative power.
Leibniz denies (3), as we have seen, and thus might reject either (1) or (2). The evidence
points strongly towards Leibniz’s denying (1). In Leibniz’s response to the argument, he
never explicitly denies (2). In fact, an important thesis of his monadological metaphysics
is that created substances can only begin to exist via creation by God and cease to exist
74
via annihilation by God. If any entities in Leibniz’s metaphysics count as real beings,
substances do. So Leibniz partially agrees with (2).
Yet in Leibniz’s response to Bayle’s argument, Leibniz does not identify
accidents with real beings. Instead Leibniz identifies accidents with modifications and
denies that the production of modifications requires creative power. Leibniz writes, “The
production of modifications has never been called creation”. So Bayle can be
understood as arguing that the production of accidents is creation because accidents are
real beings. Leibniz responds by arguing that a created substance’s producing accidents
is not creation because accidents are modifications, not real beings.
It’s important to get a grip on what philosophers meant in Leibniz’s time in
claiming that modifications are not real beings. This in turn will shed light on the
significance of Leibniz’s response to Bayle’s argument. Many notable philosophers
during Leibniz’s era distinguished between two types of beings—res and
modifications.161 Substances counted amongst the res. Importantly, a subset of
accidents— real qualities—counted amongst the res as well. According to Suarez, a sign
that an accident is a res rather than a modification was the supernatural possibility of the
accident existing without inhering in a substance. For example, the accidents of the bread
and wine of the Eucharist could exist—miraculously—without inhering in the bread and
wine. Accidents that were not real qualities were modifications of a substance. A
modification is an accident that cannot even supernaturally exist without inhering in a
161
For a lengthy overview of how late medieval and early modern philosophers understood res and
modifications, see Robert Pasnau, Metaphysical Themes: 1274-1671 (Oxford: Oxford University Press,
2011), 179-199 and 244-278.
75
substance.162 Leibniz’s response to Bayle’s argument can then be understood as follows.
Bayle claims that the production of accidents—which are res—is creation. Leibniz
denies that the production of accidents is creation because accidents are not res. Instead,
accidents are modifications.
In summary so far, Leibniz denies (3) by denying (1). Leibniz does not deny (2),
at least in responding to Bayle’s argument. Further, Leibniz’s denial of (1) is consistent
with affirming (2). What does this imply for the argument for the transference condition?
While Leibniz has reasons to affirm that the transeunt causation of res requires
transference or creative power, Leibniz lacks similar reasons to affirm that the transeunt
causation of modifications or limitations requires transference or creative power. Shortly
before responding to Bayle’s argument, Leibniz writes in the Theodicy:
God is the one principal cause of pure and absolute realities, or of perfections.
Causae secundae agunt in virtute primae. But when one comprises limitations
and privations under the term realities one may say that the second causes cooperate in the production of that which is limited; otherwise God would be the
cause of sin, and even the sole cause.163
Leibniz argues that the term “realities” has not only pure and absolute realities or
perfections (which are the same thing) in its extension, but also limitations and
privations. Recall that the transference argument frequently invokes the term “reality.” It
162
For an excellent overview on the theological significance of early modern philosophers’ rejection of real
qualities, see Stephen Menn, “The Greatest Stumbling Block: Descartes’ Denial of Real Qualites,” in Roger
Ariew and Marjorie Grene, ed., Descartes and his Contemporaries: Meditations, Objections, and Replies
(Chicago: The University of Chicago Press, 1995), 182-207.
163
(G VI.349-50: T 392). Tad Schmaltz has recently argued that Leibniz’s specific claim in this passage in
the Theodicy, that God is the principle (presumably efficient) cause of pure and absolute
realities/perfections, is consistent with holding that creatures are secondary efficient causes of pure and
absolute realities/perfections. As I argue in Chapters 3 and 4, there are plenty of reasons to deny that
Leibniz thought that created substances produce pure and absolute realities/perfections. Instead, the effects
of secondary causation are limitations. See Tad M. Schmaltz, “Moral Evil and Divine Concurrence in the
Theodicy,” in Larry M. Jorgenson and Samuel Newlands, ed. New Essays on Leibniz’s Theodicy (Oxford:
Oxford University Press, 2014), 135-152.
76
is important to see how well the transference argument succeeds given what falls under
the extension of “realities” in Leibniz’s metaphysics.
I will argue that the transference
argument succeeds only when the accident produced is a non-modal accident such as a
real quality, which would count as pure or absolute realities in Leibniz’s metaphysics.
The argument does not succeed, however, when the effect is a modification, which we
shall soon see is a limitation. This points to an important weakness in Leibniz’s
argument, as Leibniz denies the existence of non-modal accidents in his ontology and
instead argues that all accidents are modifications. This means that Leibniz’s
transference argument, I shall argue, does not succeed in establishing that the creaturely
transeunt causation of modifications—the only type of accident in Leibniz’s ontology—
requires transference. Instead, it only succeeds in establishing that the transeunt
causation of non-modal accidents that Leibniz does not even posit in his ontology—such
as real qualities—requires transference.
Let’s first look at why the transference argument does not succeed when the effect
produced is not an absolute reality but instead a limitation or privation. Here, we enter
into a topic that is still generating a lot of scholarship, as many of Leibniz’s comments
about limitations and privations are usually made in the context of discussions about
Divine concurrence and whether evil is traceable to God. Prima facie, if A is a limitation
or a privation, then it is far from obvious why s1’s transeuntly causing A to inhere in s2
requires that s1 transfer A from itself. It’s also far from obvious why s1’s transeuntly
causing A without transferring A would require that s1 create A where s1’s creating A
requires omnipotence. The reason is that s2’s gaining A is not a net gain in absolute or
positive reality.
77
To see why, let’s first look at why the transference argument does not apply to
modifications or limitations. In a 1703 letter to De Volder, Leibniz claims that
modifications are limitations and clarifies what he means by that:
. . . a modification is a varying limitation, and modes merely limit things but do
not increase them and hence cannot contain any absolute perfection which is not
in the thing itself which they modify. Otherwise, in fact, these accidents must be
thought of in the manner of substances, namely, something which stands per se.164
Leibniz here claims that modifications are limitations. As Leibniz also argues elsewhere
that all accidents are modifications, all accidents are also limitations.165 A limitation has
no absolute perfection or reality of its own, so a modification has no absolute perfection
or reality of its own. The passage also indicates that substances are absolute or positive
realities in Leibniz’s metaphysics. Leibniz argues that if modifications were not
limitations, but instead contained absolute perfection that was not the absolute perfection
of the substance being modified, then modifications, like substances, would be things per
se.
In chapter 3, I address in much greater depth why modifications could not be
transferred as that is ultimately relevant to Leibniz’s positive account of causation. Here,
however, I argue that Leibniz has no reasons to argue that the production of
modifications requires omnipotence if the modification is not transferred. For the
production of a modification is not the production of new absolute or positive reality.
There is no net gain in positive reality when a new modification comes into being in a
substance. Instead, the production of a modification is a re-arrangement of previously
164
(G II.257: L 532).
In a 1715 letter to Des Bosses, Leibniz wrote, “Whatever is not a modification can be called a
substance.” See (G II.503-4: L 614). I address in at length in Chapter 4 Leibniz’s reasons for arguing that
all accidents are modifications.
165
78
existing absolute or positive reality. As we saw earlier, Leibniz claimed that substances
produce accidents by the changes of their limits. Leibniz also claims, as we’ve seen, that
a substance’s changing its limits is analogous to a body changing shape. Further, Leibniz
claims in his letter to De Volder that a modification is not an increase in a substance or
that a modification does not increase the absolute reality of a substance. Given this
understanding of accidents as modifications or limitations, where such modifications are
analogous to re-arrangements of the absolute reality of their substances, I’m hard-pressed
to see why a substance would need to be omnipotent to produce such an accident in a
distinct substance. That is, I’m hard-pressed to see why a substance would have to be
omnipotent to re-arrange the reality of some numerically distinct substance.
Perhaps one could offer a similar argument to the transference argument when it
comes to privations— that the transeunt causation of privation requires omnipotence. If
s1 produces a privation in s2, then s1 has caused s2 to lack a reality that s2 previously had.
Leibniz could respond (in agreement with Bayle) that s1’s production of the privation was
ultimately an annihilation of a previously existing reality. But this argument also cuts
against the immanent causation of privations. For if a substance immanently causes A to
inhere in itself and A is a privation—that is, A is a lack of a reality that s1 previously
possessed—then the substance is the cause of an annihilation of some of its reality, which
is again a power only available to God.
There is a way to make sense of the immanent causation of privations without
requiring the omnipotence of the created substance. If the privation immanently caused
is a privation of absolute or positive reality, then Leibniz would argue that such causation
is a case of annihilation requiring omnipotence. But if the privation is a privation of a
79
limitation or modification, such causation would not require omnipotence. However, if a
substance’s immanently causing the privation of a modification or limitation does not
require omnipotence, I’m again hard pressed to find a reason why the transeunt causation
of a limitation or modification would require omnipotence.
The argument for the transference condition does succeed, at least, if the accident
produced is non-modal accident or res, such as a real quality. First, non-modal accidents
such as real qualities, which Leibniz denies exist, would count as absolute realities in
Leibniz’s metaphysics rather than limitations, if such accidents existed. The reason is
that such non-modal accidents or real qualities were widely held to be able to exist, at
least supernaturally, without inhering in a substance. This was important for the Catholic
belief in transubstantiation, where the accidents of the bread and wine exist without
inhering in a substance when the bread and wine is transubstantiated into the body and
blood of Christ.166 This leads to a plausible inference: If an accident can exist (even
supernaturally) without inhering in a substance, then the accident has some absolute
perfection of its own, rather than the absolute perfection of the substance the accident
inheres in. As we’ll see in Chapter 3, one reason modifications—which are not res—
could not be transferred is because they cannot exist without inhering in a substance—
unlike such non-modal accidents or real qualities.
Given that non-modal accidents are res and so are absolute realities, Leibniz can
argue that they would have to be transferred or created (requiring omnipotence) if
166
I address Leibniz’s views on Transubstantiation in Appendix B.
80
transeuntly caused.167 That is, the transference argument does apply to non-modal
accidents and the transference condition is a condition of the production of non-modal
accidents. The reason is that an accident that is a res or real quality is an accident that
can exist—at least supernaturally—without inhering in a substance. If a created
substance s1 transeuntly causes a real quality A in a different created substance s2, then s1
has produced an entity that inheres in s2 but can exist apart from s2, apart from s1, and any
other created substance. An accident that can exist apart from any created substance is
surely more than a mere re-arrangement of the absolute perfection of a substance the
accident inheres in. So there is a net gain in the absolute reality of a substance when the
substance gains an accident that is a res.
From here, the argument for the transference condition runs through as it did in
§2. If s1 transeuntly causes an accident A which is a res to inhere in a different substance
s2 and s1 does not transfer A to s2, then if s1’s transeunt causing of A in s2 satisfies the noreality-from-patient-principle, once again there is a net gain in absolute reality rather than
a mere re-arrangement of s2’s reality (if A were a modification). Therefore, A would have
to be created. To avoid that implication, one would have to invoke the transference
condition and hold that A originally belonged to s1 but no longer belongs to s1 upon its
inhering in s2. But again, Leibniz does not posit the existence of accidents which are res.
Instead, all accidents are modifications. So the argument for the transference condition
doesn’t succeed in establishing that the transeunt causation of modifications—the only
type of accident Leibniz posits—requires that they be transferred or created by a
167
The success of the argument also hinges on Leibniz’s rejection of the act/potency distinction as being
explanatory.
81
substance with omnipotence. This is a significant weakness with Leibniz’s transference
condition.
The argument, while weakened, is not a failure. Plenty of philosophers during
Leibniz’s time did posit non-modal accidents so Leibniz can be seen as presenting a
challenge to the transeunt causation of such accidents. Perhaps these are the accidents
Leibniz had in mind in his repeated statements of the transference condition. Further,
textual evidence suggests that Leibniz thought that creaturely transeunt causation would
have to consist in the transference of real beings. Recall some of the various candidates
Leibniz suggests for transference, such as material particles and monads (substances).
Leibniz—as I’ve just argued—is likely correct that real beings—in the sense of beings
that are pure or absolute or positive realities--are the only types of things that could be
transferred. But given that many (if not all) accidents are limitations, which of course
cannot be transferred (reasons which I explore in greater depth in Chapter 3), why could
not one created substance produce a limitation in another? To put the question slightly
differently: why must the effect of creaturely transeunt causation be a res? To my
knowledge, Leibniz never answers that question. So I conclude that Leibniz has good
reasons to hold that the transeunt causation of res requires transference, or else the res
would be created. But Leibniz lacks similar good reasons to hold that such causation of
limitations or modifications requires transference or omnipotence.
82
CHAPTER 3. LEIBNIZ’S TRANSFERENCE ARGUMENT, PART 2: AGAINST
TRANSFERENCE
In this chapter, I evaluate Leibniz’s support for the second premise of the Transference
argument, which I expressed thus:
(P1) Creaturely transeunt causation is possible only if the transference of
accidents is possible.
(P2) The transference of accidents is not possible.
(C) So, creaturely transeunt causation is not possible.
As will be evident throughout this chapter, Leibniz provides ample support for (P2)
throughout his career, making (P2) much easier to defend than (P1). Additionally, the
second premise enjoys much support throughout the history of philosophy, especially by
defenders of creaturely transeunt causation who vehemently denied the transference
condition.
There are two parts to this chapter. In §1, I present an overview of the idealistic
interpretation of Leibniz’s ontology that I assume in this dissertation, in which what
exists, strictly speaking, are simple substances and their accidents. In §2, I argue that
nothing in Leibniz’s ontology could be transferred from agent to patient. I devote the
bulk of this chapter to addressing why accidents could not be transferred. I first draw
83
attention to several constraints Leibniz places on accidents that forbid them from being
transferred. I then argue that accidents have those constraints because, according to
Leibniz, all accidents are modifications and all modifications are limitations. Finally, I
look at reasons why substances could not be transferred from agents to patients.
§1 Leibniz’s Substance-Accident Ontology
While I am primarily concerned with Leibniz’s transference argument insofar as it
concerns the production and transference of accidents, I will argue in this chapter for the
stronger thesis that nothing in Leibniz’s ontology could be transferred from agent to
patient in creaturely transeunt causation. On the idealistic interpretation of Leibniz’s
metaphysics assumed in this dissertation, strictly speaking there are only simple
substances and their accidents. Everything else is reducible in some sense to these simple
substances and accidents. Many facts or statements about bodies, while true, are reduced
to facts or statements about monads and their accidents.168
We can find the mature Leibniz stating this ontology in various passages. For
example, in a 1715 letter to Des Bosses, Leibniz wrote, “Whatever is not a modification
can be called a substance.”169 Here, Leibniz seems to state that the world consists in two
and only two types of entities— substances and modifications. I write “seems” because
one might argue that Leibniz is merely making a semantic claim: If x is not a
modification then x can be called a substance. However, Leibniz’s argument in the
168
The specifics of the reduction of bodies to monads and their accidents goes well beyond the scope of
this dissertation. For a more detailed treatment of Leibnizian reduction, see J.A. Cover, “Non-Basic Time
and Reductive Strategies,” Studies in the History and Philosophy of Science 28 (1997): 289-318.
169
(G II.503-4: L 614).
84
remainder of the passage strongly indicates that Leibniz is not merely making a semantic
claim, instead he is making a metaphysical claim. Leibniz considers whether there could
be a third kind of entity that is neither a modification nor a substance, and rejects that
there could be such a third entity.170 I take this to show that Leibniz’s ontology is an
ontology of substances and accidents, where accidents are modifications. That is, what
exists are substances and their modifications. As I argue later, all accidents are
modifications for Leibniz, so an ontology of substances and accidents amounts to the
same thing.
Other texts support my contention that Leibniz’s ontology is an ontology of
substances and their modifications. For example, in the fictional dialogue between
Philarete and Ariste, speaking for Leibniz, Philarete claims, “My friend [Leibniz], whose
opinion I have just related, gives enough evidence that he leans in this direction, since he
reduces everything to monads, or to simple substances and their modifications. . .”.171
Similarly, Leibniz writes:
As a result, a monad, in itself and at a moment, can be distinguished from another
only by its [a] internal qualities and [b] actions, which can be nothing but its [a’]
perceptions (that is, the representation of the composite, or what is external, in the
simple) and its [b’] appetitions (that is, its tendencies to go from one perception to
another) which are the principles of change. For the simplicity of substance does
not prevent a [a’’ and b’’] multiplicity of modifications.172
170
Leibniz writes: “We may ask whether there can be a thing which is neither a modification nor a source
of modifications – such as the Scholastics think of as accidents, which, they say, are in a subject naturally
but not essentially, since they can be without a subject by the absolute power of God. But I do not yet see
how such a thing can be explained if it is different from my substantial chain, which is truly in the subject,
though not as an accident but as what the Scholastics call a substantial form, or as a source of modifications
– if you like, after the manner of an echo.” Ibid.
171
(G VI.590: AG 265).
172
(G VI.598: AG 207).
85
Here, Leibniz identifies [a] with [a’], [b] with [b’], but then identifies [a], [a’], [b], and
[b’] with a multiplicity of modifications [a’’ and b’’]. Given the prior text where Leibniz
claims that everything is reducible to simple substances (monads) and their modifications
and that [a] and [b] exhaust the modifications of simple substances, I once again conclude
that Leibniz’s ontology is an ontology of substances and their modifications.
I’ve used the terms “modification” and “accident” interchangeably in describing
Leibniz’s ontology. I do so because according to Leibniz, all accidents are modifications.
In several places in Leibniz’s corpus, Leibniz argues that there are no accidents that are
not modifications, such as real qualities— accidents that can exist without inhering in a
substance. For example, in an earlier 1712 letter to Des Bosses, Leibniz writes:
Let us come now to the real accidents which are in this unifying thing as their
subject. You will agree, I believe, that some of them are only modifications,
which disappear when it is removed. But you ask whether there are not certain
accidents which are more than modifications. Such accidents seem, however, to
be entirely superfluous, and whatever is in such a substance other than a
modification seems to pertain to the substantial thing itself.173
Leibniz in this passage is addressing a question Des Bosses asked, mainly, whether there
are accidents that are more than modifications. In other words, are there accidents that
are not modifications, such as real qualities? Leibniz’s gives a negative answer.
According to Leibniz, anything that is not a modification pertains to the substantial thing
itself—in other words, is itself a substance.174 In the remainder of the first 1705 letter to
Des Bosses I quoted at the beginning of this section, Leibniz makes a similar claim:
173
(G II.458: L 606).
In this particular passage, the substantial thing is the vinculum substantiale, a special type of substance
Leibniz proposed in his correspondence with Des Bosses as a way to show what would be needed for
metaphysics to be compatible with transubstantiation. The vinculum substantiale is a substance-like entity
that bonds other substances, such as the monads which makeup the bread and wine during the Eucharist. I
174
86
. . . we may ask whether there can be a thing which is neither a modification nor a
source of modifications—such as the Scholastics think of as accidents, which,
they say, are in a subject naturally but not essentially, since they can be without a
subject by the absolute power of God. But I do not yet see how such a thing can
be explained if it is different from my substantial chain, which is truly in the
subject, though not as an accident but as what the Scholastics call a substantial
form, or as a source of modifications—if you like, after the manner of an echo.175
Once again, Leibniz denies that there is anything not a modification or a substance.
Anything that is not a modification is a substantial form. That is tantamount to claiming
that it is a substance.176 Therefore, all accidents, for Leibniz, are modifications.177
§2 Why nothing in Leibniz’s Ontology can be Transferred
2.1 Why accidents cannot be transferred.
I turn now to two different but related reasons why accidents/modifications in Leibniz’s
metaphysics could not be transferred from an efficient causal agent to patient. As a
starting point, both reasons can be found in Leibniz’s 1692 “Critical Thoughts on the
General Part of the Principles of Descartes”:
I do not know whether the definition of substance as that which needs for its
existence only the concurrence of God fits any created substance known to us,
unless we interpret it in some unusual sense. For not only do we need other
substances; we need our own accidents even much more. Therefore, since
substance and accident depend on each other, other marks are necessary for
distinguishing a substance from an accident. Among them may be this one: That
a substance needs some accident but often does not need a determinate one but is
content, when this accident is removed, with the substitution of another. An
argue in Appendix B that Leibniz in fact did not seriously posit the vinculum substantiale in the final
analysis.
175
Ibid.
176
Elsewhere Leibniz denies that substances are composed of substantial forms and prime matter, thus I
assume, in this dissertation, that substantial forms just are substances. For an extended discussion of this
thesis, see Cover & Hawthorne, Substance & Individuation in Leibniz, 214-226
177
In Chapter 4, I reconstruct and assess Leibniz’s argument for the thesis that all accidents are
modifications.
87
accident, however, needs not only some substance in general but that very one in
which it inheres, so that it cannot change it.178
In this passage, Leibniz claims that while substances and accidents depend on each other,
the dependence is not of the same type. Substances need some accident or other while
accidents need the very substance they inhere in. The former claim was a common view
among many scholastics and modern philosophers. For example, Peter John Olivi argued
that “. . .subjects cannot, without contradiction, be put into existence with some such
accident, although they could without this or that one.”179 More pertinent to this section
of the chapter are two constraints Leibniz places on accidents in the passage. The first
constraint I express thus:
Dependence of Accidents on Substances: Necessarily, for any accident A, A
exists only if there is some substance s such that A inheres in s.
Most philosophers would have agreed that dependence of accidents on substances is a
constraint on accidents with certain qualifications.180 With respect to real qualities, many
Catholic philosophers were committed to a less restrictive constraint:
Natural Dependence of Accidents on Substances: For any accident A, A
naturally exists only if there is some substance s such that A inheres in s.
Catholics needing to explain transubstantiation would have been committed to the natural
dependence of accidents on substances and in fact committed to denying the Dependency
of Accidents on Substances. In transubstantiation, the real qualities of the bread and wine
can supernaturally exist without inhering in a substance. So the natural dependence of
178
(G IV.364: L 389-90).
Quoted in Pasnau, Metaphysical Themes, 248-9.
180
The constraint traces back at least as far as Aristotle, where in chapter 2 of his Categories, Aristotle
wrote, “Some [things] are in a subject but are not said of any subject. (By ‘in a subject’ I mean what is in
something, not as a part, and cannot exist separately from what it is in.)” See Aristotle, Categories, 1a2025.
179
88
accidents on substances is a constraint on what is naturally or physically possible while
allowing for the supernatural or logical possibility of an accident existing without
inhering in a substance.181 Leibniz, being Lutheran, was not burdened with the need to
accommodate transubstantiation in his metaphysics of accidents.182 Hence, Leibniz
could affirm the dependence of accidents on substances, a much stronger constraint on
what is logically possible.
The second constraint found in the passage can be expressed thus:
The Ownership Thesis of Accidents: Necessarily, for any accident A and any
substance s, if A inheres in s then there is not some substance s’ such that s’ is not
identical to s and A inheres in s’.183
The ownership thesis of accidents, as formulated, has an important diachronic
implication: An accident A cannot inhere in a substance s at a time t and a different
substance s’ at a different time t’. Leibniz primarily utilizes the diachronic implication
against the possibility of accidents being transferred from agent to patient. There are
numerous passages in support of the diachronic implication of the ownership thesis of
accidents, many of which we saw in Chapter 2, including this passage in the
Monadology:
Accidents cannot be detached, nor can they go about outside of substances, as the
sensible species of the Scholastics once did. Thus, neither substance nor accident
can enter a monad from without.184
181
For two recent and extended treatments of metaphysical issues arising from transubstantiation, see
Jeffrey E. Brower, Aquinas’s Ontology of the Material World: Change, Hylomorphism and Material
Objects, 235-241 and Pasnau, Metaphysical Themes, 185-190.
182
I defend this claim at length in Appendix B..
183
I note that the Ownership Thesis of Accidents does not entail that a substance s is essentially related to
its accident A in such a way that A cannot cease to inhere in s. Instead, when A ceases to inhere in s, A
ceases to exist (which is entailed by the Dependency of Accidents on Substances), while s can still exist.
184
(G VI.607-8: AG 214).
89
Leibniz is clear: an accident cannot be detached from one substance and then enter a
different substance. Leibniz also writes in a letter to Clarke:
If space is the property or affection of the substance which is in space, the same
space will be sometimes the affection of one body, sometimes of another body,
sometimes of an immaterial substance, and sometimes perhaps of God himself,
when it is void of all other substance, material, or immaterial. But this is a
strange property or affection, which passes from one subject to another. Thus
subjects will leave off their accidents like clothes, that other subjects may put
them on. At this rate how shall we distinguish accidents and substances?185
Again, Leibniz dismisses the notion that an accident could switch substances. He also
hints at a reason via a rhetorical question: If accidents could change substances—contra
the ownership thesis of accidents —then what would distinguish accidents from
substances? The answer to the rhetorical question is “Nothing”. One distinction then
between substances and accidents is that substances can change accidents over time while
accidents cannot change substances over time.186
The diachronic implication of ownership thesis of accidents enjoys wide support
in the history of philosophy, as we saw in Chapter 2, especially by defenders of
creaturely transeunt causation.187 Even when transubstantiation complicates matters, the
diachronic implication of ownership thesis of accidents is still a constraint on accidents.
For example, according to Aquinas, when the bread is transubstantiated into the body of
Christ, the accident of the bread’s quantity continues to exist without inhering in a
substance, thus violating the dependency of accidents on substances. The remaining
185
(G VII.398: L 702).
I explore later in this chapter and in Chapter 4 what features of accidents and substances make it the
case that accidents cannot switch substances while substances can change accidents.
187
For example, Aquinas writes, “Accidents are not transferred from subject to subject, so that numerically
one and the same accident inheres first in one subject and later in another. For an accident is individuated
by its subject. Hence, it impossible for numerically one and the same accident to inhere in one subject at
one time and in another subject at another time.” See ST 1.77.1
186
90
bready accidents inhere in the quantity.188 While it seems that the accidents have
switched substances—first inhering in the bread and later inhering in the quantity, in fact
they have not. Rather, the remaining accidents switch from inhering in the bread (a
substance) to inhering in the quantity (an accident), thus not violating the ownership
thesis of accidents. So transubstantiation is just a further violation of the dependency of
accidents on substances, as the all the bread’s accidents exist while not inhering in a
substance.189
The diachronic implication of the ownership thesis of accidents is all that is
needed to argue that accidents could not be transferred from an agent substance to a
patient substance. If a created substance s1 transeuntly causes A to inhere in a different
created substance s2, then presumably A first inhered in s1 and later inhered in s2,
violating the ownership thesis. Further, if there is a temporal gap in which A exists
without inhering in s1 or s2, then s1’s causing A to inhere in s2 violates the dependency of
accidents on substances.
The diachronic implication of ownership thesis of accidents suffices to show that
accidents cannot be transferred from one substance to another. There is a second and
third implication of the ownership thesis of accidents — a synchronic and an interworldly implication. In what follows, I present these two further implications, which in
turn provide an opportunity to explore Leibniz’s reasons for endorsing the dependency of
accidents on substances and the ownership thesis of accidents. Going into further depth
188
Aquinas writes, “It is necessary to say that the other accidents that remain in this sacrament inhere, as in
a subject, in the dimensive quantities of the bread and wine that remain. One reason for this is the
following: it seems clear to the senses that something exists having size and color and which is affected by
the other accidents. Nor are the senses deceived in this regard.” See ST 3.77.2
189
See Brower, Ibid.
91
with respect to Leibniz’s reasons for both the dependency thesis and the ownership thesis
will pay dividends in later chapters, as the dependency and ownership theses are
constraints on accidents—which are always the effects of Leibnizian creaturely immanent
causation. Gaining a greater understanding of why the dependency and ownership theses
are constraints on accidents then yields a better understanding of Leibnizian accidents.
This will assist us in the chapters ahead as I address Leibniz’s positive account of
creaturely immanent causation and change.
The second implication of the ownership thesis is that it rules out an accident
inhering in more than one substance at the same time. This is the synchronic implication
of the ownership thesis of accidents: If A inheres in s at t then there is not some
substance s’ such that s’ is not identical to s and A inheres in s’ at t. Leibniz affirms the
synchronic implication of the ownership thesis as well. For example:
It cannot be said that both of them [substances], L and M together, are the subject
of such an accident; for if so, we should have an accident in two subjects, with
one leg in one and the other in the other, which is contrary to the notion of
accidents.190
Thus, unlike substances— which can have more than one accident at the same time,
accidents can only ever inhere in one substance at some time. As with the diachronic
implication, many of Leibniz’s predecessors agreed with the synchronic implication of
ownership thesis. For example, Aquinas writes, “One should reply to the second
argument that some people said, as Avicenna notes, that numerically the same relation is
190
(G VII.401: L 704). Elsewhere, Leibniz writes, “If someone maintains that the same wisdom in number
or the same heat in number is in two subjects at once, the fact that one says that the wisdom of one fails,
whereas the wisdom of the other still subsists, can refute this.” See A.VI.4a, 991.
92
in both extremes. But that cannot be, since one accident is not in two subjects.”191 In
many passages Leibniz affirms both the synchronic and diachronic implications of the
ownership thesis of accidents, suggesting that he regards them as implications of the
same constraint. For example, Leibniz writes, “For two different subjects, as A and B,
cannot have precisely the same individual affection, it being impossible that the same
individual accident should be in two subjects or pass from one subject to another.”192
Both the synchronic and diachronic implications of the ownership thesis of
accidents, at least as I have presented them, are intra-worldly. The implications restrict
an accident from inhering in more than one substance at the same time in the same world
or in different substances at different times in the same world. It’s worth addressing
whether Leibniz thought there was an inter-worldly constraint on accidents. That is, if A
inheres in s in world W, could A inhere in a different substance s’ in a different world
W’?193
There are strong reasons to believe that Leibniz gave a negative answer to the
question. Further, Leibniz’s reasons for denying that A could inhere in s in W and that A
could inhere in s’ in W’ also shed light on why Leibniz affirmed the dependency of
accidents on substances and the diachronic and synchronic versions of the ownership
thesis of accidents. The reason Leibniz affirmed an inter-worldly ownership thesis is
because, as argued above, is related to the fact that for Leibniz, all accidents are
modifications. In a letter to Des Bosses, Leibniz writes, “But a modification is connected
191
Sent I, d. 27, q. 1, art. 1, ad 2
(G VII.401: L 704).
193
Of course, the question assumes transworld identity between accidents.
192
93
essentially to that whose modification it is.”194 If a modification M is essentially
connected to the substance s that M modifies, then it is plausible to hold that there are no
worlds where M modifies a different substance s’. Leibniz’s claims that modifications
are essentially connected to their substances and thus his affirmation of inter-worldly
implications of the ownership thesis, like the diachronic and synchronic implications, had
historical precedent. Suarez frequently characterized modes as accidents that are
essentially connected to their substances— a feature of modes that distinguished them
from non-modal accidents. For example, Suarez writes:
…this mode so necessarily includes conjunction with the thing of which it is a
mode that it is unable by any power whatsoever to exist apart from that thing.195
…for the very essence of a mode demands that it cannot exist unless actually
united to the thing it modifies.196
So Leibniz was assuming a standard position on modes tracing back at least to Suarez,
with the difference being that Leibniz held that all accidents are modes, contra Suarez.
What is it about the nature of modifications that entails the dependency of
accidents on substances and the ownership thesis of accidents? Leibniz offers some
guidance. Immediately after stating that modifications are essentially connected their
substances, Leibniz clarifies with an example, “So there can be no modification without a
subject; for example, no sitting without a sitter.” It is absurd to posit the existence of a
token accident of sitting without the sitter. However, all accidents, for Leibniz, are like
the Socrates’ shape when sitting— modifications whose nature it is to modify their
194
(G II.503-4: L 614)
Francisco Suarez, On the Various Kinds of Distinctions, trans. Cyril Vollert, SJ (Milwaukee: Marquette
University Press, 2007), 32.
196
Ibid., 46-7.
195
94
subjects in some way. Leibniz’s example of sitting to support his contention that
modifications are essentially connected to their substances also had precedent in Suarez.
Again Suarez writes:
For what is purely a mode not only cannot be separated from anything of which it
is the mode, but this individual mode cannot be separated from this individual
thing; for example, this position of sitting cannot be separated from this sitter.197
Socrates’ shape when sitting, for Suarez, not only needs to modify some substance, it
needs to modify the very substance it modifies. I note that Suarez and Leibniz both use
the example of sitting as if it were sufficient to establish that modes are essentially
connected to what they modify. Given that they usually did not elaborate further, they
likely assumed that it would be obvious to anyone else that modifications such as sitting
and other figures are essentially connected to what they modify.
Given that, in the case of Leibniz, all accidents are like figure in their being
essentially connected to what they modify, it’s worth exploring what feature(s) of
modifications support their essential connection to their substances. In what follows, I
shall argue that there is a feature of a shape modifying a substance that Leibniz believes
entails that the modification is essentially connected to what it modifies. This feature,
Leibniz argues, is found in all modifications and therefore all accidents. I do not offer
on Leibniz’s behalf a tight deductive proof for the claim that modifications are essentially
connected to what they modify, from the feature. As noted above, Leibniz would have
assumed that his audience would take the fact that accidents are modifications as being
sufficient for their essential connection to their substances. However, the feature of
197
Ibid., 46.
95
modifications—as Leibniz understands them—will shed further light on that thesis as
well as the dependency of accidents on substances.
The feature, I argue, is that modifications are limitations.
Leibniz writes, “. . .a
modification is a varying limitation, and modes merely limit things but do not increase
them and hence cannot contain any absolute perfection which is not in the thing itself
which they modify.”198 Leibniz claimed throughout his corpus that modifications are
limitations. Very early in his career, in his 1676 “Two Notations for Discussion with
Spinoza,” Leibniz wrote, “Indeed there can be no active modifications of that which is
merely passive in essence, since modifications limit rather than increase or add.”199 As
we saw in Chapter 2, Leibniz claims that a created substance changes its accidents by
changing its limits.
Leibniz’s frequent example of shapes as modifications can clarify, somewhat,
how modifications are limitations. One example, especially, is found in the Theodicy.
Leibniz claims that shapes are mere limitations, using the example of a sculptor.200 When
a sculptor sculpts a statue from a block marble, the sculptor “adds” a shape to the marble.
However, the sculptor “adds” a shape to the marble by removing chunks from the block
of marble. The “addition” of the shape, then, is added by limiting the block of marble.
198
(G II.257: L 532).
L 169.
200
Leibniz writes, “The common opinion was that forms were derived from the potency of matter, this
being called Eduction. That also meant in fact nothing, but it was explained in a sense by a comparison
with shapes: for that of a statue is produced only by removal of the superfluous marble. This comparison
might be valid if form consisted in a mere limitation, as in the case of shape.” He also writes, “But
traduction and eduction are equally inexplicable when it is a question of finding the origin of the soul. It is
not the same with accidental forms, since they are only modifications of the substance, and their origin may
be explained by eduction, that is, by variation of limitations, in the same way as the origin of shapes. But it
is quite another matter when we are concerned with the origin of a substance, whose beginning and
destruction are equally difficult to explain.” See (G VI.150-2: T 88-89).
199
96
The shape is then a limitation on the block of marble— a terminus or boundary on the
marble’s extension.
A limitation—such as shape—cannot exist without limiting something. Thus, that
modifications are limitations entails the dependency of accidents on substances. Further,
Leibniz has plausible reasons to hold that a token limitation—such as the token shape of
a block of marble—could only be a limitation of the thing it limits. While two blocks of
marble could have two distinct token shapes of the same type, they could not switch the
same shape tokens. Plato could switch from standing to sitting while Socrates switches
from sitting to standing. Plato, however, could not acquire Socrates’ token accident of
sitting.201
2.2 Why substances cannot be transferred.
As we saw in chapter 2, Leibniz denied that several different kinds of entities could be
transferred— e.g., physical atoms, sensible species, substances, and accidents.202 We’ve
seen why accidents could not be transferred and can also rule out physical atoms and
sensible species being transferred, given Leibniz’s idealistic ontology. This leaves
substances. I note that while the focus of this dissertation is largely on the creaturely
causation of accidents, which I’ve dealt with at length up until this point and will
continue to address in later chapters, it’s worth briefly seeing why substances could not
be transferred, as they are one of the only two types of entities that strictly speaking, exist
201
Of course, the thesis that all accidents are modifications and all modifications are limitations raises a
number of puzzles. For example, on the idealistic reading of Leibniz that I assume in this dissertation holds
that all accidents inhering in substances are perceptions. Perceptions, however, have content. It’s not
obvious how to explain how a perception with content is a mere limitation, unlike a modification such as
shape. I do not address these topics in this dissertation.
202
See (G IV.498-9: L 460), (G II.251: AG 176), and (G VI.607: AG 213-14).
97
in Leibniz’s idealistic ontology.203 This argument will require much less space to
develop. To address this, it’s helpful to briefly consider an amended version of the initial
characterization of transference I offered in chapter 2 which is not restricted to only
accidents being transferred but entities of any type:
For any created substance s1 and any created substance s2 and any entity E, s1
transfers E to s2 only if:
(vi)
(vii)
(viii)
(ix)
(x)
s1 is not identical to s2;
If A belongs to s1 then E does not belong to s2;
If A belongs to s2 then E does not belong to s1;
E first belongs to s1;
E second belongs to s2.
Assuming that E is a substance, E could not belong to s1 by inhering in it.204 Thus, we
need some other way of understanding how E—a substance—could belong to s1. The
most likely candidate is E’s being a part of s1. Leibniz would of course reject this second
way as well, for he argued that created substances are simple and therefore lack parts.205
Yet if s1 were to transfer a substance to s2, so that the substance transferred to s2 belongs
to s2 after belonging to s1, then if the substance transferred did not inhere in s1, prima
facie the only other way for the substance to belong to s1 would be for that substance to
be a part of s1.
203
Leibniz writes, “Whatever is not a modification can be called a substance.” See (G II.503-4: L 614).
Leibniz would argue that it’s a category mistake to hold that a substance can inhere in another
substance, as substances are what are inhered in, rather than what inhere. Evidence for why Leibniz would
claim this can be found in his numerous passages dismissing the existence of real qualities—accidents that
can exist without inhering in a substance, for Leibniz claims that such accidents would in fact be substances
instead. See, for example, (G VII.398: L 702).
205
A classic statement of this doctrine can be found in Leibniz’s Monadology, where he writes, “The
monad, which we shall discuss here, is nothing but a simple substance that enters into composites—simple,
that is, without parts.” See (G VI.607: AG 213). I address in great depth Leibniz’s mereology and his
metaphysics of simple substances in chapter 4.
204
98
However, let us bracket the issue, for now, of whether or not a created substance
could have parts. If Leibniz could argue that even if substances had parts, substances
could not be transferred from agent to patient in transeunt causation, he would have an
even stronger case against transference. Suppose that s1 causes an accident A to inhere in
s2 by transferring some entity E to s2, where E is a third substance s3. How could s1’s
transference of s3 to s2 cause A to inhere in s2? Perhaps s3 just is A. In this case, when s2
receives s3, s3 itself changes from being the substance s3 to the accident A inhering in s2.
Leibniz has strong reasons to reject this option. In addition to puzzles about how a
substance could become an accident, this option blurs the distinction between substances
and accidents. For example, rather than s3 being an accident inhering in s2, given that s3
was a substance prior to its reception by s2, why couldn’t s2 be s3’s accident instead
where s3 remains a substance?206
Perhaps instead, when s1 transfers s3 to s2, s3 itself produces A in s2. However, if
that were the case, then s3’s producing A in s2 would just be another case of creaturely
transeunt causation, in which case s3 must transfer something to s2. Then the question
reemerges: how does what s3 transfers to s2 cause A to inhere in s2? This option then
leads to a vicious infinite regress. For if what s3 transfers to s2 causes A to inhere in s2,
then whatever it is that s3 transferred to s2 and caused A to inhere in s2 must itself transfer
something from itself to s2, and so on ad infinitum.
206
While Leibniz, to my knowledge, never gives this exact objection, it is consistent with a number of
criticisms Leibniz has against accidents, which while inhering in substances, can exist apart from
substances. If such accidents existed, Leibniz argued that there would be no way in principle to distinguish
them from substances. This objection takes that criticism a step further and argues that there would also be
no way to distinguish a substance from an accident.
99
Another option is that E = s1. When s1 produces A in s2, s1 does so by transferring
itself to s2.207 This option has at least one advantage. There are no worries about s1 being
composed of other substances (e.g., s3) or having parts. This option is still not plausible.
For it faces the problems of explaining how a substance can become an accident and once
again explaining the distinction between substances and accidents. That is, why is it that
s1 becomes an accident inhering in s2 rather than s2 becoming an accident that inheres in
s1 ?
That’s the least of the difficulties. No longer bracketing Leibniz’s mereological
views, Leibniz would argue that if s1 were truly a substance and s2 were truly a substance,
then s1’s causing A to belong to s2 by s1 transferring itself would result in s1 and s2 being
an unum per aggregrans, rather than s1 or s2 being an unum per se with s1 or s2 modifying
it.208 For Leibniz not only denies that substances could have other substances as parts, he
also denies that substances have any parts.209 If substances had parts, then they would
not be substances. Instead they would be aggregates. If substances cannot have parts, a
fortiori, substances cannot have other substances as parts.
With these considerations in mind, I conclude that substances cannot play the role
of E in transference. Given that in Leibniz’s ontology, what exists are either substances
207
For example, take a piece of wood acquiring the property of being white from some white paint that was
applied to it. The board gets the property of whiteness from the paint but the paint does not lose the
whiteness the board gained. In cases then where E = s1, s1 does not lose the accident it gives to the patient
substance. This example comes from Rieber, “Causation as Property Acquisition,” 56.
208
According to Leibniz, an aggregate of substances has unity only in the mind and is not a genuine
substance. He writes, “I therefore believed that I would be allowed to distinguish beings by aggregation
from substances, since these beings have their unity in our mind only, a unity founded on the relations or
modes of true substances. If a machine is one substance, a circle of men holding hands will also be one
substance, and so will an army, and finally, so will every multitude of substances.” See (G II.96-7: AG 86).
See also AG 167.
209
I explore this in great depth in chapter 4.
100
or accidents, nothing in Leibniz’s ontology could play the role of E. Therefore,
transference cannot occur. Thus we have Leibniz’s support for the second premise of the
transference argument.
101
CHAPTER 4. LEIBNIZIAN SIMPLE SUBSTANCES AND THE REALITY OF
ACCIDENTS
As we saw in Chapter 3, Leibniz maintained throughout his career that all accidents are
modifications, a thesis which played a crucial role in his defense of the second premise of
the Transference argument. Equally, if not more important, the thesis that all accidents
are modifications also plays a crucial role in Leibniz’s positive account of creaturely
intra-substantial or immanent causation, as accidents are the effects of such causation.210
In this Chapter, I examine in greater depth Leibniz’s reasons for holding that
accidents are modifications, reasoning that sheds light on the nature of modifications,
their substances, and change. To accomplish this, I do three tasks, with a section of this
chapter devoted to each of the three tasks. In §1, I reconstruct Leibniz’s argument in his
short 1688 essay “De Realitate Accidentium” (DRA), the writing where Leibniz goes to
his greatest lengths to argue against the existence of non-modal accidents.211 While less
well known than his other writings, Leibniz draws upon principles that he defended
throughout his career for the argument in DRA. Further, as I show, Leibniz’s reasoning
in DRA impacted his later metaphysical views.
After reconstructing the argument in DRA, I address an important issue with it in
§2. While Leibniz’s arguments in DRA prima facie support the conclusion that accidents
210
211
In Chapter 5, I argue that substances are the efficient causal agents in creaturely immanent causation.
A VI, 4A, 994-996.
102
are modifications, he gives a surprisingly non-committal or agnostic answer as to whether
there are any accidents at all at the end of the essay. I argue that Leibniz hesitated to
posit even modal accidents because he worried that the problems he raised with nonmodal accidents apply to modifications as well. Specifically, Leibniz’s arguments
against non-modal accidents hinge on the assumption that such accidents are parts of
substances, which is inconsistent with Leibniz’s views on the simplicity of substances.212
However, Leibniz also worried that the arguments of DRA entail that modifications, if
they existed, would be parts of substances. Not much later, Leibniz changed his mind
and posited modal accidents in his ontology, without, however, ever explicitly addressing
how such accidents could be in a substance without being a part of it. Drawing from
Leibniz’s later mereological and geometrical writings and his understanding of
modifications as limitations, I argue that Leibniz had the resources to posit simple
substances, which have a multitude of modifications at a time and change such
modifications over time. I also show that non-modal accidents, such as real qualities,
would have to be parts of their substances, if they existed, given Leibniz’s mereological
theses. In arguing for this, I fill one scholarly void in this chapter by applying Leibniz’s
developed mereological theses to not only showing how substances can have a multitude
212
Evidence that Leibniz thought that substances must be simple can be found earlier than his “De Realitate
Accidentium”, which if dated accurately was written in 1688. For example, in a 1686 letter to Arnauld,
Leibniz argued that substances must be thoroughly indivisible and naturally indestructible. Earlier, in his
chapter 9 of his Discourse on Metaphysics, Leibniz writes, “It also follows that a substance cannot begin
except by creation, nor come to an end except by annihilation; and because one substance can’t be
destroyed by being split up, or brought into existence by the assembling of parts, in the natural course of
events the number of substances remains the same, although substances are often transformed.”See AG 42
and (G II.76: AG 79).
103
of modes at a time and over time, but why simple substances cannot have non-modal
accidents, such as real qualities.
§1 Leibniz’s “De Realitate Accidentium”
Leibniz begins “De Realitate Accidentium” (DRA) with the statement, “It is worth
considering whether accidents have a reality that is more than modal, and in what that
[reality] consists.”213 The statement proposes two questions:
(Q1) Do accidents have a reality that is more than modal?
(Q2) If Q1 has an affirmative answer, in what does that reality consist?
Q1 can be understood to simply ask if there are non-modal accidents.214 What Leibniz
means in asking Q2 will become obvious as I reconstruct his argument. In the remainder
of the essay, Leibniz considers the consequences of an affirmative answer to Q1 by
exploring the possible answers to Q2.
Call the thesis that accidents have a reality that is simply modal the “MerelyModal-Reality” thesis. Leibniz’s argument in DRA is an indirect proof for the MerelyModal-Reality thesis. He begins the argument by assuming the falsity of the MerelyModal-Reality thesis and thereby assuming an affirmative answer to Q1, which supplies
us with the first premise:
(1) Suppose that the Merely-Modal-Reality thesis is false.
213
Consideratu dignum est, utrum accidentia realitatem aliquam plus quam modalem, habeant, et in qua illa
consistat. A VI, 4A, 994.
214
I note that a negative answer to Q1, the way it is worded, is consistent with both accidents having a
merely modal reality and also with a variant of nominalism in which accidents have no reality at all.
However, given that Leibniz asks “Do accidents have a reality that is more than modal?” rather than “Do
accidents have any reality at all?”, I argue that it is fair to interpret Leibniz as intending Q1 to have only
two answers: Either accidents have a reality that is more than modal or accidents have a merely modal
reality. That Leibniz intended Q1 to have only these two answers will become evident as I reconstruct
Leibniz’s argument. I am grateful to Michael Jacovides for pressing me on this.
104
If one denies the Merely-Modal-Reality thesis, Leibniz argues there are two options:
Either the accidental reality is part of the reality of the substance or the accidental reality
is not part of the reality of the substance.215 With respect to the second option, the
accidental reality is instead a new reality added to the substance. The second premise can
then be expressed thus:
(2) If the Merely-Modal-Reality thesis is false, then (A) the accidental reality is
part of the reality of the substance or (B) the accidental reality is not part of
the reality of the substance but something additional to it.
Leibniz first considers (A) and argues against it. If the accidental reality is a part of the
reality of the substance, then when a substance changes accidents, the substance thereby
loses and gains different parts (accidental realities). However, if the substance loses and
gains parts, Leibniz argues that the whole substance perishes. Leibniz here assumes a
variant of mereological essentialism, arguing that if some whole W which has part p at t
loses p and gains a different part p’ at t’, W at t is identical in name only to W at t’.
Leibniz writes:
If it is part of the reality of the substance, it follows that the substance itself
perishes in accidental change, or it becomes a new thing, and myself yesterday
exists not yet, but another although very similar to me, so that the ship which is
repaired, or the republic, or the river, are the same in name, are not really [the
same]. (Emphasis added)216
Leibniz supports this variant of mereological essentialism by appeal to the absurdities
raised by the Ship of Theseus problem:
215
Leibniz writes, “And at least if we posit the accidental reality, whether their reality is part of the reality
of the substance, or if it adds to the substance a new reality (Et quidem si accidentia ponimus realia, aut
realitas earum pars est realitatis substantiae, aut addit substantiae realitatem novam).” Ibid.
216
Si pars est realitatis substantiae, sequitur substantiam ipsam in mutationibus accidentalibus interire, seu
rem aliam fieri, et me heri nondum fuisse, sed alium mihi licet valde similem, uti navis quae reparatur, aut
resplublica, aut fluvius, nomine idem sunt, revera non sunt. Ibid.
105
For with a part destroyed, truly the same thing does not remain, even if thus far it
is denominated the same thing by a more important surviving part, otherwise it is
able to take place, so that with all of the parts little by little destroyed, which now
belong to, yet it is finally said to be the same thing, just as the ship of Theseus.217
If the mere change of accidents causes a substance to perish—which would be entailed by
Leibniz’s mereological essentialism and (A), then as all created substances continually
change accidents, Leibniz argues that there would be no created substances:
If, however it is admitted that the substance perishes and comes into existence by
change (Which is the thought of the duke of Buckingham in the ingenious writing
about true religion the Schediasmate218) they in reality remove all changeable
substance.219
Leibniz gets to this further conclusion rather quickly, writing:
For since the changes of things are perpetual, so that nothing remains in the same
state through the smallest intervals of time, it follows that no changeable
substance ever exists and actually endures a minimum time, for any moment
whatever it is born and perishes, neither is it said to properly exist, nor to act,
neither is it able to produce anything or to endure since nothing is brought about
unless enduring for some time.220
I understand Leibniz’s reasoning thus: First, substances can cease to exist—as they
would if they lost one of their parts given Leibniz’s mereological essentialism—only if
the substances first existed. Substances can exist, however, only if they exist for some
217
Nam parte sublata res vere eadem res non manet, etsi a potiore parte superstite adhuc eadem
denominetur, alioqui fieri posset, ut sublatis paulatim partibus omnibus, quae nunc insunt, tamen aliquando
res eadem dicatur, ut navis Thesei. Ibid.
218
Here, Leibniz makes reference to the book A short Discourse upon the Reasonableness of Men’s having
a Religion or Worship of God, by George Villiers, Duke of Buckingham. For more information, see
Mugnai, 515.
219
Si qui autem substantiam fatentur interire et nasci mutationibus (quae fuit sententia ducis Buckinghamii
in Schediasmate ingenioso pro religionis veritate scripto) revera tollunt omnem substantiam mutabilem.
220
Cum enim perpetuae sint rerum mutationes, ut nihil per minimum temporis spatium in eodem statu
permaneat, sequitur nullam unquam substantiam mutabilem existere et vel minimum durare, quod enim
quovis momento nascitur et perit, nec existere proprie dicendum est, nec agere, aut efficere aliquid aut pati
potest cum nihil nisi aliquo tempore durante efficiatur.
106
duration of time. I take Leibniz to rule out the possibility of a substance existing for only
an instant, instead affirming what I’ll call the No-Momentary-Substance thesis:
For any created substance s and any time t, if s exists at t then there is at least one
time t’ such that t’ is not identical to t and for any time t* between t and t’, s exists
at t*.221
Additionally, Leibniz draws upon the thesis that created substances continually change—
that no created substance ever remains in the same state.222 We’ll call this the PerpetualChange thesis:
For any created substance s and any time t and any time t’, if s exists at t and s
exists at t’ and t is not identical to t’, then for any state of a substance C, if C is a
state of s at t then it is not the case that C is a state of s at t’.223
Now Leibniz has all he needs to argue against (A). Recall that on (A), the reality of
accidents is part of the reality of created substances. Given (A) and Leibniz’s
mereological essentialism, any time a substance changes accidents, the substance ceases
to exist. But on Perpetual-Change thesis, substances continually change accidents. So no
substance exists for more than an instant from (A), Leibniz’s mereological essentialism,
and the Perpetual-Change thesis. Any created substance s that exists at a time t with
accidents A1, A2, A3, . . ., An must change its state by changing at least one of its accidents,
221
I find Leibniz endorsing a very similar constraint, if not the same constraint, later in his career. In the
Theodicy, Leibniz writes, “If the created substance is a successive being, like movement; if it does not
endure beyond a moment, and does not remain the same (during some stated portion of time) any more than
its accidents; if it does not operate any more than a mathematical figure or a number: why shall one not say,
with Spinoza, that God is the only substance, and that creatures are only accidents or modifications?” See
(G VI.350-51: T 393).
222
That created substances continually change is a non-negotiable metaphysical thesis that Leibniz defends
throughout his career. For example, in the Monadology, Leibniz writes, “I also take it for granted that
every created being, and consequently the created monad as well, is subject to change, and even that this
change is continual in each thing.” See (G VI.608: AG 214)
223
I note that while the Perpetual-Change thesis is a thesis about the states of a substance, it entails that the
accidents of a substance continually change. States consist—in some manner—of a substance’s accidents.
In fact, a state of a substance just is the sum total of that substance’s accidents at a time. Supposing a
mereological essentialism about states, when a substance s in state C1 loses an accident A1 and gains an
accident A2, s changes from state C1 to state C2.
107
as the Perpetual-Change thesis requires. But when a substance s loses at least one of its
accidents, s ceases to exist. However, recall that on No-Momentary-Substance thesis, a
substance cannot exist for only an instant. As (A), Leibniz’s mereological essentialism,
and Perpetual-Change thesis entail that no substance exists for more than an instant but
the No-Momentary-Substance thesis requires that substances exist for more than an
instant, no created substances would exist. With the added premise that created
substances do exist, Leibniz has all he needs for the third premise of his argument for the
Merely-Modal-Reality thesis:
(3) It is not the case that (A) the accidental reality is part of the reality of the
substance.
A Spinozist might agree with Leibniz’s reasoning up until the point that affirms the
existence of created substances. The Spinozist instead might take the argument as an
argument against the existence of created substances and offer substance monism as a
solution instead. What we think are created substances are instead accidents of the one
divine substance. Leibniz anticipates such a response but argues that such a substance
monist solution only pushes the problem back a step:
Truly, nor do they thus avoid [the problem], so that in this way the changes which
created substances undergo (naturally enduring) are forced to be brought over into
God, and thus neither shall God himself endure, but shall continuously perish and
be born.224
The Spinozist then has the same difficulty of explaining how accidents are related to
God, in which case the substance monist solution is not a solution.
224
Verum nec sic [effugiunt], quin hoc modo mutationes quas ademere substantiis creatis (quippe sublatis),
in deum transferre cogantur, atque ita nec deus ipse durabit, sed continue interibit et nascetur.
108
Someone might further object that there is a different way to understand (A):
Substances consist in a substantial permanent part that persists through change and an
accidental part that perishes with change. On this understanding, all that is required for a
substance to persist through change is the persistence of substance’s permanent part.
This alternative understanding of (A) has to reject Leibniz’s mereological essentialism
which holds that all the parts of a whole are necessary for the existence of the whole.
Instead, only the essential parts are necessary. When a substance changes accidents, it
only loses an accidental part. With no change to the essential permanent part, the
substance persists.
Leibniz claims that this alternative understanding of (A) is tantamount to arguing
(B): The accidental reality is not part of the reality of the substance. He writes,
“Therefore, if someone wants a permanent part of the reality and a changeable part, they
happen in their opinion, those who prefer to add to the substantial reality something from
accidents.”225 So Leibniz next considers (B), offering at least three reasons to reject it as
well.
Leibniz gives the first two reasons when he writes, “For it will be able to be asked
why those added realities are said to belong to the substance as it were in a subject, and
why it is not considered as a thing per se, even though not enduring.”226 Here, I take
Leibniz to argue that if (B), then it is difficult to explain how an accident could be an
accident of the substance that the accident allegedly belongs to and also difficult to
225
Itaque si qui partem realitatis permanentem, partem mutabilem volunt, incidunt in eorum opinionem, qui
substantiali realitati addi aliquid ab accidentibus volunt.
226
Quaeri enim poterit cur illa realitas superaddita dicatur inesse substantiae tanquam subiecto, et cur non
consideretur ut res per se, licet non-permanens.
109
explain how the accident is an accident at all. I’ll start with the first difficulty. On (B), a
substantial permanent part is one part PS of a whole W and an accidental part is a
different part PA of W. Leibniz doesn’t elaborate but I think some sense can be made of
his objection. What we have are two distinct parts, PS and PA, of a greater whole W. But
then it doesn’t make sense to say that PA inheres in or is an accident of PS any more than
one spatial part of an object, such as the left half of a sphere, is an accident that inheres in
a distinct non-overlapping spatial part of the same object, such as the right half of a
sphere.
Second, it is difficult to explain how such an accident is an accident at all, rather
than a thing per se— a substance. The second difficulty entails the first. If the accident
is not an accident, then surely the accident cannot be the accident of a substance. Rather
than there being an accident that belongs to a substance, there are two substances. It’s
worth noting that the “accident” would be, according to Leibniz, a non-enduring thing per
se. If the thing per se is a substance, then Leibniz has all the more reason to reject (B), as
(B) also entails non-enduring substances, which Leibniz adamantly denied in his
argument against (A).
The third difficulty with (B) is expressed in the following passage:
But if that inherence seems to really affect the reality of the substance, so that it
exists somehow in close union by some real it exists, it is not apparent, how the
accidental [reality] is able to perish, without change in the substantial reality it
[the accidental reality] originates from. Therefore, it itself will be divided again
into a perishing and permanent part, contrary to hypothesis.227
227
Quodsi inhaesio illa videtur realiter afficere realitatem substantialem, ut in connexione aliqua reali
consistat, non apparet, quomodo accidentalis posit interire, quin oriatur mutatio in realitate substantiali,
ergo rursus ipsa dividenda erit in partem pereuntem, et permanentem, contra hypothesin.
110
I understand Leibniz to reason as follows: The accident is related, using the term
“related” loosely, to the substance via inherence. On (B), this inherence would be what
relates the accidental part with the substantial permanent part. But by inhering in the
substantial permanent part, the accident’s union with the substance would have to
somehow affect the reality of the substantial permanent part. Otherwise, the accident
would not be the substance’s accident. However, according to Leibniz, the substantial
permanent part would then have to perish when the accident ceases to exist. Thus (B)
doesn’t avoid the problem raised by (A). If the substance is to avoid perishing,
something would have to persist. Given that something has to persist, the substantial part
would itself have to be divided into a substantial and accidental part, call them
substantial2 and accidental2. Accidental2 would be what is affected by the first accidental
part while accidental2 inheres in substantial2. While Leibniz doesn’t state it, the
argument suggests an infinite regress. The same problem would apply to substantial2 and
accidental2, requiring that substantial2 be divided into a substantial and accidental part,
call them substantial3 and accidental3, ad infinitum.
With these three difficulties with (B), Leibniz argues:
(4) It is not the case that (B) the accidental reality is not part of the reality of the
substance.
And now Leibniz has all he needs to conclude:
(5) Therefore, the Merely-Modal-Reality thesis is true.
However, Leibniz doesn’t conclude DRA with the Merely-Modal-Reality thesis and
therefore a negative answer to Q1— the position that accidental reality is merely modal.
111
In fact, Leibniz never actually gets to the conclusion (5). Instead, Leibniz surprisingly
concludes DRA with a non-committal or agnostic answer to Q1:
I say therefore that substances change, or at diverse times their attributes are
unlike; for this has no doubt, whether however in change there is something real
that perishes and is born; and whether there are diverse realities in a substance,
which are the foundations of diverse predicates, it is not necessary to ask, and if
asked, it is difficult to decide.228
Notice that in the above passage, Leibniz denies that two questions need to be answered:
(Q3) Is there something real in substances that perishes and is born when a
substance changes?
(Q4) Are there diverse realities in a substance, which are the foundations of
diverse predicates?
Leibniz ends the passage claiming that it is difficult to decide the answer to these
questions. The reason it is difficult to decide is understandable, given the complications
that Leibniz raised throughout DRA, as we’ve seen. What’s surprising, however, is
Leibniz’s response. It’s fair to wonder why Leibniz does not instead give a negative
response to Q3 and Q4 given difficulties he raised in DRA. Speculating about Leibniz’s
motivations is not the main goal of this Chapter, but perhaps Leibniz held out hope that
he could eventually arrive at an account of accidental change that avoided the difficulties
he raised in DRA.229 In fact, the later Leibniz did think he had an account or a skeleton
of an account of accidental reality, as I shall argue soon.
228
Dicam igitur substantiam mutari, seu diversis temporibus diversa eius esse attributa; hoc enim
dubitationem non habet, an autem mutatione aliqua realitas intereat, et oriatur; et an diversae sint realitates
in substantia, quae sint fundamenta diversorum praedicatorum, quaeri necesse non est, et, si quaeratur
difficile est diiudicatu.
229
Some scholars might interpret Leibniz as concluding the argument with a variant of nominalism, in
which accidents have no reality or there are no accidents. However, I argue that Leibniz concludes the
argument instead with an agnosticism about the reality of accidents. Leibniz never claims in DRA that
accidents have no reality. Instead, Leibniz claims that it not necessary to answer that question and difficult
to decie if one attempts to answer the question. This agnosticism is consistent with the variant of
nominalism that denies that accidents have any reality, but Leibniz’s agnostic conclusion does not entail it.
112
Before I present that account, it’s worth clarifying Q3 and Q4 some more, and
relating them back to Q1 and Q2. I suspect that an affirmative answer to Q3 entails an
affirmative answer to Q4 and vice versa. That is, if there is something real that perishes
and is born when a substance changes, then there are different realities in substances that
are the foundations of diverse predicates.230 If there are different realities in substances
that are the foundations of diverse predicates, then given that substances continually
change, there is something real, which is born and perishes when a substance changes.
Further, once one gives an affirmative answer to Q3 and Q4, then it would be mighty
helpful to address Q2: Just what are those diverse realities that perish and are born and
are the foundations of diverse predicates? In other words, what is the nature of those
realities?
§2 Leibniz’s Changing Stance on the Merely-Modal-Reality thesis
Leibniz’s arguments in DRA strongly suggest that a negative answer to (Q1) “Do
accidents have a reality that is more than modal?” entails a negative answer to (Q3) “Is
there something real in substances that perishes and is born when a substance changes?”
and (Q4) “Are there diverse realities in a substance, which are the foundations of diverse
predicates?” Negative answers to Q1, Q3, and Q4 further entail that one can’t address
what kind of reality accidents have (Q2).
The not much later Leibniz disagreed. In his 1692 “Critical Thoughts on the
General Part of the Principles of Descartes”, Leibniz writes:
230
Presumably Leibniz is talking about diverse predicates over time in the same subject.
113
To deny a real distinction between modes is an unnecessary change in the
accepted use of words. For until now modes have been considered as things
[original language] and have been held to different in reality, as a spherical figure
of wax differs from a square one. Certainly, the transformation of one figure into
the other is a true change, and it has therefore a real foundation.231
While Leibniz is writing about modes as opposed to non-modal accidents, he claims that
a change of modes has a real foundation. Thus, Leibniz here doesn’t assume that a
negative answer to Q1—the Merely-Modal-Reality thesis—entails a negative answer to
Q4. If a body changes shape from F to G, Leibniz claims that that is a true change with a
real foundation. If there is a real foundation to changing modes, then as I argued above,
there is also a real foundation to diverse predications. Take the following two statements,
where t’ and t are different times:
(S1) The body B is cube-shaped at t.
(S2) The body B is sphere-shaped at t’.
S1 and S2 have the same grammatical subject—“B” designating a body B—but contrary
predicates. The Leibniz of DRA, as we saw, denied that one needs to address whether or
not there are different realities in B that are the foundations of the diverse predicates of
statements such as S1 and S2. However, the not much later Leibniz now affirms that B’s
changing shape has a real foundation. As I argued above, Leibniz is then committed to
the different predicates of S1 and S2 having a real foundation. If Q3 and Q4 are interentailing, as I argued above, then Leibniz is further committed to there being different
realities in B that perish and are born. So Leibniz once again must address all the
complications raised in DRA.
231
(G IV.365: L 390). Emphasis added.
114
2.1 Different Types of Realities
I argue that Leibniz did have a solution in mind. Several years later, in his Theodicy,
Leibniz makes a three-fold distinction between different types of realities:
God is the one principal cause of pure and absolute realities, or of perfections.
Causae secundae agunt in virtute primae. But when one comprises limitations
and privations under the term realities one may say that the second causes cooperate in the production of that which is limited; otherwise God would be the
cause of sin, and even the sole cause.232
Leibniz claims above that there are three types of realities: (i) pure and absolute realities,
or perfections; (ii) limitations; and (iii) privations. For the purposes of this chapter, I’ll
focus on absolute realities and limitations, bracketing discussion of privations.233 For the
sake of the discussion at hand, I’ll also call the realities picked out in (i) “AbsoluteRealities” and the realities picked out in (ii) “Limitations” or “Limited-Realities”. In
what follows, it’s worth briefly working through Q1-Q4 given this three-fold distinction
Leibniz has introduced in his Theodicy.
For review, Q1 asks, “Do accidents have a reality that is more than modal?” As I
argued in Chapters 2 and 3, all accidents for Leibniz are modifications. So the answer to
Q1 is still “No”. Thus, Leibniz maintains, throughout his career, that the Merely-ModalReality thesis is true. However, as we’ve seen, this doesn’t prevent Leibniz from
answering Q3 and Q4 positively later in his career, contra his conclusion in DRA. One
reason is that, as we saw in Chapters 2 and 3, modifications are limitations. Given that
232
(G VI.347-48: T 392).
I will focus on limitations in what follows, noting that recent scholars argue that privations are a type of
limitation for Leibniz. For example, see Samuel Newlands, “Leibniz on Privations, Limitations, and the
Metaphysics of Evil,” Journal of the History of Philosophy 52, No. 2 (2014): 281-308.
233
115
limitations are realities, for Leibniz, Leibniz can answer Q3 and Q4, which provide him
the workings of an answer to Q2.
Recall that Q3 asks, “Is there something real in substances that perishes and is
born when a substance changes?” Leibniz’s answer to Q3 is “Yes”. What perishes and
are born, however, are not absolute-realities. Instead, what perishes and are born are
limited-realities. Accidents, which are modifications, which are limitations, are realities
that come into and go out of existence. So when a substance changes accidents,
limitations are what come into and go out of existence.
Q4 asks, “Are there diverse realities in a substance, which are the foundations of
diverse predicates?” Leibniz’s answer again is “Yes”. However, the foundations of
diverse predicates are not absolute-realities. Instead they are limited-realities. Diverse
limited-realities—different modifications/limitations—are the foundations of diverse
predicates.
Recall that Q2 asks “What is the nature of accidental realities?” Leibniz has the
beginnings of an answer: Accidental realities are not absolute-realities. Instead they are
limited-realities— accidents are limitations on their substances. This, however, still
doesn’t provide much information as to the nature of accidents. An example Leibniz
often gives of a limitation is shape.234 A shape limits what is shaped. Yet, as we saw
above, when something changes shape, that is a true change according to Leibniz.
234
For example, Leibniz writes, “I have often said, and I do not remember having deviated from the view,
that unless there is some active principle in us, there cannot be derivative forces and actions in us, since
everything accidental or changeable ought to be a modification of something essential or perpetual, nor can
it contain anything more positive than that which it modifies, since every modification is only a limitation,
shape a limitation of that which is varied, and derivative force a limitation of that which brings about the
variation.” See (G II.270: AG 180).
116
According to Leibniz, the distinction between shape s1 a body B has at t1 and shape s2 B
has at t2 is neither a mere distinction of reason nor a modal distinction. I note, however,
that this still does not provide much by way of answer to the nature of accidental reality.
For example, while it easy to grasp how an accident such as shape is a limitation of its
substance, what about an accident such as a perception—with content—of an immaterial
monad?
2.2 The Arguments of DRA, Absolute-Realities and Limited-Realities
For now, I shall set aside this issue and address, to my mind, the more pressing question I
raised above as it directly relevant to the arguments Leibniz makes in DRA: Does
Leibniz’s distinction between absolute-realities and limitations allow him to avoid the
problems he raised in his argument for the Merely-Modal-Reality thesis, specifically the
problems with affirming either (A) the accidental reality is part of the reality of the
substance or (B) the accidental reality is not part of the reality of the substance? In what
follows, I shall argue that Leibniz has a way to argue that the problems he raised with (A)
and (B) are only problems if accidents are absolute-realities. I shall do so by turning to
Leibniz’s mereology.
In what follows, I argue that the problems Leibniz found in DRA both with
accidents that have a more than modal reality and accidents that have a merely modal
reality stemmed from Leibniz worrying that either type of accident—as a reality that is
born and perishes in change and serves as the foundations of diverse predicates—is a part
of its substance. I then argue that given Leibniz’s technical mereological views, the later
Leibniz had a way to hold that accidents—as modifications and limitations—could be
realities that born and perish in change and serve as the foundations of diverse predicates
117
without beings parts of their substances. I argue that such modifications cannot be
homogenous with their substances, where homogeneity is a necessary condition of
parthood in Leibniz’s mereology. To argue this, I examine Leibniz’s notion of
homogeneity—a technical notion that has frequently been misunderstood by scholars, as I
show. After reconstructing Leibniz’s notion of homogeneity, I argue that modifications
or limitations cannot be homogenous with what they modify. I further argue that
accidents that have more than modal reality (which would count as absolute-realities in
Leibniz’s metaphysics) could be homogenous with their substances, and so parts of their
substances.
Before presenting my argument, it’s worth addressing just why I should even
bother to offer an argument on Leibniz’s behalf. Mainly, Leibniz assumed in DRA that
the problematic disjunction (A) or (B) is entailed by the denial of the Merely-ModalReality thesis. Yet holding that accidents are absolute-realities is just to affirm the
Merely-Modal-Reality thesis, which does not, according to Leibniz, entail (A) or (B).
In response to the objection, Leibniz claimed that a substance’s changing
accidents, even when the accidents are limited-realities, is a true change with a real
foundation. As I argued above, a true change’s having a real foundation entails that there
are diverse realities in a substance that come to and cease to exist when the substance
changes— Q3 entails Q4 and vice versa. The fact that Leibniz gave a non-committal
answer to Q3 and Q4 at the end of DRA strongly suggests that he thought that any
answer lead to difficult to solve puzzles. The further fact that Leibniz claims this
immediately after working out the implications of (A) and (B) is strong evidence that the
problems Leibniz had in mind were the problems with (A) and (B). So it’s worth
118
exploring whether or not Leibniz—in claiming that the realities that perish and are born
and which are the real foundation of true changes are R2 realities—can avoid the
problems he raised with (A) and (B).
2.3 Leibniz’s Mereology and Accidents
Leibniz’s mereological writings are the best place to address this challenge. I’ll
specifically focus on the account of parts and wholes Leibniz developed in his 1714 “The
Metaphysical Foundations of Mathematics.” Both (A) the accidental reality is part of the
reality of the substance and (B) the accidental reality is not part of the reality of the
substance assume that accidents are parts. Option (A) holds that accidental reality is part
of the reality of the substance.
So, on option (A), substances are wholes that have accidents as parts. Option (B),
as Leibniz developed it in DRA, is the thesis that accidental realities are also a part of the
substance, only the accidental part not identical to an essential substantial part that
persists through change. So, on (B), a substance is whole with at least two parts— the
essential substantial part and the accidental part. What’s important for my argument is
that on both (A) and (B), the accidental reality is a part. I’ll call the thesis that accidents
are part of the substances they inhere in the thesis of Substantial-Composition: For any
accident A and any substance s, if A inheres in s then A is a part of s. This leads to a
premise:
(P1) If (A) or (B) then the Substantial-Composition thesis is true.
The first premise of Leibniz’s argument in DRA held that the denial of the MerelyModal-Reality thesis entailed the disjunction (A) or (B). However, as I’ve argued above,
Leibniz now needs a reason to also argue that the affirmation of Merely-Modal-Reality
119
thesis does not entail the disjunction (A) or (B), where the Merely-Modal-Reality thesis is
understood now as the thesis that accidents are limited-realities. That is, Leibniz needs to
argue that it is not the case that the Merely-Modal-Reality thesis entails (A) or (B). P1
presents a necessary condition for the disjunction (A) or (B). So if the Merely-ModalReality thesis entails (A) or (B), then the Merely-Modal-Reality thesis entails SubstantialComposition. A reason to deny that the Merely-Modal-Reality thesis entails SubstantialComposition would then be a handy way to deny that the Merely-Modal-Reality thesis
entails (A) or (B). In what follows, I argue that Leibniz actually has the resources to
argue that if the Merely-Modal-Reality thesis is true, then it is not the case that accidents
are parts of substances. That is, Leibniz can argue the Merely-Modal-Reality thesis
entails that Substantial-Composition is false, which is a much stronger claim. For the
former claim is consistent with the Merely-Modal-Reality thesis being the case and
Substantial-Composition being the case. It only denies that the Merely-Modal-Reality
thesis is sufficient for Substantial-Composition. But if Leibniz can argue that the MerelyModal-Reality thesis entails the falsity of Substantial-Composition, which I shall argue
he can, then a fortiori he can argue for weaker claim.
How might one address whether or not the Merely-Modal-Reality thesis entails
Substantial-Composition or its negation? That is how might one address whether or not a
modification of a substance is a part of that substance? Thankfully, Leibniz has precise
criteria for when any x counts as a part of any y: “An entity which is in something and is
also homogeneous to it is called a part, and that which it is in is called a whole; or a part
120
is a homogenous ingredient of a whole.”235 Formalized, Leibniz’s criteria of part-hood
can be expressed thus:
For any x and any y, x is a part of y if and only if:
(i)
(ii)
x is an ingredient of y; and
x is homogeneous with y.236
To argue that the Merely-Modal-Reality thesis entails the falsity of SubstantialComposition, I must establish that modes of a substance are either not ingredients of a
substance or not homogeneous with their substance. However, to establish that, I need a
definition of “ingredient” and “homogeneous”, which Leibniz also supplies. I’ll start
with ingredient. Leibniz writes:
We say that an entity is in [inesse] some locus, or is an ingredient of something,
if, when we posit the latter, we must also be understood, by this very fact and
immediately, without the necessity of any inference, to have posited the entity as
well. Thus when we posit any finite line, we also posit its end points as belonging
to it.237
Leibniz’s definition of an ingredient can then be expressed thus:
For any x and any y, x is an ingredient of y if and only if for any person P, if P
posits y, P immediately posits x.
Leibniz’s example is the endpoints that are ingredients of a finite line. One cannot posit a
finite line without positing the endpoints. While Leibniz’s examples are geometric, there
is a sense in which a modification could be understood as ingredients of its substance. In
235
L 668.
I should note that two further conditions could be added: (iii) x is not identical to y and (iv) y is a
whole. Condition (iii) implies that the definition of a part is in fact a definition of a proper part. I don’t
encounter Leibniz ever developing the notion of an improper part in his mereological writings, so I’ll leave
condition (iii) aside. Condition (iv) I take to be entailed by (i), (ii), and (iii), given Leibniz’s mereological
views. That is, if x is a homogeneous ingredient of y and not identical with y, then y is a whole of which x
is a part.
237
L 667.
236
121
Chapter 3, we saw that Leibniz claims that a created substance must have some accident
or other.238 On that understanding, the positing of a created substance requires the
positing of some accident or other. However, there are two complications with
addressing whether or not accidents could be ingredients in addressing whether or not
accidents are parts. First, it is not clear if Leibniz’s criteria of ingredient-hood requires
the positing of the very individual thing posited, or just some entity of the right type. The
latter option leads to a further complication. Leibniz—as we saw in Chapter 3— also
argues that accidents require substances.239 Thus, substances—in the sense suggested by
the latter option—would be ingredients of accidents. But surely substances could not be
parts of accidents. So it is not clear if accidents are ingredients of their substances and it
is also not clear if accidents are not ingredients of their substances.
In what follows I’ll assume that accidents are ingredients.240 However, I also
argue that a more fruitful avenue for addressing whether or not accidents could be parts
of substances lies in the second necessary condition of parthood— homogeneity. Recall
that the second necessary condition of parthood is that x be homogenous with y.
Homogeneity is also a technical term for Leibniz. Prima facie, one might take Leibniz to
be adopting the notion of homogeneous parts and wholes that Aristotle articulated in the
238
Leibniz writes, “I do not know whether the definition of substance as that which needs for its existence
only the concurrence of God fits any created substance known to us, unless we interpret it in some unusual
sense. For not only do we need other substances; we need our own accidents even much more. Therefore,
since substance and accident depend on each other, other marks are necessary for distinguishing a
substance from an accident. Among them may be this one: That a substance needs some accident but often
does not need a determinate one but is content, when this accident is removed, with the substitution of
another. An accident, however, needs not only some substance in general but that very one in which it
inheres, so that it cannot change it.” (G IV.365: L 390).
239
Ibid.
240
I note that I am not arguing that accidents are ingredients of their substances. Instead, I’m assuming
they are for the sake of argument so I can explore whether or not accidents are homogeneous with their
substances.
122
Parts of Animals.241 Aristotle gives the example of blood, flesh, and bone as
homogeneous wholes. Any part of an animal’s blood is also blood, any part of an
animal’s bone is bone, and any part of an animal’s flesh is also flesh, according to
Aristotle. This leads to the question of what feature of the part and whole make the part
and whole homogeneous for Aristotle? It can’t simply be any predicate. For example,
Socrates (a whole) and Socrates’ left hand (a part) could both have the predicate “exists
in Athens before the birth of Christ” truly predicated of them, yet that doesn’t mean that
Socrates’ left hand is homogeneous with Socrates in Aristotle’s usage of the term.
Instead, the predicate must be a kind-term that indicates what kind of thing or stuff the
part and wholes are. For example, a gold brick and the left-half of a gold brick can both
have the kind-term “gold” truly predicated of them.
Is this the conception of homogeneity that Leibniz has in mind? In places,
Leibniz makes claims that suggest he is adopting Aristotle’s usage. For example, in his
1690 “Comments on Michael Angelo Fardella,” Leibniz writes, “Further, although the
aggregate of these substances constitutes body, they do not constitute it as parts, just as
points are not parts of lines, since a part is always of the same sort as the whole.”242
Leibniz here uses the term “sort”, which might be taken to mean what contemporary
metaphysicans mean by the term “sortal”. However, contemporary sortal predicates are
count-nouns, such as horse, unicorn, book, etc. But a proper part of a unicorn is not a
unicorn. Philosophers further deny that any sortal predicate that applies to some entity x
241
242
See Aristotle, Parts of Animals, Book 1, Part 1, 640b18-22.
See AG 105. Emphasis added.
123
applies must also apply to a part of x.243 Instead, Leibniz’s usage of “sort” suggests
something closer to Aristotle’s usage. A part of a body is a body, a part of a line segment
is a line segment, and so on.
This has lead many scholars to take Leibniz to be adopting Aristotle’s usage of
homogeneity, or at least write as if they take Leibniz to be adopting such usage. For
example, Robert Adams writes:
A part of a line, on this view, must be homogeneous with the line, and therefore
must be a line segment, and not a point. See Aristotle Physics Book VI, CH. 1.
Similarly, I suppose, a part of a phenomenon must be a phenomenon, and a part of
an aggregate must be a subaggregate. Specifically, a part of a body must be a
body. If a body is an aggregate of substances which are not aggregates, those
substances will not be parts of the body, and the boy will not be composed of
them if being ‘composed of’ means having as parts.244
Pauline Phemister, commenting on Leibniz’s “Comments on Michael Angelo Fardella,”
writes:
To support this conclusion [points not parts of lines, souls not parts of matter,
bodies are parts of matter], Leibniz appeals to the principle of the homogeneity of
parts and wholes. For anything to count as a part of a larger whole, it has to be of
the same nature as the whole. The converse, that those things which do not
possess the same nature cannot stand in relation of part to whole, allows Leibniz
to assert that those substances which are aggregated together as a body are not
parts of those bodies.245
243
See Richard E. Grandy, "Sortals", The Stanford Encyclopedia of Philosophy (Spring 2014 Edition),
Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/spr2014/entries/sortals/>.
244
See Robert Adams, Leibniz: Determinist, Theist, Idealist, 244. Adams also writes, “The meaning of this
passage turns on Leibniz’s conception of the parts of bodies. The statement that ‘a part is always
homogeneous with the whole’ is a key to this. A body, according to the memo, is an aggregate. Its
homogeneous parts, therefore, are subaggregates (and thus still bodies) rather than the indivisible
nonaggregates ‘of’ which it is an aggregate. The latter we might call ‘elements’ as distinct from ‘parts’ of
the corporeal aggregate (as I have suggested in Chapter 9, section 3.1), and I think nothing is said here to
preclude their being (concrete) souls or, as I put it, ‘qualified monads.” See Ibid, 276.
245
See Pauline Phemester, Leibniz and the Natural World: Activity, Passivity and Corporeal Substances in
Leibniz’s Philosophy (Dordrecht: Springer, 2005), 98. Phemester also writes, “Since mere aggregate
bodies and corporeal substances are of essentially different natures, the homogeneity principle rules out
corporeal substances as parts of bodies as effectively as it eliminated souls.” See Ibid, 98-99.
124
And Benson Mates, commenting on whether or not component concepts are parts of
complex concepts, writes:
Thus, he [Leibniz] says, the parts of lines are not points, but other lines, although,
in his use, points are ‘in’ lines. On this definition it would appear that the
component concepts of a complex concept are parts of it, assuming that all
concepts are to be considered ‘homogeneous’.246
There are, however, two problems with understanding Leibniz’s usage of homogeneity
exactly in this Aristotelian way. First, in other places Leibniz denies that there are
homogeneous parts in Aristotle’s sense. Instead, such homogeneous parts in fact turn out
to be heterogeneous:
The point is that people mistakenly take these bodies to be homogeneous or
uniform, whereas really they are more mixed than they are thought to be. When
dealing with heterogeneous bodies, one is not surprised to find differences
between individual samples: physicians know only too well how much human
bodies differ in their balance and their constitution. In short, as I have remarked
earlier, we shall never be able to find species which are logically the lowest; and
two real, i.e., complete, individuals belonging to a single species will never be
perfectly alike.247
I take Leibniz to argue for the claim that any apparent homogeneous whole W, upon
closer inspection, would turn out to be a heterogeneous whole. There are two reasons
Leibniz argues this, both found in the above passage. First, Leibniz’s reason stems in
part from his denial that no two things are ever of the exact same species, a thesis Leibniz
maintained throughout his career. Granted, Leibniz usually applies the reasoning to
246
See Benson Mates, The Philosophy of Leibniz: Metaphysics and Language (Oxford: Oxford University
Press, 1986), 61.
247
See (G V.284: NE 305). Leibniz also writes, “I believe, however, that the four bodies they call
elements, which they believe to be simple, as well as the salts, metals, and other bodies which they believe
to be perfectly mixed, with their ingredients in fixed proportions, are not unum per se either – particularly
since we should regard them as only apparently uniform and homogeneous, and even a homogeneous body
would still be an aggregation.” See (G.V.308: NE 328).
125
created substances. No two monads are of the same species. Instead, they are like
Thomistic angels. Perhaps Leibniz also applies this to corporeal substances as well (even
if, on the idealistic interpretation I assume in this dissertation, Leibniz denies that there
are corporeal substances): No two bricks of gold are really of the same exact species or
kind. Second, Leibniz perhaps is advancing an assumption based on advances of the
science of his time. To appearances, every part of some portion of blood is also blood.
So a portion of blood only has homogeneous parts. Yet if one zooms in with a
microscope, one will instead find heterogeneous parts. So a portion of blood appears to
be a homogeneous whole, but instead is a heterogeneous whole.
Second, and more importantly, as is the case with many terms Leibniz uses,
“homogeneity” is also a technical term with a precise meaning that differs in some
respects from the Aristotelian notion of homogeneity. Leibniz writes:
Two entities are homogeneous to which two other entities can be assigned which
are equal to them and similar to each other. Given A and B; if L is taken equal to
A, and M equal to B, and L and M are similar, we call A and B homogeneous.
Hence I usually also say that homogeneous entities are those which can be made
similar to each other by means of transformations, like curves and straight lines.
That is, if A is transformed into its equal L, it can be made similar to B, or to its
equal M into which B is assumed to have been transformed.248
Leibniz seems to offer two definitions of homogeneity. I’ll use the terms
“homogeneous1” and “homogeneous2”, respectively.
248
L 667.
126
The first definition can be expressed thus, using the same variables Leibniz used:
For any A and any B, A is homogeneous1 with B if and only if: There is at least
one L and at least one M, such that:
(i)
(ii)
(iii)
L is similar to M;249
L is equal to A; and
M is equal to B.
The second definition Leibniz presents in the same passage defines two entities x and y as
homogeneous2 when they can be made similar to each other by means of transformations.
Rather than offering two unrelated definitions, the text suggests that the two definitions
are inter-entailing. That is, if x is homogeneous1 with y if and only if x is homogeneous2
with y. If two entities satisfy conditions (i) - (iii) of homogeneity1, then those two entities
can be made similar to each other by means of transformations. Of course, stating the
definitions and that they inter-entail is not the same thing as explaining why they do, so I
now attempt to do just that.250
First, to unmuddy what are some still murky waters, it’s helpful to also see how
Leibniz defines the terms “similar” and “equal” in the definition of homogeneous1, as
they are also technical notions for him. I’ll start with “equal”, which Leibniz defines
when he writes, “Equals are things having the same quantity.”251 Leibniz elaborates what
he means by “quantity” when he writes, “Quantity or magnitude is that in things which
can be known only through their simultaneous compresence – or by their simultaneous
perception. Thus it is impossible for us to know what a foot or a yard is unless we
249
For reasons that will become apparent soon, similarity is symmetric relation. If x is similar to y, then y
is similar to x. Hence, I see no need to add the condition (iv) M is similar to L.
250
Roy T. Cook finds a third definition in this very passage, only to argue that the third definition is
consistent with and entailed by the first. I’ll draw upon the first definition primarily. See Roy T. Cook,
“Monads and Mathematics: The Logic of Leibniz’s Mereology,” Studia Leibnitiana (2000): 9-12.
251
L 667.
127
actually have something to serve as a measure which can be applied to successive objects
after each other.”252
What’s not clear from this passage is whether Leibniz thinks he is offering (i)
simply a handy way to know whether some property F is a quantitative property or (ii)
the definition of a quantitative property or (iii) a way to know the particular quantity—
i.e., the measure—of something. The first sentence suggests (i) or (ii). However, in the
second sentence, Leibniz seems to assume (iii): To know if some x is a foot long, one
must have some other entity y that is a foot long that can be compared to x. Fortunately,
one doesn’t have to decide on which of (i) – (iii) applies to understand how quantity
pertains to equality. According to commentators, two objects are equal in quantity if and
only if they are equal in size.253 This understanding of equality will suffice for now.
Now onto Leibniz’s definition of “similarity”. Leibniz writes, “Similars are
things having the same quality.”254 Two objects o1 and o2 are similar if there is some
quality token F1 of quality type F inhering in o1 and some quality token F2 of quality type
F inhering in o2. Leibniz also elaborates on what he means by “quality”, “Quality, on the
other hand, is what can be known in things when they are observed singly, without
requiring any compresence. Such are the attributes which can be explained by a
definition or through the various modes which they involve.”255
Leibniz’s elaboration on what he means by “quality”, at least with respect to his
definition of homogeneity, can be contrasted with what he means by “quantity”, shedding
252
Ibid.
See for example Cook, “Monads and Mathematics: The Logic of Leibniz’s Mereology,” 3-4.
254
L 667.
255
Ibid. Leibniz used this definition of similarity as far back as 1679. See, for example, L 254-55.
253
128
light on both. One can know that an object o1 has a particular quality F1 without
requiring some other object o2 to compare with o1. This is unlike knowing whether or not
an object has a particular quantity, such as length. While it’s not clear if Leibniz is
offering this as a strict definition, once again for the purposes of my argument, it suffices.
I note that some scholars argue that the quality of an object, in Leibniz’s developed
mereological views, is the shape of the object-- two objects are similar when they are the
same shape.256 However, I shall also consider a broader conception of qualitative
similarity not limited to shape in what follows.
To recap, given Leibniz’s definition of a part, a modification can be a part of a
substance only if the modification is homogeneous with its substance. A modification is
homogeneous with its substance only if the modification can be transformed in such a
way that it is similar to its substance. To avoid both (A) the accidental reality is part of
the reality of the substance and (B) the accidental reality is not part of the reality of the
substance, Leibniz needs to a way to argue that a modification is not a part of its
substance. To do so, I shall argue that a modification cannot be homogeneous with its
substance. Once I have established that, I will have provided all Leibniz needs to defend
the premise (P2) If the Merely-Modal-Reality thesis is true, then SubstantialComposition is false.
In what follows, I shall argue that modifications cannot be parts of their
substances on both the idealistic and non-idealistic interpretations of Leibniz, where by
non-idealistic I mean the interpretation that holds that there are bodies and/or corporeal
256
For example, see Graham Solomon, “Leibniz and Topological Equivalence,” Dialogue 32 (1993), 722.
129
substances in Leibniz’s ontology. I’ll start with the non-idealistic interpretation and
argue that modes cannot be homogeneous with the bodies they modify. While I assume
the falsity of the non-idealistic interpretation of Leibniz’s ontology in this project,
starting with the non-idealistic is still beneficial for the following reason: Leibniz’s stock
example of a limitation, as we’ve seen, is the shape of a body. If the shape of a body
cannot be homogeneous with the body shaped and the reason the shape cannot be
homogeneous is due to its being a limitation of that body, then ipso facto a modification
of a substance—be it corporeal or non-corporeal—cannot be homogeneous with the
substance modified. For modifications, according to Leibniz, are limitations.
One might still object to my even bothering to address whether modes could be
homogeneous with bodies. On the idealistic ontology I assume in this project, there are
no corporeal substances or bodies. Instead there are only immaterial monads and their
accidents. According to some scholars, Leibniz’s definition of homogeneity in his
definition of parts is a geometrical definition applying to shaped entities of varying
dimensions such as lines and spheres. But monads are dimension-less. So it’s a category
mistake to claim that an accident of a monad is a part of a monad. Nothing more needs to
be argued.
This objection is correct as far as it goes but it is still fruitful to run the argument
through in greater depth as it sheds light on an important related issue. Monads,
according to Leibniz, are simple even though they are both synchronically and
diachronically complex— monads have a multitude of accidents at a time and they
change accidents over time. Many of Leibniz’s predecessors would have claimed that
130
such synchronic complexity is inconsistent with monadic simplicity. If monads have
accidents, then they are composite and therefore not simple.257
Contemporary scholars have responded to this issue in one of two ways. First,
some scholars have resorted to arguing that Leibniz is ultimately committed to the
position that monads are not really synchronically and diachronically complex. That is,
monads do not have a multitude of accidents at a time and they do not endure through
change of accidents over time.258 This position, however, is in tension with Leibniz’s
numerous arguments defending the reality of monadic change.259 Additionally, Leibniz
argues in the Monadology that there can only be a plurality of monads if they differ in
quality from each other.260
Other scholars who have not been willing to jettison diachronic and synchronic
complexity, have proposed a more promising solution. Accidents are not parts of monads.
Instead, monads are simple entities that serve as parts of composite wholes consisting in
the monad and accident.261 However, this solution runs into problems that Leibniz
257
For example, Aquinas famously argued that God couldn’t have accidents because that would entail that
God is not simple. See SCG 1.23.3.
258
John Whipple defends such a view in a number of articles. See John Whipple, “The Structure of
Leibnizian Simple Substances,” British Journal for the History of Philosophy 18 (2010): 379-410; “Leibniz
on Divine Concurrence,” Philosophy Compass (2010): 865-879; and “Continual Creation and Finite
Substance in Leibniz’s Metaphysics,” Journal of Philosophical Research 36 (2011): 1-30.
259
Such as Leibniz’s Perpetual-Change thesis he appealed to in DRA that we saw earlier in this chapter.
260
In the Monadology, Leibniz writes, “However, monads must have some qualities, otherwise, they
would not even be beings. And if simple substances did not differ at all in their qualities, there would be no
way of perceiving any change in things, since what there is in a composite can only come from its simple
ingredients; and if the monads had no qualities, they would be indiscernible from one another, since they
also do not differ in quantity.” See (G VI.608: AG 214).
261
For example, Timothy Allan Hillman, in his own work on this topic, writes, “Now, accidents all by
themselves need not compromise substantial simplicity. After all, Aquinas and other Medieval
philosophers had maintained that immaterial substances—for example, angels—were singular entities
which, nonetheless, had accidents. Whenever some accident inheres in an angel, the angel’s simplicity is
not corrupted; instead, the simple substance is said to enter into a larger substantial composite of which it
and the accident are constituent parts. So, Leibniz need not fear that a monad’s possession of some mode
131
himself drew attention to in DRA above. Mainly, if monads are simple parts of
composite wholes, of which accidents are the other part, then Leibniz argues that it is
difficult to explain how the accidents are accidents of the monad that inhere in the monad
instead of entities that exist per se. In what follows, I show how Leibniz can maintain
that accidents are accidents of simple monads without being parts of monads or parts of a
greater whole composed by the accident and monad.
A final reason to show that limitations/modifications are not parts of what they
limit even in the case of immaterial monads is that some scholars interpret qualitative
similarity in Leibniz’s definition of homogeneity to not be limited to shape.262 If these
scholars are right then Leibniz’s definition of homogeneity may not be limited to
applying to bodies are corporeal substances, for Leibniz argues that the accidents of
monads are qualities.263 In what follows, I shall also offer reasons to hold that even if
qualities are not limited to shapes, qualities could not be homogeneous with what the
qualities are qualities of.
2.3.1 Shapes not Homogenous with what they Shape
I first argue that the shape of a body—a modification—is not homogeneous with
its body, and therefore cannot be a part of its body. Leibniz suggests one route to this
thesis in his “Metaphysical Foundations of Mathematics” when he writes, “It is clear
from this that a boundary is not homogeneous with what it bounds, nor a section with
or state affects its simplicity per se.” See T. Allan Hillman, “Substantial Simplicity in Leibniz: Form,
Predication, & Truthmakers,” The Review of Metaphysics 63 (2009): 120.
262
For example, in his own work on Leibniz’s technical definition of homogeneity found in Leibniz’s
“Metaphysical Foundations of Mathematics,” Hartz writes, “Whatever is homogeneous must be of a certain
kind K, ‘all the way down’ in its decomposition or ‘all the way up’ in its composition.” See Glenn A.
Hartz, Leibniz’s Final System: Monads, Matter and Animals (New York: Routledge, 2007), 69.
263
See (G VI.608: AG 214). Recall as well in Chapter 3 that qualities are modifications.
132
what it cuts.”264 Earlier in Leibniz’s career, Leibniz identified the boundary of a thing
with its figure or shape. In a 1669 letter to Jacob Thomasius, Leibniz wrote, “Here too
everything agrees remarkably if we assume that form is nothing but figure. For since
figure is the boundary of a body, a boundary is needed to introduce figure into bodies.”265
From these two passages, we have an argument: Shape is what bounds a body, but a
boundary cannot be homogeneous with what it bounds. So a shape cannot be
homogeneous with what it shapes. Therefore, a shape cannot be a part of what it shapes.
Given that shape is Leibniz’s stock example of a modification, at least one type of
modification cannot be a part of what it modifies.
There are two problems, however, with this argument. First, immediately
preceding the passage above in his “Metaphysical Foundations of Mathematics,” Leibniz
writes, “A common boundary of two things is an entity which is in them when they do
not have a part in common. Insofar as these two things are understood to be parts of a
single whole, their common boundary is called a section of the whole.”266 This passage
leads to a tension with Leibniz’s Ownership Thesis of Accidents (OTA) that we saw in
Chapter 3, if the shape of a body—a modification--is to be identified with the boundary
of a body.267 Leibniz claims that two entities can have a common boundary. However,
recall that on OTA, no modification can modify more than one substance at the same
264
L 668.
(G I.18: L 95).
266
L 668.
267
The Ownership Thesis of Accidents can be expressed thus: For any accident A and any substance s, if A
inheres in s then there is not some substance s’ such that s’ is not identical to s and A inheres in s’. See (G
IV.365: L 390).
265
133
time— the synchronic implication of OTA.268 If a shape is a modification and a
modification is a boundary, then Leibniz here is committed to holding that a modification
can modify more than one entity at the same time.
There’s a second problem. Leibniz gives the example of the endpoint of a line
segment being an ingredient of the line segment but not homogeneous with it.269 Instead,
a smaller line segment within the original line segment would be homogeneous.
Commentators have point out that Leibniz generalizes this point, and further argues that a
one-dimensional line that is an ingredient of a two-dimensional polygon is not
homogeneous the two-dimensional polygon and that a two dimensional surface is an
ingredient of but not homogeneous with the three dimensional object the surface is the
surface of, and so on.270 For an entity to be homogeneous with another entity, however,
both entities must have the same number of dimensions— a smaller line segment in a line
segment is homogeneous with the larger line segment unlike the endpoint. This leads to a
necessary condition of homogeneity: for any x and any y, x is homogeneous with y only
if the number of x’s dimensions = the number of y’s dimensions. However, the number
of dimensions of the shape of a body is identical to the number of dimensions of the
body.271 A sphere is three-dimensional and the shape of a sphere is three-dimensional.
So the shape of a body—which is mode of its body—satisfies an important condition of
268
See Chapter 3, section 2.2.
L 667.
270
See Hartz, Leibniz’s Final System: Monads, Matter and Animals, 69.
271
Assuming that the later Leibniz did not identify shape with boundary. If Leibniz did identify shape with
boundary, then there are plausible reasons to suppose that the number of dimensions of a shape is equal to
the number of dimensions of what is shaped minus one. For discussion, see Achille Varzi, "Boundary",
The Stanford Encyclopedia of Philosophy (Winter 2013 Edition), Edward N. Zalta (ed.), URL =
<http://plato.stanford.edu/archives/win2013/entries/boundary/>.
269
134
homogeneity. Given that the shape of a body is also an ingredient of its body (on one
understanding of ingredient I presented above), we’re not far from concluding that the
shape is a part of its body.
This second problem seems also to apply to immaterial substances— monads.
Monads are zero-dimensional entities.272 The modifications of monads—which on one
understanding of ingredient are also ingredients of monads-- are also zero-dimensional.
So again we’re not far from concluding—contra Leibniz’s claims—that the modification
of a monad is homogeneous ingredient of a monad, and therefore a part of a monad.
However, this second problem rests on a confusion. In fact, addressing the
confusion paves the path to establishing that the shape of a body could not be
homogeneous with its body, and further that the modification of a monad could not be
homogeneous with its monad. I’ll start with shapes. If a shape were homogeneous with
its body, then the shape would be similar (or could be transformed to be similar) to its
body. But what would it be for a shape to be similar to its body? It would be for the
shape to have the same shape as its body. But that’s absurd. We would be treating the
shape as if it were a substance with a shape modifying or inhering in it, which we then
compare to the shaped body. Instead, the body has the shape it has because the shape
modifies it and the shape only exists as modifying its body.
A slightly different way to clarify the confusion of the second problem is as
follows. Homogeneity is symmetric. If x is homogeneous with y then y is homogeneous
with x. So if a shape is homogeneous with its body then a body is homogeneous with its
272
Monads, being non-extended, are zero-dimensional. See (G VI.607: AG 213).
135
shape. If a body is homogeneous with its shape, then the body is similar to its shape or
could be made similar to the shape via transformations. Yet once again this is absurd for
the same reason. What would it be for a body to be similar to its shape? It would be for
the body to have the same shape as its shape. But once again we are treating the shaped
body as a distinct entity from the shape shaping the body, which we then compare. The
confusion then lies in supposing that two entities are being compared and then judged
similar—the body and the body’s shape, as if one glances at the body and then glances at
the body’s shape, and then judges the body to have the same shape as its shape.273
Instead, while the shape of a body is an ingredient of its body (on one
understanding of ingredienthood), it is not homogeneous with the body. What would be
both an ingredient of and homogeneous with a body B is a smaller body B’, such as B’s
left half. B’ could be transformed so that it is similar in shape to B. The transformation
would involve B’ changing from having one shape M modifying B’ to a different shape
M’ similar to B’s shape. Thus, commentators have correctly claimed that only a body
could be a part of another body in Leibniz’s metaphysics, even if the route they took to
the conclusion was not the route Leibniz takes.
2.3.2 Limitations are not Homogenous with what they Limit
Further, in addition to a shape not being homogeneous with what it shapes,
Leibniz has grounds to argue that no modification can be homogeneous with what it
modifies. In fact, I shall argue that claiming that a limitation’s being homogeneous with
what it limits leads to a regress. Recall that modifications are limited-realities—
273
Of course if they are really distinct they are two entities, but not two entities have the same shape in the
same respect that would be needed for them both to be homogeneous with each other.
136
modifications are limitations. This especially includes shape, Leibniz’s frequent example
of a limitation. A shape is a limitation of its body. The confusion that lies then in
claiming that a shape is homogeneous with its body lies can be made clearer when one
takes into consideration that a shape is a limitation. To claim a shape is homogeneous
with its body would be to claim that the shape is limited in the same way that its body is
limited. The problem is, what limits the body is the body’s shape.
Generalized, a limitation L is homogeneous with an entity x that L limits only if L
is limited in the same way that x is limited. But x is limited by L. Assuming that L
cannot be limited by L, if L were to be limited in the same way that x is limited, then L
would have to be itself limited by a different limitation L’ that is similar to L (which
limits M). But then if L’ is similar to L, then L’ in turn would have to be limited by L’’,
and so on ad infinitum. To avoid the regress, one must deny that a limitation could be
homogeneous with what it limits. Given that modifications are limitations, no
modification can be homogeneous with what it modifies. Thus, no modification could be
a part of what it modifies. Leibniz now has the second premise of his argument he
needed above:
(P2) If the Merely-Modal-Reality thesis is true then Substantial-Composition is
false.
In which case, Leibniz can conclude:
(C)
If the Merely-Modal-Reality thesis is true then it is not the case that (A) the
accidental reality is part of the reality of the substance and it is not the
case that (B) the accidental reality is not part of the reality of the
substance.
Thus, affirming the Merely-Modal-Reality thesis does avoid the problems Leibniz raised
with (A) and (B). Leibniz in fact did have good reasons to avoid the non-committal
137
answers he originally gave to Q3 and Q4. Further, Leibniz can give answers to Q3 and
Q4 while also giving an affirmative answer to Q1.
2.3.3 Absolute-Realities and Homogeneity
One important issue remains to be addressed: What reasons could Leibniz have
for holding that an accident must be a part of its substance if the accident is an absolutereality? Recall that Leibniz’s arguments against accidents being absolute-realities stem
from the consequences of the disjunction (A) or (B). Both (A) and (B) assume that
accidents—if absolute-realities—are parts of their substances. What Leibniz needs, then,
is a reason to hold that if accidents are absolute-realities, then accidents are parts of their
substances. As we’ve seen, the two necessary conditions of Leibnizian part-hood are
ingredient-hood and homogeneity. So Leibniz needs a reason then to argue that if
accidents are absolute-realities, then accidents are ingredients of and homogeneous with
their substance. As I just argued, on at least one understanding of ingredient-hood,
modifications are ingredients of their substance. However, as modifications are not
homogeneous with their substance— as I also argued above, modifications are not parts
of their substance. What needs to be established, then, is that if accidents are absoluterealities, then accidents are homogeneous with their substances.
Recall that if an accident that is an absolute-reality is to be homogeneous with its
substance, then the accident must be similar to its substance or able to be transformed so
that it would be similar to its substance. As I noted above, there are two different
conceptions of similarity—the strictly geometrical conception and a broader conception.
On the strictly geometrical conception, x is similar to y only if x has the same
138
shape/figure as y. On the broader conception of similarity, x is similar to y if x has the
same type of quality as y.
I’ll start with the geometrical conception. On the geometrical conception of
homogeneity, the accident A—if it is to be homogeneous with its substance s—would
have to be similarly shaped or capable of being similarly shaped to s. Hence, A would
have to be a smaller body in s. So Leibniz now needs reasons to argue that if accidents
are absolute-realities, then both the accidents and their substances are bodies. I can’t find
one. Leibniz can argue that if A is a smaller body within s then A is a part of s. Leibniz
can also argue that if A is smaller body in s, then A is an absolute-reality. But I find no
reason why Leibniz could argue that if A is an absolute-reality then A is a body and s is a
body.
Fortunately, there’s a different route Leibniz can take with a broader
understanding of similarity. On the broader understanding of similarity, x is similar to y
if and only if either (i) x has the same quality (type) as y or (ii) x is transformable so that
it has the same quality (type) as y. Take the following two passages conveying the same
claim. The first passage was written the same year as DRA:
“It seems that something inheres in a subject, if and only if, its reality belongs to
the reality of the subject. That is to say, . . . A is in B, if all that is immediately
required by A, is also immediately required by B.”274
Leibniz claims here that if x inheres in y, then x’s reality belongs to y’s reality. The
second passage was written several decades later:
. . . a modification is a varying limitation, and modes merely limit things but do
not increase them and hence cannot contain any absolute perfection which is not
274
A VI.iv.990.
139
in the thing itself which they modify. Otherwise, in fact, these accidents must be
thought of in the manner of substances, namely, something which stands per se.275
The second passage utilizes Leibniz’s distinction between absolute and limited realities
while the first does not. However, the first passage but can be taken to apply to absoluterealities. That is, if x inheres in y, then x’s absolute reality belongs to y’s absolute reality.
I note also that Leibniz makes a point at the end of the second passage strikingly similar
to a point he makes in DRA: If an accident contains absolute perfection not found in the
substance the accident inheres in, then the accident is a substance—something which
stands per se, and therefore not an accident.
On the broader understanding of similarity, for an accident, which is an absolutereality to be homogeneous with its substance, it would have to be something that can
have qualities rather than what simply is a quality. Here, the case is easy to make then.
An entity that is an absolute-reality is something that has absolute perfection or positive
reality of its own. This is in contrast with a modification or limitation, which has no
absolute perfection of positive reality of its own but instead is merely a limitation on an
absolute reality. So absolute-realities can have limitations. Further, there are reasons to
believe that Leibniz would require any created absolute-reality to have some limitations,
otherwise the absolute-reality would be unlimited— something privy only to God. On
this broader understanding of similarity then, an accident that is an absolute-reality could
be homogeneous with its substance by having similar limitations or capable of being
transformed so that it has similar limitations. Conjoined with the fact that the accidents
are also ingredients (on one understanding of ingredient-hood), the accidents being
275
(G II.257: L 532).
140
absolute-realities entails they are parts after all. Hence, accidents being absolute-realities
entails either (A) the accidental reality is part of the reality of the substance or (B) the
accidental reality is not part of the reality of the substance, which is what Leibniz needed.
Conclusion
In this chapter, I argued that in his “De Realitate Accidentium,” Leibniz hesitated to
argue that accidents have even a modal reality because he worried that if accidents had
any kind of reality, such accidents would be parts of their substances. Thus, positing of
any kind of reality to accidents then ran afoul of Leibniz’s views on the simplicity of
substances. However, the not later Leibniz argued that accidents have a modal reality. I
argued that the later Leibniz could affirm that accidents have a modal reality given his trifold distinction between absolute-realities, limited-realities, and privations, where
modifications are limited realities and his developed mereological views. According to
Leibniz, some x is a part of some whole y only if x is homogenous with y. After a
reconstruction of Leibniz’s technical notion of homogeneity, I argued that given that
modifications are limitations, modifications cannot be homogenous with their substances.
If accidents were absolute-realities such as real qualities, however, such accidents could
be homogenous with their substances and therefore could be parts of their substances.
Therefore, a major reason Leibniz argued all accidents are modifications and no accidents
have a more than modal reality were his views on the simplicity of substances.
141
CHAPTER 5. DETERMINISTIC AND INTELLIGIBLE LEIBNIZIAN
SUBSTANCE CAUSATION
In Chapter 3, I argued that Leibniz’s views on the nature of accidents—specifically that
accidents are modifications or limitations—led him to argue against the possibility of
creaturely inter-substantial causation or creaturely causal interaction. In Chapter 4, I
argued that Leibniz held that accidents are modifications or limitations because of his
views that created substances are mereologically simple while possessing a plurality of
modifications, which they change over time. In this chapter, I more closely examine the
role accidents play in intra-substantial or immanent causation, by addressing an
interpretative controversy concerning the causal relata in such causation. Specifically,
the controversy is over what, strictly speaking, causes a substance’s accidents— the
substance itself or its accidents.
A hallmark of Leibniz’s metaphysics is his thesis of creaturely spontaneity— the
thesis that created substances are causally responsible for their accidents.276 Given
276
Leibniz writes, “For why should God be unable to give substance, from the beginning, a nature or an
internal force that can produce in it, in an orderly way (as would happen in a spiritual or formal automaton,
but free in the case where it has a share of reason), everything that will happen to it, that is, all the
appearances or expressions it will have, without the help of any created being?” (G IV.483-4: AG 143-44).
See also (G VI.295-6: T 300).
142
creaturely spontaneity, Leibniz denies that created substances ever causally interact, as
we’ve seen.277 Any apparent causal interaction is then reducible—in some sense—to the
immanent causal activity of created substances. Furthermore, Leibniz argues that
creaturely spontaneity entails the falsity of Occasionalism— the theory that only God is
causally responsible for change.278
While this much is clear, the details are murkier the closer one looks at Leibniz’s
thesis of spontaneity, as Leibniz makes several conflicting claims about what, precisely,
causes a substance’s accidents: Either the substance itself causes its accidents or the
substance’s earlier accidents or states cause its later accidents. Leibniz’s differing
answers have understandably divided scholars. Several notable scholars endorse the view
that strictly speaking, it is the substance itself that produces its accidents, a view I’ll call
the “Efficacious-substance” interpretation. Most scholars, however, take Leibniz’s
claims that earlier accidents or states are the causes of later accidents or states to be his
genuine account of spontaneity.
277
I address in depth Leibniz’s reasons for denying creaturely causal interaction in chapters 2 and 3 and
why the thesis of spontaneity rules out causal interaction in Appendix A.
278
In “On Nature Itself,” Leibniz writes, “For who would call into doubt that the mind thinks and wills, that
we elicit in ourselves many thoughts and volitions, and that there is spontaneity that belongs to us? If this
were called into doubt, then not only would human liberty be denied and the cause of evil things be thrust
into God, but it would also fly in the face of the testimony of our innermost experience and consciousness,
testimony by which we ourselves sense that the things my opponents have transferred to God, without even
a pretense of reason, are ours. But if we were to attribute an inherent force to our mind, a force for
producing immanent actions, or to put it another way, a force for acting immanently, then nothing forbids,
in fact, it is reasonable to suppose that the same force would be found in other souls or forms, or, if you
prefer, in the natures of substances—unless someone were to think that, in the natural world accessible to
us, our minds alone are active, or that all power for acting immanently, and further, as I put it, all power for
acting vitally is joined to an intellect, assertions that are neither confirmed by any rational arguments, nor
can they be defended except by distorting the truth.” See (G IV.510: AG 161).
143
Moreover, recent scholars argue that a specific type of accident—appetitions—is
the genuine productive cause of a substance’s later accidents. I call this the “Efficaciousappetition” interpretation. Appetitions in Leibniz’s metaphysics are principles of change,
specifically tendencies or strivings for future perceptions. So appetitions just seem to be
the right kind of entity to produce later accidents. Unlike the efficacious-substance
interpretation, the efficacious-appetition interpretation is also compatible with Leibniz’s
determinism and meets his strictures on explaining change— specifically his principles of
intelligibility and sufficient reason.
In this chapter, however, I argue that the efficacious-appetition succumbs to a
serious objection originally raised by Locke and endorsed by Leibniz in his New Essays
on Human Understanding— what I’ll call the “Multiplication of Agents” objection. If
appetitions are the efficient causes of a substance’s later accidents, then there a plurality
of distinct efficient causal agents in created substances, a consequence that Leibniz
rejects because it runs afoul of his views on substantial simplicity and unity. The
efficacious-substance interpretation overcomes this objection as it only posits one
efficient causal agent— the substance.
I further argue that the efficacious-substance account can be reconciled with
Leibniz’s determinism and strictures on explanation. Leibniz utilized a distinction found
as early as Aquinas and developed at length by Suarez between two different kinds of
efficient causes— the principle quod efficient cause/efficient causal agent and the
principle quo efficient cause/power by which the agent acts. Scholastics such as Suarez
held that in most cases, substances are efficient causal agents yet they also had principled
accounts of how substances could be efficient causal agents and yet act deterministically
144
or of necessity in some sense. I argue that Leibniz had similar reasons to consistently
affirm that substances are efficient causal agents—and so avoid the Multiplication of
Agents objection—but also affirm that such substances deterministically produce their
effects in a way that satisfies Leibniz’s strictures on explanation. I do so by arguing that
while substances are the principle quod efficient causes of their accidents, appetitions are
the principle quo efficient causes or the powers by which substances produce their
accidents. Appetitions can then explain the changes substances deterministically undergo
without requiring Leibniz to posit a multiplicity of distinct efficient causal agents in each
created substance.
§1 Leibniz’s Prima Facie Inconsistent Views on the Efficient Cause of Accidents
1.1 Creaturely Spontaneity
Leibniz summarizes his thesis of spontaneity in his 1695 “A New System of Nature”:
For why should God be unable to give substance, from the beginning, a nature or
internal force that can produce in it, in an orderly way (as would happen in a
spiritual or formal automaton, but free in the case where it has a share of reason),
everything that will happen to it, that is, all the appearances or expressions it will
have, without the help of any created being?279
Similarly, in the Theodicy, Leibniz writes:
But to say that the soul does not produce its thoughts, its sensations, its feelings of
pain and of pleasure, that is something for which I see no reason. In my system
every simple substance (that is, every true substance) must be the true immediate
cause of all its internal actions and passions; and, speaking with metaphysical
rigor, it has none other than those which it produces.280
279
280
(G IV.485: AG 144).
(G VI.353-4: T 400).
145
Given the thesis of creaturely spontaneity, any accident of a created substance is
produced by the substance as opposed to a distinct created substance or God alone
(except, of course, in the case of miracles). This much is clear, as we’ve seen in previous
chapters and in the passages above. Unfortunately, when one zooms in further and aims
to understand in greater depth how accidents are produced by their substances, Leibniz
gives several seemingly conflicting answers: Sometimes he writes as if the substance
itself is, strictly speaking, the cause of its accidents while in other passages he writes as if
the substance’s earlier accidents or states produce its later accidents.
1.2 The Efficacious-Substance Interpretation
Let’s look at the former view first, in which substances themselves are what, strictly
speaking, cause their accidents. Support for this view—what I’ll call the “EfficaciousSubstance” account—is found in many passages throughout Leibniz’s career. For
example, in his New Essays on Human Understanding, Leibniz writes:
As I have already said, anything which occurs in what is strictly a substance must
be a case of ‘action’ in the metaphysically rigorous sense of something which
occurs in the substance spontaneously, arising out of its own depths; for no
created substance can have an influence upon any other, so that everything comes
to a substance from itself (though ultimately from God).281
In his Theodicy, Leibniz writes:
Bayle asserts, for instance, that by purely philosophical meditations one can never
attain to an established certainty that we are the efficient cause [la cause
efficiente] of our volitions. But this is a point which I do not concede to him: for
the establishment of this system demonstrates beyond a doubt that in the course of
nature each substance is the sole cause of all its actions, and that it is free of all
physical influence from every other substance, save the customary cooperation of
God.282
281
(G V.195: NE 210).
(G VI.295-6: T 300). See also (G V.58: NE 65), PM 100, (G IV.483-4: AG 144), and (G IV.504-5: AG
156).
282
146
And in On Nature Itself, Leibniz writes:
To the extent that I have made the notion of action clear to myself, I believe that
the widely received doctrine of philosophy, that actions pertain to supposita,
follows from the notion and is grounded in it. Furthermore, I believe that we
must grasp the fact that this also holds reciprocally, so that not only is it the case
that everything that acts is an individual substance, but also that every individual
substance acts without interruption, including even body itself, in which one
never finds absolute rest.283
Passages like the ones above have lead many scholars, such as Bobro, Clatterbaugh, and
Jorati to defend the efficacious-substance account.284
1.3 The Efficacious-Accident Interpretation
Yet in quite a few passages throughout his career, Leibniz writes as if the earlier
accidents or states of a substance produce its later accidents or states. Call this the
“Efficacious-Accident” interpretation. Notable scholars such as Robert Sleigh have
endorsed this view and for understandable reasons, as in numerous passages, Leibniz
does just seem to claim that earlier accidents or states of substances cause their later
accidents or states.285 In a 1698 letter to Arnauld, Leibniz writes, “Every present state of
a substance occurs to it spontaneously and is only a consequence of [une suite de] its
preceding state.286 In a later letter written to Arnauld, Leibniz restates the same thesis,
283
(G IV.509-10: AG 160).
See Marc Bobro and Kenneth Clatterbaugh, “Unpacking the Monad, Leibniz's Theory of Causality,”
The Monist, (1996) 79: 409–26 and Julia Jorati, “Leibniz on Causation –Part 1,” Philosophy Compass
(2015) 10: 389-397.
285
Sleigh writes, “Every non-initial, non-miraculous state of every created substance has as a real cause
some preceding state of that very substance.” See Robert C. Sleigh, Jr., “Leibniz on Malebranche on
Causality,” p. 162. Additionally, Kulstad endorsed this view in earlier works but as will become apparent,
he has since endorsed a more nuanced interpretation of monadic-causation. Kulstad writes, “I mean the
view that created substances can be real causes, or, more specifically, that each state of a created substance
arises causally from its preceding state.” See Mark Kulstad, “Causation and Pre-established Harmony in
the Early Development of Leibniz’s Philosophy,” in Causation and Early Modern Philosophy, ed. Steven
Nadler (University Park, PA: University of Pennsylvania Press, 1993), p. 96.
286
G II.47.
284
147
writing, “Everything occurs in each substance in consequence of [en consequence du] the
first state that God gave it in creating it.287 In his 1695 “Clarification of the difficulties
that Mr. Bayle found in the New System of the Union of the Soul and the Body,” Leibniz
continues to affirm the same thesis, writing that “the present state of each substance is a
natural result of its preceding state.288
However, Leibniz divided accidents into two kinds: perceptions and appetitions.
Therefore, it’s worth further examining which of the two kinds of Leibnizian accidents
are efficient causes on the efficacious-accident interpretation: the substance’s
perceptions, appetitions, or both? In many passages, Leibniz writes as if perceptions are
efficacious:289
In fact, nothing can happen to us except thoughts and perceptions, and all our
future thoughts and perceptions are merely consequences, though contingent, of
our preceding thoughts and perceptions, in such a way that, if I were capable of
considering distinctly everything that happens or appears to me at this time, I
could see in it everything that will ever happen or appear to me.290
every present perception leads to [que la suite de] a new perception.291
subsequent [perceptions] are derived [derivantur] from preceding ones.292
In an article arguing against the efficacious-perception interpretation, Bobro and
Clatterbaugh have rightly pointed out that in the above passages, Leibniz does not use the
language of efficient causation when describing how future states come from a
287
G II.91.
G IV.521.
289
Nicolas Jolley endorses the view that perceptions are efficacious. He writes, “Although Leibniz may
say that it is substances which produce their states, this is only a loose way of speaking; in strictness, it is
perceptual states which causally produce other perceptual states of the same substance.” See Nicholas
Jolley, “Causality and Creation in Leibniz,” The Monist (1998) 81, no. 4, 605.
290
Leibniz writes this in chapter 14 of his Discourse on Metaphysics. See (G IV.439-40: AG 47).
291
(G VI.356-7: T 403).
292
A 1709 letter to Des Bosses. See G II.372
288
148
substance’s earlier states.293 Instead, Leibniz uses logical terms, such as “consequences”
and “derived”. However, in other passages, Leibniz does use causal language:
But this expression which the soul has of the future in advance, although obscure
and confused, is the true cause (cause veritable) of what will happen to it and of
the clearer perception it will have afterwards, when the obscurity is lifted, since
the future state is a result of the preceding one.294 (Emphasis added)
Notice that Leibniz claims that a soul’s expression—which just is a perception—is the
true cause of its later perception. In other passages, Leibniz uses not only causal
language but efficient causal language:
The representation of the present state of the universe in the soul … will produce
(produira) in it the representation of the following state of the same universe, just
as the objects in the preceding state actually produce (produit) the following state
of the world. In the soul the representations of these causes are the causes of the
representations of these effects.295 (Emphasis added)
The present state of body is born from the preceding state through the laws of
efficient causes; the present state of the soul is born form its preceding state
through the laws of final causes. The one is the place of the series of motion, the
other of the series of appetites; the one is passed form cause to effect, the other
from end to means. And in fact, it may be said that the representation of the end
in the soul is the efficient cause of the representation in the same soul of the
means.296
Recently, however, scholars such as Rutherford, Carlin, and Bolton have argued that a
substance’s appetitions are the efficient causes of a substance’s later accidents.297 Their
reasons, which I soon present in greater depth, are that appetitions—being more akin to
293
See Bobro and Clatterbaugh, “Unpacking the Monad, Leibniz's Theory of Causality,” 415.
(G II.91: AG 82).
295
G IV.532-3.
296
Quoted in Carlin, 226
297
See Martha Brand Bolton, “Change in the Monad,” in Eric Watkins, ed. The Divine Order, the Human
Order, and the Order of Nature: Historical Perspectives (Oxford: Oxford University Press, 2013), 178;
Laurence Carlin, “Leibniz on Final Causes,” Journal of the History of Philosophy (2006) 44: 231; and
Donald Rutherford, “Laws and Powers in Leibniz,” The Divine Order, the Human Order, and the Order of
Nature: Historical Perspectives. Ed. Eric Watkins. (Oxford: Oxford University Press, 2013), 167 and
“Leibniz on Spontaneity.” Leibniz: Nature and Freedom. Eds. Donald Rutherford, and Jan A. Cover,
(Oxford: Oxford University Press, 2005), 167.
294
149
causal powers—are better suited to cause accidents than perceptions. This interpretation,
which I’ll call the “Efficacious-appetition” interpretation, finds support in the following
passages from Leibniz’s “Principles of Nature and Grace Based on Reason” and
Monadology:
As a result, a monad, in itself and at a moment, can be distinguished from another
only by its internal qualities and actions, which can be nothing but its perceptions
(that is, the representation of the composite, or what is external, in the simple) and
its appetitions (that is, its tendencies to go from one perception to another), which
are the principles of change.298
The action of the internal principle which brings about the change or passage
from one perception to another can be called appetition; it is true that the appetite
cannot always completely reach the whole perception toward which it tends, but it
always obtains something of it, and reaches new perceptions.299
As this brief survey should have made obvious by now, appealing to texts alone won’t
settle this debate, as there is ample textual support for all three interpretations: the
efficacious-perception, efficacious-appetition, and efficacious-substance view. Other
considerations are needed if this debate is to be resolved. While I defend a variant of the
efficacious-substance interpretation in this paper, I defend one that incorporates the
specific strengths of the efficacious-appetition view, so it’s worth seeing just what those
strengths are. Before I do so, however, I need to address two preliminary issues.
1.4 Why not both Efficacious-substance and Efficacious-Accidents?
Someone might respond that if Leibniz claims that both substances and their accidents
are efficacious, then Leibniz must have meant that both are efficacious. That is, it is true
that created substances are efficient causes and it is also true that accidents are efficient
298
299
(G VI.598: AG 207).
(G VI.609: AG 215).
150
causes. In fact, such a view had historical precedent, for many scholastics held that both
substances and accidents are efficacious. For example, on Suarez’s metaphysics, it is not
just substances that are efficacious, but res, which are efficacious.300 The category of res,
however, is not limited to substances but also includes certain kinds of accidents such as
real qualities.301
However, two reasons count against appealing to scholastic affirmations of the
efficacy of both substances and accidents in attempting to make sense of Leibniz’s own
claims. First, Scholastics were pushed to affirm the causal efficacy of accidents because
of their commitment to transubstantation, where the wine and bread at a Eucharistic mass
undergo a substantial change into the blood and body of Christ. When the wine and
bread become the blood and body of Christ, the accidents of the wine and bread continue
to exist without inhering in the substance of the wine and bread (as it no longer exists)
and also without inhering in Christ’s body and blood. Such accidents are efficacious
because they can be seen, felt, and tasted by the recipients of communion.302 Leibniz,
however, was a Lutheran who ultimately did not affirm transubstantiation, and so would
not have that as a reason for affirming the efficacy of both substances and accidents.303
Second, the accidents, at least on Suarez’s view, which are efficacious are real
qualities. A sign of an accident’s being a real quality is its separability from its
substance—such as the accident of redness in wine, which can exist apart from the wine
300
See DM 18.4.3 and 18.4.7. Stephan Schmid points this out in his recent work on Suarez on efficient
causality. See Stephan Schmid, “Efficient Causality: The Metaphysics of Production,” in Jakob Leth Fink
ed. Suarez on Aristotelian Causality (Brill, 2015), 104-105.
301
I explore at length Leibniz’s own views on whether accidents are real qualities in Chapter 4.
302
See DM 18.3.13
303
I address Leibniz’s views on transubstantiation in the appendix to chapter 3.
151
in which it originally inhered upon transubstantiation.304 Leibniz, however, is adamant
that such real qualities do not exist.305 Instead, all accidents, for Leibniz, are modes that
are inseparable from their substances.306 Leibniz argued that if real qualities did exist,
they would not be accidents at all but rather substances (and so not accidents).307
A second issue concerning the response that both substances and accidents are
efficacious concerns the precise contribution each would make in the production of a new
accident. From the get go, one can rule out both the substance and accident being
sufficient for the newly produced accident, as this would result in the future substances of
accidents being causally overdetermined. While Leibniz never explicitly rules out the
possibility of causal overdetermination, surely he would reject widespread and systematic
overdetermination, where every non-initial accident of a substance has more than one
sufficient cause— substances and their earlier accidents.308
Perhaps Leibniz meant that substances and accidents are each partial causes of the
future accidents of substances. This account avoids the problems with holding that
304
See Ibid. For Suarez’s full treatment of the distinction between separable real qualities and inseperable
modes, see Suarez, On the Various Kinds of Distictions, Cyril Vollert, SJ trans. (Milwaukee: Marquette
University Press, 2007).
305
Leibniz writes, “Let us come now to the real accidents which are in this unifying thing as their subject.
You will agree, I believe, that some of them are only modifications, which disappear when it is removed.
But you ask whether there are not certain accidents which are more than modifications. Such accidents
seem, however, to be entirely superfluous, and whatever is in such a substance other than a modification
seems to pertain to the substantial thing itself.” See (G II.458: L 606)
306
Concerning the inseparability of accidents from their substances, Leibniz writes, “An accident, however,
needs not only some substance in general but that very one in which it inheres, so that it cannot change it.
See (G IV.364: L 390).
307
Leibniz writes, “. . . we may ask whether there can be a thing which is neither a modification nor a
source of modifications—such as the Scholastics think of as accidents, which, they say, are in a subject
naturally but not essentially, since they can be without a subject by the absolute power of God. But I do not
yet see how such a thing can be explained if it is different from my substantial chain, which is truly in the
subject, though not as an accident but as what the Scholastics call a substantial form, or as a source of
modifications—if you like, after the manner of an echo.” See (G II.504: L 614).
308
In the Appendix A, I argue that Leibniz has reasons to reject even non-widespread and non-systematic
overdetermination.
152
substances and accidents are each sufficient causes, but has its own difficulties. Mainly,
if substances and accidents are partial causes, what do substances and accidents
contribute in the production of new accidents? Unfortunately, looking to Leibniz yields
very little by way of an answer. For now, I shall assume that it is not the case that both
substances and accidents are efficient causes of a substance’s later accidents in a univocal
sense of efficient cause. Later, however, I argue for a nuanced interpretation of the
efficacious-substance interpretation in which appetitions are efficacious, but not in the
same sense in which substances are efficacious.
1.5 Against the Efficacious-Perception Interpretation
While I ultimately argue for a nuanced interpretation of the efficacious-substance
interpretation of Leibniz that incorporates elements of the efficacious-appetition view,
it’s worth briefly addressing why, in spite of numerous passages where Leibniz seems to
claim that perceptions are efficacious, it is not the case that perceptions are genuinely
efficacious. As we’ve seen, Leibniz posits two types of accidents in his ontology—
perceptions and appetitions.309 These two types of accidents could, with some caveats,
be understood in contemporary parlance as categorical properties and dispositional
properties where perceptions are categorical properties and appetitions are dispositional
309
Leibniz writes, “As a result, a monad, in itself and at a moment, can be distinguished from another only
by its internal qualities and actions, which can be nothing but its perceptions (that is, the representation of
the composite, or what is external, in the simple) and its appetitions (that is, its tendencies to go from one
perception to another) which are the principles of change. For the simplicity of substance does not prevent
a multiplicity of modifications.” See (G VI.598: AG 207).
153
properties.310 If perceptions, however, are the genuine efficient cause of later accidents,
then Leibniz has categorical properties serving as genuine efficient causes.
I find two problems with such a view, however. First, as mentioned above and
what I develop in much greater depth soon, there are plausible reasons to hold that
appetitions—which are a lot like causal powers—are efficacious. If both appetitions and
perceptions are efficacious in the same sense, then we are lead back to the difficulties we
faced in supposing that both substances and accidents are efficacious in the same sense.
If both appetitions and perceptions are efficacious in the same sense, then they are either
both sufficient causes or partial causes of a substance’s later accidents. If they are both
sufficient causes, then the same difficulties with overdetermination arise—mainly that
such a picture results in systematic and widespread overdetermination. If they are both
partial causes, then a detailed account—which is lacking in Leibniz’s corpus—is needed
as to what each distinctly contribute.
Second, let’s assume for the moment that only perceptions are efficacious.
According to Leibniz, appetitions are “tendencies from one perception to another”311 and
perceptions are representational entities.312 Thus, with some caveats that I address later
in this paper, perceptions are akin to categorical properties while appetitions are more
like dispositional properties. However, if perceptions are solely efficacious, Leibniz has
a scenario where categorical properties—perceptions—do all the causal work. Earlier
310
I address the differences between appetitions and normal Aristotelian powers or dispositions later in this
chapter.
311
(G VI.598: AG 207).
312
In the 14th chapter of his Monadology, Leibniz writes, “The passing state, which involves and represents
a multitude in the unity or in the simple substance is nothing other than what one calls perception. . .” (G
VI.608: AG 214). See also (G VI.598: AG 207), G II.311, G III.622, G VII.529, and 566.
154
perceptions produce later perceptions. I find two problems with this scenario. First, the
efficacious-perception view leaves little work for appetitions. If perceptions are solely
efficacious, then it is not clear what role appetitions play in Leibniz’s metaphysics. The
appetitions would be superfluous. Second, this picture is at odds with Leibniz’s causal
views at large, as Leibniz is a full-fledged realist about causal powers.313 However, on
the efficacious-perception interpretation, categorical properties—perceptions—do all the
causal work, including producing appetitions—which are akin to dispositional properties
or powers. Such a picture--which would turn Leibniz into a Humean--is further at odds
with texts where Leibniz claims that appetitions produce perceptions. Hence, I’ll assume
for the remainder of this paper that while perceptions count amongst the causal relata as
effects, they do so as effects, not causes. Instead, I shall focus on the efficacioussubstance and efficacious-appetition view.
§2 The Argument for the Efficacious-Appetition Interpretation
This leaves the efficacious-substance interpretation and the efficacious-appetition
interpretation. While I defend the efficacious-substance interpretation, I present the
motivations for the efficacious-appetition interpretation first, as the efficacious-substance
account I defend is one that incorporates the strengths of efficacious-appetition
interpretation. Donald Rutherford gives the strongest and lengthiest argument for the
efficacious-appetition interpretation against the efficacious-substance interpretation,
which I turn to now. Rutherford’s argument consists in two moves. First, Rutherford
313
Indeed, the essence of substance consists in an active force or entelechia, which is responsible for
change. See, for example (G IV.478-9: AG 139), and (G IV.504-16: AG 155-67).
155
argues that the efficacious-substance interpretation cannot be reconciled with Leibniz’s
determinism and his strictures on the explanation of monadic change— Leibniz’s
requirement that change be intelligible. Second, Rutherford argues that the efficaciousappetition interpretation does meet Leibniz’s strictures for explaining monadic change,
and thus is the true interpretation.
2.1 Rutherford on the Efficacious-substance Interpretation, Determinism and
Intelligibility
Recall that on the efficacious-substance interpretation, substances—not their earlier
accidents—are the efficient causes of their later accidents. A substance s changes from
being in state N to state N+1 because s—the efficient causal agent—produced the
accidents which makeup N and then produced the accidents which makeup N+1.
However, Leibniz is also determinist who held that a substance s is determined to be in
state N+1 given its immediate predecessor state N. For example, in his Monadology,
Leibniz states that “every present state of a simple substance is naturally a consequence
of its preceding state.”314
Additionally, Leibniz has strict requirements for the intelligibility of monadic
change.315 There has to be an intelligible reason why a monad changes from N at t to
N+1 at t+1 where by “intelligible”, Leibniz means that the explanation for the
314
(G VI.610: AG 216).
Leibniz writes, “Whenever we find some quality in a subject, we ought to believe that if we understood
the nature of both the subject and the quality we would conceive how the quality could arise from it. So
within the order of nature (miracles apart) it is not at God’s arbitrary discretion to attach this or that quality
haphazardly to substances. He will never give them any that are not natural to them, that is, that cannot
arise from their nature as explicable modifications.” A.VI.vi.66.
315
156
substance’s change from N to N+1 is found within the substance’s own nature and is
understandable by finite substances.316
While Rutherford doesn’t bring it up, a further stricture on explanation can be
drawn from Leibniz’s Principle of Sufficient Reason (PSR). While Leibniz gives several
formulations of PSR throughout his career, I have in mind the version in which any
explanation must involve contrastive reasons. For example, Leibniz writes:
And that of sufficient reason, by virtue of which we consider that we can find no
true or existent fact, no true assertion, without there being a sufficient reason why
it is thus and not otherwise, although most of the time these reasons cannot be
known to us.”317
According to Leibniz, for any state N of a substance s, there is a sufficient reason why s is
in state N as opposed to a different state N’. Further, if s changes from N to N+1, there is
a sufficient reason why N+1 rather than N’+1 is a consequence of N. In what follows, I
will focus primarily on Leibniz’s determinism and principle of intelligibility, since that is
what Rutherford focuses on in his argument.318 Later, however, I will argue that my
nuanced interpretation of the efficacious-substance account can be reconciled with PSR.
According to Rutherford, the efficacious-substance interpretation cannot be
reconciled with Leibniz’s determinism nor his requirements for intelligibility. Claiming
316
The principle of intelligibility is wielded by Leibniz against a number of targets. For example, Leibniz
argues that the principle of intelligibility rules out occult qualities such as mental properties in material
substances and Newtonian gravitation, as such qualities are not explainable by their substance’s nature.
Leibniz also utilizes it against occasional accounts of creaturely change. If God is the sole efficient cause
of any accident that comes to inhere in a created substance, then the explanation for the accident is not
found in the creature but rather in God’s will. Thus, if Occasionalism is true, then no change is
intelligible—a consequence Leibniz insists is false, in which case Occasionalism is also false.
317
(G VI.612: AG 217). See also (G VI.602: AG 209-10) and (G VI.127: T 44).
318
In an earlier paper, Rutherford distinguishes between the principle of intelligibility and PSR. However,
the version of PSR that Rutherford distinguishes from the principle of intelligibility is not the contrastive
version I address above. Instead, it’s the axiom that “nothing happens without a reason.” See Donald P.
Rutherford, “Leibniz’s Principle of Intelligibility,” History of Philosophy Quarterly 9, no. 1 (1992): 35.
157
that a substance s is in state N at t because s produced all the accidents that makeup N
does not explain why s produced the accidents that make up N. Claiming that s changes
from N to N+1 because s efficiently caused all the accidents that makeup N and later
produced all the accidents that makeup N+1 also does not explain why s first produced
the accidents that make up N and then produced the accidents that make up N+1.
Rutherford draws attention to an unpublished passage in support of his argument,
where Leibniz writes, “Saying that the soul’s God-given force is the only source
[principe] of its particular actions is not sufficient to give the explanation for those
actions.”319 The force referred to in this passage is the substance’s primary active force,
which just is the substance’s nature or substance itself.320 Thus, the efficacious-substance
interpretation, it seems, runs afoul of Leibniz’s principle of intelligiblity. As Rutherford
writes:
The concern is that there is nothing in the concept of an ‘active power,’ or even
one that is more entelechia than dunamis, that would allow us to understand why
that power should give rise to one succession of states rather than another.
Arguably, we have here the same sort of violation of the principle of intelligibility
that Leibniz elsewhere inveighs against it.321
2.2 Why the Law-of-the-Series does not help the Efficacious-Substance Interpretation
Defenders of the efficacious-substance interpretation might appeal to the substance’s lawof-the-series as providing an explanation of monadic change that meets Leibniz’s own
strictures on explanation and is consistent with Leibniz’s determinism. The substance’s
law of the series is tantamount to the substance’s essence or substantial form.322 As
319
G IV.542. See also WF 100. Quoted in Rutherford, “Laws and Powers in Leibniz,” 162.
See Rutherford, “Leibniz on Spontaneity,” 163.
321
Rutherford, “Laws and Powers in Leibniz,” 162.
322
As Rutherford notes, the law of the series is identified with the individual nature of the substance. See
320
158
Rutherford writes, “Here, what is important about the law of the series is that it involves,
in some unspecified sense, a complete history of a substance’s states.”323 Given that the
law-of-the-series involves somehow a complete history of a substance’s states, we might
then have an explanation for why a substance changes from state N to state N+1— it is
because N+1 follows N in the substance’s law of the series. This also provides an
explanation for s changes from N to N+1 rather than N’+1—it is because N+1 follows N
in the substance s’s law of the series. Further, given that the law-of-the-series is
identified with the nature of a substance, which involves active power, one might be
further tempted to give the law-of-the-series a causal role. A substance s just is its law of
the series, and s changes from N to N+1 because s produces all the accidents that makeup
N+1. The reason s changes from N to N+1 instead of from N to N’+1 is because N+1
follows N in s’s law of the series. Thus, the law-of-the-series plays both the efficient
causal role and the role of explainer or determiner with respect to monadic change.
Rutherford writes:
There is no doubt that Leibniz invests the law of the series with a causal aspect.
Insofar as this law is identified with the individual nature of a substance, and that
nature involves an active power that is the spontaneous source of all of a
substance’s states, the law of the series can be seen as ‘determining’ the
succession of those states.324
However, according to Rutherford, this just pushes the problem back a step. For we still
do not have an explanation for why N+1 must follow from N in a substance’s law of the
series instead of N’+1 following from N. Rutherford writes:
Ibid., 164. Rutherford is not recognizing that Leibniz identifies the law of the series with the individual
nature of the substance or the substance itself. See Cover and O’Leary-Hawthorne, Substance and
Individuation in Leibniz, 219-226.
323
Rutherford, “Laws and Powers in Leibniz,” 164.
324
Ibid., 163-64.
159
The law of the series is said to contain this information, but it provides (so far as
we can understand it) no explanation of that order. It encapsulates a complete
history of all that a substance will do, but it does not render intelligible, as an
instance of natural change, the transition from one state of the substance to
another.325
Rutherford continues:
Knowing a substance’s law of the series (which only God can know), one would
know all the states of the substance, in the order in which they occur. What one
wouldn’t know, however, is why if a given substance is in state Sn, it will
thereafter, as a matter of natural necessity, be in state Sn+1. Such an explanation
requires a generality that is missing in the law of the series, each example of
which pertains uniquely to a single substance. It must explain why if any
substance is in a state Sn, characterized in suitable theoretical terms, it will
thereafter, given the laws of nature, be determined to be in state Sn+1.326
2.3 Rutherford on the Efficacious-Appetitions, Determinism, and Intelligibility
Rutherford’s second move is to argue that in explaining monadic change, the efficaciousappetition account succeeds where the efficacious-substance account fails. In support of
both appetitions being explainers of monadic change and their efficacy, Rutherford first
draws attention to some key passages such as the following in Leibniz’s “Principles of
Nature and Grace” which we’ve seen already but is worth repeating:
A monad, in itself and at a moment, can be distinguished from another only by its
internal qualities and actions, which can be nothing but its perceptions (that is, the
representation of the composite, or what is external, in the simple) and its
appetitions (that is, its tendencies to go from one perception to another) which are
the principles of change.327
That appetitions are tendencies and principles of change is strong evidence in favor of the
efficacious-appetition interpretation, according to Rutherford. Additionally support
comes from passages where Leibniz claims that appetitions are forces—specifically
325
Ibid., 165.
Ibid.
327
(G VI.598: AG 207).
326
160
derivative forces which modify the monad’s primitive force.328 More importantly,
however, especially considering that there are also many passages in support of the
efficacious-substance account, the efficacious-appetition account avoids the issues that
plagued the efficacious-substance account. Specifically, Rutherford argues the
efficacious-appetition account makes sense of how monadic change can be deterministic
and also meets Leibniz’s strictures on explanation.
Rutherford first draws attention to a definition of change that Leibniz offered
throughout his career, where change is defined as “nothing but a complex of two states
which are immediate and opposite to each other, together with a force or reason for the
change, which reason itself is a quality.”329 It’s worth unpacking this definition. A
monad’s change involves two states— N and N+1. N and N+1 are ordered as prior and
posterior and the reason for N’s being prior to N+1 and N+1 following N is due to a
quality of N which is a force or reason for N+1. That force, Rutherford points out, just is
the appetitions in N.330 Such appetitions are forces or reasons for change— specifically
appetitions for future states, namely, perceptions. Thus, the ordering is not arbitrary.
N+1 does not follow N merely because N+1 follows N in the monad’s law of the series.
Instead, N+1 follows from N because N has appetitions for the accidents that make up
N+1. This explanation also explains why N+1 follows N in the monad’s law of the series
rather than N’+1— it is because the appetitions which makeup N are appetitions for the
accidents of N+1 rather than N’+1.
328
Rutherford, “Laws and Powers in Leibniz,” 166. See also (G II.562: L 533), (G II.270: L 537), and (G
II.200-1: NE 216).
329
Ibid., 167, quoting C 9/MP 134.
330
Ibid., 167.
161
So, according to Rutherford, the efficacious-appetition account meets Leibniz’s
intelligibility requirements. When a substance s changes from N to N+1, the explanation
for the change from N to N+1 comes from s’s own nature. Specifically, s’s nature as
modified.331 The substance in state N has the appetitions—which are modifications or
accidents of itself—for the accidents of N+1 and those accidents produce N+1.
Therefore, the reason for the substance’s change from N to N+1 is the substance’s nature
rather than some other created substance or merely God’s will. That the substance’s state
N has appetitions for N+1 and so produce the accidents of N+1 rather than N’+1 also
makes sense of how such change is deterministic and in line with Leibniz’s principle of
sufficient reason.
According to Rutherford, s changes from N to N+1 because the appetitions in N
are appetitions for the accidents of N+1 and so efficiently cause the accidents which
makeup N+1.332 Therefore, it is the appetititions, rather than the substance, which are
genuinely efficacious. Rutherford has provided a powerful argument for the efficaciousappetition interpretation that incorporates the strengths of the efficacious-substance
account while also presenting a strong case against the efficacious-substance
interpretation.
331
Rutherford writes, “Changes in the states of a monad are explained in terms of its own nature—
however, crucially, it is that nature as modified. While all the states of a substance depend ontologically on
the primitive active force produced by God, since they exist only as modifications of primitive force,
changes in the substance are explained by appeal to prior states that determine the existence of new states.”
See Ibid., 166-67.
332
Rutherford writes, “Monadic states themselves (or the appetitive forces associated with those states) are
causally efficacious in the production of new states, which in turn are productive of new states, and so on.”
Ibid.
162
§3 Leibnizian Substance Causation
However, I argue that the efficacious-substance interpretation can be formulated in a way
that incorporates the strengths of the efficacious-appetition interpretation. In fact, such
an understanding will prove to be necessary for Leibniz himself, I soon show, raises a
fatal objection to the efficacious-appetition account— at least as it has been developed by
recent defenders such as Rutherford, et al. I shall argue that substances are the efficient
causal agents but the substances appetitions still explain why a substance changes from N
to N+1. I do so by drawing upon a widely held scholastic distinction that is found as
early as Aquinas and developed at length by Suarez between the principle quod efficient
cause and a principle quo efficient cause. The principle quod cause is the efficient causal
agent and the principle quo cause is the efficient causal power by which the efficient
causal agent acts. I shall argue that substances, for Leibniz, are the principal quod cause
and their appetitions are the principle quo causes. This will make sense of texts where
Leibniz says substances—specifically their primary active forces—are principles of
change and other texts where Leibniz says that their appetitions are principles of change.
On this formulation of the Efficacious-substance account, if one asks “Is it the
substance or the appetition that efficiently causes a substance’s later accidents?” the
answer is “Yes”. If one asks “Is it the substance or the appetition which is a principle of
change?” the answer is also yes. One would have to further specify whether they mean
the principle quod or principle quo cause to get an answer of only the substance or only
its accidents. A lot of the debate in the secondary literature has hinged on the assumption
that any time Leibniz says that substances, perceptions, states, or appetitions cause later
163
states, Leibniz always means that they are efficient causes in the same sense— the
principal quod or efficient causal agent.
3.1 Suarez on the Efficient Principle Cause Quod and Quo
Scholastics often distinguished between the agent in efficient causation and the power by
which the agent acts when efficiently causing some effect. This distinction traces back at
least as far as Aquinas, who in the Summa Theologica writes, “In every action two things
are to be considered, the ‘suppositum’ acting, and the power whereby it acts; as, for
instance, fire heats through heat.”333 Suarez utilized this distinction throughout his
Disputation Metaphysica, especially disputations 17-22 on efficient causation.
According to Suarez, both the agent which efficiently causes some effect and the power
by which the agent causes the effect are efficient causal principles. The agent is the
principle quod or suppositum of the act and the power by which the agent acts is the
principle quo. Suarez writes that, “. . . a suppositum is said to act immediately as a
principle quod when it acts through a power inherent in itself in such a way that the
action proceeds immediately from that power as a principle quo.”334
As notable Suarez scholars have pointed out, both the agent and the power by
which the agent acts are efficient causal principles for Suarez. Freddoso writes, “Suarez
distinguishes an efficient principle ut quod, that is, the substance which exercises a power
333
ST 1a q36 a1. This distinction appears throughout the Summa. For example, Aquinas writes, “Now
actions belong to supposits and wholes and, properly speaking, not to parts and forms or powers, for we do
not say properly that the hand strikes, but a man with his hand, nor that heat makes a thing hot, but fire by
heat, although such expressions may be employed metaphorically.” For example, see ST 2a2ae, q58, a2c.
334
DM 22.1.19. Suarez also writes, “. . . there is one sort of principal cause which operates and another
sort which is a principal principal of operating—they are commonly called, respectively, a principal cause
quod and a principal cause quo.” See DM 17.2.7. Suarez also writes, “For the principle quod is the
suppositum, just as in other actions.” See DM 18.2.1.
164
and to which the resulting acting is ultimately attributed, from an efficient principle ut
quo, that is, the power or faculty by which such a substance operates.”335 In a recent
article on Suarez’s account of efficient causation, Schmid explains the implications of
this distinction in even greater depth:
But what sorts of things are causes if they are not events? In treating this
question, we have to be careful since asking ‘what are efficient causes?’ is
ambiguous. Taken in one way, Suarez explains, this question addresses the
principle-quod (or ‘the principle cause which operates’), that is, the thing or
suppositum from which a certain action arises. Taken in another way, the
question refers to the principle-quo (or the principal principle of the operation’),
that is, the principle by virtue of which a certain agent performs its action.336
This distinction can help clear much confusion that may arise in a first time reader of
Suarez’s disputations on efficient causation, where in quite a few sections Suarez
develops accounts of how various non-substances—such as the substantial form or a
substance’s powers—are efficient causal principles. As Freddoso writes, “I mention this
[the distinction] in part because several of the questions concerning efficient causality
that Suarez deals with in Disputations 17-19 center around the principle ut quo, and it is
important to understand from the beginning that Suarez takes the principle ut quo to fall
under his general characterization of an efficient causal principle.”337
335
Freddoso clarifies this in his lengthy exposition of Suarez’s account of efficient causation in the
introduction to his translation of Suarez’s DM 20-22. See Suarez, On Creation, Conservation, and
Concurrence: Metaphysical Disputations 20-22, xxix-xxx.
336
And in agreement with what I’ve written so far, Schmid continues, “The principle-quo of an action is
nothing but the power by means of which a suppositum can perform a particular action.” See Schmid,
“Efficient Causality: The Metaphysics of Production,” 103-4.
337
See Suarez, On Creation, Conservation, and Concurrence: Metaphysical Disputations 20-22, xxix-xxx.
165
3.2 Why the distinction applies to Leibniz
I argue that the distinction applies to Leibniz. That is, substances are the agent or
principle quod efficient cause of accidents for Leibniz and powers are the principle quo
efficient cause of accidents. Specifically, appetitions are the principle quo efficient
causes or powers by which the substance acts.
3.2.1 Appetitions are powers
As scholars who defend the efficacious appetition account, such as Kulstad, Carlin,
Bolton, and Rutherford, have argued, appetitions just are powers.338 Leibniz claims
several times in his mature writings that appetitions are tendencies. For example, Leibniz
writes to Samuel Masson in 1716:
But the tendency of which I speak is of another nature; it is internal to the soul,
which is not a point. It is the progress of one thought to another, and since
thoughts (though in a soul not composed of parts) represent things composed of
parts, it is only in this sense that these perceptions are called composite, as are
their tendencies or appetites—that is, they contain a multitude of modifications
and relations all at once.”339
Earlier, in his “Principles of Nature and Grace”, Leibniz claims as well that appetitions
are tendencies from one perception to another.340
Leibniz identifies also appetitions as derivative forces, which entails that
appetitions are powers. In the New Essays, Leibniz writes, “There are other efforts,
arising from insensible perceptions, which we are not aware of; I prefer to call these
‘appetitions, rather than volitions, for one describes as ‘voluntary’ only actions one can
be aware of and can reflect upon when they arise from some consideration of good and
338
See Bolton, 119, Carlin, 228, Kulstad, 133-34, Rutherford, “Leibniz on Spontaneity in Leibniz: Nature
and Freedom,” 163.
339
(G VI.627: AG 228).
340
(G VI.598: L 636).
166
bad; though there are also appetitions of which one can be aware.”341 So appetitions are
efforts. Not much earlier in the New Essays, Leibniz identifies efforts with derivative
force, writing, “Force would divide into ‘entelechy’ and ‘effort’; for although Aristotle
takes ‘entelechy’ to generally that it comprises all action and effort, it seems to me more
suitable to apply it to primary acting forces, and ‘effort’ to derivative ones.”342 In
Leibniz’s 1695 “A Specimen of Dynamics,” however, Leibniz claims that derivative
forces are powers. Specifically, Leibniz argues that there are two types of active force,
primitive and derivative active forces. Both kinds of active forces, however, “might not
inappropriately be called power”, as Leibniz claims.343
3.2.2 Substances and Appetitions are Principles of Change
Leibniz also claims that both substances and their appetitions are principles of change.
The second thesis is explicitly stated in a famous passage in his “Principles of Nature and
Grace” that we’ve already seen but is worth repeating:
It follows that one monad by itself and at a single moment cannot be distinguished
from another except by its internal qualities and actions, and these can only be its
perceptions—that is to say, the representations of the compound, or of that which
is without, in the simple – and its appetitions—that is to say, its tendencies from
one perception to another – which are the principles of change.344 (Emphasis
added)
The first thesis requires a bit more work to develop but is still well supported. In the
Theodicy, Leibniz identifies the soul with Entelechy and he identifies the Entelechy as the
active principle, writing, “Moreover, it is true that the soul is the Entelechy or the active
principle, whereas the corporeal alone or the mere material contains only the passive.
341
(G V.158-9: NE 173).
(G V.155-6: NE 169-70).
343
AG 219.
344
(G VI.598: L 636).
342
167
Consequently the principle of action is in the soul, as I have explained more than once in
the Leipzig Journal.”345 Leibniz also identifies the entelechy with the monad.346 So the
monad or created substance just is the entelechy or active principle. Which just is to say
that the created substance is a principle of change.
It’s worth noting that Leibniz uses the term “principle” roughly in the same way
that Suarez and Aquinas use it, a usage which traces back to Aristotle’s Metaphysics,
where a principle plays a metaphysical role as a beginning or origin rather than the
contemporary usage as rule or axiom. Aquinas writes that “the word ‘principle’ signifies
only that whence another proceeds: since anything whence proceeds in any way we call a
principle; and conversely.”347 While Suarez has a broad notion of principle that
designates the first element in any ordering, he has a narrower usage that is closer to
Aquinas’s in which the principle is not merely the first element in an ordering but also is
a thing from which another follows via some sort of connection.348 Given that the
principle under discussion is an active principle or principle of change, I submit that
Leibniz’s usage here closely follows Aquinas’s and Suarez’s.349
345
(G VI.89-90: T 69).
(G II.193-4: L 521-22).
347
ST 1 q33 a1
348
Suarez writes, “Therefore, ‘principle’ of the thing’ can be used either only on account of an order and
whatever connection or on account of some intrinsic disposition towards.” (DM 12.1.4) Freddoso,
commenting on Suarez’s account of principle, writes, ““The term ‘principle’, he tells us, can be used in a
wide sense to designate the first element in any sort of ordering, real or merely conceptual, and in this sense
it is obviously more inclusive than the term ‘cause’. However, ‘principle’ is used most properly in a
narrower metaphysical sense to designate ‘that which truly and directly communicates (influens) some sort
of being (esse) to that of which it is the principle,’ or, in other words, that on which a real entity depends in
some way for its existence.” (Freddoso, xxv-xxvi)
349
A principe may either be a source, foundation, or rule of acting. Petit Robert; dictionnaire de la langue
francaise, ed. A. Rey and J. Rey-Debove, 3rd Ed, (Paris: Dictionnaires Le Robert 2003).
346
168
3.2.3 Substances are Principle Quod Causes and the Appetitions are Principle Quo
Causes
So as we’ve seen, Leibniz claims that substances are efficient causes in numerous
passages but in other passages Leibniz claims that appetitions—which are modifications
of substances—are also efficient causes. We’ve also seen that both substances and
appetitions are principles of change. However, given that appetitions are powers,
appetitions are best understood as principle quo efficient causes and substances are the
principle quod efficient causes. In other words, substances are the efficient causal agents
that produce their accidents by means of their powers or appetitions.
Support for substances being the principle quod or agent or suppositum which
acts can be found in several important passages. For example, in “On Nature Itself”,
Leibniz writes, “To the extent that I have made the notion of action clear to myself, I
believe that the widely received doctrine of philosophy, that actions pertain to supposita,
follows from the notion and is grounded in it.”350 Here, Leibniz explicitly endorses the
thesis we’ve seen Aquinas and Suarez defend as above, that substances or supposita act.
Relatedly, in Leibniz’s fifth letter to Clarke, Leibniz writes, “Properly speaking, motives
do not act on the mind as weights do on a balance, but it is rather the mind that acts by
virtue of the motives, which are its dispositions to act.”351 The mind acts in virtue of its
motives, where such motives are a species of appetitions. Here, Leibniz is again using
the language of principle quo efficient causes.
350
351
(G IV.509: AG 160).
(G VII.392: L 698).
169
The strongest support for this thesis that substances are principle quod causes and
appetitions are principle quo causes, however, is found in a telling section of the New
Essays on Human Understanding. As Theophilus, Leibniz writes, “Faculties or qualities
do not act; rather, substances act through faculties.”352 The context of this passage is key,
for Leibniz is responding—as the character of Theophilus—to an argument in Locke’s An
Essay Concerning Human Understanding against the thesis that powers are efficient
causal agents. I’ll call this the Multiplication of Agents Objection. Locke writes:
But the fault has been, that Faculties have been spoken of, and represented, as so
many distinct agents. For it being asked, what is was that digested the meat in our
stomachs? It was a ready, and very satisfactory answer, to say, that it was the
digestive faculty. What was it that made any thing come out of the body? The
explusive faculty. What moved? The motive faculty: and so in the mind, the
intellectual faculty, or the understanding, understood; and the elective faculty, or
the will, willed or commanded: which is in short to say, that the ability to digest,
digested; and the ability to move, moved; and the ability to understand,
understood. For faculty, ability, and power, I think are but different names of the
same things: which ways of speaking, when put into more intelligible words, will,
I think, amount to thus much; that digestion is performed by something that is
able to digest, motion by something able to move; and understanding by
something able to understand….”353 (2.21.20)
In what follows, I’ll focus on Philalethes’s version, as Philalethes is Leibniz’s voice for
Locke in Leibniz’s New Essays on Human Understanding:
The ordinary way of speaking is, that the understanding and will are two faculties
of the same soul; a word proper enough, if it be used as all words should be, not to
breed any confusion in men’s thoughts,’ as I suspect has happened in this matter
of the soul. And when we are told that ‘the will is superior faculty of the soul;
that it is, or is not free; that it determines the inferior faculties; that it follows the
dictates of the understanding, may be understood in a clear and distinct sense’, yet
352
Ce ne sont pas les facultes ou qualites, qui agissent, mais les Substances par les facultes. (G V.160: NE
174).
353
John Locke, An Essay Concerning Human Understanding (Oxford: Oxford Unviersity Press, 1975),
243-44.
170
I am afraid that they have misled many people into a confused idea of so many
agents acting separately in us.354
Leibniz, writing as Theophilus, responds:
The question of whether there is a real distinction between the soul and its
faculties, and whether one faculty is really distinct from another, has long
exercised the Scholastics. The realists have said Yes, the nominalists No; and the
same question has been debated concerning the reality of various other abstract
beings which must stand or fall with faculties. But I do not think that we need
here plunge into the brambles in an attempt to settle this question, despite the fact
that Episcopius, I remember, attached such importance to it that he thought that if
the faculties of the soul were real beings then human freedom would be untenable.
However, even if they were real, distinct beings, it would still be extravagant to
speak of them as real agents. Faculties or qualities do not act; rather, substances
act through faculties.”355
From Leibniz’s response, Leibniz can be taken to understanding Locke as arguing for the
following conditional, which I’ll call the “Multiplication of Agents” Objection:
If (A) there is a real distinction between the soul and its faculties (powers), then
(B) there would be a plurality of agents acting separately in us.356
Leibniz denies the consequent (B), as evinced by his last sentence in which he claims that
faculties do not act. His strategy, however, is not to deny (A). While he takes no stance
on the truth of (A) in this passage, in fact he argues for (A) elsewhere— Leibniz claims
that a substance is really distinct from its accidents.357 Given that appetitions count
354
(G V.160: NE 174).
Ibid.
356
Leibniz identifies faculties with powers in several passages in his New Essays. At NE 379, Leibniz
offers a line of reasoning that supports the thesis that appetitions are faculties, writing, “Primary powers are
what make up the substances themselves; derivative powers, or ‘faculties’ if you like, are merely ‘ways of
being’ – and they must be derived from substances. . .” As appetitions are derivative forces and derivative
forces are powers, appetitions can be called ‘faculties’. At NE 169-70, Leibniz argues that “The active
power can be called ‘faculty’. . .” As I argued earlier, appetitions are derivative active powers.
357
In his Theodicy, Leibniz writes, “It is true that God is the only one whose action is pure and without
admixture of what is termed ‘to suffer’: but that does not preclude the creature’s participation in actions,
since the action of the creature is a modification of the substance, flowing naturally from it and containing a
variation not only in the perfections that God has communicated to the creature, but also in the limitations
that the creature, being what it is, brings with it. Thus we see that there is an actual (distinction réelle)
355
171
amongst a substance’s accidents, Leibniz is committed to a substance being really distinct
from its appetitions, and so a substance is really distinct from its derivative powers.358 So
Leibniz needs instead to deny that (A) entails (B), which is just what he does. It doesn’t
follow, Leibniz argues, that if a soul (or substance) is really distinct from its powers, then
those powers are agents. Instead, substances are the agents which act or cause through
their powers. Here, Leibniz just is utilizing the principle quod/quo distinction, or the
distinction between agents and the powers by which agents act, in arguing that (A) does
not entail (B).
Leibniz’s response to the Multiplication of Agents Objection spells trouble for the
efficacious-appetition account, if appetitions are efficient causal agents on that account.
For if individual appetitions are efficient causal agents, then given that there is a plurality
of appetitions inhering in a substance in any state the substance is in, there is a
multiplication of agents in a substance in any state it is in. It is no surprise why Leibniz
would deny (B), for the simplicity and unity of created substances is a crucial aspect of
his monadological metaphysics.359 Holding that there are in fact a plurality of efficient
causal agents that are responsible for monadic change would seriously threaten such
simplicity and unity.
Those who endorse the efficacious-appetition account, however, can avoid the
Multiplication of Agents Objection if appetitions are understood as principle quo efficient
causes rather than principle quod efficient causes. However, the efficacious-substance
distinction between the substance and its
modifications.” (G VI.121: T 32).
358
Ibid.
359
I address mereological issues that arise in Leibniz affirming a plurality of accidents in simple substances
in Chapter 4.
172
account—where substances are the principle quod efficient causes or the efficient causal
agents—is consistent with the efficacious-appetition account once appetitions are
understood as principle quo efficient causes.
3.3 The Efficacious-Substance Interpretation and Determinism, Intelligibility, and
PSR
The principle quod/quo distinction helps the efficacious-appetition account avoid the
Multiplication of Agents objection and also shows that in fact the efficacious-substance
account is consistent with the efficacious-appetition account, when understood in a
certain way— substances are principle quod efficient causes and appetitions are principle
quo efficient causes. It still remains to be seen whether affirming that substances are
principle quod efficient causes and appetitions are principle quo efficient causes is
consistent with Leibniz’s determinism and able to meet his strictures on explanation—
Leibniz’s principle of intelligibility and his principle of sufficient reason. I argue that it
is consistent and that it can meet his strictures.
Recall that Leibniz defines change as “nothing but a complex of two states which
are immediate and opposite to each other, together with a force or reason for the change,
which reason itself is a quality.” There is nothing in this definition that requires the force
or reason for the change—the appetition in the case of monadic change—to be the
efficient causal agent. Instead, this definition is consistent with the monad being the
agent. A substance S in state N is the efficient causal agent that produces state N+1 and
does so through its appetitions in N, which constitute its force or reason for N+1. In
other words, the powers by which the substance produces N+1.
173
This account is consistent with Leibniz’s determinism as well. A substance in
state N can only act through the powers it has in state N, where such powers are its
appetitions for future perceptions. A substance in state N will not produce the
perceptions which make up N+1 unless it has appetitions for the perceptions which make
up N+1. This is so because appetitions are appetitions for particular, specific
perceptions, rather than random perceptions. It’s worth noting a similarity between
scholastic powers and Leibnizian appetitions. Fire produces heat in its patients through
fires active powers, rather than coldness or wetness, because fire’s powers are powers for
heat. That is, fire is determined to certain kinds of effects because of its powers.
This account is also consistent with Leibniz’s Principle of Sufficient Reason. The
reason s changes from N to N+1 instead of N’+1 is because s had the appetitions for the
perceptions of N+1 instead of N’+1. That is, s at N had the powers to produce the
accidents which makeup N+1, as opposed to N’+1. Finally, this account also is
consistent with Leibniz’s requirement that change be intelligible. A substance’s changing
from N to N+1 is still explained by the substance’s nature. Further, in agreement with
Rutherford, it’s the substance’s nature as modified. For a substance s’s appetitions when
s is in state N are modifications of s. This does not require that such appetitions are the
efficient causal agents that produce the accidents which makeup N+1, however. Instead,
it’s worth repeating that the appetitions are the powers by which s causes N+1.
3.3.1 Scholastic Substance Causation and Determinism
In further support of my argument that Leibnizian substances, rather than their accidents,
can be principle quod efficient causes while deterministically producing their effects,
scholastics such as Suarez also held that—with the exception of the free actions of
174
intelligent substances—substances which were principle quod efficient causes acted
necessarily in some sense. For a digression, I take a closer look at scholastic accounts of
how effects could be necessitated by their causes when their principle quod causes are
substances. I’ll focus primarily on Suarez’s account but note that his account was just a
more developed version of a broadly scholastic-Aristotelian account.
Before I do so, however, there are five caveats. First, as mentioned earlier,
Suarez is not strictly a substance-causal theorist but rather a res causal theorist. That is,
any res—be it a substance or a real quality—could be a principle quod efficient cause.
Given that Leibniz denied real qualities, I’ll focus only cases where the substance rather
than a real quality is a principle quod efficient cause. The second caveat is that when the
effect is due to a free action of an intelligent substance, the cause does not act necessarily.
However, in all other natural efficient causation, such as fire heating water, the causes act
necessarily. Thus, I’ll further restrict my focus to cases where the substances are nonintelligent.360 A third caveat, related to the second, is that Scholastics such as Suarez
restricted causes that always act necessarily to corporeal substances— substances which
do not exist, speaking strictly, in Leibniz’s ontology.
A fourth caveat is that the standard scholastic cases of efficient causes that act
necessarily are cases of transeunt efficient causation between distinct created
substances.361 Leibniz, of course, denies any transeunt efficient causation between
360
Suarez writes, “For the present, we are asserting merely that every faculty which altogether lacks the use
of reason exercises its operations by natural necessity.” DM 19.1.12
361
Jeff Brower has argued that according to scholastics, any deterministic immanent causation would in
fact be formal causation and they would thus argue that all monadic change is also due to formal causation.
This might be troublesome for my account if a substance s’s formally causing some accident A entailed that
s did not efficiently cause A. However, as Robert Pasnau has shown, many scholastics also held that formal
175
created substances and instead holds that all creaturely efficient causation is immanent
causation within a substance. The fourth caveat points to a fifth. The principle quod
causes—substances—which act of necessity in scholastic cases have powers which often
need an external excitation for the substance to act. Leibnizian powers, in contrast, act
unless impeded.362 In spite of these five caveats, the scholastic account of principle quod
efficient causes that act necessarily and are also substances provides Leibniz with ample
reasons to also affirm deterministic substance causation. The reasons hinge mainly on
the principle quod/quo distinction and God’s concurrence. I’ll first explain the scholastic
account and then show it sheds light on how, with some important revisions, Leibnizian
substance causation can also be deterministic.
Suarez’s DM 19.1 is dedicated to efficient causes that act necessarily. Suarez
writes, “among created causes there are many that operate necessarily once all the things
they require for operating are present.”363 For much of DM 19.1, Suarez articulates just
what he takes those requirements to be, ultimately concluding that there are nine:
For any agent substance sA and any accident A, sA efficiently causes A if and only if:
(i)
(ii)
(iii)
sA has the full and sufficient power to act;364
There is a patient substance sP such that sP is susceptible and sufficient close
to sA;365
Any medium between sA and sP is suitable for and susceptible to sA’s
action;366
causes efficiently cause accidents. See Robert Pasnau, “Form, Substance, and Mechanism,” Philosophical
Review 113, no. 1 (2004): 31-88. On this understanding, Leibnizian immanent efficient causes are also
formal causes, but their being formal causes does not entail that they are not efficient causes. For a recent
paper arguing that all secondary causation is formal but not efficient causation, see Sukjai Lee, “Leibniz on
Divine Concurrence,” Philosophical Review 113, no. 2 (2004): 203-48.
362
I clarify this difference shortly.
363
DM 19.1.1
364
DM 19.1.2
365
Ibid.
366
Ibid.
176
(iv)
(v)
(vi)
(vii)
(viii)
(ix)
There is nothing impeding the action with an equal power to resist it;367
The patient sP does not already have the accident that sA would cause;368
Any action required beforehand which is presupposed for sA’s causing A in sP
is already completed;369
sA is not a free cause;370
God concurs with sA’s causing A;371
sA is not indifferent with respect to more than one effect;372
It’s important to address the relation between conditions (i) through (ix) and some agent
substance sA efficiently causing an accident A in a patient substance sP. Conditions (i)
through (ix), according to Suarez, constitute the total cause. Further, there is a necessary
relation between the total cause and sA’s efficiently causing A in sP. The total cause
necessitates the effect.373 Whatever the type of necessity is involved, it is quite strong.374
As Suarez writes, “Still, if the matter is considered carefully, even God himself does not
seem to be able to bring it about in the composed sense (as they call it) that a cause which
367
Ibid.
This fifth condition rules out causal overdetermination.
369
DM 19.1.3
370
DM 19.1.4
371
Ibid.
372
DM 19.1.5. Suarez argues that the ninth condition is in fact entailed by or equivalent to one of the
previous conditions. While Suarez doesn’t specify which condition, precisely, entails or is equivalent,
textual evidence points in favor of it being the seventh condition. He writes, “Nonetheless, one should, it
seems, reply that this [ninth] condition has rather to be traced back to one of the conditions posited above;
that is, to the absence of one of those conditions. For the condition of indifference, taken just by itself, is in
some sense incompatible with the proper determination of natural agents, since it is proper for them to be
determined to one effect. How, then, can they have this indifference of themselves.” See DM 19.1.6
373
I’m less concerned with the exact type of necessity involved and more concerned with establishing that
there is some sort of necessity involved between requirements (i) – (ix) and sA’s efficiently causing A in sP.
Given that what is at issue is the particular effects of particular substances, natural necessity is the most
likely candidate. However, Walter Ott has recently argued that the necessity is in fact logical necessity.
That is, the effect is logically necessitated by the Total cause because it is a contradiction to posit the Total
cause and deny the effect. Ott writes, “Now, once the requisite active an passive powers are in place,
‘natural causes cannot prevent the action of a necessary agent, since they do not have the power to change
the nature of things or to remove wholly intrinsic properties.’ Note what it would take for a natural cause
to prevent the action of such an agent, i.e., to change the course of events: one would have to alter its
intrinsic properties. In other words, one would have to bring it about that fire was not fire.” See Walter
Ott, “Causations, Intentionality, and the Case for Occasionalism,” Archiv f. Gesch. d. Philosophie, 90
(2008): 175.
374
Assuming a difference between logical and physical or natural necessity, where the former is stronger
than the latter.
368
177
by its nature acts necessarily should fail to act once all the things required for acting have
been posited.”375 So the only way for sA to not efficiently cause A in sP would be for one
of (i) – (ix) is not the case. Not even God could bring prevent sA’s efficiently causing A
in sP if (i) – (ix) are in place.376 God would have to remove one of (i) – (ix).377
For an example of Suarezian substance causation where the substance acts of
necessity, take the example of fire burning some wood.378 Fire—the agent substance—
has the power to burn some wood— the patient substance. Fire burns the wood via its
heat—a causal power of fire. When the fire is appropriately situated near some wood,
either through direct contact or with the right kind of medium between the fire and the
wood—such as air, the fire will burn the wood as long as God concurs and no other
substances impedes the fires burning the wood, such as a bucket of water spilled on the
fire. So in this case, the fire—a substance—produces an effect—burnt wood—of
necessity. While Suarez does not use the exact term, if conditions (i) through (ix) obtain,
the fire is determined to burn the wood.
It’s again important to stress the principle quod efficient cause—the efficient
causal agent—is the fire, a substance in Suarez’s metaphysics. Yet a power of the
substance—specifically, the fire’s heat—plays a crucial part in explaining the substance’s
efficiently causing the wood’s being burnt. The fire’s powers also play a crucial part of
the explanation for why the fire’s efficiently burning the wood is necessitated. The fire’s
375
DM 19.1.14.
Suarez writes, “Still, if the matter is considered carefully, even God himself does not seem to be able to
bring it about in the composed sense (as they call it) that a cause which by its nature acts necessarily should
fail to act once all the things required for acting have been posited.” (DM 19.1.14)
377
Suarez continues, “Therefore, it is not the case that God brought it about that the fire did not act even
though all the required things had been posited; instead, he removed one of those things.” (DM 19.1.14)
378
I use fire because it is a frequent example of Suarez’s.
376
178
powers are one of the necessary ingredients in the total cause. It’s just that the fire’s
powers are not efficient causal agents. Instead, they are the agent’s powers.
3.3.2 Deterministic Leibnizian Substance Causation
I argue that if the scholastics can consistently hold that substances rather than their
powers are the efficient causal agents that act of necessity, then Leibniz can consistently
hold that monads are efficient causal agents which deterministically cause their effects.379
There are, of course, important differences between the deterministic efficient causal
activity of Leibnizian and scholastic substances. In Leibniz’s metaphysics, the efficient
causal agents are immaterial monads, which immanently efficiently cause their
accidents—perceptions and appetitions—by means of their derivative powers—
appetitions. So the Leibnizian model is going to differ in some important ways from the
scholastic model. For a start, Suarezian conditions (ii) and (iii) will not apply to
Leibniz’s model.
Another important difference that calls for further elaboration concerns how a
Leibnizian power would be impeded. It’s easy to imagine cases of corporeal scholastic
substances being impeded from exercising their causal powers. Fire would be impeded
from burning some wood if a different substance—such as some cold water—were
splashed onto the wood. However, Leibniz’s account is burdened with answering how an
immaterial monad would be impeded from exercising its powers since it doesn’t causally
interact with other created substances and is instead causally responsible for all its
miraculous accidents. Of course, God could impede such a power, say by withholding
379
I address soon the type of necessity involved in deterministic monadic efficient causation as Leibniz is
adamant throughout his writings that such deterministic monadic efficient causation is contingent in some
sense.
179
his concurrence, in which case the impediment would be miraculous. But more pertinent
here are natural impediments.
To address this question, it’s worth first noting that according to Leibniz, a key
difference between monadic powers and the powers of scholastic substances is that
monadic powers always act unless impeded, rather than needing a stimulus of some sort:
Active force differs from the mere power familiar to the schools, for the active
power or faculty of the scholastics is nothing but a proximate principle of acting,
which needs an external excitation or a stimulus, as it were, to be transferred into
action. Active force, in contrast, contains a certain act or entelechy and is thus
midway between the faculty of acting and the act itself and involves a conatus, It
is thus carried into action by itself and needs no help but only the removal of an
impediment.380
So a monad’s powers will always be exercised unless impeded. But again, what impedes
a monad’s powers? Following Mark Kulstad, let’s draw a distinction between the overall
derivative force of a substance s in state N and the individual derivative forces which
makeup s’s state N.381 The individual derivative force which make up state N are s’s
appetitions AN1, AN2, AN3, . . . , ANx. Recall that appetitions are always appetitions for
various accidents. As causal powers, appetitions are powers for s to produce such
accidents.
Yet these appetitions frequently conflict.382 The perception PN1 which appetition
AN1 is an appetition for will conflict with the perception PN2 which appetition AN2 is an
appetition for. How such appetitions conflict takes us into Leibniz’s metaphysics of final
380
L 433.
Mark Kulstad, “Appetition in the Philosophy of Leibniz,” Mathesis rationis: Festschrift für Heinrich
Schepers (1990): 133-151.
382
Leibniz writes, “Various perceptions and inclinations combine to produce a complete volition: it is the
result of the conflict amongst them.” (G.V.178: NE 192).
381
180
causation, a subject of much recent scholarly debate.383 As I do not need to take a stand
in this paper, I will only canvas the options found in Leibniz’s writings and the secondary
literature. AN1 conflicts with AN2 if (i) the goodness of PN1 is greater or lesser than the
goodness of PN2 or (ii) the apparent goodness of PN1 is greater or lesser than the apparent
goodness of PN2.384 The goodness (or apparent goodness) of PN1 which is greater or
lesser than the goodness (or apparent goodness) of PN2 is either (ia/iia) the goodness of s
or (ib/iib) the goodness of the whole universe.385 So two or more appetitions conflict
when they differ with respect to (i) or (ii) and one appetition will impede another when it
is stronger than the other in terms of either (i) or (ii). So what impedes an individual
appetition is either God or a different appetition.
The total derivative force of s in state N is the result of the outcome of the various
conflicting appetitions which makeup N. In a sense, the total force could be understood
as an appetition or tendency for the next complete state N+1. But the total force is made
up of the individual appetitions and the result of the conflict between them, it would not
be impeded by an individual appetition. Rather, I suggest that the only thing that could
impede the total derivative force of a monad—its tendency for the complete next state
N+1—is God.386
383
See Rutherford, “Laws and Powers in Leibniz,” 169-174 and “Leibniz on Spontaneity,” 166-174;
Bolton, “Change in the Monad,” 175-196; Jorati, “Leibniz on Causation –Part 2,” 398-405.
384
Following Rutherford, call (i) the “natural” teleology view and (ii) the “desire” teleology view.
385
Rutherford explains both of these views in greater depth in “Laws and Powers in Leibniz,” 170.
386
With Kulstad, I think there is a strong and intended analogy in Leibniz’s metaphysics between the
conflicting appetitions of a monad and the conflicting forces of a physical body. See Kulstad, 139.
181
3.3.3 The necessity involved in Leibnizian Deterministic Substance Causation
There is one last issue that needs to be addressed— the necessity involved in Leibnizian
deterministic substance causation. I address this because in the account of scholastic
causal necessitation I presented above, some scholars, such as Walter Ott, have argued
that there is a logically necessary connection between the total cause and its effect, even
in the many cases where the principle quod efficient cause is a substance. Leibniz,
however, argues throughout his career that created substances deterministically but
contingently produce their effects.
I address this with some irony, since as we’ve seen throughout this paper, the
efficacious-substance interpretation has been challenged for being incompatible with
whatever sort of necessity is required for deterministic causation. Yet if the account of
scholastic substances acting of necessity that I presented above is accurate and if the
scholastic account sheds light on how Leibnizian substance causation can be
deterministic, then prima facie it saddles Leibnizian substance causation with a type of
necessity that he would deem too strong for deterministic substance causation— logical
necessity.
This worry is misplaced, however, once one takes into account the distinction
between the total cause and the principle quod efficient cause. Recall that Suarez writes,
“Still, if the matter is considered carefully, even God himself does not seem to be able to
bring it about in the composed sense (as they call it) that a cause which by its nature acts
necessarily should fail to act once all the things required for acting have been posited.”
According to Suarez, a substance will act necessarily or produce an effect when all the
182
things required for acting have been posited, where all the things required for acting are
all of the elements of the total cause, instead of just the principle quod efficient cause.
Leibniz’s distinction between absolute and hypothetical necessity is important
here, for what Leibniz specifically denies is that what is determined is absolutely
necessary.387 Some truth is absolutely necessary when its denial entails a contradiction.
A created substance s being in state N would not be absolutely necessary on Leibniz’s
account, as the denial of the proposition “s is in state N” does not entail a contradiction.
The claim “s’s being in state N efficiently causes the accidents which makeup state N+1”
is also not absolutely necessary, as it’s denial would not entail a contradiction. However,
Leibniz would likely agree with Suarez, however, that there is a contradiction in any
conjunction affirming a total cause but denying the effect of the total cause. For the total
cause is not limited to the created principle quod efficient cause modified in some way
but also God or God’s concurrence and the laws of monadic change.388
So Leibniz’s account of deterministic causation differs in some important ways
from the scholastics’ accounts. Yet, none of these differences entail that Leibniz can’t
consistently hold that substances deterministically produce their effects while scholastics
can hold that substances deterministically produce their effects. Any objection to
Leibniz’s account would affect scholastic accounts as well. However, even if there are
serious reasons to doubt the plausibility of deterministic substance causation, it’s valuable
to at least document the historical precedent to the interpretation I’m defending.
387
See (G IV.436-7: AG 45-6), (G III.400-1: AG 194), (G V.161: NE 176), (G VI.163-4: NE 178-9), (G
VI.123-4: T 37), (G VI.131-2: T 53), (G VI.341-2: T 381), and (G VI.351: T 395).
388
But the necessity involved when the effect follows from the Total Cause is hypothetical, rather than
absolute necessity. I note further that my account of contingent but deterministic causation differs little
from Rutherford’s own account. See Rutherford, “Laws and Powers in Leibniz,” 165-66.
183
Conclusion
In this chapter, I argued for a nuanced interpretation of the Efficacious-Substance
Interpretation-- account of the causal relata in creaturely causation in Leibniz’s
metaphysics. This interpretation has the advantage of having historical precedent in
many Scholastic accounts of efficient causation. It is also supported by numerous
passages in Leibniz’s writings. More importantly, the interpretation I defended
overcomes a serious challenge to the efficacious-appetition interpretation— Locke’s
“Multiplication of Agents” objection. Finally, the interpretation I defended showed how
appetitions play an important explanatory role in monadic change and is consistent with
appetitions determining monadic change even though such appetitions are not efficient
causal agents.
APPENDICES
184
Appendix A. Leibniz’s Missing Overdetermination Premise
In at least two passages written early in his career, Leibniz argues that created substances
are not transeunt causes because everything that happens to a created substance is the
result of its own notion or complete concept. In “Primary Truths”, Leibniz writes:
Strictly speaking, one can say that no created substance exerts a metaphysical
action or influx on any other thing. For, not to mention the fact that one cannot
explain how something can pass from one thing into the substance of another, we
have already shown that from the notion of each and every thing follows all of its
future states.389 (Emphasis added)
In chapter 14 of his Discourse on Metaphysics, Leibniz also writes:
We could therefore say in some way and properly speaking, though not in
accordance with common usage, that one particular substance never acts upon
another particular substance nor is acted upon by it, if we consider that what
happens to each is solely a consequence of its complete idea or notion alone, since
this idea already contains all its predicates or events and expresses the whole
universe.390 (Emphasis added)
Leibniz gives same argument against creaturely transeunt causation in both passages.
Using the wording from the first passage, a first-pass of the argument can be expressed
thus:
(P) From the notion of each and every thing follows all of its future states.
(C) So, no created substance exerts a metaphysical action or influx on any other
thing.
Scholars have rightfully pointed out that (P) alone is not sufficient for (C). Even if all of
a substance’s future states follow from it’s complete concept— if the substance’s
389
390
AG 33.
(G IV.439: AG 47)
185
complete concept is the sufficient cause of all the substance’s future states, it is still
possible that some states are also caused by different substances.391 To make the move
from (P) to (C), Leibniz needs a further premise (P2) ruling out causal
overdetermination.392 Unfortunately, Leibniz never supplies an argument for (P2) in the
vicinity of the above passages, nor does he even state (P2). Worse, an argument for (P2)
in Leibniz’s corpus seems to be lacking.393 In this appendix, I shall argue that given
Leibniz’s transference condition, Leibniz does have reasons to rule out causal
overdetermination of the sort needed to allow him to make the move from (P) to (C).
Before I present the argument, some justification is needed for why one should
even bother with Leibniz’s argument from (P) to (C), as several objections can be raised.
It is doubtful that Leibniz ever held the view that the complete concept does any efficient
391
See Donovan Cox, “Leibniz on Divine Causation: Creation, Miracles, and the Continual Fulgurations,”
Studia Leibnitiana, (2002): 189; Mark Kulstad and Laurence Carlin, "Leibniz's Philosophy of Mind", The
Stanford Encyclopedia of Philosophy (Winter 2013 Edition), Edward N. Zalta (ed.), URL =
http://plato.stanford.edu/archives/win2013/entries/leibniz-mind/; Nicholas Jolley, “Leibniz: Truth,
Knowledge, and Metaphysics,” in G.H.R. Parkinson edited, Routledge History of Philosophy, Volume IV,
The Renaissance and Seventeenth-Century Rationalism, 382; Christia Mercer, Leibniz’s Metaphysics: It’s
Origins and Development (Cambridge: Cambridge University Press, 2001), 227-230; and R.C. Sleigh, Jr.,
Leibniz & Arnauld: A commentary on their Correspondence (New Haven: Yale University Press, 1990),
144.
392
Kulstad and Carlin write, “Even if conceptual considerations about substances were sufficient to explain
their apparent causal activity, it does not seem to follow that substances do not interact—unless one is
assuming that causal overdetermination is not a genuine possibility. Leibniz seems to be assuming just that,
but without argument.” See Ibid.
393
Sleigh draws attention to one potential statement against causal overdetermination. In a letter to
Arnauld, Leibniz wrote, “Anything capable of having many causes is never a complete entity.” (G II.72)
However, in the context, Leibniz is discussing the causation of substances. A substance—being a complete
being—could not have multiple causes. Yet that doesn’t allow Leibniz to make the further move that
accidents can’t have multiple causes. In his Discourse on Metaphysics, Leibniz writes, “Since this is so, we
can say that the nature of an individual substance or of a complete being is to have a notion so complete
that it is sufficient to contain and to allow us to deduce from it all the predicates of the subject to which this
notion is attributed. An accident, on the other hand, is a being whose notion does not include everything
that can be attributed to the subject to which the notion is attributed.” See (G IV.432-3: AG 41) and Sleigh,
144.
186
causal work.394 Even if the early Leibniz did hold such a view, the mature Leibniz did
not. Instead, the law-of-the-series—an active entity--is responsible for a substance’s
states, as opposed to the complete concept or notion—a static entity.395 Further, the lawof-the-series just is the substance.396 So the mature Leibniz held that the substance is
responsible for all of its states, a thesis I soon unpack and which I’ll call the Spontaneity
Thesis (ST). Finally, the mature Leibniz didn’t conclude that created substances are not
transeunt causes because of the Spontaneity Thesis. Instead, the mature Leibniz reasoned
to the Spontaneity Thesis in part because he argued that created substances cannot be
transeunt causes, as addressed in chapters 2 and 3.397
However, I argue that there are two reasons why it’s worth addressing whether
Leibniz had reasons to deny causal overdetermination, allowing him to infer (C) from
(P). First, while the later Leibniz didn’t infer (C) from (P), he also never explicitly
denied that you could make the inference. Second, whether or not Leibniz had reasons to
394
Rather than the complete concept being the efficient cause of the substance’s states, J.A. Cover argues
that Leibniz is merely expressing in the formal mode something intimately tied to what Leibniz expresses
in the material mode. Cover writes, “What Leibniz expresses in the formal mode, as the thesis that every
predicate true of a substance has a reason or foundation in its individual concept, is intimately tied to what
he expresses in the material mode, as the thesis that each state of a substance has its causal origin in
preceding states.” A better understanding, then, is that the substance is the cause of its states, rather than
the concept. See J.A. Cover, “Non-Basic Time and Reductive Strategies: Leibniz’s Theory of Time,” Stud.
Hist. Phil. Sci., 28, No. 2 (1997): 308.
395
Donald Rutherford addresses the progression of Leibniz’s thought away from the complete concept as
expressive of the substance’s nature towards the law of the series. See Donald Rutherford, Leibniz and the
Rational Order of Nature (Cambridge: Cambridge University Press, 1995), 138-154.
396
In this section, I rely on J.A. Cover and John O’Leary Hawthorne’s argument that the law of the series is
the substance, rather than a component of the substance. See J.A. Cover and John O’Leary-Hawthorne,
Substance & Individuation in Leibniz (Cambridge: Cambridge University Press, 1999), 214-226.
397
For example, in his 1695 “A New System of Nature,” Leibniz writes, “Therefore, since I was forced to
agree that it is not possible for the soul or any other true substance to receive something from without,
except by divine omnipotence, I was led, little by little, to a view that surprised me, but which seems
inevitable, and which in fact has very great advantages and rather considerable beauty. That is, we must
say that God originally created the soul (and any other real unity) in such a way that everything must arise
for it from its own depths, through a perfect spontaneity relative to itself, and yet with a perfect conformity
relative to external things.” See (G IV.484: AG 143).
187
affirm or deny the possibility of causal overdetermination is a question worth pursuing on
its own in a project whose central focus is Leibniz’s metaphysics of causation.
Before laying out the argument, I should also remark about the methodological
approach I adopt in this appendix. In this appendix, more than any other part of the
dissertation, I utilize “philosophical” history of philosophy. As in other parts, my usage
of “philosophical” history of philosophy is in the service of “exegetical” or “explanatory”
history of philosophy. I’m interested in whether or not Leibniz is justified in making an
inference that he made in important writings of his early and mid-career— whether or not
Leibniz can infer (C) from (P). In Chapter 1, I distinguished three ways the historian of
philosophy can see look for connections between premises in an argument by a historical
figure such as Leibniz when Leibniz doesn’t explicitly offer the justification: (i) by
searching through Leibniz’s writings for theses he did explicitly defend which can link
(P) to (C), even though Leibniz does not himself show that or even state that such theses
link (P) and (C); (ii) by looking broader at theses widely held by historical figures whom
Leibniz was familiar with that could link (P) and (C) and which Leibniz would have no
reason to reject; or (iii) by engaging in metaphysical reasoning oneself to rationally
reconstruct a link between (P) and (C). In this appendix, I help myself to (i) and (iii) in
addressing if Leibniz has reasons to reject causal overdetermination, therefore providing
him the missing premise that allows him to infer (C) from (P). Since this rational
reconstruction will go significantly beyond the text, I’ve dealt with this topic in an
appendix to the chapter instead of the main body.
Thus, I’m going to proceed with the argument. Further, I’m going to do so by
actually offering an argument that allows Leibniz to get the denial of creaturely transeunt
188
causation from either the Spontaneity thesis (that I soon present) or (P). In keeping with
the overall focus on this dissertation, I’ll start with the Spontaneity thesis. By assuming
the Spontaneity Thesis, I will show that Leibniz has an argument he can give from his
mature metaphysics to the denial of creaturely transeunt causation via reasons Leibniz
had to deny overdetermination. The argument, once given, can then be re-applied to the
older argument that assumes (on some interpretations) that the complete concept is doing
real causal work— the view that the complete concept is the genuine cause of all of a
substance’s states. So Leibniz has the resources to deny that creatures are transeunt
causes from (P), even if the mature Leibniz believes (P) is false.
I now turn to the argument. First, as indicated above, given that the mature
Leibniz likely did not hold that the complete concept of a created substance did any
efficient causal work. Instead, the created substance itself is the natural cause of all of its
accidents. Leibniz maintains this throughout his later writings. For example, in his 1695
“A New System of Nature,” he writes, “That is, we must say that God originally created
the soul (and any other real unity) in such a way that everything must arise for it from its
own depths, through a perfect spontaneity relative to itself, and yet with a perfect
conformity relative to external things.”398 A few lines further, he elaborates, writing:
For why should God be unable to give substance, from the beginning, a nature or
an internal force that can produce in it, in an orderly way (as would happen in a
spiritual or formal automaton, but free in the case where it has a share of reason),
everything that will happen to it, that is, all the appearances or expressions it will
have, without the help of any created being?399
398
399
Ibid.
(G IV.485: AG 144).
189
In his New Essays on Human Understanding, Leibniz continues to affirm the spontaneity
of created substances, writing, “Anything which occurs in what is strictly a substance
must be a case of action in the metaphysically rigorous sense of something which occurs
in the substance spontaneously arising out [arrive] of its own depths.” (New Essays A vi,
6, 210) In his Theodicy, Leibniz also defends the thesis of creaturely spontaneity against
the occasionalists and skeptics of the efficacy of creatures such as Bayle:
Bayle asserts, for instance, that by purely philosophical meditations one can never
attain to an established certainty that we are the efficient cause [la cause
efficiente] of our volitions. But this is a point which I do not concede to him: for
the establishment of this system demonstrates beyond a doubt that in the course of
nature each substance is the sole cause of all its actions, and that it is free of all
physical influence from every other substance, save the customary cooperation of
God.400
Therefore, it’s worth replacing (P) with a different premise stating what I’ll call Leibniz’s
“Spontaneity Thesis” (ST):
For any created substance s and any accident A, if A naturally inheres in s and it is
not the case that A initially inhered in s then s is the natural sufficient efficient
cause of A’s inhering in s.
I draw attention to five features of the Spontaneity Thesis. First, the Spontaneity Thesis
applies—as indicated—only to non-initial accidents. I formulate the Spontaneity Thesis
that way in order to leave open the possibility that either God is the efficient cause of
initial accidents (upon the first moment of the creation of a substance).401 I note,
400
(G VI.295-6: T 300).
190
however, that the Spontaneity Thesis is consistent with the created substance efficiently
causing initial accidents (on some accounts where substantial forms are like functions,
where God created the function and supplies the first argument but the substance/function
is what outputs the state and thus efficiently causes the first state).402 Second, and
relatedly, I include the term “natural” in the antecedent to focus on non-miraculous
accidents. For the purposes of this appendix, I understand a natural accident as a noninitial accident that inheres in a substance and is not solely caused by God.
Third, note the term “sufficient” in the Spontaneity Thesis, where s is a sufficient
efficient cause of A’s inhering in s. By sufficient, I mean what many early moderns and
Scholastics meant by a “total” cause, something that is contrasted with a “partial” cause.
Take two persons pulling a boat by rope, where each person is individually too weak but
together the two are jointly strong enough to pull the boat. Each person individually is
then a partial cause of the boat’s motion. The two persons taken together are a sufficient
cause of the boat’s motion. This notion of “sufficient” is what contemporary
401
However, ST conjoined with other theses from Leibniz’s metaphysics is inconsistent with a different
created substance causing a non-initial accident in s. The two other considerations are that first, Leibniz
holds that created substances can only come into existence through creation by God. Second, we saw
Leibniz’s claims in chapter 2 that created substances must have some accident or other. So God cannot
create a substance without that substance having some accident or other. If God cannot create a substance
without that substance having some accident or other, then we must address what caused the accident.
There are three options: God, s, or a different created substance s’. The most obvious answer is God but
the second option, s, also has merit. God could create s and s cause A. This would not be a temporal
progression. Instead, s is prior in nature to A. So it is not the case that s exist at some time without having
an accident inhere in it.
But suppose that s’ cause A to inhere in s, where A is s’s first accident. Given the thesis that a
substance must have some accident or other, s’ cannot cause A to inhere in s at a time t’ later than the first
moment of s’s existence. So we have God creating a substance s and a different substance s’ causing A to
inhere in s at the moment of s’s creation. I don’t think this picture works. S’ has to have something to work
with to cause A to inhere in s, but that presupposes that s exist prior to s’ causing A to inhere in s. That is,
if s’ can cause A to exist in s only if s exists. But prior to the creation of s, s does not exist.
402
Therefore, this option is consistent with a model of the substance proposed and defended at length by
J.A. Cover and John-O’Leary Hawthorne. See Cover and O’Leary-Hawthorne, Substance & Individuation
in Leibniz, 214-252.
191
philosophers have in mind when discussing overdetermination. Overdetermination
occurs when there are two causes sufficient for some effect. I’ll specify more precisely
the sort of overdetermination at play soon.
Fourth, I use the term “naturally” in the consequent in order to make the
Spontaneity Thesis consistent with divine concurrence. Divine concurrentists such as
Leibniz hold that God is also an efficient cause of accidents inhering in created
substances.403 Thus, the Spontaneity Thesis as expressed is the thesis that the created
substance is the only created cause needed for an accident to inhere in itself.
Fifth and finally, I note that the Spontaneity Thesis itself is in fact consistent with
a different created substance s2 also efficiently causing A to inhere in s1.404 That is, while
ST holds that a substance causes all its natural non-initial accidents, those accidents could
also be caused by a different created substance as well. Thus, the Spontaneity Thesis
needs a premise against overdetermination, seemingly lacking in Leibniz’s corpus, if it’s
to get (C) – just as the first formulation of (P) needs a premise against overdetermination
as well.
I’ll replace (P) with the Spontaneity Thesis then as the first premise of the
argument. In keeping with the theme of my dissertation, I’ll replace or revise (C) with a
conclusion denying creaturely transeunt causation. The denial of creaturely transeunt
causation can be expressed as follows:
403
The details are complicated and go beyond the topic of this appendix, so I shall bracket the topic for
now. I note that in this appendix, I am concerned primarily with the possibility of creaturely causal
overdetermination, where two or more created substances are sufficient causes of the same effect.
404
ST is in fact consistent with a different created substance causing an initial or non-initial accident. I’ve
presented problems with a created substance causing an initial created accident in another substance in a
footnote above.
192
For any created substance s and any accident A, if A inheres in s then it is not the
case that there is some created substance s’ such that s is not identical to s’ and s’
is an efficient cause of A’s inhering in s.405
As expressed above, Leibniz could only get the denial of creaturely transeunt causation
from the Spontaneity Thesis if there were an additional premise against
overdetermination. Before I get to the main argument, I need to specify exactly what
kind of overdetermination Leibniz would have to rule out, as there are several kinds of
overdetermination. Specifically, Leibniz needs a premise against what Eric Funkhouser
calls independent causal overdetermination or what other philosophers call coincident
overdetermination.406 As an example of independent causal overdetermination, take two
assassins who both shoot the same person at the same time in the heart, where each bullet
individually is sufficient for the person’s death. The person’s death, event E, is then
overdetermined. Three features of E make E an effect that is independently
overdetermined. First, both assassins or assassins’ shootings are sufficient for E. Both
bullets struck the heart in such a way that the person would have died instantly without
the other bullet. This is in contrast to a scenario in which the assassins or assassins’
shootings are jointly sufficient and individually necessary but not individually sufficient,
where, say, each assassin individually would seriously harm but not kill the person.
Second, both causes of E are distinct from each other. This needs little
elaboration. Assassin A1 is not numerically identical to Assassin A2. Third, both causes
of E are independent of each other. I don’t mean that the assassins have different
405
I note that this denial of creaturely transeunt efficient causation applies to both initial and non-initial
accidents.
406
Eric Funkhouse, “Three Varieties of Causal Overdetermination,” Pacific Philosophical Quarterly 83
(2002): 335-351.
193
employers. Instead, neither assassin or assassin’s shooting—as a cause—was necessary
for the existence of the other cause. This is best understood in contrast with certain types
of non-independent and non-coincident overdetermination entailed by certain views on
mental causation where two sufficient causes are distinct but dependent on each other.407
Take a multiply realizable mental state M2 that has as a sufficient cause the previous
mental state M1. On some understandings of the relation between the mental and the
physical, M2 also has as a sufficient cause the brain state B1, where B1 is distinct from M1.
However, while M1 and B1 are distinct, M1 is dependent on B1. That is, the token mental
state M1 cannot exist without the token brain state B1. Thus, in contrast to the assassins,
M1 and B1 are non-coincident/non-independent overdeterminers of M2.
Given that what is at issue in the argument from the Spontaneity Thesis to the
denial of creaturely transeunt causation is whether one created substance could be the
efficient cause of an accident in another, the type of overdetermination ruled out then
needs to be independent causal overdetermination, where there cannot be two sufficient,
distinct, and independent causes (in this case created substances) of an accident inhering
in a created substance. So what reason(s) would Leibniz have to rule out
overdetermination of that sort? Prima facie, Leibniz has no good reasons, nor does
anyone else. Non-independent/non-coincident overdetermination is at least initially
problematic, given that it would be widespread and systematic. Appealing to one
understanding of Ockham’s razor, if there is a theory requiring lawful overdetermination
of event type E and a theory that doesn’t require lawful overdetermination of event type
407
This is a view of mental causation criticized by Jaegwon Kim’s Causal Exclusion Argument. Kim gives
this argument in several publications, including Jaegwon Kim, “Mechanism, Purpose, and Explanatory
Confusion,” Philosophical Perspectives 3(1989): 77-108.
194
E, the latter theory is to be preferred to the extent that it posits less sufficient causes of
the same effect. But coincident/independent overdetermination is not systematic and
widespread in that way. Why couldn’t two assassins overdetermine E where E is the
person’s death? Ted Sider brings up a reason to rule out such independent
overdetermination:
Metaphysical objection: overdetermination is metaphysically incoherent. Here is
a picture. Causation is a kind of fluid divided among the potential causes of an
effect. If one potential cause acts to produce an effect, that fluid is used up, and
no other potential cause can act. Atoms causing the shattering of a window would
use up the available causal fluid, leaving none of the baseball composed of those
atoms.408
But Sider rightly dismisses such a reason, writing, “This, of course, is a bad picture. It
takes seriously a view of causation that no one accepts.”409 I agree with Sider that the
picture above is a picture of a theory no one accepts. However, with some revisions, a
similar picture can be developed that does capture how Leibniz understood creaturely
transeunt causation.410 I grant that it is a mistake to think that when one cause C1 causes
an effect E in a patient substance P, C1 drains or uses up all the causal fluid so that the
other causes C2, C3, . . . , CN no longer have any causal fluid to give. However, keeping
the fluid analogy, I argue that one should think of P as a container.411 When C1 causes E
in P, if C1 is a sufficient cause of E in P, then P is filled. So while C2, C3, . . . , CN may
have fluid to give, they can’t give it to P because P is already full.
408
Theodore Sider, “What’s So Bad About Overdetermination?”, Philosophy and Phenomenological
Research 67 (2003): 719-726.
409
Ibid.
410
Additionally, the picture I develop captures how many contemporary metaphysicians understand
causation, as I show in the final section of this appendix.
411
Jonathan Jacobs first introduces the analogy of effects with containers. See Jonathan D. Jacobs, “Causal
Powers: A Neo-Aristotelian Metaphysic” (PhD diss., Indiana University, 2007), 29-30.
195
The above picture fits in with Leibniz’s understanding of creaturely transeunt
causation given his transference condition for such causation. As I argued in chapter 2,
in spite of strongly worded denials from scholastics, Leibniz thought that if it were
possible for a created agent substance to transeuntly produce an accident in a distinct
patient substance, such causation would have to consist in the transference of the accident
from the agent to the patient.412 Leibniz, using the language of figures such as Suarez,
thought such causation—if possible—would have to consist in a literal “flow” or “influx”
from the agent to the patient. In other places, he describes it as an accident “passing” 413
from one subject to another or “detaching” from the agent.414 The patient substance
receives the effect—the accident—that the agent transfers.415 This gives Leibniz a reason
to rule out causal overdetermination at least in cases where a created substance produces
an accident in a patient, where such production consists in transference. If a created
substance s1 produces the token accident A in s2 and s1 is a sufficient cause of A’s
412
See, for example, (G IV.498-500: L 459-60).
Leibniz of course dismisses the possibility of creaturely transeunt causation because he thinks such
passing is not possible, but the argument reveals that he thinks such passing is necessary if creaturely
transeunt causation is possible, as I argue in Chapter 2. Leibniz writes, “I am not surprised that you
encounter insurmountable problems when you seem to be entertaining something as inconceivable as an
accident’s passing from one subject to another; but I see no reason why we have to suppose such a thing. It
is almost as strange as the Scholastics’ notion of accidents which are not in any subject; though they are
careful to attribute theirs solely to the miraculous workings of divine omnipotence.” See (G V. 208: NE
224). See also L 269.
414
Leibniz writes, “Monads have no windows, through which anything could come in or go out. And
accidents cannot detach themselves and stroll about outside of substances, as the Scholastics' sensible
species used to; so neither substance nor accident can come into a monad from outside” See (G VI.607-8:
AG 214).
415
Leibniz does not think created substances can be acted on because, in part, he doesn’t think such created
substances can “receive” accidents from other created substances. But this shows that Leibniz thinks such
“receiving” is a necessary condition for creaturely transeunt causation. Leibniz writes, “Further, the action
of one substance upon another is not an emission or a transplanting of some entity, as is commonly
supposed; and it can be understood reasonably only in the way just shown. It is true that we can easily
conceive of both the emission and the reception of the parts in matter and can in this way reasonably
explain all the phenomena of physics mechanically. But since material mass is not a substance, it is clear
that the action of substance itself can be only what I have just described.” See (G IV.498-9: L 459). See
also (G IV.432-3: AG 58).
413
196
inhering in s2, then while s3, s4, . . . , sn may have accidents of the same type as A to give
to s2, s2 already has A and so there is no “room” left.
A related analogy is to compare causation to work, a view defended by Ned Hall,
amongst others.416 Jonathan Jacobs presents reasons to hold that the work view of
causation doesn’t allow overdetermination as described above:
On the causal work analogy, if I did the full amount of work required to get
something done, there’s simply no work left for you to do. And if you contribute
work, either I did less than I otherwise would have, or we finished the task more
quickly, or we produced something better than I would have if you hadn’t
contributed your work.417
It’s fair to attribute such an understanding of efficient causation in general to Leibniz.
Recall, as we saw in Chapter 1, that the efficient cause, for Leibniz, is the “active
cause”418 and is the cause which produces.419 Take two created substances s1 and s2
which are hot rods and a third created substance s3 which is a pan of water to be heated.
On the transference understanding of creaturely transeunt causation, if s1 causes a change
in s3, say by heating s3 to 200F, s1 does so by transferring a number of accidents to s3. If
s1 is the sufficient cause of s3’s being 200F, then there’s nothing left for s2 to do. If s2
were also a cause, then, as Jacobs argues, either s1 transmits less, or if s1 transmits the
same amount, then s3 heats up to 200F quicker or s3 heats up to a hotter temperature.
The above arguments give Leibniz reasons to argue that an effect which is
transeuntly caused by a creature could not be causally overdetermined, whereby causally
416
Ned Hall, “Two Concepts of Causation,” in J. Collins ed. Causation and Counterfactuals (Cambridge:
MIT Press, 2004).
417
See Jacobs, “Causal Powers: A Neo-Aristotelian Metaphysic,” 29-30.
418
This is how Leibniz defines the efficient cause in his Table of Definitions, as we saw in Chapter 1.
Leibniz writes, “efficiens est causa activa.” C 472. See also A.VI.2.490.
419
(G V.211: NE 228).
197
overdetermined, I mean independently causally overdetermined. Let “CTEC” stand for
“Creaturely Transeunt Efficient Causation” and “COD” stand for “Causal
Overdetermination”:
(1) If CTEC then Transference.
(2) If Transference then ~COD.
(3) So, if CTEC then ~COD.
With the argument above, Leibniz now has the resources to get the denial of Creaturely
Transeunt Efficient Causation from the Spontaneity Thesis. Suppose that the Spontaneity
Thesis is the case. If the Spontaneity Thesis is true, then for any created substance s1, s1
is the sufficient efficient cause of all of s1’s non-initial natural accidents. So if A inheres
in s1 and A is a non-initial natural accident of s1, then s1 is the sufficient natural efficient
cause of A. But if the Spontaneity Thesis is true, and so s1 is the sufficient natural
efficient cause of all of it’s non-initial naturally inhering accidents, then it cannot also be
the case that some created substance s2, not identical to s1, is also the sufficient efficient
of at least some of s1’s non-initial naturally inhering accidents. For if s2 were to
sufficiently efficiently cause some of s1’s non-initial naturally inhering accidents, then as
that is a case of creaturely transeunt causation and the effect of such causation —given
transference—could not be overdetermined, s1 could not also be the cause of the
accident’s that s2 causes to inhere in s1. That is, if s1 were to also to sufficiently
efficiently cause the accidents that s2 causes, then that would be independent causal
overdetermination, which can’t happen given creaturely transeunt causation, as I argue.
So for any case of the creaturely transeunt causation of an accident, where the cause is a
sufficient cause, that accident cannot also be efficiently caused by the patient substance.
Leibniz now has reasons to argue from Creaturely Transeunt Efficient Causation to the
198
denial of the Spontaneity Thesis. But if Leibniz can argue from creaturely transeunt
causation to the denial of the Spontaneity Thesis, then Leibniz can also argue from the
Spontaneity Thesis to the denial of creaturely transeunt causation via contraposition.
Thus, Leibniz has a way to argue from the Spontaneity Thesis to the denial of creaturely
transeunt causation, with a premise against overdetermination as the link.420
Further, on the Spontaneity Thesis—in which case for any created substance s1, s1
is the sufficient cause of all of its non-initial natural accidents, then s2 could not even be a
partial cause of A’s inhering in s1. Recall the work analogy: If s1 does all the work
needed, then there is no work left for s2 to do. Suppose s1 is a pan of water. If s1
sufficiently causes itself to heat up to 200F, s2—a hot rod—could not also be an cause of
s1’s heating up. Given transference, s2 would heat s1 by transferring heat from itself to s1.
But if s2 were to partially cause s1 to heat up by s2’s transmitting heat from itself to s1,
then either: (i) s2 caused s1 to heat up quicker than it would have in the absence of s2; (ii)
in the absence of s2, s1 would have reached a lower temperature. With respect to (ii), s1
420
It’s worth presenting the argument in a more formal manner. Below is one way the argument could be
developed:
CTEC ~COD. [Premise]
If ~COD, then not (CTEC and ST) [Implication of ~COD]
So, if CTEC then not CTEC and ST. [HS 1 and 2]
Suppose CTEC.
a. So, ~COD. [MP 1 and 4]
b. So, not CTEC and ST. [MP 4a and 2]
c. So, not CTEC or not ST. [DeMorgans 4b]
d. Not not CTEC. [DN 4]
e. So, not ST. [DS 4d and 4c]
(5) So, if CTEC then not ST. [CP 4 and 4e]
(6) So, if ST then not CTEC. [Contraposition 5]
(1)
(2)
(3)
(4)
199
would arguably have a different accident other than being-200F. So I argue that if s1 is a
sufficient cause of its being 200F, s2 could not even be a partial cause.421
Given the above reasoning-- that an immanently and sufficiently produced
accident cannot also be transeuntly caused, Leibniz can also argue from (P) to the denial
of the possibility of creaturely transeunt causation. For if the complete concept/notion of
a created substance is sufficient for all the substance’s future states and so in some sense
is the cause of those states, then a different created substance producing a state in the
patient substance would overdetermine those states. However, a different created
substance producing a state in a patient substance is creaturely transeunt causation, but
such the effects of such causation cannot be overdetermined. So, for very similar
reasons, Leibniz can argue that (P) entails the denial of creaturely transeunt causation,
even though the mature Leibniz thought the antecedent—or at least some construal’s of
(P) in which the complete concept actually efficiently causes a substance’s states—is
false.
421
With respect to (i), there is a potential complication that needs to be addressed. One could object that (i)
counts against s2’s being a partial cause of s1’s being 200F when s1 is a sufficient cause of s1’s being 200F
only if the relatum on the effect side of the causal relation is fine-grained rather than coarse-grained. With
respect to the example I’ve used, a coarse-grained relatum would be s1’s being hot while the fine-grained
relatum would be s1’s being 200F. One could argue that while the same coarse-grained relatum could be
caused by s1 alone or s1 and s2, a different fine-grained relatum would be caused depending on whether s1
alone or s1 and s2 causes s1 to be 200F. Again-- s1 alone would take a longer amount of time. However, all
one needs to do is reject that the relata are fine-grained and then they could reject (i).421
The best way to respond is to argue that for Leibniz, the relata would have to be fine-grained. On
the effect side, the relatum is an accident: the accident A produced. Accidents—such as being 200F,
however, are fine-grained. If the effect was an event, such as the event of s1’s increasing in temperature,
then it could be coarse-grained. But as I’ve argued throughout this project, the effects of efficient causation
in Leibniz’s metaphysics are accidents (when the cause is a created substances) or substances (when the
cause is God). Jonathan Schaffer canvasses the various implications of holding that the relata in causation
are coarse-grained or fine-grained in Jonathan Schaffer, "The Metaphysics of Causation", The Stanford
Encyclopedia of Philosophy (Summer 2014 Edition), Edward N. Zalta (ed.), URL =
<http://plato.stanford.edu/archives/sum2014/entries/causation-metaphysics/>.
200
In summary, I’ve argued that Leibniz can make the move from the Spontaneity
Thesis to the denial of creaturely transeunt causation, and the way in which he can do so
is via a premise denying causal overdetermination. The reason is that Leibniz’s views on
creaturely transeunt causation provide reasons to deny causal overdetermination. The
same reason also gives Leibniz the resources to move from (P) to (C), as he did several
times in his early career.
201
Appendix B. Leibniz on Transubstantiation.
According to the Catholic view of the Eucharist, codified in the Council of Trent (15451563) and developed at length earlier by influential Catholic philosophers such as
Aquinas, during the Eucharistic mass, when the priest recites Christ’s words of the
institution, the bread and wine are converted or transubstantiated into the body and blood
of Christ.422 While the bread and wine are transubstantiated into the body and blood of
Christ, the accidents of the bread and wine continue to exist.423 That the accidents
continue to exist is evident to the senses, for the there is something that looks, feels, and
tastes like bread. 424 However, the accidents no longer inhere in the bread and the wine
because it has been converted into Christ’s body and blood. The accidents do not inhere
in Christ’s body and blood, because of Christ’s impassibility. Further, the accidents do
not inhere in a third type of substance, such as the surrounding atmosphere.
According to Aquinas’s influential account of transubstantiation, the reason the
accidents do not swap substances, say from the bread and wine to the atmosphere is
422
The words of the institution, according to Christian tradition, are “This is my body, which is for you, do
this in remembrance of me” and “This cup is the new covenant in my blood; do this, as often as you drink
it, in remembrance of Me.” See 1 Corinthians 11.
423
According to the Council of Trent, the accidents “which present themselves to the eyes or other senses
exist in a wonderful and ineffable manner without a subject. All the accidents of bread and wine we can
see, but they inhere in no substance, and exist independently of any; for the substance of the bread and wine
is so changed into the body and blood of our Lord that they altogether cease to be the substance of bread
and wine.” Catechism of the Council of Trent for Paris, Priests, trans. John McHugh and Charles Callan
(New York: Joseph Wagner, 1962), 228-9, quoted in Daniel Fouke, “Dynamics and Transubstantiation in
Leibniz’s Systema Theologicum,” Journal of the History of Philosophy 32, No. 1 (1994): 49
424
Jeffrey E. Brower, Aquinas’s Ontology of the Material World: Change, Hylomorphism, and Material
Objects (Oxford: Oxford University Press, 2014), 235-258.
202
because he, like many Catholic philosophers, affirmed a variant of the Ownership Thesis
of Accidents:
Accidents are not transferred from subject to subject, so that numerically one and
the same accident inheres first in one subject and later in another. For an accident
is individuated by its subject. Hence, it is impossible for numerically one and the
same accident to inhere in one subject at one time and in another subject at
another time.425
Instead, the accidents of the bread and wine exist but without inhering in any substance.
Later Scholastics such as Suarez distinguished between two kinds of accidents—
modifications and real qualities—to explain transubstantiation. Real qualities were
accidents that could exist—at least with the miraculous help of God—without inhering in
a substance.426 The accidents of the bread and wine that continued to exist during
transubstantiation were among the real qualities. Modifications, on the other hand, could
not exist even miraculously without a substance. That is, not even God could hold a
modification existence without that modification inhering in a substance.
Three of Leibniz’s metaphysical theses that I have argued in this dissertation are
crucial for his metaphysics of causation are in tension with the Trenton understanding of
transubstantiation— the dependency of accidents on substances, the ownership thesis of
accidents, and the thesis that all accidents are modifications. The dependency of
accidents on substances, recall, is thesis that accidents only exist when inhering in
substance. The ownership thesis of accidents is the thesis that an accident can only ever
inhere in its original substance. The accident cannot swap substances or inhere in two or
more substances at the same time. The reason for both the dependency and ownership
425
426
ST 1 q77 a1
See Suarez, DM 7.
203
theses is that Leibniz believed that all accidents are modifications and all modifications
are limitations, as I argued in Chapter 3.
Given these theses, Leibniz could not consistently affirm the Trenton
understanding of transubstantiation where the accidents of the bread and wine continue to
exist without inhering in a substance. As Leibniz was a Lutheran, one might think that he
did not have to explain transubstantiation or be committed to the Trenton view that
accidents can exist—at least supernaturally—without a substance. However, at several
stages in his career, Leibniz developed accounts of the metaphysics of transubstantiation,
leading to a tension with the dependency and ownership theses and the thesis that all
accidents are modifications.
Early in Leibniz’s career, during the 1660s and 70s, Leibniz in fact affirmed
transubstantiation and argued that the Lutheran view of the Real Presence was the same
as the Catholic Trenton view of transubstantiation. His understanding of
transubstantiation in these earlier writings was importantly different from Scholastic
understandings, but nonetheless, in several works, Leibniz explicitly argued that his own
(earlier) metaphysical views were not only consistent with but explained
transubstantiation.427
For example, in his 1668 (?) De Transsubtantiatione, Leibniz argued that bodies,
such as the bread and wine, were substances only when united with a mind.428 A human
body was a substance only because it is united with a human mind.429 Substances of
427
For a helpful overview of Leibniz’s views on transubstantiation in the 1660s and 70s, see Daniel C.
Fouke, “Metaphysics and the Eucharist in the Early Leibniz,” Studia Leibnitiana (1992): 145-159.
428
L 116.
429
Ibid.
204
bodies that lack reason, such as the bread and wine in the Eucharist, are substances when
united with God. When considered apart from their union with a mind (Divine or
human), bodies are merely accidents or appearances. When the bread and wine are
transubstantiated into the body and blood of Christ, the accidents are united with Christ’s
mind rather than the general concourse, which God’s mind has with all bodies (which are
not united with human minds).430 This account developed by the young Leibniz is in
tension with Leibniz’s Ownership Thesis of Accidents. For the bread and wine on this
account, considered apart from their union with a mind, are accidents that swap
substances. However, Leibniz abandoned this defense of Transubstantiation by 1690.431
More pertinent to my project, however, are accounts of transubstantiation Leibniz
developed in his later writings. Specifically, during his multiple-years spanning
correspondence with Des Bosses, Leibniz expended a great deal of ink describing what
would have to be the case in order for his metaphysics to be compatible with
transubstantiation. If the mature Leibniz genuinely believed in transubstantiation and
such transubstantiation involves accidents existing without a substance or swapping
substances, then this results in a deep tension within his mature metaphysics. As I argued
in chapters 3 and 4, the dependency thesis, ownership thesis, and theses that accidents are
modifications play crucially important roles both in his positive account of causation and
change and his criticisms of competing accounts.
Early on in their correspondence, Des Bosses wanted to know how Leibniz
reconciled transubstantiation with his metaphysics of non-corporeal non-causally
430
431
Ibid.
Fouke, Ibid., 159
205
interacting substances, writing “But principally it would have been helpful to have known
how you would defend by your principles the real presence of Christ in the Eucharist, a
matter which I believe you have discussed to some extent in your anti-Bayle work.”432
Throughout the remaining letters, Leibniz went to great lengths to sketch out what would
have to be the case for the possibility of transubstantiation within his metaphysics by
introducing an entity, the vinculum substantiale. According to Leibniz, bread—being
corporeal—is not a substance but an aggregate of simple substances or monads.433 To
unite the collection of monads so that the aggregate is an unum per se or genuine
substance, however, an additional entity would have to be added— the vinculum
substantiale.434 The substantiality of the bread would then consist of this uniting entity.
During the Eucharist, a different bond, substituted by God, would replace the original
vinculum substantiale—uniting the monads into the bread—.435 While the bond is
substituted, however, the situation would be phenomenally equivalent to the previous
union of monads.436 Thus, the phenomena remains constant, as do the monads
themselves, but the substantiality of the bread is changed given that the substantial bond
has changed.
However, two reasons count against concluding Leibniz meant for his account of
the vinculum substantiale—which he developed to explain transubstantiation—to be a
part of his strict metaphysics. These reasons also point towards the mature Leibniz
denying transubstantiation. First, early in his correspondence with Des Bosses, after
432
G II.388.
G II.399.
434
Ibid.
435
(G II.461: L 607).
436
G II.399.
433
206
initially presenting the notion of a vinculum substantiale in response to Des Bosses’s
question about how Leibniz would explain transubstantiation, Leibniz wrote, “But we
who reject transubstantiation do not need such a thing [vinculum substantiale].437 Hence,
even after introducing the idea of the vinculum substantiale, Leibniz explicitly denies
transubstantiation.
Second, as Brandon Look has shown in his own work on the topic438, the
vinculum substantiale is a (i) a principle of action439 insofar as it is a source of
modifications440; (ii) which unites monads441; but (iii) which can exist independently of
the monads it unites.442 Thus, the vinculum substantiale ought to count as a substance in
Leibniz’s metaphysics.443 The only difference between the vinculum substantiale and
other substance is that the former doesn’t have perceptions.444
What’s especially key here is the role of the vinculum substantiale as unifying
what would otherwise be a mere aggregate of monads into an unum per se. Leibniz is
437
G II.399.
Brandon Look, “Leibniz and the Substance of the Vinculum Substantiale,” Journal of the History of
Philosophy 38, No. 2 (2002): 203-220.
439
Leibniz writes, “This vinculum will be the principle of action of the composite substance.” L 613 (G II
503: L 613).
440
Leibniz argues that the vinculum substantiale is a source of modifications. (G II 503-4).
441
See, for example, (G II.516: AG 202).
442
Leibniz writes, “A vinculum substantiale superadded to the monads is in my opinion something
absolute, such that although it corresponds accurately, in the course of nature, to the affections of the
monads, that is, to their perceptions and appetites, and can therefore be taken to be within the monad in
whose body its body is, it can nevertheless be independent of the monads in a supernatural sense and can be
removed and adapted to other monads while its former monads remain.” (G II.474: L 608).
443
As opposed to just a relation between the monads it unites, which Leibniz rules out. Leibniz writes,
“For orders, or relations which join two monads, are not in one monad or the other, but equally well in both
at the same time, that is, really in neither, but in the mind alone. You will not understand this relation
unless you add a real vinculum, that is, something substantial which is the subject of the predicates and
modifications joining them together.” See (G II.517: AG 203)
444
Look, 220.
438
207
adamant that without the vinculum substantiale, the monads would be a mere aggregate
and bodies (such as the bread and wine) would have merely phenomenal reality:
If that substantial bond [the vinculum substantiale] of monads did not exist, all
bodies together with all of their qualities, would be nothing but well-founded
phenomena, like a rainbow or an image in a mirror, in a word, continual dreams
perfectly in agreement with one another, if a body is a substance, it is a making
real of the phenomena over and above their agreement.445
However, in order to unify the various monads into a genuine substance or unum per se,
it must be capable of exercising causal powers on those monads.446 For the vinculum
substantiale is not a mere relation between the various monads of the bread and wine.
Instead the vinculum substantiale is a bond and a unifying reality that makes the monads,
which would otherwise be an aggregate, into an unum per se.447 Given that Leibniz is
adamant that no created substance (such as the vinculum substantiale uniting the monads
prior to transubstantiation) can causally affect other substances, as we saw in Chapter 2, I
argue that Leibniz should not have posited the vinculum substantiale in his metaphysics.
Given Leibniz’s claim that those who reject transubstantiation (a group he counts
himself in) do not need the vinculum substantiale, I further conclude that Leibniz in fact
did not posit it in his metaphysics. Instead, I argue that Leibniz’s accounts of the
vinculum substantiale in his letters to Des Bosses should be viewed as a conditional with
both an antecedent and consequent that Leibniz in fact rejected: Transubstantiation is
445
(G II.435-6: AG 198-99).
Look, 219-20.
447
Leibniz writes: “. . . either bodies are mere phenomena and so extension will also be only a
phenomenon and the monads alone will be real, the union will be provided by the operation of the
perceiving mind on the phenomena, or, if faith compels us to accept corporeal substances, we must say that
the substance consists in that unifying reality that adds something complete (and therefore substantial),
though in flux, to those things that are to be united.” (G II.435: AG 198).
446
208
explainable in Leibniz’s monadological metaphysics only if there is a vinculum
substantialie.
209
Appendix C. Leibniz’s “De Realitate Accidentium”448
It is worth considering, whether accidents have a reality that is something more than
modal, and in what that [reality] consists. And at least if we posit the accidental reality,
whether their reality is part of the reality of the substance, or if it adds to the substance a
new reality. If it is part of the reality of the substance, it follows that the substance itself
perishes in accidental change, or it becomes a new thing, and myself yesterday exists not
yet, but another although very similar to me, so that the ship which is repaired, or the
republic, or the river, are the same in name, are not really [the same]. For with a part
destroyed, truly the same thing does not remain, even if thus far it is denominated the
same thing by a more important surviving part, otherwise it is able to take place, so that
with all of the parts little by little destroyed, which now belong to, yet it is finally said to
be the same thing, just as the ship of Theseus. If a true part is understood to always
remain, it at least will be the same, whereas the whole arrangement itself will not be.
Therefore, if someone wants permanent part of the reality and a changeable part, they fall
into their opinion, those who prefer to add to the substantial reality something from
accidents. If however it is admitted that the substance perishes and comes into existence
by change (Which is the thought of the duke of Buckingham in the ingenious writing
about true religion the Schediasmate) they in reality remove all changeable substance.
For since the changes of things are perpetual, so that nothing remains in the same state
448
A.VI.iv 994-996.
210
through the smallest intervals of time, it follows that no changeable substance ever exist
and actually endures a minimum time, for any moment whatever it is born and perishes,
neither is it said to properly exist, nor to act, neither is it able to produce anything or to
endure since nothing is brought about unless enduring for some time.449 It follows
therefore that all enduring things are by nature changeable substances, which by reason
we fall into the doctrine of Spinoza and of the Averroists, and the certain long established
tradition, which considers God alone or nature as a substance, creatures otherwise have
no reality other than as a mode of God.450 Truly, nor do they thus avoid [the problem], so
that in this way the changes which created substances undergo (naturally enduring) are
forced to be brought over into God, and thus neither shall God himself endure, but shall
continuously perish and be born. Whence it follows that in the end nothing exists
altogether, for if each thing perishes once, so that from here it follows, nothing will be
because nothing revives it; for out of nothing comes nothing, and nothing is produced
freely from itself. Therefore, it is necessary that something in things persist through
change. But if now a part of the divine reality remains, and a part perishes, we return
back to those who add accidental realities to the substances, and why do they not admit it
in creatures, because now we say [it is] in God, and indeed we relinquish created
substance?
449
Mugnai’s Translation: Given that things change perpetually so that nothing remains in the same state for
even the smallest amount of time, it follows that there is not a changeable substance and that it [i.e. the
substance] does not endure even a minimal amount of time. What in any moment is born and then perishes,
one cannot say it exists in a proper sense, for it does not act or undergo anything, because everything needs
time to exist. (A VI, 4A, 995)
450
Mugnai’s Translation: [. . .] they [Spinoza and the Averroists] consider only God as substances or as
nature, and regard creatures as the modes of God. (A VI, 4A, 995)
211
We now come to those who think that substances have a two-fold reality, one
substantial and the other accidental. These [views] do not themselves also lack their own
difficulties. For it will be able to be asked why those added realities are said to belong to
the substance as it were in a subject, and why it is not considered as a thing per se, even
though not enduring.451 But if that inherence seems to really affect the reality of the
substance, so that it exists somehow in close union by some real it exists, it is not
apparent, how the accidental [reality] is able to perish, without change in the substantial
reality it [the accidental reality] originates from. Therefore, it itself will be divided again
into a perishing and permanent part, contrary to hypothesis.452
Nor thus far do I see another way to avoid these obstacles, how if abstracta are
considered as things, but as shorthands of speaking, as when I call heat, it is not useful in
order that I bring about naming of any wandering subject; or if in order that I say that
something is hot, and so far I am a nominalist, at least through caution.453 I say therefore
that substances change, or at diverse times their attributes are unlike; for this has no
doubt, whether however in change there is something real that perishes and is born; and
whether there are diverse realities in a substance, which are the foundations of diverse
predicates, it is not necessary to ask, and, if asked it is difficult to decide.454 It’s enough
451
Mugnai’s Translation: Why does one believe that the added reality inheres in the substances, as in a
subject, and why does one not consider it to be a thing in itself, even though it does not persist?
452
Mugnai’s Translation: If this inherence, being some kind of real connection, affects the substantial
reality, it is not clear how the accidental reality may perish without causing any change in the substantial
one. The entire substance would, therefore, divide again into perishing and persisting parts, contrary to the
hypothesis.
453
Mugnai: It seems to me that, till now, the only way to avoid these obstacles was to consider the abstract
terms not as [corresponding to] things, but as a kind of shorthand for discourse [. . .] and it is exactly on this
point that I am a nominalist, even though it is only to be precautious.
454
Mugnai: I will therefore say that the substance changes, meaning that, in different times, its attributes
are different, since this is beyond any doubt. It is not necessary, however, to ask whether there is
212
for substances alone to be posited as things, and to say truths about them.455 Geometers
don’t actually use abstract definitions, but they reduce them to the concrete; thus, Euclid
doesn’t use the definition of ratio which he has, but he explains those things in the ratio,
which are said to have equal, greater, or smaller ratios.
something real that perishes and is born, which corresponds to a given change, nor whether there are
different realities in a substance that are the foundations of different predicates. If someone were to pose
these questions, it would be very difficult to answer them.
455
Mugnai: “It suffices alone to consider the [individual] substances as things and to state truths about
them” (A VI, 4A 996)
BIBLIOGRAPHY
213
BIBLIOGRAPHY
Adams, Robert Merrihew. Leibniz: Determinist, Theist, Idealist. New York: Oxford
University Press. 1994.
____________. “Moral Necessity.” In Leibniz: Nature and Freedom, edited by
Rutherford, Donald & J.A. Cover. Oxford: Oxford University Press, 2005. 18388.
Al-Ghazālī'. The Incoherence of the Philosophers. Trans. Michael E. Marmura. Provo:
Brigham Young University Press, 1997.
Aquinas, Thomas. Basic Writings of Saint Thomas Aquinas: Volume One. [ST] Ed. &
Trans. Anton Pegis. Indianapolis: Hackett Publishing Company, Inc., 1945.
Cited by Book, Question, and Article.
___________. Commentary on Aristotle’s Metaphysics. [In Meta] Trans. John P.
Rowan. Notre Dame: Dumb Ox Books, 1995. Cited by book, chapter, and
section.
___________. On the Power of God [QDP] Trans. English Dominican Fathers.
Westminster: The Newman Press, 1952. Cited by question and article.
___________. Aquinas on Matter and Form and the Elements: A Translation of the De
Principiis Naturae and the De Mixtione Elementorum of St. Thomas Aquinas.
[DPN] Trans. Joseph Bobik. Notre Dame: Notre Dame University Press, 1998.
Cited by Chapter and section.
___________. Summa Contra Gentiles. [SCG] Trans. Anton Pegis. Notre Dame: Notre
Dame University Press, 1975. Cited by book, chapter, and section.
___________. Truth. [QDV] Trans. James V. McGlynn, Robert W. Mulligan, & Robert
W. Schmidt. Chicago: Henry Regnery Company, 1954. Cited by book, question,
and article.
Aristotle. The Complete Works of Aristotle, vols I & II. Ed. Jonathan Barnes. Princeton:
Princeton University Press, 1984. Cited by Book, chapter, and Bekker number.
214
Bacon, Bacon. Roger Bacon’s Philosophy of Nature. [RB] Ed. & Trans. D. Lindberg.
Oxford: Oxford University Press, 1983. Cited by page.
Bobro, Marc. “Leibniz on Concurrence and Efficient Causation.” Southern Journal of
Philosophy, 46 (2008): 317–38
Bobro, Marc & Kenneth Clatterbaugh. “Unpacking the Monad, Leibniz's Theory of
Causality.” The Monist, (1996) 79: 409–26
Bodelschwingh, J. Von. “Leibniz on Concurrence, Spontaneity, and Authorship.”
Modern Schoolman, 88(2011): 267–297.
Bolton, Martha Brand. “Change in the Monad,” in The Divine Order, the Human Order,
and the Order of Nature: Historical Perspectives, edited by Eric Watkins.
Oxford: Oxford University Press, 2013. 175-196.
Brower, Jeffrey E. Aquinas’s Ontology of the Material World: Change, Hylomorphism
and Material Objects. Oxford: Oxford University Press, 2014.
Carlin, Lawrence. “Leibniz on Final Causes.” Journal of the History of Philosophy,
44 (2006): 217–33.
Castaneda, Hector-Neri. “Causes, Causity, and Energy.” Midwest Studies in Philosophy
9 (1984): 17-27.
____________. The Causation Debate in Modern Philosophy, 1637-1739. New York:
Routledge, 1999.
Cook, Roy T. “Monads and Mathematics: The Logic of Leibniz’s Mereology.” Studia
Leibnitiana (2000): 1-20.
Cover, J.A. “Non-Basic Time and Reductive Strategies: Leibniz’s Theory of Time.”
Studies in History and Philosophy of Science 28 (1997): 289-318.
Cover, J A., & John O'Leary-Hawthorne. Substance and Individuation in Leibniz.
Cambridge: Cambridge University Press., 1999.
Cox, Donovan. “Leibniz on Divine Causation: Creation, Miracles, and the Continual
Fulgurations.” Studia Leibnitiana, (2002): 185-207.
Descartes, Rene. The Philosophical Writings of Descartes, vols. I & II. [CSM] Ed. &
Trans. John Cottingham, Robert Stoothoff, and Dugald Murdoch. Cambridge:
Cambridge University Press, 1985. Cited by volume and page.
215
____________. The Philosophical Writings of Descartes, Volume III. [CSMK] Ed. &
Trans. John Cottingham, Robert Stoothoff, Dugald Murdoch, and Anthony
Kenny. Cambridge: Cambridge University Press, 1991. Cited by page.
Di Poppa, Francesca. “God Acts from the Laws of His Nature Alone: From the Nihil Ex
Nihilo Axiom to Causation as Expression in Spinoza’s Metaphysics.” PhD Diss.,
University of Pittsburgh, 2006.
Ehring, Douglas. Causation and Persistence: A Theory of Causation. Oxford: Oxford
University Press., 1997.
Fair, David. “Causation and the Flow of Energy.” Erkenntnis 14 (1979): 210-250.
Fouke, Daniel. “Dynamics and Transubstantiation in Leibniz’s Systema Theologicum.”
Journal of the History of Philosophy 32, No. 1 (1994): 45-61.
____________. “Metaphysics and the Eucharist in the Early Leibniz.” Studia Leibnitiana
(1992): 145-159.
Funkhouse, Eric. “Three Varieties of Causal Overdetermination.” Pacific Philosophical
Quarterly 83 (2002): 335- 351.
Futch, Michael. Leibniz’s Metaphysics of Time and Space. Springer, 2008.
Grandy, Richard E. "Sortals." In The Stanford Encyclopedia of Philosophy, edited by
Edward N. Zalta. Spring 2014 Edition. URL =
<http://plato.stanford.edu/archives/spr2014/entries/sortals/>.
Hall, Ned. “Two Concepts of Causation.” In Causation and Counterfactuals, edited by
John Collins, Ned Hall, & L.A. Paul. Cambridge: MIT Press, 2004. 225-276.
Hartz, Glenn A. Leibniz’s Final System: Monads, Matter and Animals. New York:
Routledge, 2007.
Hillman, Timothy Allan. “Leibniz on Monadic Action & Divine Concurrence.” PhD
Diss. Purdue University, 2008.
____________. “Substantial Simplicity in Leibniz: Form, Predication, & Truthmakers.”
The Review of Metaphysics 63 (2009): 91-138.
Ishiguro, H. “Pre-established harmony versus constant conjunction.” Proceedings of the
British Academy, 63(1977): 239–63
Jacobs, Jonathan D. “Causal Powers: A Neo-Aristotelian Metaphysic.” PhD diss.,
Indiana University, 2007.
216
Jolley, Nicholas, ed. The Cambridge Companion to Leibniz. Cambridge: Cambridge
University Press, 1995.
____________. “Causality and Creation in Leibniz.” The Monist. 81(1998): 591–611.
____________. “Leibniz: Truth, Knowledge, and Metaphysics.” In Routledge History of
Philosophy, Volume IV, The Renaissance and Seventeenth-Century Rationalism,
edited by G.H.R. Parkinson. New York: Routledge, 2005. 384-423.
Jorati, Julia. “Leibniz on Causation –Part 1.” Philosophy Compass (2015) 10: 389-397.
Kim, Jaegwon. “Mechanism, Purpose, and Explanatory Confusion.” Philosophical
Perspectives 3(1989): 77-108.
Kulstad, Mark. “Appetition in the Philosophy of Leibniz,” Mathesis rationis: Festschrift
für Heinrich Schepers (1990): 133-151.
____________. “Causation and Preestablished Harmony in the Early Development of
Leibniz's Philosophy.” In Causation in Early Modern Philosophy, edited by
Steven Nadler. University Park: Pennsylvania State Press, 1993. 93–118.
Kulstad, Mark & Laurence Carlin. "Leibniz's Philosophy of Mind." In The Stanford
Encyclopedia of Philosophy, edited by Edward N. Zalta. Winter 2013 Edition.
URL = http://plato.stanford.edu/archives/win2013/entries/leibniz-mind
Lee, Sukjae. “Leibniz on Divine Concurrence.” Philosophical Review. 113(2004): 203–
48.
____________. "Occasionalism." The Stanford Encyclopedia of Philosophy (Spring 2014
Edition). Ed. Edward N. Zalta. URL
=<http://plato.stanford.edu/archives/spr2014/entries/occasionalism/>.
Leibniz, G.W. Die philosophischen Schriften von Gottfried Wilhelm Leibniz [G] Ed. C.I
Gerhardt. Berlin: Weidman, 1875-90. Reprint, Hildesheim: Georg Olms, 1965.
Cited by volume and page.
____________. Philosophical Papers and Letters. [L] 2nd Edition (2nd Printing). Ed. &
Trans. Leroy Loemker. Kluwer Academic Publishers, 1989. Cited by page.
____________. G.W. Leibniz: Philosophical Essays. [AG] Ed. & Trans. Roger Ariew &
Daniel Garber. Indianapolis: Hackett Publishing, 1989. Cited by page.
____________. G.W. Leibniz: Philosophical Texts. [WF] Ed. & Trans. R.S. Woolhouse
& Richard Francks. New York: Oxford University Press, 1998.
217
____________. New Essays on Human Understanding. [NE] Eds. & Trans. Peter
Remnant & Jonathan Bennett. Cambridge University Press, 1996. Cited by page.
____________. Opuscules et fragments inédits de Leibniz. [C] Ed. Louis Couturat.
Paris: Felix Alcan, 1903. Reprint, Hildesheim: Georg Olms, 1966. Cited by page.
____________. Philosophical Writings. [PW] Ed. & Trans. Mary Morris & G.H.R.
Parkinson. London: Dent, 1995. Cited by page.
____________. Sämtliche Schriften und Briefe. [A] Darmstadt and Berlin: Berlin
Academy, 1923-1. Cited by series, volume, and page.
____________. Textes inédits. [Grua] Ed. Gaston Grua. Paris: Presses Universitaires de
France, 1948. Cited by page.
____________. Theodicy. [T] Trans. E.M. Huggard (Yale University Press, 1952.
Reprint, LaSalle, Il: Open Court, 1985). Cited by page.
Locke, John. An Essay Concerning Human Understanding. Oxford: Oxford University
Press, 1975.
Look, Brandon. The Continuum Companion to Leibniz. London: Continuum., 2011.
____________. "Gottfried Wilhelm Leibniz". In The Stanford Encyclopedia of
Philosophy, edited by Edward N. Zalta. Spring 2014 Edition. URL
=http://plato.stanford.edu/archives/spr2014/entries/leibniz/
____________. “Leibniz and the Substance of the Vinculum Substantiale.” Journal of
the History of Philosophy 38 (2000): 203-220.
____________. “Leibniz’s Metaphysics and Metametaphysics: Idealism, Realism, and
the Nature of Substance.” Philosophy Compass 5, No. 11 (2010): 871-79.
Malebranche, Nicolas. The Search after Truth: With Elucidations of The Search after
Truth. [LO] Ed. Thomas M. Lennon & Paul J. Oscamp. Cambridge: Cambridge
University Press, 1997. Cited by page.
__________. Malebranche: Philosophical Selections, edited & translated by Steven
Nadler. Indianapolis: Hackett Publishing Company, Inc, 1992.
Mates, Benson. “Individuals and Modality in the Philosophy of Leibniz.” Studia
Leibnitiana. 1972: 81-118.
____________. The Philosophy of Leibniz: Metaphysics and Language. New York:
Oxford University Press., 1986.
218
McDonough, Jeffrey K. “Leibniz’s Two Realms Revisited.” Nous, 42 (2008): 673-696.
Menn, Stephen. “The Greatest Stumbling Block: Descartes’ Denial of Real Qualities,” in
Descartes and His Contemporaries: Meditations, Objections, and Replies, edited
by Roger Ariew & Marjorie Grene. Chicago: The University of Chicago Press,
1995. 182-207.
Mercer, Christia. Leibniz’s Metaphysics: It’s Origins and Development. Cambridge:
Cambridge University Press, 2001.
Mugnai, Massimo. “Leibniz on Substance and Changing Properties.” Dialectica 59, No.
4 (2005): 503-516.
Nadler, Steven, ed. Causation in Early Modern Philosophy. University Park: The
Pennsylvania State University Press, 1993.
Newlands, Samuel. “Leibniz on Privations, Limitations, and the Metaphysics of Evil.”
Journal of the History of Philosophy 52, No. 2 (2014): 281-308.
O’Neill, Eileen. “Influxus Physicus.” In Causation in Early Modern Philosophy, edited
by Steven Nadler. University Park: Penn State Press, 1993. 27-56.
Ott, Walter R. “Causation, intentionality, and the case for occasionalism.” Archiv für
Geschichte der Philosophie 90, No. 2 (2008): 165-87.
____________. Causation and Laws of Nature in Early Modern Philosophy. Oxford:
Oxford University Press, 2009
Pasnau, Robert. Aquinas on Human Nature. Cambridge: Cambridge University Press,
2002.
____________. “Form, Substance, and Mechanism.” Philosophical Review 113, no. 1
(2004): 31-88.
____________. Metaphysical Themes: 1274-1671. Oxford: Oxford University Press,
2011.
Phemester, Pauline. Leibniz and the Natural World: Activity, Passivity and Corporeal
Substances in Leibniz’s Philosophy. Dordrecht: Springer, 2005.
Pinkston, Brian Russell. “Francisco Suarez on Efficient Causation.” PhD Diss.,
University of Virginia, 2012.
219
Richardson, Kara. “Avicenna and Aquinas on Form and Generation.” In The Arabic,
Hebrew and Latin Reception of Avicenna's Metaphysics, edited by Dag Nikolaus
Hasse & Amos Bertolacci. Boston: Walter de Gruyter GmbH & Co., 2011. 25174.
Rieber, S.D. “Causation as Property Acquisition.” Philosophical Studies: An
International Journal for Philosophy in the Analytic Tradition 109, No. 1 (2002):
53-74
Rota, Michael. “Causation.” In The Oxford Handbook of Aquinas, edited by Brian
Davies & Eleonore Stump. Oxford: Oxford University Press, 2012. 104-114.
Rutherford, Donald. “Laws and Powers in Leibniz.” In The Divine Order, the Human
Order, and the Order of Nature: Historical Perspectives, edited by Eric Watkins.
Oxford: Oxford University Press, 2013. 149-174.
____________. Leibniz and the Rational Order of Nature. Cambridge: Cambridge
University Press., 1998.
____________. “Leibniz on Spontaneity.” In Leibniz: Nature and Freedom, edited by
Donald Rutherford & Jan A. Cover. Oxford: Oxford University Press, 2005. 156180.
____________. “Leibniz’s Principle of Intelligibility.” History of Philosophy Quarterly
9, no. 1 (1992): 35-49.
Salmon, Wesley. “Causality Without Counterfactuals.” Philosophy of Science 61 (1994):
297-312.
Schaffer, Jonathan. "The Metaphysics of Causation." In The Stanford Encyclopedia of
Philosophy, edited by Edward N. Zalta. Summer 2014. URL =
<http://plato.stanford.edu/archives/sum2014/entries/causationmetaphysics/>.
Schmid, Stephen. “Efficient Causality: The Metaphysics of Production.” In Suárez on
Aristotelian Causality, edited by Jakob Leth Fink. Leiden: Brill, 2015. 84-120.
Schmaltz, Tad. M. Descartes On Causation. Oxford: Oxford University Press, 2008.
Sider, Theodore. “What’s So Bad About Overdetermination?” Philosophy and
Phenomenological Research 67 (2003): 719-726.
Sleigh, R C. Leibniz and Arnauld: A Commentary On Their Correspondence. New
Haven: Yale University Press.,1990.
220
____________. “Leibniz on Malebranche on Causality.” In Central Themes in Early
Modern Philosophy: Essays Presented to Jonathan Bennett, edited by J.A. Cover
& Mark Kulstad. Indianapolis: Hackett, 1990. 161-194.
____________. “Moral Necessity in Leibniz’s Account of Human Freedom.” In
Metaphysics and the Good: Themes from the Philosophy of Robert Merrihew
Adams, edited by Samuel Newlands & Larry M. Jorgensen. Oxford: Oxford
University Press, 2009.
252-71.
Spinoza, Baruch. Ethics. [E] Trans. Shirley Samuel. Indianapolis: Hackett Publishing
Company, 1992. Cited by part.
Solomon, Graham. “Leibniz and Topological Equivalence.” Dialogue 32 (1993): 72124.
Suarez, Francisco. On Efficient Causality: Metaphysical Disputations 17, 18, and 19.
[DM] Ed. & Trans. Alfred J. Freddoso. New Haven: Yale University Press, 1994.
Cited by disputation, chapter, and section.
__________. On Creation, Conservation, and Concurrence: Metaphysical Disputations
20-22. [DM] Ed. & Trans. Alfred J. Freddoso. South Bend: St. Augustine’s
Press, 2009. Cited by disputation, chapter, and section.
____________. On Various Kinds of Distinctions. [DM] Trans. Cyril Vollert, SJ.
Marquette: Marquette University Press, 1947. Cited by Disputation, chapter, and
section.
Tuozzo, Thomas M. “Aristotle and the Discovery of Efficient Causation.” in Efficient
Causation: A History, edited by Tad M. Schmaltz. Oxford: Oxford University
Press, 2014. 23-47.
Whipple, John. “Continual Creation and Finite Substance in Leibniz’s Metaphysics.”
Journal of Philosophical Research 36(2011): 1-30.
____________. "Leibniz's Exoteric Philosophy." In The Stanford Encyclopedia of
Philosophy, edited by Edward N. Zalta. Summer 2013 Edition. URL =
<http://plato.stanford.edu/archives/sum2013/entries/leibniz-exoteric/>.
____________. “Leibniz on Divine Concurrence.” Philosophy Compass 10, no. 5
(2010): 865- 879.
____________. “The Structure of Leibnizian Simple Substances.” British Journal for
the History of Philosophy 18(2010): 379-410.
221
Wippel, John. The Metaphysical Thought of Aquinas. Washington: The Catholic
University of America Press, 2000.
Varzi, Achile. "Boundary." In The Stanford Encyclopedia of Philosophy, edited by
Edward N. Zalta. Winter 2013 Edition. URL =
<http://plato.stanford.edu/archives/win2013/entries/boundary/>.
VITA
222
VITA
Davis Kuykendall received his BA in philosophy (summa cum laude with departmental
honors) from the University of North Carolina at Charlotte in 2009. As a PhD student in
philosophy at Purdue, he was awarded the Ross Fellowship and a Purdue Research
Foundation Dissertation fellowship. His main interests are early modern philosophy and
metaphysics.