Duns Scotto

John Duns Scotus on Parts, Wholes, and Hylomorphism
Investigating Medieval Philosophy
Managing Editor
John Marenbon
Editorial Board
Margaret Cameron
Simo Knuuttila
Martin Lenz
Christopher J. Martin
VOLUME 7
The titles published in this series are listed at brill.com/imp

John Duns Scotus on Parts,
Wholes, and Hylomorphism
By
Thomas M. Ward
LEIDEN | BOSTON
Library of Congress Cataloging-in-Publication Data
Ward, Thomas M.
John Duns Scotus on parts, wholes, and hylomorphism / by Thomas M. Ward.
pages cm. -- (Investigating medieval philosophy, ISSN 1879-9787 ; VOLUME 7)
Includes bibliographical references and index.
ISBN 978-90-04-27831-8 (hardback : alk. paper) -- ISBN 978-90-04-27897-4 (e-book) 1. Duns Scotus, John,
approximately 1266-1308. 2. Hylomorphism. I. Title.
B765.D64W37 2014
111’.1092--dc23
2014018273
This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering
Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities.
For more information, please see www.brill.com/brill-typeface.
issn 1879-9787
isbn 978-90-04-27831-8 (hardback)
isbn 978-90-04-27897-4 (e-book)
Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands.

Koninklijke Brill nv incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided
that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive,
Suite 910, Danvers, ma 01923, usa.
Fees are subject to change.
This book is printed on acid-free paper.

For Katie
∵
It all seemed too good to be true. Hither and thither through the meadows
he rambled busily, along the hedgerows, across the copses, finding every-
where birds building, flowers budding, leaves thrusting—everything happy,
and progressive, and occupied.
~Kenneth Grahame, The Wind in the Willows
…
Contents
Acknowledgements ix
Abbreviations x
Introduction 1
1 The Purpose of Prime Matter 6

i Distinguishing Matter from Form 6
ii Motivating Matter: The Argument from Change 8
iii Why Must Matter Persist through Change? 12
iv Matter as Passive Power 18
v Obediential Potency and the Subject of Passive Power 23
2 The Ontology of Prime Matter 27

i Matter as Subjective Potency 27
ii A Shifting Opinion about Matter’s Separability from Form 30
iii Particular and Total Separability 34
3 How Matter and Form Compose a Substance—Part I 41

i Forms as Parts 43
ii Matter and Form as Essential Parts 47
iii Degrees of Unity 48
iv Existence is not Enough 51
v Making a Difference 52
4 How Matter and Form Compose a Substance—Part II 60

i Producing Substance from Matter and Form 60
ii Causal and Co-Causal Relations 62
iii Innovating Aristotle’s Principle about Relational Change 66
iv The Identification of Parthood Relations with Causal Relations 68
v The Causality of Matter and Form 70
vi Dispensing with Total Separability? 72
vii Conclusions: Composing, and Composing 73
5 Scotistic Pluralism about Substantial Form—Part I 76

i Unitarianism and Pluralism about Substantial Form 78
ii Scotus against Unitarianism 81
iii Scotus against Standard Pluralism 84
iv An Inconsistent Position about the Form of Corporeity? 90
viii
6 Scotistic Pluralism about Substantial Form—Part II 94
i The Special Potency Question 97
ii Essential Orders 98
iii Essentially Ordered Unity 103
iv The Role of Soul in Scotus and Two Unitarians 106
7 Contingent Supposits and Contingent Substances 110
i Three Modes of per se Being 112
ii Ockham on the Distinction between Substance and Supposit 115
iii What’s Special about Supposits? 118
iv Arbitrary Part-Substances? 121
8 The Mereological Status of the Elements in a Mixture 125
i Mixed Opinions about Mixtures 127
ii The Argument from Quantitative Forms 132
iii The Generation and Corruption Argument 137
iv The Violence Argument 139
v Generation from the Elements 140
vi Mixtures and Organic Parts 142
9 Why the World is not a Substance 145
i Motivating Monism 148
ii The Argument from the Distinguishing of Forms 151
iii The World/Organism Analogy 158
iv Properties of the Whole that are not Properties of the Parts 161
10 Scotistic Hylomorphism and the Problem of Homonymy 165
i Scotistic Hylomorphism Among Other Varieties 165
ii Ackrill’s Problem 169
iii Aquinas’s Response 171
iv The Standard Pluralistic Response 173
v Scotus’s Response 177
vi Scotus, Aquinas, and the Ultimate Subject of Substantial Change 178
vii Faulty Metaphysics or Faulty Chemistry? Scotistic Hylomorphism

and the Four Elements 179
Bibliography 183
General Index 189
Acknowledgements
I would like to thank the midwives of this project, Calvin Normore and Marilyn
McCord Adams. Normore supervised my dissertation at ucla, on which this
book is based; Adams supervised my M.Phil thesis at Oxford and then offered
supererogatory support as a member of my dissertation committee. They are
primarily responsible for forming me into the kind of philosopher I have
become—deficiencies in the product are the fault of the material rather than
the efficient causes! Thanks also go to the other members of my dissertation
committee, John Carriero, Brian Copenhaver, and Debora Shuger. Richard
Cross deserves special mention, both for his personal encouragement of some
aspects of the project and for his scholarly work on Scotus, which has been
indispensible for developing my own ideas about Scotus, as all readers of this
book could easily discover for themselves. I have had helpful feedback from
Joshua Blander, Sean Kelsey, Gavin Lawrence, jt Paasch, Robert Pasnau, Martin
Tweedale, Scott Williams, and from my colleagues here at lmu, in particular
Jim Hanink, Christopher Kaczor, and Eric Perl. John Marenbon invited me to
consider revising my doctoral thesis for publication in Brill’s Investigating
Medieval Philosophy Series, of which he is the general editor; I thank him for
suggesting the idea and for his uplifting feedback. I am grateful to audiences
at the Metaphysics: Aristotelian, Scholastic, Analytic conference in Prague, the
Ontology of Relations: the Medieval Contribution workshop in Lausanne, the
Unum, Verum, Bonum conference in Lisbon, the University of St. Thomas in
St. Paul, mn, and in my own department, for feedback on different portions of
the book. Thanks also go to Terence Sweeney for help with the index. For dif-
ferent reasons I am grateful to The Johns Hopkins University Press for permis-
sion to use my “Animals, Animal Parts, and Hylomorphism: John Duns Scotus’s
Pluralism about Substantial Form,” Journal of the History of Philosophy 50:4
(2012), pp. 531–558, which is almost totally reprinted here as Chapters 5 and 6.
For the privilege of having philosophy as my labor I am grateful to the Mustard
Seed Foundation, the ucla Department of Philosophy, the ucla Graduate
Division, and the lmu Bellarmine College of Liberal Arts, which have funded
the research for this project at different times over the years. I am grateful to the
librarians of the Pontifical Gregorian University in Rome for permission to use
the Library for last-minute corrections to the proofs. Finally, and closer to home,
I want to thank my grandparents and parents for their support, patience and
constant encouragement; and Katie, Edith, and Sophia, for love and distraction.
T. M. Ward
Los Angeles
June, 2013
Abbreviations
Reportatio Duns Scotus, Reportatio Parisiensis

QMet Duns Scotus, Quaestiones super Libros Metaphysicorum Aristotelis
OTh William Ockham, Opera Theologica
OPh William Ockham, Opera Philosophica
Vatican Duns Scotus, Opera Omnia, published by Typis Vaticanis
Wadding Duns Scotus, Opera Omnia, edited by Luke Wadding
Bonaventure n Volume n of Duns Scotus, Opera Philosophica, published by
The Franciscan Institute, St. Bonaventure University
Barnes n Volume n of Aristotle, Complete Works, edited by Jonathan Barnes
Wolter Duns Scotus, De Primo Principio, edited and translated by Allan
B. Wolter, O.F.M.
Introduction
Let me introduce you to Mr. Mole, or Mole, for short. He is an animal, an

individual member of some species of talpid. He is therefore a material object.
He has material parts: paws, heart, liver, eyes, fur, snout, and so on. According
to Duns Scotus, if you add all these material parts together you have not tallied
the sum of Mole’s parts. In addition to these he has another part, a soul, and
this part has a special role to play. It endows all Mole’s material parts with a
certain kind of unity—substantial unity—that they would not have were it not
for that soul. In short, Mole’s soul makes the difference between there being,
on the one hand, some talpid parts and, on the other, the living, breathing
Mole. Call all of Mole’s material parts his matter, or hyle, and his soul his form,
or morphé. Mole is therefore a hylomorphic composite, a composite of matter
and form.
Medieval philosophers would have classified hylomorphism as part of a
theory of physics, since physics, according to them, is the discipline which
studies things undergoing change and only things undergoing change are hylo
morphic composites. Nowadays we would classify hylomorphism as a meta
physical theory. What exactly we would mean by calling it a metaphysical
theory is not clear, because we contemporary folks do not have a consensus
about what metaphysics is. Part of what we would mean, however, is that hylo
morphism, if true, is a theory about material objects that could be neither con
firmed nor denied by any empirical science. We could say that hylomorphism
is consistent with, and even presupposed by, all or some of the empirical sci
ences; but we would not hold our breath for any empirical science to deliver a
final verdict on its truth. That comes, if at all, through a different sort of inves
tigation, one which takes as data the deliverances of commonsense as reason
ably chastened by scientific knowledge but which does not restrict the range
of what is knowable to what empirical science can discover and which makes
free use of the techniques of a priori investigation: logic, and analysis or
distinction-making.
Anyone who is sympathetic to the ideas that (a) at least some complex,
structured wholes—paradigmatically living things—cannot be reduced to
their simpler material parts, and that (b) structures or organizations or pat
terns are or might be real features of the sensible world but themselves are not
or would not be physical objects, should be sympathetic to hylomorphism.
Thus, while this is a book about John Duns Scotus’s hylomorphism and will
therefore be of interest primarily to scholars of the Middle Ages and in particu
lar to scholars of medieval philosophy, the topic should have (relatively)
© koninklijke brill nv, leiden, 2014 | doi 10.1163/9789004278974_002

2 Introduction
broader appeal. Hylomorphism’s history and Duns Scotus’s place in it should

matter to those who are interested in hylomorphism for purely philosophical
reasons. During the Middle Ages hylomorphism was debated at a high level of
refinement. The Aristotelian consensus among medieval intellectuals from the
thirteenth century onward fostered a focused and collaborative “research proj
ect” on various problems arising from Aristotle’s hylomorphism, and none con
tributed to the debates with as much sophistication as Duns Scotus. So we
should be optimistic that medieval hylomorphisms in general and Scotus’s in
particular can provide insights to contemporary debates.
I remain optimistic about hylomorphisms, but in this book I do not defend
Scotus’s version—or any version—of hylomorphism. I have for the most part
shied from endorsing metaphysical theses in this book, Scotistic or otherwise,
because on the philosophical topics the book addresses I simply have not set
tled on a position of my own. It is enough to let Scotus do the talking. Here I am
content, primarily, to develop interpretations that make some sense out of dif
ficult but suggestive texts, and, secondarily, to suggest some ways in which
Scotus’s philosophical achievements (and failures) can be useful to us in our
own philosophical endeavors. That said, the final chapter is a modest endorse
ment of one aspect of Scotus’s hylomorphism as compared with other versions
of hylomorphism. If hylomorphism is true, then I would think a roughly
Scotistic version is true.
Some of Scotus’s arguments and assumptions are bound to seem quite
strange to readers not already immersed in medieval metaphysics, physics,
chemistry, cosmology, and theology. I have tried my best simultaneously to do
justice to Scotus, by not bowdlerizing the bits that we now find weird or clearly
wrong, and to be considerate to readers, by motivating Scotus’s arguments and
assumptions as well as I can. I do not take for granted that all readers will be
interested in, e.g., medieval theories of mixture for their own sakes; but since
we’ve got to know something about theories of mixture in order to understand
Scotus’s version of hylomorphism, I have tried to keep this and similar curiosi
ties as philosophically focused as possible.
Scotus’s version of hylomorphism is hylomorphism from the ground up. By
this I mean to emphasize Scotus’s consistent doctrine that parts are prior to
their wholes: essential parts—prime matter and substantial form—are prior
to and can exist independent of material substances; integral parts like organs
are prior to substances like organisms; elements are prior to the substances
mixed from them; and things in the world are prior to the world they compose.
A ground up approach to parts and wholes theorizing requires some principle
of unity for each kind of composite whole, composed of things capable (if only
by divine power) of existing independent of the whole of which they are parts.
Introduction 3
Somewhat surprisingly, when Scotus seeks to offer principles that explain why
some things do or do not, can or cannot, compose another thing, the meta
physical apparatus he most often reaches for is not form but essential order. By
no means does Scotus give up on form, and he still thinks that form ultimately
explains how the parts of a substance compose one substance. But Scotus
turns to essential order to explain how matter and form themselves unite to
compose substance, how several substances can be together in potency to
receive a substantial form, why the elements are not able to exist in a mixture,
and in what sense the world itself is a cosmos. Most chapters of the book,
therefore, draw on Scotus’s understanding of essential order.
The first two chapters consider Scotus’s understanding of prime matter. The
first asks what theoretical role or function prime matter performs in Scotus’s
natural philosophy. Unsurprisingly, its role is broadly Aristotelian, and when
arguing that prime matter is something real and really distinct from form,
Scotus explicitly acknowledges the Philosopher’s influence. None of this is
worth writing a chapter about. What is, however, is the way in which Scotus
subtly distinguishes two theoretical roles for prime matter to play, which many
would be tempted to conflate. On the one hand, Scotus argues, matter is a pas
sive power in virtue of which a substance can undergo change. On the other
hand, matter is a subject that persists through change. For Scotus, to show that
matter is one is not thereby to show that it is the other. I examine Scotus’s rea
soning for both claims.
The second chapter considers the development of Scotus’s understanding
of prime matter’s ontology. It is well known that Scotus holds that prime mat
ter is, at least by divine power, totally separable from form, such that it could
exist without being informed by any form. Less well known, however, is that
Scotus did not always think this and at one time denied it. I trace the develop
ment of Scotus’s mature view, and closely examine a few of Scotus’s arguments
for total separability, defending a couple against recent criticism.
The third and fourth chapters consider Scotus’s understanding of how mat
ter and form compose a composite substance. Scotus assumes that there is
some difference in things between matter and form’s merely existing and mat
ter and form’s existing and composing a substance. So Scotus develops an
account of what happens when matter and form do compose a substance. He
argues that some change must occur for matter and form to compose a sub
stance, that the only sort of change that fits the bill is the production of a new
entity, and goes on to argue that this new entity is and can only be a substance.
Scotus therefore holds that the real distinction of a whole (substance) from its
(essential) parts is necessary for an adequate account of how matter and form
compose.
4 Introduction
The fifth and sixth chapters first appeared together as “Animals, Animal
Parts, and Hylomorphism: John Duns Scotus’s Pluralism about Substantial
Form,” Journal of the History of Philosophy 50:4 (2012), pp.531–558. They are
the first philosophical exploration of a little-known feature of Scotus’s
hylomorphism: Scotus held that the extended parts of some material sub
stance, paradigmatically the organs of organisms, are themselves substances—
part-substances, as I will call them—with their own specific substantial forms.
Most scholars assume that Scotus thought that the extended parts of a body
are unified by a form of corporeity, and that the compound of material parts
with the form of corporeity constitutes a subject in potency to some higher-
order form, such as a soul. I argue against this opinion, showing that Scotus
thinks that the form of corporeity is nothing over and above the forms of the
plurality of extended parts. I develop an alternate account according to which
a plurality of substances potentially compose a substance when they are essen
tially ordered to one another in the orders of final- and efficient-causality.
Chapter 7 explores some implications of the view of I attribute to Scotus in
the fifth and sixth chapters. Scotus distinguishes between substances and sup-
posits and says that a supposit is a substance that (a) is not a part of another
substance and (b) belongs to a complete determinate natural kind. I argue that
Scotus’s doctrine of part-substances, together with some other philosophical
and theological commitments, entails that any created supposit is only contin
gently a supposit since any created supposit could become a part-substance
and thereby cease to be a supposit. More radically, Scotus thinks that at least
some substances are only contingently substances in the non-trivial sense that
they can cease to be substances without ceasing to exist.
Chapter 8 considers Scotus’s reasons for denying that the four chemical ele
ments actually exist in the bodies of which they are the basic ingredients, in
light of the fact that he admits in principle that substances can compose sub
stances. The chemistry of Scotus’s day identified the basic elements of physical
reality as four different kinds of material substance: earth, air, fire, and water;
medieval thinkers described material substances as mixtures of these ele
ments. Scotus and most medieval philosophers in the thirteenth century and
beyond were agreed that the elements cease to exist when they are mixed.
Since Scotus holds in general that substances can be parts of more complex
substances, it is initially surprising that he holds with the majority view about
the ontological status of the elements in mixtures. I argue that Scotus’s funda
mental reason for holding the majority view is that elemental bodies cannot
compose a unity of order: they are neither efficient- and final-causally related
to one another, since elements act for the sake of reaching their proper sublu
nary region, and they naturally corrupt elemental bodies of other kinds if they
Introduction 5
are able. Any combination of elemental bodies therefore lacks the unity requi
site for a plurality of substances potentially to compose a complex substance.
Whereas the eighth chapter considers Scotus’s reasons for thinking that the
composition of substances by substances does not extend all the way down to
the four elements, Chapter 9 considers his reasons for denying that such com
position extends all the way up, to the world as a whole. I argue that, for Scotus,
there is no significant metaphysical difference between the sort of substance
that naturally composes a substance, such as a heart, and the sort of substance
that does not, such as Mole. Or, to be more precise, there is no difference
between these sorts of substances that is relevant to questions about composi
tion. This means that there is no logical or metaphysical impossibility that all
the substances in the world should in fact compose a substance, or that several
substances which do not ordinarily compose a substance should in fact com
pose a substance. Moreover, Scotus thinks that every substance in the world is
essentially ordered to every other substance. In spite of this, Scotus denies
that there is one world-substance. This chapter examines Scotus’s reasons for
this denial.
The final chapter applies the results of the preceding chapters to show how
Scotus would respond to a problem with the application of hylomorphism to
living things. First, hylomorphism seems to entail that matter is contingently
formed by its form. Second, the principle of homonymy, endorsed by Aristotle
and Aquinas, holds that some substance, s, is a member of a kind, K, if and only
if it is able to perform the characteristic function or functions of the Ks. By the
first thesis, Socrates’s corpse is numerically identical with Socrates’s body,
since Aristotle identifies Socrates’s body as the matter of Socrates. By the sec
ond thesis, Socrates’s corpse is not numerically identical with Socrates’s body,
since the corpse cannot perform the characteristic functions of a human body.
This contradiction has been called “the fundamental problem about hylomor
phism.” I argue that Scotus’s pluralism about substantial forms entails rejec
tion of the homonymy principle, since at least some organic parts, such as
hands, eyes, and kidneys, are distinct kinds of substances, which depend on an
organism only for their natural operations and not for their existence. Therefore
Scotus’s hylomorphism solves the fundamental problem.
Chapter 1
The Purpose of Prime Matter
i Distinguishing Matter from Form
Experience acquaints us with bodily things in wide variety: earth, air, fire,
and water; rocks, flowers, and trees; worms, birds, and lizards; cats and
mice, lions and lambs; sun, moon, and stars; and, of course, human
beings. Two things cry out for explanation: difference and change.1
For hylomorphists, that in reality which corresponds to our experience of the

difference of things is form. The form of the lion is different in kind from the
form of the lamb, and the presence in matter of lion form is that which explains
why this chunk of matter exhibits all of the essential characteristics and char-
acteristic activities of a lion, while the presence in matter of lamb form is that
which explains why that chunk of matter exhibits all of the essential character-
istics and characteristic activities of a lamb. Remove the form, remove that
feature that makes this chunk different in kind from that chunk.
That in reality which corresponds to our experience of change is, for the
hylomorphist, matter. A lion comes into existence and goes out of existence.
Under normal conditions it does not simply pop into existence from nothing;
it comes into existence from something else, and when it goes out of existence
something else comes to be from it. Matter is simply that which is capable of
receiving different forms. The matter, at one time under sperm form and egg
form, becomes, when nature follows its course, matter under lion form (or lion
embryo form).
Most hylomorphists recognize two related senses of matter, however. In the
first sense, matter, or proximate matter, is simply whatever fits as an answer to
the question, “What is it made of?” The table is made of wood and is under the
form of table; the statue is made of bronze and is under the form of statue; the
lion is made of organic parts and is under the form of lion. In contemporary
terminology, matter is simply that which constitutes a material object, while a
form (of some kind, K) is that which structures matter in such a way that what
the matter constitutes is a K.
1 Marilyn McCord Adams, Some Later Medieval Theories of the Eucharist: Thomas Aquinas,
Giles of Rome, Duns Scotus, and William Ockham (Oxford: Oxford University Press, 2010), p. 4.

The Purpose Of Prime Matter 7
In the second sense, matter, or prime matter, is an unobservable theoretical

entity that partially constitutes every material object but itself is not consti-
tuted by anything. It is the last possible answer to the question, “What is it
made of?” The statue is made of bronze; the bronze is made of copper and tin
(or arsenic); these are made of some mixture of elements (earth, air, fire,
water); the elements are made of prime matter, and there is nothing of which
prime matter is made. By its nature it is not composed of form, though it can
compose with form every possible kind of material object. The intuitive basis
for positing a fundamental level of matter, something that is itself not made of
anything but which can become anything through reception of forms, is the
same that undergirds cosmological arguments for an unmoved mover; it is the
apparent need to avoid an infinite regress of ingredient entities. The answers to
“What is it made of?” questions must stop somewhere, the thinking goes, and
medieval hylomorphists all agreed that these questions, for any material sub-
stance, all stop at the same answer: prime matter.2 Precisely because any proxi-
mate matter is itself a composite of at least one form and (proximate or prime)
matter, it is numbered among the things that come into existence from some-
thing else and from whose going out of existence something else comes to be.
And precisely because prime matter is not so composed, it is not numbered
among these; God alone can produce prime matter.
Scotus makes free use of both senses of matter, part and parcel as they were
of his philosophical milieu. This chapter analyzes Scotus’s argument for the
existence of prime matter and its persistence through substantial change. To
identify Scotus’s motivation for positing this theoretical entity is a fortiori to
identify his motivation for hylomorphism: fundamentally, Scotus thinks, we
analyze material substances as composites of form and matter in order to
explain how certain kinds of change are possible. After analyzing Scotus’s
argument, I move on to consider in some detail Scotus’s account of how matter
performs its function in change. This account involves, first, Scotus’s character-
ization of matter as a passive power for receiving forms; and second, Scotus’s
argument that matter persists through change, and that the persistence of mat-
ter through change is somehow necessary for there to be any change at all. It is
tempting to think that matter’s being a passive power simply amounts to its
ability to persist through change, but I argue that for Scotus it does not amount
to this. Instead, Scotus’s argument that matter is a passive power proceeds
2 This answer is consistent with the view that prime matter is itself divisible or even that it has
actual parts. The point is that prime matter (if divisible) is not divisible into anything that is
not prime matter, and that prime matter (if composed of parts) is not composed of anything
that is not itself prime matter.
8 Chapter 1
independent of assumptions about matter’s persistence, and his argument for

matter’s persistence through change involves drawing out a necessary conse-
quence of matter’s role as a passive power for receiving forms.
I also raise and attempt to provide a solution to a problem about the relation
between prime and proximate matter. The problem concerns the role of forms
in substantial and accidental change. On the one hand, matter is potency and
form is act. A form strives to conserve its existence in matter whereas matter is
receptive to new forms; change can therefore be characterized as the success
of an efficient cause in driving out one form in matter and replacing it with a
new. From this perspective, therefore, it is tempting to think that any proxi-
mate matter is in potency to form only because it has prime matter as a con-
stituent. On the other hand, the forms under which some prime matter
currently exists determine to some extent the range of additional or alterna-
tive forms to which it is in potency. For example, an embryo is in potency to the
soul when its organs are suitably developed and when it is properly disposed
(i.e., when it is neither too hot nor too cold). From this perspective, therefore,
it is tempting to think that the whole of proximate matter—prime matter and
form(s)—is in potency to form. I argue for the first hand, viz., that the passive
potency of proximate matter is due to its prime matter, by considering Scotus’s
notion of obediential potency, which is the sort of potency that prime matter
has to be immediately informed by any form through the action of a non-
natural agent such as God.
ii Motivating Matter: The Argument from Change
What is prime matter supposed to do? What philosophical problem does it

solve? Unlike a conviction that there are things like cows and oak trees, which
we might call a pretheoretical conviction, the conviction that there is prime
matter is a theoretical conviction—it is posited as part of a philosophical the-
ory, and in particular a theory that tries to accommodate commonsense con-
victions about how things like cows and oak trees come to be and pass away.
Cows and oak trees have beginnings and endings, but they seem to come from
other things and when they pass away other things seem to come to be from
them. For simplicity’s sake it is tempting to move in one of two revisionist
directions. On the one hand, one might deny that there really are things like
cows and oak trees by denying that anything really comes to be or passes away:
there is just one thing (Parmenides), or perhaps myriad atoms in the void
(Democritus). On the other hand, one might deny that things like cows
and oak trees really do come from other things: perhaps God zaps them into
existence at the moment it looks as if they come from another (al-Ghāzāli,

Malebranche). Hylomorphism is a philosophical attempt to avoid the revisions
and preserve the phenomena.
In Lectura. II, d. 12, q. un, Scotus offers an Aristotelian argument for the exis-
tence of prime matter based on certain convictions about the nature of
change.3 The argument attempts to show that without some “patient on which
an agent acts” there could be no substantial change:
[Aristotle’s] argument is formed in this way:
[A1] every natural agent requires a patient on which it acts (this is clear
to the senses);
[A2] the patient, on which an agent acts, is changed from opposite to
opposite;
[A3] this opposite does not come from that opposite in such a way that
nothing common to both remains (as whiteness does not become
blackness);
[A4] therefore just as in the case of accidental change, in which the agent
of change moves a thing from opposite to opposite, while that thing
remains the same under both opposites, so also it is necessary in the
case of generation that the generating agent changes something
from form to form,4 which remains the same under both [forms];
that is said to be matter.5
Scotus takes [A1] to be obvious on empirical grounds. Natural agents produce

their effects from existing things. A beaver builds his dam from trees; a baker
3 See for example Physics I 190a 13–21 (Barnes I, p. 324); On Generation and Corruption I 319b
6–320a 4 (Barnes I, pp. 522–523); Metaphysics XII 1069b 3–21 (Barnes II, p. 1689).
4 It is interesting that in [A4] Scotus uses a forma in formam, introducing the term “form” for
the first time in this paragraph, instead of ab opposito in oppositum, as he uses twice earlier in
the argument. Unless Scotus commits a blatant fallacy of four terms, he must be using the
terms “opposite” and “form” in exactly the same way—in this text at least.
5 Lectura. II, d. 12, q. un (Vatican XIX, p. 72); Formatur eius ratio sic: [A1] omne agens naturale
requirit passum in quod agens agit (hoc patet ad sensum); [A2] illud passum, in quod agens
agit, transmutatur ab opposito in oppositum; [A3] hoc opppositum non fit illud oppositum,
ita quod nihil commune remaneat utrique (sicut albedo non fit nigredo); [A4] sicut igitur in
transmutatione accidentali transmutans aliquid, movet illud ab opposito in oppositum
manens idem sub utroque oppositorum, ita oportet in generatione quod generans transmu-
tet aliquid a forma in formam, manens idem sub utraque; illud dicitur esse materia. All trans-
lations from Latin are my own unless otherwise indicated, in this and every other chapter.
10 Chapter 1
bakes his cake from eggs, flour, and water. By ‘natural agent’ Scotus apparently
means ‘non-divine’ agent.6 He would likely include angels among the natural
agents since, although they are immaterial, they nevertheless produce their
material effects in pre-existing materials.7 A divine agent is unlimited in its
causality and therefore can and does produce its effects from nothing.
The idea expressed in [A2] is that the two terms of a change are opposed in
the sense that they are incompossible. One and the same thing cannot be a
forest and a beaver dam simultaneously; the egg must be cracked for the cake
to come into being. This incompossibility may be due to either metaphysical or
merely logical reasons. A man cannot be musical and non-musical simultane-
ously, so the change by which a man becomes musical is the very change by
which he ceases to be non-musical. But there is not something, non-musicality,
that is corrupted when the man acquires musicality, so the opposition between
these properties is merely logical. Nor can a man be pale all over and tan all
over simultaneously. The change by which a man becomes tan is the very
change by which he ceases to be pale, but in this case there is something, pale-
ness, that is corrupted when the man gets his tan. The opposition between
these properties is something more than logical opposition; tanness and pale-
ness just cannot exist together, so we describe their opposition as metaphysi-
cal. For Scotus’s purposes in [A2], either sort of opposition is salient.
[A2] leaves open the possibility that the patient on which the agent acts is
identical with one of a pair of opposites. For all [A2] says, for example, it may
be that when a baker pulls his cake out of the oven, there is no part of the flour,
water, eggs, etc. that is now a part of the cake; perhaps every part of the ingre-
dients is destroyed in the process of producing the cake. And [A2] must leave
open this possibility; otherwise the argument for matter from change would
include as a premise that something in addition to an opposite is required for
change—in other words, it would include its conclusion as a premise. [A3]
rules out this possibility. The pallor of a pale thing is not the reason for or cause
6 Scotus elsewhere contrasts a natural power with a rational power, where a natural power is a
power that is determined to produce its effect and a rational power is a power that is not (cf.
Quaestiones super Libros Metaphysicorum Aristotelis IX, q. 15, Bonaventure IV, pp. 675–699).
But Scotus does not seem to have this distinction in mind in [A1] since a non-divine agent,
whether natural or rational in the sense intended in Quaestiones super Libros Metaphysicorum
Aristotelis IV, q. 15, cannot produce something ex nihilo.
7 In Reportatio Parisiensis II-A, d. 8, q. un, n. 4 (Wadding XI, p. 299), Scotus speculates that
angels locally move bodies through will power (per imperium voluntatis). I have not been able
to find a text where Scotus speculates on angels’ abilities to effect qualitative or substantial
change.
of that thing’s ability to become tan (except insofar as being pale entails being
not-tan, which of course is a necessary condition for a thing to have the ability
to become tan). When pale-Socrates becomes tan-Socrates, the change in color
is not due to any passive or active power in the pallor of Socrates; pallor itself
cannot become what it is not, but a pale thing can; and, strictly speaking, pale-
Socrates cannot cease to be pale, but Socrates can cease to be pale-Socrates.
The color change is explained by divvying up pale-Socrates into Socrates him-
self and his pallor. Socrates is the substance that is the subject of pallor, and
pallor is simply an accident of that substance. The substance itself has the abil-
ity to take on many different colors.
[A4] draws the analogy between accidental change, such as Socrates’s color
change, and substantial change, the sort of change by which one substance is
corrupted and another is generated. Earth cannot become fire just insofar as it
is earth. The substantial change by which earth becomes fire is explained by
divvying up earth into a substantial form (of earth), and something else, prime
matter, in virtue of which earth can be transformed, and which “remains the
same” first under the form of earth and then under the form of fire.
It is worth saying a little more about what it means for one form to be
opposed to another. First, according to Scotus, a form “is not naturally corrupt-
ible nor is inclined to corruption.” It “strives to preserve itself,” and there is in it
no “natural aptitude of itself to not existing.” If a substance were form alone
without matter, it would follow that it is “intrinsically incorruptible.”8 Form, in
other words, cannot become what it is not. But a material substance can.
Substances are intrinsically corruptible and apt to not existing (as we know
from experience), so something besides form makes a substance to be this way.
What Scotus calls matter is just that part of a substance that has these (passive)
powers that form lacks. Whereas form strives to preserve itself, matter has an
appetite for the corruption of the substance of which it is a part and for the
generation of a new substance:
If there were not matter, remaining the same under both contraries,
no passive generation would be natural, since if there were not matter,
8 Lectura II, d. 12, q. un, n. 19 (Vatican XIX, pp. 75–76), [S]i non esset materia in composito, sed
tantum forma, quodlibet creatum intrinsece esset aequaliter incorruptibile. Nam actus tan-
tum et simplex non est naturaliter corruptibilis nec inclinatur ad corruptionem: nam forma
aëris nititur se salvare, nec est in aliqua aptitudine naturali ex se ad non essendum, ut videtur;
si igitur aër esset tantum forma, sicut tu dicis, et quodlibet creatum, ita quod nullum ens habe-
ret in se diversa principia (potentiale et actuale), sequitur quod quodlibet aequaliter sit incor-
ruptibile instrinsece, non habens magis instrinsice aër quod sit corruptibilis quam caelum.
12 Chapter 1
then no appetite for the term generated would preexist, nor would there
be something that naturally is inclined to the term generated, for the
form does not desire to be corrupted, since then it would desire its own
corruption.9
Since the generation of a new substance requires both an efficient cause and
a patient on which that cause acts, generation is both active and passive—
active generation is the action of the efficient cause which generates, and
passive generation is the passion of the patient which is generated. It is natural
for a substance to be generated from another substance because the latter
substance includes a passive component, prime matter, the nature of which is
to desire or be inclined toward a new generated term.
Second, there are at least two senses in which one form is opposed to
another, only one of which is relevant to the argument from change. In the first
sense, any two forms, whether substantial or accidental, are opposed in the
sense that one cannot become another. But this sense of opposition is not rel-
evant to the argument for matter, since many forms coinhere in one and the
same substance. A whiteness is opposed in this sense to Socrates’s substantial
form, but both forms are compossible because both can inform one and the
same subject simultaneously. In the second and relevant sense, however, some
forms are opposed in such a manner that, for some individual matter, the
inherence of one form in that matter is incompatible with the simultaneous
inherence of the other form in that matter. For such pairs of forms, a necessary
condition for a subject’s becoming informed by one is that subject’s ceasing to
be informed by the other.
iii Why Must Matter Persist through Change?
The conclusion of the argument from change holds that the patient on which
an agent acts “remains the same” under one form and then another. I take this
to mean that matter persists through change, and Scotus understands it this
way, too. But the argument from change does not actually entail that matter
9 Lectura II, d. 12, q. un, n. 17 (Vatican XIX, p. 74), Si non esset materia, manens eadem sub
utroque contrariorum, nulla generatio passiva esset naturalis, quia si non esset materia, tunc
nullus appetitus praecessit ad terminum generatum nec esset ibi aliquid quod naturaliter
inclinaretur ad terminum generatum, nam forma corrumpenda non appetit, quia tunc
appeteret sui ipsius corruptionem.
persists through change.10 Instead, it merely entails that there must be

something whose nature it is to be that by which a substance is corruptible and
by which something new can be generated from that substance. Scotus explic-
itly reasons for matter’s persistence in a second argument in Lectura II, d. 12,
q. un. This argument begins with the claim, evident from the argument from
change, that an efficient cause requires matter to produce its effect, and goes
on to argue that this entails that matter must persist through change. In this
section I analyze this argument.
The argument is an indirect proof which shows how an obviously false con-
clusion follows from the assumption that both the form and matter of the
patient on which an agent acts are corrupted. He begins by presenting the view
he means to reject, namely that
[B1] In the instant of corruption the agent does not presuppose the
patient on which it acts.11
The claim here is that while the patient of change has some role in the agent’s
production of a new substance, that role does not involve one of its parts (viz.,
the matter) persisting through the corruption of the patient (viz., the compos-
ite of matter and form). But if the agent does not presuppose the patient in this
way, Scotus reasons, then
[B2] In the instant of corruption the agent generates ex nihilo,12
an activity traditionally thought to be God’s alone. But [B2] only follows from
[B1] if generation ex nihilo is understood in a special sense. Scotus distinguishes
two ways of understanding production ex nihilo:
[ex nihilo1] The agent’s generation of a new substance does not depend
on the active or passive power of anything besides the agent,
and
10 David Ebrey argues that Aristotle agrees. David Buckley Ebrey, “Aristotle’s Motivation for
Matter,” unpublished doctoral dissertation (ucla, 2007).
11 Lectura II, d. 12, q. un, n. 12 (Vatican XIX, p. 73). [D]icitur a quibusdam quod agens naturale
agit in passum corrumpendum, et illud passum corrumpendum praesupponit in quod
agat; sed in instanti corruptionis non praesupponit, sed tunc totum vertitur in totum, ex I
De generatione.
12 Lectura II, d. 12, q. un, n. 13 (Vatican XIX, p. 73).
14 Chapter 1
[ex nihilo2] There is no part of the generated thing which was first a part
of something else.13
[B1] does not entail that the agent generates both ex nihilo1 and ex nihilo2, how-
ever. The defender of [B1] would deny that a natural agent produces its effect
ex nihilo1. But he would assert
[B2*] In the instant of corruption the agent generates ex nihilo2.
For Scotus, however, the sense in which an agent’s generation of a new sub-
stance depends on a patient just is that the agent requires a part from the
patient—the patient’s matter—in order to generate its effect. So if Scotus’s
opponent’s position is to be salvaged, there must be some alternative way of
understanding the dependence of an agent on a patient. But there is not,
argues Scotus. In other words, Scotus claims that
[B3] No agent can generate ex nihilo2 without also generating ex nihilo1.
Since no one would be willing to attribute generation ex nihilo1 to a merely

natural agent (and since [B2*] follows from [B1]), showing [B3] furnishes
Scotus with a successful ad hominem argument against [B1].
He argues for [B3] in the following way. If, in the instant of the corruption,
there is no part of the generated thing which was first a part of the patient (that
is, if [B2*]), then
[B4] The agent generates every part of its effect, including both the mat-
ter and form of the effect.
If an agent generates every part of its effect, what then is the patient contribut-
ing to the agent’s causal activity? Scotus discerns no role for a patient, given
[B4]. But he also thinks that a patient of change in some sense resists the
efficient causality of the agent, and therefore thinks that an agent able to
13 Lectura II, d. 12, q. un, n. 14 (Vatican XIX, p. 73), Ad hoc dicitur quod “aliquid produci ex
nihilo” potest intelligi dupliciter: vel sicut de termino et initiative, vel sicut de parte et
subiective. Primo modo non est verum quod generans generat ex nihilo, sed requiritur
aliquid corrumpendum; et ideo generat de aliquo initative vel in termino a quo. Sed
secundo modo verum est quod generans generat ex nihilo, quia nihil corrupti manet in
generato. See also Reportatio Parisiensis II-A, d. 12, q. 1, nn. 2–4 (Wadding XI, pp. 315–316).
generate every part of its effect would actually be diminished in its causal
efficacy while it is in proximity to such a patient:
[A]n agent which has every part of the effect in its power is no less able to
produce when that which weakens rather than strengthens its power is
removed; but according to you the generator has in its active power every
part of the effect, since in the instant of generation it presupposes noth-
ing of [that which weakens rather than strengthens an agent’s power].
Therefore the generator is able to produce the thing to be generated,
when that which weakens rather than strengthens its power is removed.
But [an agent’s] power is weakened rather than strengthened through
action on the contrary thing to be corrupted. Therefore a natural agent,
having taken away any patient, is able to produce its effect.14
Scotus supposes that if [B4] were true, then any patient on which the agent
acts in bringing about all the parts of the effect will hinder rather assist the
agent’s efficient causal activity. So he concludes that given a commitment to
[B2*], not only is there no way for an agent to depend on a patient in producing
its effect, but the agent would actually be better off without the patient! Scotus
reinforces his argument by appealing to the Aristotelian efficient causal axiom
[B5] A natural15 agent necessarily produces its effect unless there is an

impediment,
and continues,
[A] natural agent, such as fire, has in its power a whole effect, from what
was given; therefore if not impeded, it will produce the whole effect.
But it is not impeded through some natural agent, nor through the
14 Lectura II, d. 12, q. un, n. 15 (Vatican XIX, pp. 73–73), Contra: agens quod habet in virtute
sua totum effectum, non minus potest producere amoto quocumque quo posito magis
debilitatur virtus eius quam fortificetur; sed per te generans habet in virtute sua activa
totum effectum, quia nihil eius praesupponit in instanti generationis; igitur generans
potest producere genitum amoto quocumque quo posito magis debilitatur virtus eius
quam fortificetur. Per actionem autem in contrarium corrumpendum, debilitatur virtus
eius activa et non fortificatur; igitur agens naturale amoto quocumque passo potest pro-
ducere effectum.
15 The qualification “natural” in [B5] is intended to exclude rational agents, which do not
produce their effects necessarily but have a power for opposites. Cf. Quaestiones super
Libros Metaphysicorum Aristotelis IX, q. 15.
16 Chapter 1
absence of some contrary patient (but rather it is weakened through its

presence); therefore absent water and any contrary agent, it will produce
fire, and this from nothing at all.16
It is clear that the argument depends on an understanding of “impediment”

in [B5] according to which impediments to an agent’s productivity come in
degrees. That is, a patient may be said to impede an agent when that agent
cannot produce its effect, but a patient may also be said to impede an agent
when it resists but does not prevent an agent’s productivity. Scotus says that
the latter sort of impediment “weakens” an agent. This broader understand-
ing of impediment is rooted in Scotus’s understanding of the opposition of
forms, expressed in [A3]. Scotus says that form “strives to preserve itself ”;17 it
is in the nature of form to keep on existing, and in this sense form resists
activity which corrupts the subject which it informs. For example, fire burns
a dry log rapidly, a damp log less so, and a saturated log even less so. The
water in the log is opposed to and resists the fire. Understanding [B5] in this
broader sense yields
[B6] Any patient impedes or weakens an agent, even if it does not actu-
ally prevent the agent from producing its effect.
Since Scotus holds that an agent not only acts on a patient but changes one of
its parts (the matter) into a part of the substance it generates, Scotus can simul-
taneously hold that a patient both weakens an agent’s power but is required for
an agent’s production of a new substance. Remove this requirement, as Scotus’s
opponent does, and what is left, by [B6], is a patient whose presence merely
impedes the agent’s activity. By [B5], the agent would be better off without the
patient. So in the absence of an alternative explanation of how an agent
depends on a patient, and in light of [B6], Scotus, thinks [B3] holds and there-
fore denies [B1]. Scotus concludes, then, that in addition to being that by
which a material substance has passive power, matter also persists through
16 Lectura II, d. 12, q. un, n. 16 (Vatican XIX, p. 74), omne agens naturale, potens in aliquem
totum effectum, de necessitate illum faciet non impedimentum, ex IX Metaphysicae cap.
4, sicut etiam de necessitate producit formam, si non impeditur; sed agens naturale, ut
ignis, habet in virtute sua totum effectum, ex datis; igitur non impeditum, sic producet
totum effectum. Sed non impeditur per aliquem agentem, nec per absentiam alicuius
contrarii passi (sed magis debilitatur per eius praesentiam); igitur absente acqua et
quocumque contrario agente, producet ignem—et ita de nihilo omnino.
17 Lectura II, d. 12, q. un, n. 19 (Vatican XIX, pp. 75–76).
substantial change such that for any two substances, where the first is substan-
tially changed into the second, the matter of the first becomes the matter of
the second.
Here is the argument for matter’s persistence in summary form. Suppose for
reductio that
[B1] In the instant of corruption the agent does not presuppose the
patient on which it acts,
where this is understood to mean that no part of the patient persists through
the generation of the agent’s effect. From [B1] it follows that
[B2] In the instant of corruption the agent generates ex nihilo.
But there are two senses of generation ex nihilo, namely,
[ex nihilo1] The agent’s generation of a new substance does not depend
on the active or passive power of anything besides the agent,
and
[ex nihilo2] There is no part of the generated thing which was first a part
of something else.
Since Scotus and his opponent would concur that no natural agent generates
ex nihilo1, [B2] can be stated more precisely as
[B2*] In the instant of corruption the agent generates ex nihilo2.
From [B2*] it follows that
[B4] The agent generates every part of its effect, including both the mat-
ter and form of the effect.
But given that
[B6] Any patient impedes or weakens an agent, even if it does not actu-
ally prevent the agent from producing its effect;
and
18 Chapter 1
[B5] A natural agent necessarily produces its effect unless there is an

impediment,
an agent that did not produce its effect from a part of a patient would more
easily produce its effect without any relation at all to the patient, and would
therefore not depend on the patient for its effect. From this it follows that
[B3] No agent can generate ex nihilo2 without also generating ex nihilo1.
No one would hold that natural generation can be ex nihilo1, however, so [B3]
entails, ad hominem, that not-[B1].18
iv Matter as Passive Power
Matter is supposed to be a partial explanation of why some change occurs, but

being able to persist does not explain change. It only asserts that, when some
change occurs, some part of the corrupted substance is not itself destroyed and
can become a part of the generated substance. For the explanation of change
we need something by nature cooperative with any efficient cause, something
that, as Scotus says, has an appetite for and is naturally inclined to the term of
change—call this matter.19 Matter’s nature is to be the passive power of a sub-
stance to undergo corruption and to be that from which something else comes
to be. That this passive power persists, and must persist, is a consequence of a
created agent’s dependence on preexistent materials in producing its charac-
teristic effect, but being able to persist is not what it is to be a passive power.
What matters for matter is not merely that it persists but that it is a special kind
of power. In this section I characterize this additional feature of matter in vir-
tue of which it plays the role it does in change, namely, that it is essentially a
passive power for receiving forms.
The text we have been examining, Lectura II, d. 12, q. un, describes matter
several times as a passive power. Scotus develops his understanding of matter
18 Versions of the argument tracked in [B1] through [B6] can also be found in Quaestiones
super Libros Metaphysicorum Aristotelis VII, q. 5, nn. 8–12 (Bonaventure IV, pp. 132–133)
and Reportatio Parisiensis II-A d. 12, q. 1, nn. 3–4 (Wadding XI, pp. 314–315).
19 Lectura II, d. 12, q. un, n. 17 (Vatican XIX, p. 74), Si non esset materia, manens eadem sub
utroque contrariorum, nulla generatio passiva esset naturalis, quia si non esset materia,
tunc nullus appetitus praecessit ad terminum generatum nec esset ibi aliquid quod natu-
raliter inclinaretur ad terminum generatum, nam forma corrumpenda non appetit, quia
tunc appeteret sui ipsius corruptionem.
as passive power in QMet IX, so in this section we will look to this text for
guidance.
In QMet IX, qq. 1–2, Scotus distinguishes two meanings of potentia or
potency: potency as a mode of being, and potency as a principle.20 As a princi-
ple, potency is power—that by which a thing can come to be. As a mode of
being, potency is possible existence, such as the possible existence of a white-
ness in Socrates or the possible existence of Antichrist (to borrow Scotus’s
example). In this section I consider potency as a principle, and in the first sec-
tion of Chapter 2 I consider it as a mode of being.
Scotus divides powers into two kinds, active and passive, identifying the for-
mer with the efficient cause and the latter with matter. Scotus thinks of a
causal power, whether active or passive, as essentially an absolute (i.e., non-
relational) essence, which is by nature a power for something or other, and
which is the foundation of a relation to whatever it is that comes about as a
result of the exercise of that power:
[…N]othing pertains to the essential notion of power except some abso-

lute essence in which some relationship to the effect brought about by
that power is immediately founded, such that no relation precedes in act
the activity itself by which the effect is brought about; no relation is that
through which a power is determined to produce its effect. Instead, the
absolute effect is from something absolute, without any preceding rela-
tion; once the effect is posited, an actual mutual relation of the effect to
the power follows, naturally posterior to the effect, which could not exist
in one if it were not posited in the other.21
20 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq. 1–2, n. 14 (Bonaventure IV,
p. 512), Uno modo potentia dicit modum quendam entis. Alio modo specialiter importat
rationem principii. Cui autem istorum fuerit nomen prius impositum, et inde ad aliud
translatum, dubium est. Si tamen primo imponebatur ad significandum modum quen-
dam entis, cum iste non conveniat enti tali nisi per aliquod eius principium per quod
potest esse, convenienter potest nomen potentiae transferri ad principium tamquam ad
illud quo possibile potest esse, non ‘quo’ formaliter, sed causaliter—Similiter, si primo
imponebatur principio per quod res potest esse, potest transferri ad significandum gener-
aliter modum essendi similem illi quem habet principiatum in principio.
21 Quaestiones super Libros Metaphysicorum Aristotelis IX, q. 5, n. 13 (Bonaventure IV, p. 563),
[…N]ihil est de ratione potentiae nisi absoluta aliqua essentia, in qua immediate fundatur
aliquis respectus ad principiatum, ita quod nullus respectus praecedit in actu ipsam prin-
cipiationem per quam quasi determinetur ad principiandum. Sed ab absoluto, since
omni respectu praecedente, est effectus absolutus; quo posito, posterius natura sequitur
20 Chapter 1
And what holds of power in general holds of passive power in particular:
[Passive power], according to its essential notion, is a part of the

composite and is perfected by form (which is the other part), accord-
ing to which notion it is naturally prior to its effect as such a princi-
ple. But this is precisely under an absolute notion; for in this case
it is immediately a part of the composite and is perfected by form,
since if some relation were of the notion of passive power insofar
as it is a part of the composite, that relation would also be of the
notion of the composite, and thus no material thing would be essen-
tially absolute.22
Scotus’s arguments that powers in general and passive powers in particular are
absolute (non-relative) are aimed at Henry of Ghent’s opposing view, that
powers are essentially relative. For Henry, to ascribe a power to something
absolute, whether form, matter, or the composite, is to attribute to it a relation
to a term.23 As Henry says,
Concerning the nature of a power (in as much as it is a power), it is spo-

ken of with reference to activity, so it is not some absolute thing, but
rather just this relatedness [to the activity in question] that is based on
something absolute.24
relatio actualis mutua principiati ad principium, quae in neutro esse potuit, altero
extremo non posito […].
[Potentia passiva] secundum illam rationem essentialem, est pars compositi et perficitur
a forma, quae est altera pars, secundum rationem est prior naturaliter principiato ut tale
principium. Hoc autem est praecise sub ratione absoluti; sic enim immediate est pars
compositi et perficitur a forma. Quia si aliqua relatio esset de ratione eius in quantum est
pars compositi, illa etiam relatio esset de ratione compositi, et ita nullum materiale esset
essentialiter absolutum. Also, Quaestiones super Libros Metaphysicorum Aristotelis IX,
qq. 1–2, n. 15 (Bonaventure IV, p. 512).
23 Quaestiones super Libros Metaphysicorum Aristotelis IX, q. 5, nn. 4–6 (Bonaventure IV
pp. 560–561); the editors reference several texts of Henry in the notes. See J.T. Paasch’s
discussion in Divine Production in Late Medieval Trinitarian Theology: Henry of Ghent,
Duns Scotus, and William Ockham (Oxford: Oxford University Press, 2012), pp. 117–122; and
see Scott Williams, “Henry of Ghent on Real Relations and the Trinity: The Case for
Numerical Sameness without Identity,” Recherches de Théologie et Philosophie Médiévales
79:1 (2012), pp. 109–148.
24 Henry of Ghent, Summae Quaestionum Ordinarium a. 35, q. 2, ed. Jodicus Badius (Paris,
1520; reprinted by The Franciscan Institute, St. Bonaventure, ny, 1953). “[D]e ratione
Prima facie, Henry’s view has the ring of truth, since a power is only intelligible
as a power for something or other. But Scotus objects that powers naturally
precede their functioning, such that the subject of a power is powerful prior to
the production of that power’s effect. To make Scotus’s argument against Henry
perspicuous, let x be the subject of a power, P, and let’s say that x φs when x
exercises P and produces its effect, E.
(Scotus himself thinks that there are some powers that are, as it were,
their own subjects, such that x and P would be identical. In such cases, as
he puts it, that which (quod) is powerful is the same as that by which (quo)
it is powerful. Prime matter itself is just such a power. I will revisit this
point later on.)
Scotus argues that if P were simply a relation (of x to E), one of three disjuncts
would be true: x is related by P to E (i) simultaneous with, or (ii), posterior to, or
(iii) prior to, E’s existence (and therefore simultaneous with, posterior to, or
prior to, x’s φing). (iii) is false because a real (causal) relation requires the actual
existence of the effect—P is a real relation of x to E only if E actually exists. If we
suppose instead, trying to salvage (iii), that P is a relation to potential-E, then P
itself would be merely a potential relation. But what is potential does not remain
potential when it is reduced to act. Thus, if P is a potential relation of x to poten-
tial-E, P ceases to be at the moment E begins to be, and therefore (since E is the
effect brought about by P) P ceases to be at the moment x φs.25
This argument against (iii) should sound fishy. Why suppose that, if P is a
potential relation, P ceases to be when E is actualized? Why not suppose
instead that if P is a potential relation of x to potential-E, then P simply
becomes actual when potential-E becomes actual E? Scotus’s answer lies in his
claim,
[pr] Whatever belongs to the nature of P prior to x’s φing belongs to the
nature of P when x φs.26
It follows from [pr] that if P is a potential relation prior to x’s φing then P is
a potential relation when x φs. And it follows from this that x is potentially
potentiae inquantum potentia, est quod quod dicatur ad actum, ita quod nihil sit absolu-
tum, sed solus respectus fundatus in re super aliquo absoluto.” The translation is from
Paasch, Divine Production in Late Medieval Trinitatian Theology, p. 118.
25 Quaestiones super Libros Metaphysicorum Aristotelis IX, q. 5, nn. 9–10 (Bonaventure IV,
pp. 561–562).
Manet autem in principio quando principiat quidquid est de ratione eius in quantum est
prius naturaliter principiato.
22 Chapter 1
(causally) related to its actual effect E, which is infelicitous. While Scotus does
not argue for [pr], it is not hard to motivate it. A power is supposed to be that
because of which something comes to be. If P is a power for bringing about E,
then P is supposed to explain how E becomes actual, i.e., moves from potency
to act. But if both P and E are potential entities, then P cannot explain how E
becomes actual.
(i) and (ii) will not work for the obvious reason that powers are prior to
the effects brought about by means of powers. It is evident by the meaning
of the terms that if x has P for producing E, then x can produce E. But if x
has P only when x is φing or has φd, then x can φ if and only if it is φing or
has φd. But presumably at least some statements with the form, “x can φ
but is not φing,” are true.27 For example, a generator that is turned off is not
generating electricity but it can generate electricity (if it is turned on).28
But according to (i) and (ii), from statements of the form, “x can φ but is
not φing,” together with the biconditional, “x can φ if and only if it is φing
or has φd,” one can always derive a contradiction. Scotus therefore con-
cludes that P is not a relation but something absolute. Of course, showing
that powers are not relations does not entail that there are absolute
powers. One might agree with Scotus’s analysis of Henry’s theory but
reject powers altogether, as many philosophers have done, for example,
27 For clarity’s sake, I am not invoking Scotus’s innovative idea about synchronic possibility,
that x can φ at the temporal instant at which x is not φing. Even if Scotus is wrong and x
can φ only after it ceases to not-φ, “x can φ but is not φing,” can still be true. For some of
the scholarship on Scotus’s metaphysics of modality see Simo Knuuttila, Modalities in
Medieval Philosophy (London and New York: Routledge, 1993); Simo Knuuttila,
“Interpreting Scotus’s Theory of Modality: Three Critical Remarks,” in Via Scoti:
Methodologia ad mentem Joannis Duns Scoti, vol. 1, ed. Leonardo Sileo (Rome: Edizioni
Antonianum, 1995), pp. 295–303; Stephen Dumont, “The Origin of Scotus’s Theory of
Synchronic Contingency,” The Modern Schoolman 72 (1995), pp. 149–167; Calvin Normore,
“Scotus, Modality, Instants of Nature and the Contingency of the Present,” in John Duns
Scotus: Metaphysics and Ethics, ed. Ludger Honnefelder, Rega Wood, and Mechthild
Dreyer (Leiden: E.J. Brill, 1996), pp. 161–174; Calvin Normore, “Duns Scotus’s Modal Theory,”
in The Cambridge Companion to John Duns Scotus, ed. Thomas Williams (Cambridge:
Cambridge University Press, 2002), pp. 129–160; Douglas Langston, “Scotus and Possible
Worlds,” in Knowledge and the Sciences in Medieval Philosophy, vol. 2, ed. Simo Knuuttila,
Reijo Työrinoja, and Stan Ebbesen (Helsinki: Yliopistopaino, 1990), pp. 240–247; Nicole
Wyatt, “Did Duns Scotus Invent Possible Worlds Semantics?” Australasian Journal of
Philosophy 83 (2000), pp. 196–212.
28 The example is from Paasch, Divine Production in Late Medieval Trinitatian Theology,
p. 144.
David Hume29 and David Lewis.30 But Scotus and Henry inhabit a power-
saturated world.
Scotus’s powers are, then, essentially absolute entities. In a text quoted
above, Scotus says:
[…N]othing pertains to the essential notion of power except some abso-

lute essence in which some relationship to the effect brought about by
that power is immediately founded, such that no relation precedes in act
the activity itself by which the effect is brought about; no relation is that
through which a power is determined to produce its effect. Instead, the
absolute effect is from something absolute, without any preceding rela-
tion; once the effect is posited, an actual mutual relation of the effect to
the power follows, naturally posterior to the effect, which could not exist
in one if it were not posited in the other.31
Here Scotus clearly distinguishes between a power, the exercise of that power,
and the relation of that power to the effect brought about by that power.
Neither the power nor the exercise of that power is a relation, and this should
be clear from the argument against (iii), above: there are no real relations to
nonexistent terms.
v Obediential Potency and the Subject of Passive Power
Socrates has passive power to receive a form of whiteness. Socrates, of course,

is a full-fledged substance, a composite of prime matter and several substantial
29 David Hume, An Enquiry Concerning Human Understanding VII, ed. Eric Steinberg
(Indianapolis, in: Hackett, 1993), p. 41, “When we look about us towards external objects,
and consider the operation of causes, we are never able in a single instance, to discover
any power or necessary connexion; any quality, which binds the effect to the cause, and
renders the one an infallible consequence of the other. We only find, that the one does
actually, in fact, follow the other.”
30 David Lewis, “Causation,” Journal of Philosophy, 70, pp. 556–567, and the updated theory
in David Lewis, “Causation as Influence,” Journal of Philosophy, 97, pp. 182–197. Lewis’s
basic idea is to analyze causal dependence counterfactually and as a relation between
events (not substances or their properties), such that an event e1 causally depends on an
event e2 (or e2 influences e1), if and only if, if e1 occurs e2 occurs and if e1 does not occur e2
does not occur.
31 Quaestiones super Libros Metaphysicorum Aristotelis IX, q. 5, n. 13 (Bonaventure IV,
p. 563), […N]ihil est de ratione potentiae nisi absoluta aliqua essentia, in qua immediate
24 Chapter 1
forms. And Socrates himself is the proximate matter that is in potency to a

whiteness. What is it about Socrates in virtue of which he has this potency? Is
it his prime matter alone, or his prime matter together with his forms?
The short answer is that it is his prime matter alone. But the question is
harder to answer than it first appears. On the one hand, Scotus says that in
substances composed of matter and form, the substance has passive power
because it has a part—prime matter—that is passive power. In Scotus’s terms,
the substance is that which (quod) is passively powerful, but the matter of the
substance is that by which (quo) it is passively powerful.32 Prime matter does
not have, but just is, passive power.33 In QMet IX, q. 12, Scotus says of prime
matter that,
[It] has of itself a natural inclination to be a part of any composite, and to

be perfected by any form which can be a part and by which [prime mat-
ter] can be perfected.34
On the other hand, under natural conditions, where prime matter is informed
by one or more forms, prime matter is not apt to receive just any form, but only
a limited range of kinds of forms. There is a “necessary order” in the succession
of forms in matter, Scotus says, such that the forms, whether accidental or sub-
stantial, to which a particular material substance has passive potency at one
time, are determined by the accidental and substantial forms which partially
compose that substance at that time.35 Although Scotus expresses doubt about
fundatur aliquis respectus ad principiatum, ita quod nullus respectus praecedit in actu
ipsam principiationem per quam quasi determinetur ad principiandum. Sed ab absoluto,
since omni respectu praecedente, est effectus absolutus; quo posito, posterius natura
sequitur relatio actualis mutua principiati ad principium, quae in neutro esse potuit,
altero extremo non posito […].
32 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq. 3–4, n. 19 (Bonaventure IV,
p. 542).
33 For an opposing interpretation of Scotus according to which passive power is a property
of matter, see Peter King, “Duns Scotus on Possibilities, Powers, and the Possible,” in
Potentialität und Possibilität: Modalaussagen in Der Geschichte Der Metaphysik, ed.
T. Buchheim, K. Lorenz, and C.H. Kneepkens (Stuttgart: Frommann-Holzboog, 2001),
pp. 175–199, esp. p. 189. Richard Cross, on the other hand, has rightly recognized that for
Scotus matter is identical with passive power. See his “Identity, Origin, and Persistence in
Duns Scotus’s Physics,” History of Philosophy Quarterly 16 (1999), pp. 1–18, esp. p. 10.
34 Quaestiones super Libros Metaphysicorum Aristotelis IX, q. 12, n. 6 (Bonaventure IV, p. 612).
Et de hoc videtur dicendum quod cum formae naturales habeant ordinem necessario in
our ability to have any deep insight into the reasons why this order is the way
it is, as an example we might say that a composite of prime matter and the
form of bronze is in immediate potency to the form of a statue, but not in
immediate potency to an animal soul. This suggests that having passive
potency is more than just having a part that is passive potency; forms seem to
play some role in a substance’s passive potency.
Nevertheless, Scotus takes this “necessary order” of forms to be necessary
only with respect to created agents:
But because that order is not necessary except insofar as matter is

changed by a natural agent to whose power that order is not subject, thus
obediential potency has a place here, according to which matter can
receive any form immediately after any other, through the transmutation
of an agent to whose power that aforementioned order is subject.
Nevertheless this potency of matter is not properly obedient to form or
the composite, but to the agent from which it receives form. For ‘obedi-
ence’ properly signifies a subjection to an agent able to make the
obedient thing do what it wants it to do.36
With respect to God’s power, Scotus reasons, matter really is able immediately
to be informed by any form, regardless of the forms it now is under. The Prince,
for example, cannot immediately be transformed into a frog by any natural
agent, since any natural agent would be subject to the necessary order of the
succession of forms; but perhaps God could turn the Prince into a frog. The
moral of the story is that prime matter of itself is in potency to any form, and
that a sufficiently powerful agent can actualize this potency in matter all by
itself. A natural agent’s inability to transform the bronze immediately into a liv
ing thing is due solely to its own limitations and not to any restriction of prime
matter’s potency by the form of bronze.
succedendo sibi invicem in materia (secundum quod manifestum est ad sensum […])
oportet dicere quod materia, ut sub una forma, non est nata transmutari immediate ad
quamcumque, sed ad determinatam.
Sed quia ordo iste non est necessarius nisi in quantum materia transmutatur ad agente
naturali, cuius virtuti non subicitur iste ordo, ideo hic habet locum potentia oboedi-
entialis, secundum quam materia cuiuscumque formae capax est immediate post
quamcumque, per transmutationem ab agente, cuius virtuti subest dictus ordo. Non
tamen est haec potentia materiae ad formam vel compositum proprie oboedientialis, sed
ad agens a quo sic recipit formam. ‘Oboedientia’ enim proprie significat subiectionem
respectu agentis potentis de oboediente facere quod vult.
26 Chapter 1
Scotus’s doctrine of the obediential potency of matter to any form should

influence the way in which we interpret passages in which he describes the
conditions under which a subject is in potency to a form. For example, Scotus
thought that several conditions must be met before an embryo could receive a
soul:
For the animation by the intellective soul there is needed a heart and
liver of determinate heat, a brain of determinate frigidity, and so on for
individual organic parts. But such a disposition ceasing, there can still
remain some species or quality which stays with the form of corporeity,
although not that which is required for the existence and operation of
the intellective soul in matter.37
If the embryo’s prime matter is always in obediential potency to receive any

form, then it can receive the intellective soul without these conditions being
met. When merely natural agents are involved and these conditions are
met, therefore, these additional forms in matter do not limit or modify the
potency of prime matter; instead they enable a natural agent to actualize
the potency of matter. What God could do all by himself under any conditions,
a natural agent can do only under special conditions.
I conclude, then, that the passive potency of proximate matter to receive
form is just the passive potency of one its parts—prime matter. The apparent
determinations of prime matter’s potency by forms are ultimately explained as
limitations on natural agents’ abilities to produce form in prime matter.
37 Ordinatio IV, d. 11, p. 1, a. 2, q. 1, n. 284 (Vatican XII, p. 267), Istud patet exemplo, quia ad
animationem ab intellectiva requiruntur cor et hepar determinate calida, cerebrum
ceterminate frigidum, et sic de singulis partibus organicis; tali autem dispositione ces-
sante, potest adhuc manere aliquas species vel qualitas, quae stat cum forma mixtionis,
licet non illa quae requiritur ad esse et operationem intellectivae in materia.
Chapter 2
The Ontology of Prime Matter
Scotus said that we can think of matter either as a power or as a possible being.1 In
the previous chapter I examined Scotus’s characterization of prime matter as a cer-
tain kind of power, the passive power of a substance to undergo accidental and sub-
stantial change. This chapter focuses on the sense in which matter is a possible being
and analyzes Scotus’s arguments that prime matter can exist by divine power all on
its own, without any form, or, put another way, that there can be free-floating pow-
ers, powers that are not the powers of anything. This is a strange thought, and it is all
the stranger given that for Scotus matter as such is not corporeal or extended; these
features are due to forms, substantial form and the form of quantity, respectively.
Matter as such does not have parts outside of parts, although it is able to have parts
(habet partibilitatem).2 Like Scotus, William Ockham holds that potentia is not a rela-
tion founded on matter, but is instead matter itself.3 Unlike Scotus, however, Ockham
denies that there are distinct quantitative forms, holding that matter de se has parts
outside of parts.4 For Ockham then the basic power for undergoing change is identi-
fied with a fundamental extended substratum which is determined or shaped by
substantial and qualitative forms. Traditional analogues of prime matter, such as
bronze or clay, are therefore useful for getting a grip on Ockham’s conception of mat-
ter. But for Scotus passive power is essentially independent from and prior to being
extended; the traditional metaphors are therefore misleading for understanding
Scotus’s theory of matter. Transitioning here to Scotus’s ontology of matter, it will be
useful to keep in mind that Scotus’s investigation concerns whether or not a certain
kind of power, passive power, can exist and not be the power of any subject. An
Ockhamist enquiry into matter’s ontology on the other hand would investigate
whether there can be an extended thing that did not have determinate dimensions.
i Matter as Subjective Potency
Aristotelians say that prime matter is a being in potency. Aquinas took this to
mean that matter was not of itself an actual being; it depends for its actuality
1 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq.1–2, n.14 (Bonaventure IV, p.512).
2 Reportatio Parisiensis II-A, d.12, q.2, n.7 (Wadding XI, p.322).
3 Ockham, Summula philosophiae naturalis I, c.10, ll.62–63 (OPh VI, p.183), potentia non est
relatio fundata in materia, sed est ipsamet materia et non fundatur in ea.
4 Ockham, Summula philosophiae naturalis I, c.13 (OPh VI, pp.191–194).

28 Chapter 2
on form, such that apart from form matter cannot exist, even by divine power.5
Scotus disagrees and argues that in order for matter to do all it is supposed to
do, it must be an actual being off its own bat. It is a power, a cause, it persists
through change, it is a part of a substance, and so on.6 So Scotus offers a differ-
ent take on what it means for matter to be a being in potency.
Scotus distinguishes two senses of possible existence, the possible existence
of something that can be made, such as Antichrist, and the possibility for an
existing subject, such as prime matter, to be perfected by form. He calls the first
objective potency and the second subjective potency:
[Objective potency] is [the potentiality] of any substantial or accidental

essence to its own existence, and it is founded on that essence whose
proper existence it is. For thus the essence of an accident or a whiteness
is in potency to its own existence, just as the essence of the soul to be cre-
ated is in potency to its existence.7
[Subjective potency] is not [the potentiality] of just any being, since it

does not exist except in that which, in addition to its own existence, is
able to receive some being from another; and thus when it does not have
that, it is in potency to it. For example, a non-white body is in potency to
existing not simpliciter but existing as white, which is its existence in a
qualified and extrinsic way.8,9
5 Aquinas, Quaestiones disputatae de potentia dei q.3, a.5, ad.5 (Turin, 1953); Summa theologiae
Ia, q.45, a.4 (Turin, 1952); Scriptum super sententiis II, d.17, q.1, a.2, corp. (Parma, 1856). These
and subsequent references to Aquinas are taken from Opera Omnia, ed. Robert Busa, S.J.
Milan: Editoria Elettronica Editel, 1992. Revised edition on the internet, by Enrique Alarcón,
http://www.corpusthomisticum.org /iopera.html, 2005.
6 Lectura II, s.12, q.un, n.29 (Vatican XIX, p.79).
7 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq.1–2, n.41 (Bonaventure IV, p.524–
525), Prima est cuiuscumque essentiae substantialis vel accidentalis ad proprium esse, et fun-
datur in illa essentia cuius est illud proprium esse. Ita enim essentia accidentis vel albedinis est
in potentia ad prorium esse suum, sicut essentia animae creandae est in potentia ad suum esse.
8 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq.1–2, n.42 (Bonaventure IV, p.525),
Secunda non est cuiuslibet entis, quia non est nisi illius quod, praeter esse proprium, natum
est recipere aliquod esse ab alio; et ita quando non habet illus, est in potentia ad illud. Verbi
gratia, corpus non-album est in potentia ut sit, non simpliciter sed ut sit album, quod est esse
eius secundum quid et extrinsecum. Et ista potest dici “subiectiva.”
9 Scotus also distinguishes between objective and subjective potencies in his commentaries
on Sentences II, d.12, elaborating the idea that matter is a being in potency by saying that it
has subjective instead of objective potency. See Reportatio Parisiensis II-A, d.12, q.1, n.11
(Wadding XI, p.317), and Lectura II, d.12, q.un (Vatican xix, p.80).
The Ontology Of Prime Matter 29
In Lectura II, d.12, q.un, Scotus draws the distinction between subjective and
objective potency to carve out a way of describing prime matter both as an
actual entity and as potential entity.10 The sense in which matter is a potential
entity is not that it has objective potency, since objective potency is just the
sense of potency in which potency is opposed to act. But subjective potency is
not itself opposed to act. Matter’s subjective potency for receiving a form, F, is
of course opposed to matter’s actually being informed by F. But matter’s sub-
jective potency for receiving F (or any form) is not opposed to matter’s actual
existence—in fact, it presupposes it.
The two kinds of potency are related in such a way that that to which some-
thing has subjective potency is itself in objective potency. For example, if
Socrates has subjective potency to become white, then a whiteness has objec-
tive potency. But something can be in objective potency without a correlation
to something in subjective potency. For example, for Scotus an angel is an
immaterial form whose natural condition is to exist without informing mat-
ter.11 Therefore before an angel is created it has objective potency and there is
nothing which has subjective potency to it. Scotus speculates that an essence
having objective potency could be ordered to a subject having subjective
potency, even if the latter does not actually exist. Matter has subjective potency
to be perfected by substantial form, and Scotus says that even if both matter
and substantial form did not exist, matter would still have subjective potency
to be perfected by substantial form.12 Ordinarily, however, we are interested in
10 Lectura II, d.12, q.un, nn.30–32 (Vatican XIX, p.80).

11 Scotus, Quodlibet IX, a.2, n.7, in Obras del Doctor Sutil Juan Duns Escoto (edicion biligue):
Cuestiones Cuodlibetales, trans. (with Latin edition) by Felix Alluntis (Madrid: Biblioteca
de autores cristianos, 1968), pp.344–345.
12 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq.1–2, nn.54–57 (Bonaventure
IV, pp.529–530); in n.57 Scotus speculates that if both matter and form were in objective
potency, matter would have objective and subjective potency, but the objective potency
would be actualized prior to the actualization of the subjective potency. Incidentally this
resolves a disagreement between Richard Cross and Simona Massobrio over whether or
not matter has objective potency. Cross claims, “There is no evidence that Scotus would
want to claim that matter exhibits objective potentiality as well as subjective potentiality.”
Massobrio claims that matter’s privation with respect to all substantial forms “represents”
objective potentiality. According to Scotus, however, matter could have objective potency
and simultaneously have subjective potency, even though it (apparently) usually actually
exists (and therefore does not have objective potency) and has subjective potency.
Privation in matter is not, pace Massobrio, a suitable candidate for representing or being
objective potency, because privation is matter’s lack of e.g., some form, F, cum power to
become F, whereas objective potency is e.g., F’s lack of existence cum power to exist.
30 Chapter 2
cases in which the subject of subjective potency exists, either as prime or prox-
imate matter.
The term of a subjective potency is identical with the term of an objective
potency. If the essence of Antichrist is in objective potency then (presuming
he arrives on the scene as most men do, through biological reproduction) there
is some matter in subjective potency to become Antichrist. Matter’s subjective
potency is strictly speaking not to form but to the composite of matter and
form, the substance which matter together with form can become, or the qual-
ified substance which a substance together with accidental form can become.13
ii A Shifting Opinion about Matter’s Separability from Form
In Lectura II, d.12 Scotus is unequivocal: matter and form are res et res.14 This is
unsurprising, since in substantial change, as we have seen, matter persists but
form does not. Therefore some particular matter is separable from—and
therefore really distinct from—some particular form.15 However, Scotus recog-
nized that it is one thing to claim that matter can exist without some particular
form, and quite another to claim that matter can exist without any form at all.
Matter’s persistence through substantial change substantiates the first but not
the second claim. Scotus’s mature position on matter’s real distinction from
form, which holds both the first and second claims, has been well docu-
mented;16 my own contribution here is to document what appear to be devel-
opments in Scotus’s views on matter’s separability from form.
Scotus thought that matter and form were really distinct throughout his
teaching career and also that matter could exist apart from whatever form was
Richard Cross, The Physics of Duns Scotus (Oxford: Oxford University Press, 1998), p.18,
n.17, and Simona Massobrio, “Aristotelian Matter as Understood by St. Thomas Aquinas
and John Duns Scotus” (unpublished doctoral dissertation, McGill University, 1991), p.205.
13 Quaestiones super Libros Metaphysicorum Aristotelis IX, qq.1–2, n.46 (Bonaventure IV, p.526).
14 Lectura II, d.12, q.un, n.49 (Vatican XIX, p.88).
15 Separability implies real distinction, but the biconditional may not hold. Arguably, for
Scotus, the three persons of the Trinity are really distinct from each other but are insepa-
rable. See Ordinatio I, d.2, p.2, qq.1–4, n.421 (Vatican II, p.366), and Richard Cross, Duns
Scotus on God (Burlington, vt: Ashgate, 2005), pp.153–155.
16 Etienne Gilson, Jean Duns Scot: Introduction a ses Positions Fondamentales (Paris: J. Vrin,
1952), pp.432–444; Marilyn McCord Adams, William Ockham v.2 (South Bend, in:
University of Notre Dame Press, 1987), pp.633–647; Simona Massobrio, “Aristotelian
Matter as Understood by St. Thomas Aquinas and John Duns Scotus,” pp.241–244; Richard
Cross, The Physics of Duns Scotus, pp.13–33.
informing it at some time.17 But we do find a difference in views presented in

Scotus’s various texts on the issue of whether matter could exist without being
informed at all. The Vatican editions of Ordinatio II and Lectura II, d.12 suggest
that Scotus did not always think that God could conserve matter denuded of
form, since in these texts Scotus simply doesn’t discuss the issue (where it
would be natural to do so). Luke Wadding had printed a second question in his
edition of Ordinatio II d.12, under the heading, “Whether through some power
matter can exist without form?” in which Scotus argues that matter can be
conserved without any form,18 but the Vatican edition leaves out from its
Ordinatio not only this second question but all of d.12—the text skips from d.11
to d.13. The editors claim that in those manuscripts and editions in which
Ordinatio II is printed with d.12, the questions contained therein are taken
either from “other works of Scotus” or from Alnwick’s Additiones Magnae II.19
At the end of Ordinatio II, d.11, the Vatican editors refer the reader to Reportatio
II-A d.12—known to be among Scotus’s latest works—for questions on Book II,
d.12. I will examine this text later on.20
While Lectura and Ordinatio appear to be silent on the issue of matter’s sep-
arability, another early work, Quaestiones super Praedicamenta Aristotelis, indi-
cates that Scotus at one point denied that matter was totally separable from
form. In q.15, which asks whether the parts of substance (namely, matter and
form) are substances, Scotus presents an argument for the view that they are,
which he goes on to reject. The argument that Scotus rejects is this:
What is divided is predicated of the things resulting from the division;

from On the Soul II [412a 7–9]: ‘Substance is divided into matter and form
and the composite.’ Therefore any one of these is substance.21
17 Lectura II, d.12, q.un, nn.52–81 (Vatican XIX, pp.89–101); Reportatio Parisiensis II-A, d.12,
q.1, nn.19–24 (Wadding XI, pp.319–320); Quaestiones super Libros Metaphysicorum
Aristotelis VII, q.5 (Bonaventure IV, pp.131–139).
18 Wadding VI, pp.680–698.
19 Vatican VIII, p.224.
20 Vatican VIII, p.225, note 17*. Reportatio Parisiensis II-A is the current designation of what
Wadding printed as Reportata Parisiensia II (Wadding XI). A second Paris Report on Book II,
known as II-B, is unedited. Cf. Thomas Williams, “Introduction: The Life and Works of John
Duns the Scot,” in The Cambridge Companion to Duns Scotus, p.11. Given the uncertainty sur-
rounding Wadding’s edition of Ordinatio II, d.12, I have focused on the Lectura II and Reportatio
Parisiensis II-A versions instead, along with the philosophical commentaries where helpful.
21 Quaestiones super Praedicamenta Aristotelis q.15, n.7 (Bonaventure I, pp.384–385), Item,
divisum praedicatur de divientibus; II De anima: “substantia dividitur in materiam et for-
mam et compositum”; igitur quodlibet istorum est substantia.
32 Chapter 2
Scotus’s refutation begins by noting that what is divided is predicated of the

things resulting from the division only when the division is one of genus into
its several species. But Scotus denies that the sort of division Aristotle has in
mind in the text quoted from De anima is a division of genus into species, and
his reason is fascinating in light of his later views about matter:
I say that that division of substance is not a division of genus into species,
because then matter would have a difference distinguishing it from form,
and it would be a being in act of itself, distinct from form, which is not true.22
Scotus has in mind here the traditional Aristotelian taxonomy according to

which a species is distinct from other species of the same genus in virtue of a
unique difference. For example, humanity is animality with the difference of
rationality. If matter is a species of substance then whatever is true of sub-
stance is true of matter. But a substance in its primary sense is a being per se, so
if the matter and form of a substance are substances, then they too are beings
per se. And this is precisely what Scotus thinks is not true of matter.
A text from QMet also indicates that Scotus at one time did not think that
matter could exist without form. Scotus devotes a full question to whether or
not matter is a being (QMet VII, q.5), but gives no indication one way or another
about its separability from form. But in QMet VII, q.6, Scotus seems to be
explicitly against the view that matter can exist on its own. He writes,
But that nature of substance is to exist per se separably from others, not
depending on others. But this does not pertain to matter except through
form.23
The meaning of this text is not transparent but one possible reading is that
only through form does matter exist per se and separably from others, and
therefore matter cannot exist apart from an inhering form. The meaning is
not transparent because it is not immediately clear how matter, per formam,
can exist per se separably from others. Scotus could have in mind here a seman-
tic distinction of which he elsewhere makes use, the distinction between
22 Quaestiones super Praedicamenta Aristotelis q.15, n.15 (Bonaventure I, p.387), Ad illud de II

De anima, dico quod illa divisio substantiae non est divisio generis in species, quia tunc
materia haberet differentiam distinguentem ipsam a forma, et esset ens actu de se dis-
tinctum a forma, quod non est verum.
23 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.6, n.16 (Bonaventure IV, p.144),
Sed illa ratio substantiae est “per se existere separabiliter, ab aliis non dependens ab aliis.”
Sed istud non convenit materiae nisi per formam.
denominative and essential predication.24 Essential predication predicates

some essential feature of a subject, whereas denominative does not. “Matter is
a substance” is false essentially, since matter does not exist per se separably
from others; but it can be true denominatively, for example when matter is a
part of a material substance. Denominative predication permits predicating of
a part of a substance a feature of a whole substance. “The musical man builds
a house” predicates housebuilding denominatively of the musical man and not
essentially, since per se a musical man is not a builder, but together with the
skill of building he is nevertheless a builder. Likewise “matter exists per se sepa-
rably from others” predicates per se separable existence denominatively of
matter and not essentially, since only when informed by substantial form
and made thereby a part of a composite does matter exist per se separably
from others. This distinction provides a way of understanding what Scotus
could mean when he says that matter through form exists per se separably from
others. But it reinforces the reading of the passage according to which Scotus
thought that matter could not exist apart from form.
This QMet text taken with Quaestiones super Praedicamenta Aristotelis q.15, n.7
establish, I take it, that at some point in his career, to the extent that Scotus had a
view about whether or not matter is separable from form, it was that it is not.
But at which point? And for how long? Quaestiones super Praedicamenta
Aristotelis is regarded as a very early work, but the editors of the critical edition
of QMet have argued that QMet V–IX actually belong to a quite late period
of Scotus’s career.25 There is a complication, however: QMet does not yield a
consistent doctrine of matter. Whereas QMet VII, q.5 is indeterminate about
the precise nature of matter’s separability from form, and QMet VII, q.6 seems
to argue against matter’s separability, there is at least one text elsewhere in
QMet, namely, VIII, q.4, nn.30–31, in which Scotus asserts, without argument,
24 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.5, nn.21–22 (Bonaventure IV,
pp.136–137). In this text Scotus actually distinguishes two kinds of denominative predica-
tion, accidental denominative predication and another kind of denominative predica-
tion, where the distinction is based on what kind of part the subject of the denominative
predication is. Matter is an essential part of a substance and unites with substantial form
in an essential unity, whereas a categorical accident is not and does not.
25 Bonaventure III, p.xliii; see also Stephen Dumont, “The Question on Individuation in
Scotus’ Quaestiones super Metaphysicam,” in Via Scoti: Methodologia ad mentem Ioannis
Duns Scoti, vol. I, ed. Leonardo Sileo (Rome: Edizioni Antonianum, 1995), pp.193–227; and
Timothy Noone, “Scotus’ Critique of the Thomistic Theory of Individuation and the
Dating of the Quaestiones in libros Metaphysicorum VII, q.13,” in Via Scoti: Methodologia
ad mentem Ioannis Duns Scoti, vol. I, ed. Leonardo Sileo (Rome: Edizioni Antonianum,
1995), pp.391–406.
34 Chapter 2
that matter can exist without any form at all.26 This inconsistency suggests to
me that the composition of books V–IX of QMet was a stop-and-go affair; it is
more than odd that Scotus would deliver a negative opinion on matter’s sepa-
rability in VII, q.6, and then simply assert the contrary in VIII, q.4, if these
books had been written as a single sustained work of philosophy. The inconsis-
tency seems less odd, however, on the supposition that Scotus composed
Reportatio II-A d.12, q.3—the text in which he argues at length for matter’s total
separability—after writing QMet VII, q.6 but before writing QMet VIII, q.4.
Whether this conjecture can ultimately be sustained or not, the doctrine of
matter assumed in at least one question of QMet and argued for in Reportatio
II-A d.12, q.2 clearly represents a dramatic departure from Scotus’s early views
about the nature of matter.
iii Particular and Total Separability
Scotus undoubtedly thought later in his career that matter could exist without
being informed. The texts for this view are Reportatio II-A, d.12, q.2 and what is
printed as Ordinatio II, d.12, q.2 in Wadding VI. Here I focus exclusively on the
first, given the dubious status of the second. Scotus gives four arguments for
the conclusion that “matter can exist without every form.”27 It is useful up front
to distinguish more precisely two independent theses about matter’s separa-
bility from form, what I will call for ease of reference particular separability
and total separability:
[Particular separability] For any individual matter x and for any indi-
vidual form y, it is possible that x exists without being informed by y.
[Total separability] For any individual matter x it is possible that x exist

and there is not a form y such that y informs x.28
26 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q.4, nn.30–31 (Bonaventure IV,
p.498). I discuss this text in Chapters 3 and 4.
27 Reportatio Parisiensis II-A, d.12, q.2, n.4 (Wadding XI, p.321), materia potest esse sine omni
forma.
28 Particular and total separability are logically equivalent to Cross’s (6*) and (6**), respec-
tively, presented on p.23 of The Physics of Duns Scotus. (6*) says, “For every individual
matter m and every individual form f, it is possible that m exists without f. ” (6**) says, “For
every individual matter m, it is possible that, for every individual form f, m exists without
f.” Total separability and (6**) are equivalent since ~∃xφ is equivalent to ∀x~φ by quanti-
fier negation. Expressed formally, where “Mn” abbreviates “n is an individual matter,”
Scotus’s expression of his view that matter can exist without every form (omni
forma) is ambiguous between particular and total separability. Of the four
arguments Scotus presents, the first would, if sound, establish only particular
separability, while the last three would establish total separability.
The first argument, [C], runs as follows:
[C1] For any absolute (i.e., non-relational) thing, x, and for any absolute
(i.e., non-relational) thing, y, if x is prior to y, then x can without
contradiction exist without y.
[C2] Matter and form are absolute things.
[C3] In a relevant sense matter is prior to form.
[C4] Therefore, matter can without contradiction exist without form.29
[C] is a valid argument for particular, but not total, separability. [C2] is uncon-
troversial. As for [C3], there is a sense in which form is prior to matter—accord-
ing to Scotus form is prior to matter in the essential order of eminence or
perfection. But the relevant sense of priority in [C3] is an essential order of
dependence. In De Primo Principio II.32, Scotus says, “As for one being inde-
pendent of the other, it would seem that matter is prior, for the contingent and
informing cause seems to depend upon the permanent and informed, since we
think of what can be formed before we think of its form.”30 But [C1] is more
controversial. We can assume that the sense of priority intended in [C1] is
exactly the same as that intended in [C3]—otherwise the argument would be
invalid. So the sense of priority intended in [C1] is an essential order of depen-
dence. Scotus distinguishes several kinds of essential orders of dependence
“Fn” abbreviates “n is an individual form,” and “Rnm” abbreviates “n informs m,” particular
separability says ∀x(Mx→(∀y♢~(Fy∧Ryx))). On the same scheme of abbreviation, total
separability says ∀x(Mx→(♢~∃y(Fy∧Ryx))).
29 Reportatio Parisiensis II-A, d.12, q.2, n.4 (Wadding XI, p.321), Ideo dico quod compositum
generabile per se habet duas partes realiter diversas, et materia potest esse sine omni
forma. Quod probo. Primo sic; quia absolutum prius absoluto alio, potest sine contradic-
tione esse sine illo; materia est absolutum aliud a forma, et prius; igitur sine contradic-
tione potest esse sine illa. “So I say that the generable composite as such has two really
distinct parts, and matter can exist without every form. Which I prove, first, in this way.
An absolute thing [which is] prior to another absolute thing can without contradiction
exist without that other. Matter is an absolute thing other than form, and it is prior [to
form]. Therefore, without contradiction [matter] can exist without [form].”
30 De Primo Principio II.32 (Wolter, p.26), Videtur tamen materia prior secundum indepen-
dentiam, quia contingens et informans videtur dependere a permanente et informato,
quia informanti praeintelligitur formabile. (Wolter’s translation.).
36 Chapter 2
and, in De Primo Principio I.8, says that in at least some—and perhaps all31—
that which is essentially prior cannot exist without that which is essentially
posterior to it.32 This does not on the surface look consistent with [C1]. But [C1]
does not say that the prior can exist without the posterior; instead it says that
the prior can without contradiction exist without the posterior. The suggestion,
then, is that it could turn out to be fully consistent that x cannot exist without y
and that x can without contradiction exist without y. The suggestion may not at
first glance have the sparkle of cogency, but the fact that Scotus thinks this is so
is perfectly corroborated by another line from De Primo Principio, taken from
the same paragraph from which the citation above was taken. Here Scotus says
that, while the prior produces the posterior necessarily and consequently
could not exist without it, nevertheless, on the assumption that the posterior
does not exist, the prior will be, without the inclusion of a contradiction (prius
erit sine inclusione contradictionis). The phrase, Without the inclusion of a con
tradiction, suggests that the relevant modality here is logical possibility: it is
logically possible that the prior exist without the posterior. But when Scotus
also says, in the very same paragraph, that the prior cannot exist without the
posterior, he must be thinking of something other than logical impossibility—
otherwise the text is gibberish. I hope it is not gibberish, but it is indisputably
hard to understand, and a fully satisfactory treatment of it would involve delv-
ing very deep into the mines of Scotus’s modal theory. Here we can say (evad-
ing the really hard work), operating on the principle of charity, that in argument
[C] the sense of possibility used in the premises is exactly the same as in the
conclusion, so [C] shows that it is at least logically possible for matter to exist
separate from form. This argument only establishes particular separability,
however—the argument would still hold if it turned out that matter could per-
sist through a succession of substantial forms but always needed to be substan-
tially informed.33
31 I say, “perhaps all,” because the text, quoted below, is vague about this.
32 De Primo Principio I.8 (Wolter, pp.4,5), [L]icet prius necessario causet posterius et ideo
sine ipso esse non possit, hoc tamen non est quia ad esse suum egeat posteriori, sed e
converso; quia si ponatur posterius non esse, nihilominus prius erit sine inclusione con-
tradictionis. “Although the prior necessarily were to produce the posterior and so could
not exist without it, nevertheless this is not because it requires the posterior for its exis-
tence, but the converse. For if the posterior be posited not to exist, still the prior will exist
without the inclusion of a contradiction.” (My translation. Wolter’s makes for nicer read-
ing but it does not draw out the parallel between this text and Reportatio Parisiensis II-A,
d.12, q.2, n.4. on which I am here relying.).
33 Richard Cross, The Physics of Duns Scotus, pp.24–25.
The second, third, and fourth arguments aim to establish total separability.
We will examine the second and third.34 Here is the second argument, [D]:
[D1] For any x, if God creates x and there is some y such that y is not part
of the essence of x and y is ordinarily a secondary cause of x, then it
is possible that God creates x without the secondary causality of y.
[D2] God creates matter, and form is not part of the essence of matter,
and form is ordinarily a secondary cause of matter.
[D3] Therefore it is possible that God creates matter without the sec-
ondary causality of form, that is, it is possible that God creates mat-
ter informed by no form at all.35
[D] is also a valid argument and it is an argument for total separability.

Whatever the secondary causality of form over matter amounts to, it involves
form’s informing matter; so to show that God can create matter without form’s
secondary causality is to show that God can create matter without any form at
all. The three conjuncts of [D2] are uncontroversial within Scotus’s milieu.
[D1] says that God can do all by himself whatever he normally does in coopera-
tion with something else, and this too is uncontroversial. But one might object
that [D2] is incompatible with [C3], since if ordinarily form causes matter
(according to [D2]) and is therefore prior to matter, and if ordinarily matter is
prior to form (according to [C3]), then Scotus seems to commit himself to a
vicious circle of priority relations. Fortunately for Scotus, the senses of priority
here are different. Matter is prior to form, Scotus tells us, in the sense that it is
the subject in which form comes to be and without which form would not
(naturally) come to be. Form cannot be a (secondary) efficient cause of matter,
and so not in this way prior to matter, since (a) efficient causing is the action of
an agent by which matter and form are united, so form cannot be an efficient
34 See Richard Cross, The Physics of Duns Scotus, pp.23–24 for an excellent critique of
Scotus’s fourth argument, which I find persuasive. Cross’s insight is that Scotus’s argu-
ment relies on an invalid inference of total separability from particular separability.
35 Reportatio Parisiensis II-A, d.12, q.2, n.5 (Wadding XI, p.322), Secunda ratio ad idem,
quidquid Deus creat respectu absoluti per causam secundam, quae non est essentia rei,
potest Deus immediate sine causa secunda causare; sed esse in materia causat per for-
mam, et ipsa non est de essentia materiae; igitur potest esse causare in materia immedi-
ate sine forma. “The second reason for the same concusion: regarding absolute things,
whatever God creates through a secondary cause, which is not of the essence of a thing,
God can cause immediately without a secondary cause. But God causes existence in mat-
ter through form, and form is not of the essence of matter; therefore God can cause exis-
tence in matter immediately without form.”
38 Chapter 2
cause, and (b) Scotus thinks that matter exists prior to its receiving form. But
form is prior to matter in the sense that matter depends on form for its being
the matter of some kind of composite substance. So while form does not
explain matter’s existence, it does explain matter’s being such. Therefore mat-
ter’s priority to form and form’s priority to matter are two sorts of priority, and
therefore Scotus is committed to no vicious circle of priority between form and
matter.
The third argument, [E], is as follows.
[E1] For any x, if God immediately creates x then God can immediately
conserve x.
[E2] God immediately creates matter.
[E3] So God can immediately conserve matter.36
Given a proper understanding of “immediate creation,” [E] is a valid argument

for total separability. The efficient causality of every natural agent presupposes
matter, as we saw in Chapter 1. But this entails that no natural agent can effi-
ciently cause matter. The creation of matter “is not under the power of created
nature, because created nature can produce nothing except by having presup-
posed something.”37 And, of course, Scotus thinks that matter is a created
thing; it is not co-eternal with God. So it is something produced and only God
can produce it. But if God creates matter all by himself, Scotus reasons, there is
no good reason to think that he cannot continue to conserve matter all by him-
self, where conservation is the activity by which something continues to exist
once it has been created. Arguments [D] and [E] together entail that God could
produce matter without producing form, and conserve that matter on its own.
Richard Cross has analyzed Scotus’s arguments for total separability and has
found them wanting.38 He rejects Scotus’s fourth argument since it involves an
36 Reportatio Parisiensis II-A, d.12, q.2, n.6 (Wadding XI, p.322), Tertia ratio ad idem, quod
Deus immediate creat, immediate potest conservare; sed materiam immediate creat, quia
materia est quid creatum: non enim est ens omnino increatum, et non est subest virtuti
naturae creatae, quia nihil potest natura creata producere, nisi aliquo praesupposito; igi-
tur Deus potest immediate materiam conservare sine entitate alia absoluta. “The third
argument for the same conclusion: that which God immediately creates, he is able imme-
diately to conserve. But he immediately creates matter, because matter is a created thing;
for it is not a totally uncreated being, and it is not under the power of created nature,
because created nature can produce nothing except by having presupposed something.
Therefore God can immediately conserve matter without another absolute entity.”
38 Richard Cross, The Physics of Duns Scotus, p.26.
invalid inference from particular to total separability. He notes that the first
argument only establishes particular separability. But he also rejects the sec-
ond and third arguments, which rely on assumptions about God’s power (in
particular, that God can create on his own whatever he can bring about through
secondary causes, and that God can conserve whatever he can create), about
creaturely power (in particular, that no created agent can produce matter), and
about the nature of matter (in particular that its essence does not include
form), to argue that God can create and conserve matter without form, and
therefore that matter and form are totally separable. Cross writes of the third
argument,
The argument is deceptively appealing. It is clearly the case that individ-

ual matter cannot be produced by any created agent at all; this is why the
claim that God creates individual matter directly looks convincing. But it
overlooks the possibility that the existence of form might be (as Aquinas
would reason) a necessary condition for the existence of individual mat-
ter, even if form does not itself have any efficiently causal role in the pro-
duction of individual matter. In this case, it would be false to claim that
God creates individual matter ‘directly.’39
For Aquinas, matter is not a per se term of creation;40 strictly speaking it is

concreated with form as a part of a composite substance,41 and it has actuality
insofar as it is under form.42 Aquinas could accept that matter and form, to the
extent that they can be defined, have distinct essences, and he could accept
that God can bring about on his own whatever he brings about through
secondary causes, and still deny that matter is totally separable from form. But
this is because Aquinas denies that matter is an actual thing really distinct
from form. It is no wonder, therefore, that Aquinas would object to Scotus’s
second and third arguments for total separability, which presuppose Scotus’s
earlier arguments that matter and form are really distinct things. Cross’s objec-
tion to Scotus therefore seems just a tad unfair, insofar as it does not take into
account Scotus’s and Aquinas’s broader ontologies of matter and how these
relate to the arguments for total separability. A more fundamental dispute
39 Richard Cross, The Physics of Duns Scotus, pp.25–26. Cross does not discuss the second
argument, but his conclusion about Scotus on the separability of matter from form is
clearly intended to apply to Scotus’s whole effort to establish total separability.
40 Aquinas, Quaestiones disputatae de potentia dei q.3, a.5, ad.5 (Turin, 1953).
41 Aquinas, Summa theologiae Ia, q.45, a.4 (Turin, 1952).
42 Aquinas, Quaestiones disputatae de potentia dei q.3, a.5, ad.3 (Turin, 1953).
40 Chapter 2
about matter between Aquinas and Scotus concerns Scotus’s claim that there
is a sense in which matter is both potency and act: it is an actual being whose
nature it is to be in subjective potency to be perfected by any form, whereas for
Aquinas matter is pure potency whose actuality is derived entirely from form.43
I conclude then that given Scotus’s understanding of the nature of matter
and God’s omnipotence, the second and third arguments for the total separa-
bility of matter from form are successful.
To sum up: I have argued, first, that matter is an entity whose nature is to be
passive power. Second, Scotus uses the distinction between objective and sub-
jective potency to describe the way in which matter can be in potency to
receive any form, and yet have an actuality of its own, prior to and indepen-
dent of form. Third, because matter is in act it can persist through substantial
change, and, fourth, because its nature is passive power in subjective potency
to any form, it is the sort of persisting thing required for substantial change.
I have argued that all four of these conclusions about the nature of matter were
in place in Scotus’s philosophy of matter before he penned his arguments for
total separability, and even that Scotus advanced these conclusions while he
still denied total separability. As we will see in Chapters 3 and 4, however, the
total separability of matter from form furnishes Scotus with a crucial premise
in the argument that a substance is really distinct from matter and form.
43 Aquinas, Scriptum super sententiis II, d.17, q.1, a.2, corp. (Parma, 1856).
Chapter 3
How Matter and Form Compose a

Substance—Part I
Like most hylomorphists, Scotus thinks that matter and form—e.g., Mole’s
body and his sensitive soul—are parts of a substance. But he also recognizes
what we would think of as ordinary parts, or spatially extended parts, things
like Mole’s heart, ears, and paws. He calls the former the essential parts, and the
latter the integral parts, of a substance. This chapter and the next focus on
essential parts, and Chapters 5 and 6 focus on integral parts.1
In this chapter and the next I explore what it means for matter and form to
compose a substance. I have two aims. First, I show that for Scotus, matter and
form compose a substance by causing a substance. I examine Scotus’s reasons
for thinking this and I explain what this means. Second, I show that for Scotus,
the parthood relations of matter and form to substance just are causal rela
tions—causal relations of a peculiar sort.
Scotus’s basic idea is that the parts of a substance are united in some special
way in which the parts of other kinds of wholes are not, and that this special
unity demands that a substance be really distinct from its material part, its
formal part, and from both of these taken together. Each of matter and form
can exist on its own (if only by divine power), and thus satisfy the existence
conditions of an aggregate—a whole that is identical with its parts—but not
compose a substance.2 So, Scotus asks, what happens when matter and form do
compose a substance? Parthood is a relation, and Scotus thinks that in general
(with important exceptions discussed below) changes in relative properties
must be explained by changes in absolute (non-relative) properties.3 So for a
1 For a careful yet entertaining introduction to these senses of part in medieval philosophy, see
Calvin Normore, “Ockham’s Metaphysics of Parts,” The Journal of Philosophy 103:12 (2006),
pp.737–754. Unless otherwise specified, in this chapter by part I mean essential part.
2 In this chapter by form I mean substantial form unless otherwise noted. As Scotus discusses
the unity of matter and form, he seems to mean by matter, prime matter.
3 For Scotus, a relative property or relation is an accident whose nature is to be ad aliquid,
toward something. An absolute entity is anything falling under the categories Substance,
Quality, and Quantity, and prime matter (and God). Fregean logic gives us polyadic proper-
ties whose logical form suggests that one polyadic property belongs to two (or more) sub-
jects at once: for example, the property R referred to in the formula Rab is a property of or
between a and b. Scotus and the medievals thought of relations quite differently. With some

42 Chapter 3
thing to become a part—to acquire a parthood relation—some absolute

change must occur. He argues that the only absolute change that can explain
how matter and form acquire parthood relations is the production of a new
entity, and goes on to argue that this new entity is and can only be a substance.
Developing the Aristotelian idea that matter and form exercise certain kinds of
causality, Scotus argues that matter and form are the causes that bring about
the substance of which they (once the substance is brought about) are parts.
Finally, Scotus asserts that what it is for matter and form to be related to a sub-
stance as essential parts just is to be causally related to a substance—in slogan
form, parthood relations are causal relations.
The arguments for this view are scattered through several texts, including
QMet VIII, q.4, Ordinatio III, d.2, q.2, Lectura III, d.2, q.2, and Reportatio III-A
d.2, q.1. Philosophically these are consistent with one another. The three ques-
tions from the Sentences commentaries have a very similar argumentative
structure, which Richard Cross has admirably analyzed.4 But the QMet ques-
tion is structured rather differently. Overall it is less organized but philosophi-
cally richer than its counterparts in the Sentences commentaries. Taking these
texts together, but letting the QMet VIII, q.4 account set the dialectical agenda,
I present what I take to be the most interesting reasons for thinking that matter
and form compose a substance only when they cause a substance that is really
distinct from them, and that the parthood of matter and form simply consists
in the causality of matter and form.5 Readers of Scotus on this topic have the
benefit of Ockham’s criticisms of Scotus’s views. I have found it helpful for
exceptions, a real relation inheres in a subject and is toward its term. In cases of mutual real
relations, for example similarity relations between two substances, there is one similarity
relation in one substance, and another in the other substance. For further background on
medieval theories of relations, see my “Relations without Forms: Some Consequences of
Aquinas’s Metaphysics of Relation,” Vivarium 48:3–4 (2010), pp.279–301, and especially Mark
Henninger, Relations: Medieval Theories 1250–1325 (Oxford: Oxford University Press, 1989).
4 Richard Cross, “Duns Scotus’s Anti-Reductionistic Account of Material Substance,” Vivarium
33:2 (1995), pp.137–170. Cross briefly discusses Quaestiones super Libros Metaphysicorum
Aristotelis VIII, q.4 in an appendix, in which his purpose is to establish the consistency of the
Quaestiones super Libros Metaphysicorum Aristotelis account with that of Ordinatio, against
the opinion of Mauritius de Portu (Wadding IV, p.757). For comparison of Scotus and
Ockham on the sense in which matter and form compose a substance, see Richard Cross,
“Ockham on Part and Whole,” Vivarium 37:2 (1999), pp.143–167.
5 Scotus thinks that most material substances actually have more than one substantial form. In
this chapter I will write as though one substantial form is all it takes to make a complete
substance, but this should be understood merely as shorthand, to be enriched by the subse-
quent discussion of Scotus’s pluralism about substantial forms in the following chapters.
How Matter and Form Compose a Substance—Part I 43
understanding Scotus to draw heavily on Ockham’s criticisms, and my account

of Scotus here reflects that engagement with Ockham.
Scotus simply takes it for granted that matter and form are parts, but it is not
obvious that a hylomorphist should think of both matter and form as parts. To
make straight the way of the Subtle Doctor, therefore, I start by motivating and
then presenting an argument for the conclusion that forms are not parts at all,
showing how Scotus would respond to it. Then I will present two ways in which
it makes sense to think of matter and form as essential parts of a substance.
i Forms as Parts
One way of getting a handle on the sense in which matter and form are parts of
a substance is, first, to have some sense of the reasons why one might be
inclined to distinguish form and matter in the first place. In addition to the
argument from change analyzed in Chapter 1, there is the heuristically more
useful idea that a hylomorphist draws the distinction between form and mat-
ter whenever there is good reason to distinguish between a thing and what a
thing is made of.6 When the brazen sphere becomes a brazen cube, for exam-
ple, bronze persists but sphericity does not. The sphere is therefore not identi-
cal with its bronze. A hylomorphist explains the distinction between the
sphere and what it is made of by positing that the sphere has something in
addition to its bronze. The hylomorphist calls what a sphere is made of its mat-
ter, and the something extra the sphere’s form. Similar (but more complicated)
reflections yield a division of form and matter for organisms. When Mole dies
his body remains as a corpse, so Mole is not identical with his body; the body
is Mole’s matter, his sensitive soul his substantial form.7
The matter and form of a substance are its essential parts. Essential
parts were contrasted with what were called integral parts, which are,
roughly, extended parts.8 The integral parts of Mole are his organs, blood,
6 As I will argue in Chapter 9, however, this second way of motivating hylomorphism is reduc-
ible to the argument from change.
7 By the Aristotelians’ lights, Scotus’s included, a brazen sphere or cube is not a genuine sub-
stance, since the forms of sphere or cube are not substantial forms; they are accidental modi-
fications of the bronze. I use the example, as Aristotle and many of his commentators use it,
merely heuristically. The application of this hylomorphic analysis to paradigmatic
Aristotelian substances, viz., living organisms, is much more complicated than this para-
graph suggests. I consider these complications in detail in Chapters 5 and 6.
8 The origin of the distinction is probably from Aristotle, Metaphysics VII 1034b 34–36 (Barnes
II, p.1634) where the Philosopher contrasts quantitative with substantial parts.
44 Chapter 3
limbs, nerves, and so on. His essential parts are, on the part of matter, the body,
and, on the part of form, soul, the something extra in virtue of which the body
is living.
In Aristotelian terminology, the form of Mole is that which actualizes the
potentiality of the things that can be mole parts,9 specifically their potentiality
to compose a mole. So form is one answer to the question—the Special
Composition Question, as Peter Van Inwagen puts it—about how things
become parts, or how things compose.10 But it can quickly be shown that, for
Scotus, form cannot be the only answer to the question. He assumes that form
itself is a part. If it becomes a part by (together with matter) being actualized
by an additional form, then we must ask how this second form becomes a part.
Applying the same answer will generate an infinite regress, so we should deny
in general that forms become parts through additional forms.
If not an additional form, then, in virtue of what do form and matter com-
pose a material substance? One could evade the question altogether simply by
denying that forms are parts—what is not a part does not compose, in the
sense of compose intended in the above question. Mark Johnston does just this
in his recent version of hylomorphism. Johnston uses an argument very similar
to the one just given to conclude that a form is not a part of the thing whose
form it is:
Of any item that has parts we may inquire as to what principle unifies
those parts into the whole that is the complex item. The principle had
better not be merely another part, for the question would remain:
Consider that part along with the other parts; what relation is such that
its holding of all these parts gives us the whole? And that would be the
principle we seek.11
Notice, however, that Johnston’s argument only has teeth on the assumption
that form must be invoked whenever we attribute unity to two or more things,
and hence that only form can explain how things become parts of another
thing. For Johnston the assumption makes sense because his forms just are
9 I am using the cumbersome expression “things that can be mole parts” rather than “mole
parts” to make it clear that, despite ordinary usage, in which we would say that a detached
paw is still a mole part, strictly speaking something is a part only if it is part of a whole,
and a detached paw is not a part of a whole mole (even if it is still a paw, as Scotus thinks,
as I argue in Chapters 5 and 6).
10 Peter Van Inwagen, Material Beings (Ithaca: Cornell University Press, 1990), p.21.
11 Mark Johnston, “Hylomorphism,” Journal of Philosophy 103:12 (2006), p.652.
unifying principles. But Scotus recognizes several different kinds of unity, and
thinks that what it takes for two or more things to be unified is relative to a
kind of unity.12 Moreover Scotus thinks that two or more things are parts of the
same thing just insofar as they are unified in the relevant sense. As I argue
below, all it takes for things to be parts of what Scotus calls an aggregate is that
they exist.13 To be parts of what Scotus calls a unity of order, two or more things
must have essential dependence relations. (Essential dependence is a techni-
cal notion for Scotus, which I elaborate in detail in Chapter 5, but for now all
we need to know is that if y essentially depends on x, then y cannot exist if x
does not exist.) And, to foreshadow the claims of later chapters, to be parts of
an organism, organs must be unified in two senses: they must be both essen-
tially ordered and informed by a soul: essential order explains how organs
compose one body and soul explains how a body is one living body. Thus Scotus
recognizes several senses of unity according to which the things so unified are
not unified by form, and hence several ways in which things can become parts
without being unified by a form. So, while form and matter cannot be unified
through an additional form, on pain of infinite regress, they may be able to be
unified, and so be parts, in some other way. And, of course, Scotus thinks there
is some other way: matter and form become parts of a substance by being caus-
ally related to a substance.
Here it is worth forestalling a likely objection to my account thus far, namely
that Scotus actually makes the very mistake that Johnston avoided making: the
mistake of trying to make forms do the job of uniting matter and form. In the
Sentences commentaries Scotus calls the composite of matter and form
the “form of the whole” (forma totius).14 On the surface it is natural to think of
the forma totius as a kind of super-form, the form whose “matter” is a substan-
tial form and prime matter. And this indeed would fall prey to Johnston’s infi-
nite regress objection—don’t we need an additional forma totius uniting the
first forma totius with its matter and substantial form?
For Scotus, the answer is non, because the forma totius is not a form at all.
The expression forma totius has a history, having been used by Aquinas and
Avicenna, apparently as a gloss on one way in which Aristotle talks about form.
12 Ordinatio I, d.2, p.2, qq.1–4, n.403 (Vatican II, pp.356–357).

13 Lectura III, d.2, q.2, n.80 (Vatican XX, pp.102–103).
14 Ordinatio III, d.2, q.2 (Vatican IX, pp.139–160); Lectura III, d.2, q.2 (Vatican XX, pp.92–110).
This expression is used just once in the Quaestiones super Libros Metaphysicorum
Aristotelis account of the unity of matter and substantial form, but it is used to express a
view which Scotus himself rejects. Quaestiones super Libros Metaphysicorum Aristotelis
VIII, q.4, n.17 (Bonaventure IV, p.495).
46 Chapter 3
Aristotle at times equates form with essence, and says that the essences of
material things include matter.15 But Aristotle also talks about form as that
which together with matter composes the essence of a material thing.16 So we
have one use of form which simply means essence, and another use which
means part of an essence. Since Aristotle says that the essence of material
things includes matter, form in the sense of essence includes, as one of its parts,
form in the sense of part of an essence. Readers of Aristotle were then as now
confronted with the challenge of explaining the confusing use of terms. Form
as used of the essence of a thing came to be called forma totius, while form as
used of that which unites with matter to compose a substance came to be
called forma partis or, more commonly, simply the substantial form. One’s
view about what Aristotle meant by identifying essence with form determined
how one thought about the relation between forma totius and forma partis. For
example, Averroës interpreted Aristotle as holding that matter was no part of
the essence of a material substance. The essence instead is form alone. Averroës
seems to have held that the form of the whole is identical with the forma partis
or substantial form, and given his understanding of essence this is not surpris-
ing. As Aquinas presents him, Averroës thought that the expressions forma
totius and forma partis denote the same thing, a substantial form. But they
have different senses: the forma totius is the substantial form considered as
that by which a whole composite falls under a species, e.g., humanity, whereas
the expression forma partis bears the sense of the substantial form considered
as that which perfects matter and makes it something actual, e.g., the soul.
According to Aquinas’s Avicenna, on the other hand, matter is included in the
essence of a material substance. Accordingly, the relation between forma totius
and forma partis is not identity but is instead that of whole to part; forma partis
is the substantial form, and forma totius is the whole composite resulting from
the union of substantial form and matter.17 Aquinas himself seems to have
waffled between Averroës’s and Avicenna’s views, sometimes identifying forma
totius with the substantial form,18 and sometimes not.19 Ockham does not use
the term to express his own view.
15 Aristotle, Metaphysics V 1013a 26 (Barnes II, p.1600); Metaphysics VI 1025b 28–32 (Barnes
II, p.1620).
16 Aristotle, Metaphysics VII 1035a 32–34 (Barnes II, p.1635).
17 Aquinas, Sententia metaphysicae VII, l.9, n.8 (Turin, 1950).
18 Aquinas, Summa theologiae Ia, q.76, a.8, corp. (Turin, 1952); Summa contra gentiles II, c.72,
n.3 (Turin, 1961).
19 Aquinas, De ente et essentia, c.I, n.2 and c.II, n.5 (Turin, 1957).
Scotus is firmly on the Avicennan side in this respect: he includes matter in

the essence of a composite thing.20 Moreover, since Scotus thinks that matter
is really distinct from form, having its own actuality, he cannot hold that sub-
stantial form makes matter actual.21 For Scotus, then, forma totius is simply the
whole substance itself composed of matter and form—it is not a form at all.
Thus, we need not worry that Scotus has made the mistake of using forms to
unite form and matter. In this chapter I plan to avoid the medieval confusion
about the term form by using expressions like “the whole substance” and
“the substance itself” instead of “form of the whole.” This seems especially jus-
tified given the way in which Scotus distances himself from the expression in
QMet VIII, q.4.
ii Matter and Form as Essential Parts
There are at least two prima facie plausible ways of understanding how essen
tial is modifying parts in the expression essential parts. The first treats essential
as a certain sense of necessary. This view is suggested by Scotus’s claim that a
material substance essentially depends for its existence on the causality of mat-
ter and form, where essential dependence entails that it is not possible for the
substance to exist without the causality of matter and form.22 So we might
think of essential parts as the sort of parts that are necessary for the existence
of the whole they compose.
But this is not quite right. While it is true that matter and form are necessary
for the existence of a material substance, there are other parts of substances that
are not called essential parts but are, nevertheless, necessary for the existence of
the substance they compose (e.g., the blood, heart, brain, etc., of a human).
A better understanding of essential parthood is what we might call a defini
tional understanding. On this understanding, essential parts are parts of the
essence, where the parts of the essence in some sense correspond to the parts
of a real definition.23 Aristotle says, with Scotus’s endorsement, that the parts
20 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.16 (Bonaventure IV,

pp.311–326).
21 Lectura II, d.12, q.un (Vatican XIX, pp.69–101); Quaestiones super Libros Metaphysicorum
Aristotelis VII, q.5 (Bonaventure IV, pp.131–139).
22 A substance essentially depends on other things in addition to its matter and form, but
I am ignoring these for now.
23 A real definition (as opposed to a merely nominal definition) is an expression that picks
out a real essence.
48 Chapter 3
of a definition correspond to the parts of an essence.24 Also, Scotus argues that

there is real identity between an individual essence (such as Socrates) and an
essence as such (such as humanity),25 such that real definition is of the indi-
vidual substance and not just the essence as such. Finally, Aristotle himself
seems to say, and Scotus interprets him as saying, that a composite substance
is defined with reference to both matter and form.26 For Scotus, then, the
essential parts of a substance are just those entities posited to account for, in
rebus, the notes of a real definition, even if there is not a strict isomorphism
between the logical parts of a definition and the essential parts of a composite
substance. According to Scotus, the essential parts of a human are, on the side
of matter, a body composed of different kinds of organic substances such as a
heart substance, brain substance, and so on, and, on the side of form, one ratio
nal soul.27 In a full definition of a human, e.g., rational animal, where animal
can be decomposed into sensitive animate corporeal substance, Scotus argues
that one thing (the rational soul) is all it takes to account for—in the individual
human—all that is expressed by rational, sensitive, and animate.28
iii Degrees of Unity
Motivating Scotus’s thought on how matter and form compose material sub-
stance is the conviction that material substances are unified in some special
way in which, for example, the parts of an aggregate are not. Scotus’s label for
this special sort of unity is “unity per se.” Scotus also gives this label to things
24 Aristotle, Metaphysics VII, 1034b 20–21 (Barnes II, p.1633), “Since a definition is a formula,
and every formula has parts, and as the formula is to the thing, so is the part of the for-
mula to the part of the thing […].” Scotus quotes this text in Quaestiones super Libros
Metaphysicorum Aristotelis VII, q.16, n.22 (Bonaventure IV, p.316). In Scotus’s Latin, defini
tion is definitio (horismos in Aristotle’s Greek), formula is ratio (logos in Aristotle), and
thing is res (pragma in Aristotle).
25 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.7 (Bonaventure IV, pp.147–
156). For a brief but helpful discussion of this, see p.138 ff. of Richard Cross, “Duns Scotus’s
Anti-Reductionistic Account of Material Substance,” Vivarium 33:2 (1995), pp.137–170.
26 Aristotle, Metaphysics VIII, 1043a 14–19 (Barnes II, p.1646). Quaestiones super Libros
Metaphysicorum Aristotelis VII, q.16, n.29 (Bonaventure IV, p.315).
27 My account of body should strike readers familiar with Scotus as odd. I expand and
defend the claim in Chapters 5 and 6.
28 Ordinatio IV, d.44, q.1, n.4 (Wadding X, p.98); see also Richard Cross, “Philosophy of Mind,”
in The Cambridge Companion to Duns Scotus, ed. Thomas Williams (Cambridge:
Cambridge University Press, 2003), pp.263–284.
that are genuinely simple, such as angelic forms, and also to bare matter and
bare forms. Scotus denies that there could be a genuinely simple (i.e., indivisi-
ble) material object, since every material object is extended and therefore
divisible, even if by divine power alone. He thinks that simple per se unities are
more unified than complex per se unities, and any per se unity is more unified
than any unity per accidens, such as an aggregate or the composite of a sub-
stance and an accidental form. Scotus thinks therefore that unity comes in
degrees; a genuinely simple, formally identical object enjoys a high degree of
unity, whereas an aggregate, which Scotus defines as an object which is identi-
cal with its parts, has the lowest degree.
Scotus uses several Latin words to name a whole which is identical with its
parts: aggregatio, congregatio, acervus, and cumulus. These words have differ-
ent (though related) meanings but I translate each of them as “aggregate” for
convenience’s sake and for the following additional reasons. “Aggregate” is free
of connotations of a certain sort of arrangement or structure of the things
composing the aggregate, and Scotus’s uses of these four Latin terms suggests
that he thinks that the spatial arrangement of the parts of an aggregate adds
nothing over and above the parts themselves. “Heap” or “mound,” by contrast,
do connote a certain sort of arrangement. You wouldn’t call my briefcase and
the Eifel Tower a heap or a mound, for example. But an aggregate, in Scotus’s
sense, is indifferent to every spatial arrangement: the same stones can form a
pile and then form a wall (a dry stone wall, to keep the example simple), and
the same geese can form a gaggle or not. Scotus’s arguments for his thesis that
a material substance is some third thing, not identical with matter and form,
require a use of “aggregate” which is indifferent to the particular spatial
arrangements of parts. If Scotus used aggregatio or its semantic cousins in a
way that connoted the arrangement of parts, however, then an aggregate is not
identical with its parts since the parts can exist without their arrangement
(disassemble the dry stone wall and you still have the stones). Scotus’s aggre-
gates, therefore, seem to be what would nowadays be called mereological sums,
whose existence conditions are no more stringent than that their parts exist.29
But there are kinds of wholes besides aggregates or mereological sums, and
one way that Scotus describes the difference between kinds of wholes is to say
that relative to each other, different kinds of wholes are more or less unified.
The difference between the extremes of an aggregate and a genuine simple is
clear and it seems a plausible use of the comparatives more and less to say that
29 For an introduction to twentieth century mereology, including its historical development,

see the early chapters of Peter Simons, Parts: A Study in Ontology (Oxford: Clarendon
Press, 1987).
50 Chapter 3
the first is more unified than the second and that the second is less unified
than the first. But Scotus also holds that among genuinely complex objects
there are different degrees of unity; in particular, Scotus holds
[SO1] A substance is more unified than an aggregate.30
I think that there is some intuitive sense to the idea that one complex thing can
be more unified than another. One might begin crudely by thinking that parts
are not very unified if they compose an aggregate, such as the sum of the Eiffel
Tower and my briefcase, are slightly more unified if the object they compose is
non-arbitrary but ephemeral, such as the water vapors composing a cloud, or
the grains of sand composing a sandcastle, and are even more unified if the
object they compose is sturdy, such as the parts of an automobile or the parts
of Mole. On this picture one might think that a degree of unity corresponds to
the spatial proximity of parts, or to the difficulty of separating (without simply
destroying) the parts. But when it comes to the greater degree of unity of a
substance relative to an aggregate, Scotus’s ideas are rather different from
these. For Scotus, the unity of a substance consists in its being a thing in addition
to its parts, and in this sense it is more unified—more a thing that is one—than
an aggregate. He writes,
But it is shown that the entity of a whole [substance] is other than the
entity of the parts, matter and form, such that it is a third entity: because
otherwise there would not be a difference between unities which consti-
tute one per se and unities which constitute one by aggregation, which is
against the Philosopher in Metaphysics VIII, where he says that that
which is one like an aggregate is nothing other than its parts.31
Scotus thinks then that [SO1] implies
30 In what follows I use numbered premises to make Scotus’s reasoning clear. I also intro-
duce a number of premises shared by Ockham and Scotus, and some held by Ockham and
not Scotus. Shared premises begin “SO,” premises that are Ockham’s alone begin “O,” and
Scotus’s own begin “S.”
31 Lectura III, d.2, q.2, n.80 (Vatican XX, pp.102–103), Quod autem entitas totius sit alia ab
entitate partium materiae et formae, ita quod sit tertia entitas, ostenditur, quia aliter non
esset differentia inter unita quae constituunt unum per se et inter unita quae constituunt
unum aggregatione—quod est contra Philosophum VIII Metaphysicae, ubi vult quod
illud quod est unum sicut cumulus, nihil aliud est quam partes […]; Aristotle, Metaphysics
VIII 1045a 7–33.
[S] A substance is something in addition to its essential parts.
In the quoted text above, Scotus seems to think that [S] directly follows from
[SO1]. This does not seem right to me and it is certainly not obvious. But there
is a longer route to [S] that is more compelling, and to this I now turn.
iv Existence is not Enough
Ockham is happy with [SO1] but denies [S]. He argues that
[O] A substance is identical with its essential parts.
How then does Ockham distinguish between the unity of substances and the
unity of aggregates? His answer is that a material substance is composed of
essential parts one of which is in potency and the other is in act, where these
parts are the essential parts composing a thing falling under just one genus, i.e.,
the genus of substance. By contrast, wholes that have less unity than
substances have parts such that each part is itself a whole under some
genus, whether (a) the parts are of the same genus (e.g., an aggregate of several
substances) or (b) the parts are of different genera (e.g., an accidental unity
of a substance and its accident).32 So, while Scotus would agree with
Ockham that
[SO2] Matter is essentially potency and form is essentially act,33
the former does not but the latter does make [SO2] the reason for the greater
unity of a substance. For Ockham, no further explanation of the unity of sub-
stance is needed, even though he agrees with Scotus that
[SO3] It is possible that matter exists and form exists and that they do
not compose a substance.34
32 Ockham, Quaestiones Variae VI, a.2, ll.150–162 (OTh VIII, pp.213–214).

33 Ockham like Scotus thinks that matter is not pure potency; it has an act of its own, an act
whose essential feature is to be in potency to form.
34 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q.4, n.14 (Bonaventure IV,
p.493), […M]ateria et forma, si intelligantur non unita, non est contradictio quod
utrumque intelligatur in se esse, et compositum non erit.
52 Chapter 3
More precisely, Scotus and Ockham hold that there is nothing repugnant about
matter’s existing without being informed at all, and of form’s existing without
being in any matter; in other words, both think that matter and form are totally
separable.
Since matter and form are essentially potency and act, respectively, neither
loses its essential nature when each exists totally separately. [SO2] holds,
therefore, even when matter and form do not compose a substance, and Scotus
takes this as sufficient evidence that [SO2] cannot be the reason why matter
and form are more unified than the parts of an aggregate. Against [O] Scotus
argues that if a substance were identical with its essential parts then the only
way for it to be corrupted would be for either matter or form or both to be
destroyed. But this is inconsistent with [SO3]: if the matter and form of some
particular substance were separated from each other and each continued to
exist, the substance would be corrupted but neither essential part would be,
just as a bicycle would cease to exist if its parts were disassembled. Likewise, if
a substance were identical with its essential parts, then the only way for a sub-
stance to be generated would be for either matter or form or both to be pro-
duced. But this too is inconsistent with [SO3].35
The argument is not quite fair to Ockham’s position, however. Ockham can
and does hold both that a substance is identical with its essential parts and
also that there is some genuine difference between matter and form’s compos-
ing a substance and matter and form’s merely existing. Matter and form exer-
cise their natural aptitudes when form actualizes the potency of matter; they
do not exercise their natural aptitudes when form does not actualize the
potency of matter. Ockham’s commitment to [O], [SO2], and [SO3] is therefore
unproblematic provided he has some account of what explains why matter
and form sometimes exercise their aptitudes and sometimes do not.
v Making a Difference
Unlike Scotus, Ockham thinks that what needs explanation is not the unity of
the essential parts, but their disunity, when it happens that matter and form
35 Ordinatio III, d.2, q.2, nn.74–75 (Vatican IX, pp.149–150); Lectura III, d.2, q.2, nn.81–82
(Vatican XX, p.103); Reportatio Parisiensis III-A, d.2, q.1, n.5 (Wadding XI, p.428). Ockham
would not accept the arguments from generation and corruption, however, because he
thinks that strictly speaking there is not generation or corruption unless there is the pro-
duction or destruction of an essential part, de novo. See Quaestiones Variae VI, a.2, ll.192–
201 (OTh VIII, p.215).
exist, do not compose a substance, and are not the matter or form of any other
substance. But both accept total separability and therefore both think that one
needs some account of what changes when matter and form compose a sub-
stance, or when matter and form cease to compose a substance (but go on
existing).
Scotus argues that
[S1] The sort of change that occurs when matter and form begin to com-
pose a substance is the production of some entity (other than mat-
ter and form).
Ockham does not accept [S1], because he understands [S1] to mean that the
produced entity will either be a part of a substance or the substance itself, both
of which are inconsistent with [O]. (Scotus understands [S1] in this way, too.)
However, Ockham does accept a modification of [S1] which [S1] entails and
which Scotus therefore should also accept:
[SO4] The sort of change that occurs when matter and form begin to
compose a substance is either the production of some entity or
the destruction of some entity.
Ockham argues for [SO4] in the following way. He begins by noting that change
occurs in three different ways: through temporal passage, local motion, and
the production or destruction of an entity. If the first two are ruled out Ockham
has [SO4]. Temporal passage is not relevant, but Ockham is attracted to the
idea that local motion is sufficient for bringing about the union or disunion of
matter and form, suggesting that Aristotle, following “natural reason,” would
hold simply that being in the same place is sufficient for matter and form to
compose a substance. But he considers two theological cases that close off
this—according to him—otherwise reasonable position. The first is that dur-
ing Christ’s three days in the tomb, Christ’s intellective soul was taken to be in
the same place as his body, and yet (because he was dead) not inform it and so
not compose a substance with it.36 The second is that, according to the Bible,
the resurrected Christ passed through a closed door and was therefore in the
same place as the door, but the soul of Christ neither informs the door nor
composes a substance with it.37 Ockham concludes by elimination: “thus it is
36 Ockham, Quaestiones Variae VI, a.2, ll.250–253 (OTh VIII, p.217); for this view also see
Ockham, Quodlibet IV, q.31, ll.48–56 (OTh IX, p.453).
37 Ockham, Quaestiones Variae VI, a.2, ll.254–257 (OTh VIII, pp.216–217); John 20:26.
54 Chapter 3
necessary that when from a non-composite a composite comes to be, or vice

versa, something positive is produced or destroyed.”38
Scotus seems to agree with Ockham’s theological points.39 But he has strictly
philosophical resources with which to rule out local motion as the sort of
change able to bring about the composition of matter and form. In at least one
text, Scotus thinks that totally separated matter and totally separated form
would literally not be anywhere.40 The reason seems to be that being in a place,
according to Scotus, involves having an accidental form of a certain kind. But
having separable accidents belongs not to matter and form individually but to
the substance they compose. So nothing is in a place unless and until it is or is
in a substance or an integral part of a substance.
What then is produced or destroyed when a substance comes to be or ceases
to be? Scotus and Ockham agree about the truth of
[SO5] The change that occurs when matter and form begin to compose
a substance cannot be the production of a relative form,
and
a substance cannot be the production of an absolute form.
[SO6] holds for reasons considered at the beginning of the chapter, namely
that if the change were the production of an absolute form, then we would
need some account of how matter and form and the additional form become
parts of the same substance. If this account involves the production of yet
another form, then an infinite process ensues. If this account does not involve
the production of a third form, then there was no reason for positing the sec-
ond form in the first place.41 The argument for [SO6] turns on the assumption
that if the change that occurs when matter and form begin to compose a sub-
stance were the production of an absolute form, then this absolute form would
be, along with matter and form, a part of the whole substance.
38 Ockham, Quaestiones Variae VI, a.2, ll.58–60 (OTh VIII, pp.217), Et ideo necesse est quod
quando de non composito fit compositum vel econtra quod aliquis positivum producatur
vel destruatur.
39 Lectura III, d.22, q.un, n.71 (Vatican XXI, p.94).
41 Ordinatio III, d.2, q.2, nn.80 (Vatican IX, p.152); Lectura III, d.2, q.2, n.78 (Vatican XX, p.101).
The argument for [SO5] is more complicated. Since Scotus thinks that most
relations are genuine entities really distinct from their relata, it would have
been open to him to argue for [SO5] in exactly the same way as he argued for
[SO6]. But he does not.42 Instead, in arguing for [SO5] Scotus assumes that if
the change that occurs when matter and form begin to compose a substance
were the production of a relative form, then this relative form would not be a
part of the whole substance but would be the whole substance. He writes,
The quiddities of all absolute things, as including matter and form, and as
definable (because as they are definable, they are species of an absolute
genus), are not formally merely relative entities—which, nevertheless,
would be necessary if the proper entity of the whole were a relation.43
It is not obvious just why Scotus thinks that if the generation of a relation
explains matter and form’s composing a substance, then that relation is “the
proper entity of the whole.” Presumably Scotus does not think that the com-
posite substance would just be a relation, since it does not follow that if the
production of a relation explains how matter and form compose a substance,
then the substance composed of matter and form just is a relation. Perhaps
what Scotus has in mind is that if the difference between matter and form
composing and not composing is a relation, then the relation would enter into
the definition of the substance, such that a substance would be essentially
relative.
Scotus argues for [SO5] in a different way in QMet VIII, q.4:
A new relation cannot exist, neither in one term nor in the other, if no
change was made in something absolute. [But] in neither matter nor
form does [an absolute] change occur if they remain [after] having been
separated. Therefore the relation of these is the same as it was before the
separation.44
42 And he would not, for he elsewhere rejects Bradley’s Regress-style arguments against his
realism about relations. QMet V, q.11, nn.51–55 (Bonaventure III, pp.584–585). Also, see the
discussion in Marilyn McCord Adams, William Ockham vol. I, pp.215–233.
43 Ordinatio III, d.2, q.2, n.78 (Vatican IX, pp.150–151), [Q]uidditates omnium absolutorum,
ut includentes materian et formam, et ut definibiles (quia ut sic, sunt species generis
absoluti), non sunt tantum formaliter entia respectiva—quod tamen oporteret si entitas
propria totius esset respectus.
44 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q.4, n.23 (Bonaventure IV,
p.496), […R]elatio non potest esse nova, nulla mutatione facta in aliquo absoluto, et hoc
56 Chapter 3
Here, Scotus presumes that the produced relation would be a part of the sub-
stance together with matter and form. He starts from the Aristotelian thesis
that a change in relation occurs only in virtue of a change in something abso-
lute.45 If two things are related they are related in respect of something abso-
lute, for example, Mole and Rat are similar with respect to genus,46 a white
stone and a black stone are opposed with respect to color, and Badger is larger
than Mole with respect to quantity. (Of course, x and y could be, say, taller than
z, and therefore both would be similar with respect to being taller than z. In
some sense therefore things can be related in respect of something relative,
but this is not relevant to Scotus’s argument.) So if matter and form begin to
compose a substance because a relative entity is produced, and likewise cease
to compose a substance when the same relative entity is destroyed, then there
must be some absolute change(s) in virtue of which that relation is produced
or destroyed. But, according to Scotus, matter and form do not change in any
absolute way when they first compose and then do not compose a substance.
So if matter and form are related while composing a substance, this relation
will remain after they cease to compose a substance. The absurdity would fol-
low that the matter and form do not compose a substance but have exactly the
same relations to each other in virtue of which they compose a substance. So
the relation between matter and form cannot explain their composing a sub-
stance—it cannot be the difference in things that makes matter and form into
parts of a substance.
Scotus does not deny that real relations are produced when matter and form
compose a substance. In fact he thinks that two pairs of relations are produced:
relations of matter and form to each other, and relations of matter and form to
the substance they compose. But neither pair nor both together explain why
matter and form compose, because each is naturally posterior to the composite.
Scotus and Ockham agree about [SO5] and [SO6] but here there agreement
ends. Ockham thinks that [SO5] and [SO6] cross off every relevant item on the
inventory of the things there are, and therefore feels entitled to claim
[O1] (=not-[S1]) It is not the case that the sort of change that occurs
when matter and form begin to compose a substance is the produc-
tion of some entity (other than matter and form).
nec in uno extremo relationis nec in alio. Materia et forma, si manent separata, in nullo
illorum fit mutatio; ergo eadem est ipsorum relatio quae fuit ante separationem.
45 Aristotle, Physics V 225b 11–13 (Barnes I, p.381).
46 They are similar with respect to genus in the Porphyrian-Tree sense, not the modern zoo-
logical sense: they are both sensitive, animate, material substances.
But from [O1] and
the destruction of some entity,
Ockham can infer that
[O2] The sort of change that occurs when matter and form begin to com-
pose a substance is the destruction of some entity.
Ockham is not explicit about his reasons but seems to think it is obvious that
[O3] What is destroyed when matter and form begin to compose a sub-
stance is a real relation of distinction.
Ockham distinguishes between two sorts of real distinction. In the first sort,
two things are distinct because “this is this, and that is that.” No real relative
entity is required to explain how two individuals are really distinct from each
other in this first way; if it were, then something would be individual because
of its real relations of distinctions to everything other than it. But according to
Ockham things are individuals prior to their relations to other things. According
to the second sort of real distinction, two things are really distinct if and only
if they are not united but are things which naturally are united, such as form
and matter, which are such that naturally the first informs and the second is
informed.47 If matter and form exist but do not compose a substance, some
explanation in things is required to explain their being really distinct. Ockham
posits a real relation of distinction—of the informative to the informable—as
just the sort of entity whose existence blocks the union of matter and form.
When this relative entity is destroyed, then form and matter are united as
informing and informed, with no real relations between them.48 “A whole sub-
stance,” then, signifies matter and form and connotes the negation of the real
relation of distinction.49
47 Whatever is really distinct in the second way is really distinct in the first, but not
conversely.
48 Ockham, Quaestiones Variae VI, a.2, ll.62–84 (OTh VIII, pp.209–210).
49 Ockham, Quaestiones Variae VI, a.2, ll.87–88 (OTh VIII, pp.210), ita quod totum significat
ipsas partes et connotat negationem illius distinctionis praedictae.
58 Chapter 3
Interestingly, Ockham does not consider that he need not think that the
composition of every material substance involves the destruction of a real rela-
tion of distinction. Naturally forms are produced simultaneously with the sub-
stances of which they are forms, so only when the intellective soul is separated
from the body is there a naturally occurring real relation of distinction between
form and matter. It follows then that, naturally, the production of an absolute
(substantial) form, without the production of a real relation of distinction, is
all that is required for the production of a material substance, assuming, with
Ockham, that matter is not producible or destructible by natural powers. The
union of matter and form must be explained as the destruction of a real rela-
tion of distinction only in miraculous cases such as resurrection.
Had Scotus been able to respond to Ockham, he might have argued against
[O3] on the following grounds. Following Aristotle, a real relation is only
destroyed when something absolute changes. So if a real relation of distinction
is destroyed for the bringing about of the composition of matter and form,
something absolute changes. Ockham denies that being in the same place is
sufficient for matter and form to compose a substance, so he cannot claim that
the relation of distinction is destroyed through local motion. Moreover, nei-
ther matter nor form has any absolute accidents the destruction of which
would result in the destruction of their distinction relation. So the relation of
distinction between matter and form could only be destroyed when either
matter or form or both are destroyed; but then of course there could not be a
substance composed of matter and form.
Ockham would likely respond that God could directly destroy the relation
without causing any absolute change(s) in matter and form. Whether or not he
can shore up his own position on logical grounds, however, Ockham’s account
of how matter and form compose a substance remains dissatisfying. Except in
a few miraculous cases, Ockham denies that relations are things distinct from
their foundations. As in Trinitarian metaphysics so here in the metaphysics of
substance, Ockham appeals to real relations against his philosophical impulses
to explain otherwise—according to him—unexplainable data. But the Scotus-
defender or the Scotus-sympathizer might reasonably wonder: what is so bad
about Scotus’s thesis that a substance is really distinct from its matter and
form, such that an ad hoc appeal to miraculous relative entities should be intel-
lectually more satisfying? We cannot appeal to Ockham’s razor, since Ockham
in effect just replaces one sort of uncouth entity—a substance really distinct
from its parts—with another—a relation. True, there turn out to be fewer of
these uncouth relations in Ockham’s theory than there are uncouth substances
in Scotus’s, since Ockham’s relations account for exceptional cases when mat-
ter and form exist but do not compose, while Scotus’s substances account for
all non-exceptional cases when matter and form exist and do compose. So
there is a greater parsimony in the number if not the kinds of entities posited
in Ockham’s theory. Still, insofar as we want parsimony as part of a deeper
commitment to theoretical simplicity, then it’s hard to see how Ockham’s the-
ory has the advantage. And it’s especially hard to see how Ockham could find
his own position rational, given how he elsewhere criticizes realism about
relations.50
If [O3] is false, then [O2] is true only if there is some absolute entity the
destruction of which is sufficient for bringing about the composition of matter
and form. Ockham apparently did not think there was, and I cannot think of
any candidates. So as [O3] falls so falls [O2]. But Ockham posited [O2] as his
way of offering an account of the change necessary for matter and form to
compose a substance which was consistent with [O]. Without [O2], then,
Ockham cannot establish [O].
But [O2] follows from [O1] and [SO4]. So the negation of [O2] entails not-
[O1], which is equivalent to Scotus’s [S1], which claims that the sort of change
that occurs when matter and form begin to compose a substance is the produc-
tion of some entity.
In the following chapter we will examine Scotus reasons for thinking that
the entity that matter and form produce when they compose is a substance,
and analyze his rather complex account of just how matter and form produce
a substance.
50 Ockham, Ordinatio I, d.30 (OTh IV, pp.281–395); see discussion in Marilyn McCord Adams,
William Ockham vol.1, pp.215–276.
Chapter 4
How Matter and Form Compose a Substance—

Part II
i Producing Substance from Matter and Form
The previous chapter examined Scotus’s reasons for thinking that the differ-
ence between matter and form merely existing and matter and form compos-
ing a substance is the result of a certain sort of change, or
[S1] The sort of change that occurs when matter and form begin to
compose a substance is the production of some entity (other than
matter and form).
To seal his case for [S], Scotus argues for the additional premise
pose a substance is the production of a substance.
Ockham’s attempt to demonstrate
[O2] The sort of change that occurs when matter and form begin to com-
pose a substance is the destruction of some entity,
proceeded by trying to show that the change that brings about the composi-
tion of matter and form could not be the production of any entity, or
when matter and form begin to compose a substance is the produc-
tion of some entity (other than matter and form),
and then inferring from [O1] and
the destruction of some entity,

How Matter and Form Compose a Substance—Part II 61
that this change must be the destruction of some entity. According to Scotus,
however, Ockham was not entitled to assert [O1] because he failed to consider
one more item on the list of the things there are—material substances them-
selves. So Scotus argues for [S2] in the following way: [SO4], but not-[O2],
therefore [S1]. But
and
a substance cannot be the production of an absolute form,
so the relevant change is not the production of any form; other things, such as
prime matter, angels, or God, are not eligible candidates. So, through the elimi-
nation of all other contenders, [S2].
From [S2] we cannot derive [S], however, because [S] says not just that a
whole substance is really distinct from matter and form, but that it is really
distinct from its essential parts. So the matter and form which compose a sub-
stance when a substance is produced must themselves produce the substance
they compose. For all [S2] says it could be that the substance produced by this
matter and this form is composed of that matter and that form. Neither Scotus
nor Ockham, nor anyone else as far as I know, discussed this bizarre option, so
I will not pursue it further, except to say
[SO7] Matter and form compose a substance if and only if matter and
form are the causes of the whole substance of which they (once
they cause the substance) are the essential parts.
[SO7] simply makes the necessary clarifications that matter and form are the
causes of the substance and that they compose what they produce, and Scotus
simply assumed that to arrive at [S2] was as good as arriving at [SO7].
[SO7] does not entail [S], because [SO7] leaves it open that the substance
that matter and form produce is not a thing in addition to matter and form but
simply the matter and form conjunctim, taken together. And indeed, given the
way in which matter and form were understood to be causes within the
Aristotelian tradition, namely, as intrinsic causes, this would be the natural
way to understand the causal claim made in [SO7]. But Scotus has a bold and
startling understanding of intrinsic causality, which includes the claim:
62 Chapter 4
[S3] The effect of two intrinsic causes is something really distinct from
those causes, taken individually or together.
[SO7] supplemented by [S3] gives Scotus what he needs to establish [S]. But it
is not obvious why anyone would think that the effect of intrinsic causes is
really distinct from its intrinsic causes. To motivate this idea, consider the fol-
lowing. We know that matter and form do compose substance. But without
[S3], Scotus would be out of options for an explanatory account of how matter
and form come to compose a substance. He has argued that a change must
occur; he has argued that this must be an absolute change by which a new
entity is produced; he has argued that this absolute change cannot be the pro-
duction of a form; he takes it for granted that other sorts of things such as
prime matter, angels, or God, either cannot be produced or wouldn’t be at all
relevant, if produced, to the composition of matter and form; substance is the
only sort of thing left. Either a substance is produced by matter and form or it
is produced by something else. If a substance is produced by something else
then it’s not clear what that production would have to do with the composition
of matter and form. So it must be the case that matter and form produce the
substance they compose.
The argument is not by itself satisfactory, however, because the conclusion
it establishes, [S3], looks at first glance simply like a misunderstanding of what
Aristotle meant by including material and formal as two of the four kinds of
causes. Ockham himself seems to have thought this about Scotus; we will
return to Ockham’s worry below. First, however, we will proceed as Scotus him-
self does: not by arguing directly for the claim that the effect of intrinsic causes
is really distinct from those causes, but by showing how this assumption yields
a coherent (if extraordinarily complicated) account of how matter and form
come to compose a substance and considering two important challenges to
the coherence of this account.
ii Causal and Co-causal Relations
Recall from the previous chapter that one of Scotus’s defenses of
involved appeal to the Aristotelian principle about relational change, accord-

ing to which changes in relations are caused or explained by absolute changes.
To hold, then, that matter and form compose a substance by gaining new rela-
tions (such as parthood relations) is at best nonexplanatory—matter and form
don’t just happen to gain or lose parthood relations; they gain or lose them
because of some absolute change. Scotus’s own account of how matter and
form compose involves identifying this absolute change. The basic solution is
that the generation of the substance itself is the absolute change that explains
matter and form’s new relations. Since the substance is the joint effect of
matter and form, new relations in matter and form to the substance, and new
relations in the substance to matter and form, necessarily arise once the sub-
stance is produced. These relations are, of course, causal relations—relations
of cause to effect in matter and form, and relations of effect to cause(s)
in substance.
So far, so good. But at this point things get complicated. Scotus holds that
when matter and form cause a substance they are related to each other as
causes of the same effect, and describes these “co-causal” relations as the rela-
tions of union between matter and form. But matter and form are united, says
Scotus, before they produce their joint effect. Therefore the generation of the
substance cannot be the absolute change in virtue of which matter and form
are united in their causing. Therefore either Scotus must appeal to some addi-
tional, prior absolute change, or risk violating the Aristotelian principle about
relational change. But he doesn’t appeal to some additional absolute change.
Instead, he thinks that an efficient cause causes matter and form to cause sub-
stance simply by newly relating them:
To concur and not to concur [in causing] changes nothing about the
absolute nature of some cause […] Thus, [two] causes [concurrently]
cause at one time and don’t at another only because of the relation of
both causes to each other (though the reason for causing is in neither one
nor the other of these). So here especially, concerning these two causes,
namely matter and form: because they are united (that is, concurring to
cause) they cause, whose concurrence comes about through the action of
an agent. If they are not concurring, they do not cause.1
1 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q. 4, nn. 31, 32 (Bonaventure IV,
pp. 498–499), Concurrere autem et non concurrere nihil variant circa absolutam naturam
alicuius causae […] Itaque sola relatione alia et alia causarum ad invicem, quae tamen non
est eis nec alicui earum ratio causandi, causae quandoque causant quandoque non. Ita hic de
duabus causis specialiter, scilicet materia et forma, quia unita, hoc est concurrentia ad cau-
sandum, causant (qui concursus fit eroum per actionem agentis); non concurrentia, non
causant.
64 Chapter 4
Thus, while matter and form gain causal/parthood relations to substance natu-
rally posterior to the generation of the substance, matter and form gain co-
causal/union relations to each other naturally prior to the generation of the
substance.2 How, then, does Scotus not violate the Aristotelian principle about
relational change?
QMet VIII, q. 4, unfortunately, does not answer this question. However,
Scotus’s own theorizing about relations developed elsewhere furnishes a plau-
sible answer.3 The gist is that Scotus does not think that all relations only arise
from absolute changes; only what Scotus calls intrinsic relations do. Intrinsic
relations are such that given the absolute features of two or more objects, rela-
tions arise necessarily; the absolute terms of the relation are the “necessary
cause” of the relation.4 For example, given that Plato is six feet tall and Socrates
is five and a half feet tall, the relation taller than necessarily arises in Plato, and
shorter than in Socrates. Changes in intrinsic relations require absolute changes
in relata, since these relations essentially depend on the absolute makeup of
their relata (at some time). Extrinsic relations, by contrast, are such that the
absolute features of their relata do not necessitate the advent of the relation.
Specify every absolute feature of the subject of the relation, and every absolute
feature of the term of the relation, and it’s still a contingent affair whether or
not that relation exists. Scotus says explicitly that Aristotle’s principle does not
apply to changes of extrinsic relations,5 from which it follows that an absolute
change is not required to produce or destroy external relations. In Ordinatio III,
d. 1, p. 1, q. 1, n. 59, Scotus offers two examples of extrinsic relations: the rela-
tions of soul to body and of a quantitative form to its subject. Both items of
each pair could exist and not be related, i.e., not compose a human, and not
compose an accidental unity, respectively. Says Scotus, “If they are newly
united, no new absolute is in either term, but that relation exists contingently,
such that it can both exist and not exist in the posited term.”6
2 In Quaestiones super Libros Metaphysicorum Aristotelis IX, qq. 3–4, n. 25 (Bonaventure IV,
pp. 545–546), Scotus repeats in more general terms that intrinsic causes (i.e., matter and form)
are related both to each and to their effect. In this text, however, he does not specify which
pair of relations is prior to the other—quae essentialius vel prius, non est modo quaestio.
3 For other ways in which Scotus violates the relational change principle, see Richard Cross,
“Relations, Universals, and the Abuse of Tropes,” Aristotelian Society Supplementary Volume
79:1 (2005), pp. 53–72.
4 Ordinatio III, d. 1, p. 1, q. 1, n. 58 (Vatican IX, p. 27).
5 Ordinatio III, d. 1, p. 1, q. 1, n. 60 (Vatican IX, p. 28).
6 Ordinatio III, d. 1, p. 1, q. 1, n. 59 (Vatican IX, pp. 27–28), […] si de novo uniantur, nullum abso-
lutum est in altero extremo, sed ista relatio contingenter se habet, etiam ut possit esse et non
esse extremis positis.
Scotus therefore thinks that no absolute changes explain or cause the advent
of union relations between matter and form. But he does not think that matter
and form compose a substance simply in virtue of being united, and therefore
he does not think that being united, that is, being related to each other as
co-causes of the same effect, is that in which the parthood of matter and form
consists. Parthood in general is a relation to a whole, not to other parts, and
therefore parthood relations must exist naturally posterior to the existence of
the whole.
Once matter and form produce the substance, matter and form gain new rela-
tions to the substance, and these relations are necessary in the sense that given
matter and form and given that matter and form are the causes of the substance,
matter and form must be causally related to their effect. Causal relations are
thus intrinsic relations. Not only can causal relations not arise without some
absolute change (the coming to be of the effect), they must arise, given the effect.
Composing a substance is therefore a temporally simultaneous series of
naturally ordered instants of nature, where two items, A and B, are naturally
ordered and therefore exist at different instants of nature if and only if expla-
nation of A requires reference to B (or vice versa).7 At the temporal instant at
which matter and form compose substance Scotus distinguishes six naturally
ordered instants, (i–vi), where (i) is naturally prior to (ii), (ii) is naturally prior
to (iii), and so on:
(i) An agent efficiently causes matter and form to co-cause the substance.
(ii) Matter and form are related to each other as co-causes (i.e., they have co-
causal relations to each other).
(iii) Matter and form co-cause the substance.
(iv) The substance exists.
(v) Matter and form are related to the substance as parts to whole (i.e., they
have parthood/causal relations to the substance).
(vi) The substance is related to matter and form as whole to parts (i.e., it has
totality relations to matter and form).
(Scotus thinks that (vi) is naturally posterior to (v),8 but I myself have no intu-
itions that these are naturally ordered instants at all.)
7 Here I am following Calvin Normore’s analysis of instants of nature. See Calvin Normore,
“Duns Scotus’s Modal Theory,” in The Cambridge Companion to Duns Scotus, pp. 130–137 and
especially p. 134.
8 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q. 4, n. 45 (Bonaventure IV, p. 502).
66 Chapter 4
iii Innovating Aristotle’s Principle about Relational Change
Having distinguished this order Scotus sets himself a troubling objection.

If the union/co-causal relations of matter and form to each other are extrinsic
relations, do not depend for their existence on an absolute change (in this case
the generation of the substance), and are essentially prior to the substance,
then why should the corruption of the substance result in the ceasing of these
relations? Shouldn’t it follow that the substance could be corrupted and matter
and form cease to compose, and matter and form still be united? Non, says
Scotus; but to explain just why not demands a heavy qualification to the
Aristotelian principle about relational change:
[W]hen some things are related to each other and it is impossible that
they are thus related unless one or both of these is related to some third
thing, their mutual relation may well be corrupted without the corrup-
tion of anything absolute in either of these, solely by the corruption of an
absolute posited in that third thing to which either [or both] is referred.
So here: it is impossible that matter and form be united unless each is a
part of the composite. Therefore in the composite, something absolute
[i.e., the substance] which was the foundation of the totality relation in it
having been corrupted, the totality relation is corrupted; and by conse-
quence the relation of part in these [i.e., matter and form]; and third the
mutual relation in each of these [i.e., matter and form], which cannot
remain without those [relations] to the third. And then this is false:
‘A relation is not corrupted unless something absolute in either of the
relatives is corrupted,’ unless there is added, ‘or in some third, to which
either of the relatives is necessarily referred,’ such that without such a
relation it would not be referred.9
9 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q. 4, nn. 45–46 (Bonaventure IV,
p. 502), […Q]uando aliqua mutuo refereuntur, et incompossibile est illa sic referri nisi
alterum illorum referatur ad aliquod tertium vel ambo, bene potest corrumpi relatio eorum
mutuo sine corruptione alicuius absoluti in altero illorum, sola corruptione absoluti posita in
illo tertio ad quod alterum illorum dicitur. Sic hic: imcompossibile est materiam et formam
esse unita nisi utrumque sit pars compositi. Ergo in composito, corrupto aliquo absoluto
quod erat fundamentum relationis totalitatis in ipso, corrumpitur relatio totalitatis; et ex
consequenti relatio “partis” in istis; et tertio relatio mutua in utroque istorum, quae non
potest stare sine illa ad tertium. Et tunc ista est false “relatio non corrumpitur nisi corrupto
aliquo absoluto in altero relativorum,” nisi addatur “vel un aliquo tertio, ad quod necessario
dicitur alterum relatorum,” ita quod sine hoc tali relatione non referretur.
Here Scotus allows that relational changes can cause relational changes, pro-
vided that there is some special order between the relations (or pairs of rela-
tions, in this case), such that the second cannot exist without the first. The
claim is that union relations cannot exist without parthood relations, so the
destruction of the parthood relation entails the destruction of the union rela-
tion. He does not argue for this claim, but simply presents it as an obvious
counterexample to the Aristotelian principle—new data demands modifica-
tion of the theory.
In offering this counterexample Scotus is likely relying on more than just
intuition. The claim is backed up by deep-structure theoretical commitments.
First, Scotus denies that being essentially prior to entails being able to exist
without. In De Primo Principio he speculates that in some essential orders, the
essentially prior produces its essentially dependent effect necessarily and
therefore cannot exist without it.10 Assuming then that union/co-causal rela-
tions are essentially prior to the substance (surely a safe assumption), it is open
for Scotus to hold that union/co-causal relations are essentially prior to but
nevertheless cannot exist without the substance. Second, and closely related to
the first, Scotus thinks that, for essential causal orders, a cause is strictly speak-
ing a cause only at the moment at which it is causing.11 Thus only at the (tem-
poral) instant at which the effect begins to exist is the cause its cause. Remove
the effect, therefore, remove the cause’s causing of the effect. Similarly, if two
cause concurrently (as do matter and form), then their co-causing, and hence
their co-causal relations, obtain at the (temporal) instant at which their joint
effect begins to exist. Remove the effect, therefore, remove their co-causing
(and hence their co-causal relations).
Given this innovation of Aristotle’s principle of relational change, it’s worth
questioning whether this innovation is consistent with one of Scotus’s reasons
for holding [SO5]. [SO5] denies that the production of relations can explain
the composition of matter and form, and one reason Scotus thinks this is that
relations cannot be produced without some absolute change. Scotus’s innova-
tion is consistent with this reason for holding [SO5], it seems to me. Scotus’s
innovation does not really allow that relational change can occur without
absolute change. Instead, he thinks that an absolute change can explain the
destruction of a relation in a transitive way, through the destruction of another
relation. For example, the destruction of a substance (an absolute change)
entails the destruction of parthood relations, and the destruction of parthood
relations entail the destruction of union/co-causal relations, but by transitivity
10 De Primo Principio I.8 (Wolter, p. 5).

11 De Primo Principio III.11 (Wolter, p. 47).
68 Chapter 4
we can infer that the absolute change entails the destruction of the union rela-
tions. So Scotus does not, as it might first appear, refuse with one hand what he
takes with another. On his innovative view, it remains the case that an absolute
change is needed to explain relational change; but absolute changes can
explain relational changes both directly (e.g., the corruption of a substance
entails the destruction of parthood relations in matter and form) and transi-
tively, through the destruction of different relations (e.g., the destruction of
parthood relations in matter and form entail the destruction of the union rela-
tions of matter and form).
iv The Identification of Parthood Relations with Causal Relations
I have been assuming so far that the causal relations of matter and form to
substance are the parthood relations of matter and form to substance, that
being an essential part of a composite substance simply amounts to being a
material or formal cause of a composite substance. Scotus does not directly
argue for the identification of causal with parthood relations, but that he
thinks this is clear throughout QMet VIII, q. 4. To show this we can start by
examining three texts. The first is a portion of n. 45, part of the text just quoted
in which Scotus modifies the Aristotelian principle about relational change:
[I]t is impossible that matter and form be united unless each is a part of
the composite. Therefore in the composite, something absolute [i.e., the
substance] which was the foundation of the totality relation in it having
been corrupted, the totality relation is corrupted; and by consequence
the relation of part in these [i.e., matter and form]; and third the mutual
relation in each of these [i.e., matter and form], which cannot remain
without those [relations] to the third.
The second comes from n. 41:
The entity of the composite is some third entity from matter and form,
and is caused by these. In that [substance] a change of corruption comes
about, because it is not after it was, and both were said to be causing it per
se, although not at first. Thus a change of relation follows in these upon
the corruption of the relation and the foundation in the composite.12
12 Quaestiones super Libros Metaphysicorum Aristotelis VIII, q. 4, n. 41 (Bonaventure IV,

p. 501), Sic hic: entitas compositi est aliqua entitas tertia ab entitate materiae et formae,
The third comes from n. 49. Scotus writes that when a substance is corrupted,
First the foundation in the effect is corrupted, and thus its relation, and
third the relation in the cause.13
All three texts express the “domino effect” of relational corruptions that result
from the corruption of the substance, and all are presented to reinforce the
claim that the substance itself is the absolute item whose generation explains
how matter and form become parts and whose corruption explains how they
cease to be parts. In n. 45 the corruption of the substance results in the corrup-
tion of, first, totality relations, then, parthood relations, and finally mutual rela-
tions. (Presumably the mutual relations are the co-causal/union relations of
matter and form, since Scotus identifies no other mutual relations between mat-
ter and form.) In n. 41, upon the corruption of the substance, causal relations in
matter and form to substance are corrupted (and perhaps co-causal/union rela-
tions are implied here as well). In n. 49 the corruption of the substance leads first
to a corruption of its relation (presumably the totality relations, since these are
the only sort of relation in the whole that Scotus identifies) and then to a corrup-
tion of the relations in the causes (presumably the causal relations of matter and
form to substance). I take these texts to be making philosophically the same
move using different labels for the same relations. At no point does Scotus
distinguish causal from parthood relations, and in n. 45 he endorses the claim
of an objector, expressed in n. 44, that when matter and form compose a
substance,
[I]n the two absolute essences of matter and form four relations are
founded, having proper primary correlatives: namely, two mutual rela-
tions and two to the composite.14
et causata ab eis. In illa fit mutatio corruptionis, quia non est postquam fuit, et ad illam
dicebantur ambo causantia per se, licet non primo. Ideo sequitur mutatio relationis in eis
ad corruptionem relationis et fundamenti in illo composito.
p. 503), Nam primo corrumpitur fundamentum in effectu, et ideo relatio eius, et (quasi
tertio) relatio in causa.
pp. 501–502), Et ita in duabus essentiis absolutis materiae et formae fundantur quattuor
relationes, habentes propria correlativa prima: diae scilicet mutuae, et duae ad
compositum.
70 Chapter 4
If causal and parthood relations were distinct types of relations, we would

expect Scotus not to endorse this portion of the objection, as he explicitly
does: “the conclusion of this last argument can be conceded.”15
Finally, for indirect evidence, while Ockham denies that matter and form
have parthood relations to the substance they compose, and denies that the
substance is really distinct from matter and form (taken together), he never-
theless thinks that what being a material or formal cause of a substance
amounts to is simply being an essential part a substance: “the causality of
[matter and form] is nothing other than to be essential parts of any compos-
ite.”16 There is no reason to doubt that Scotus thought the same thing (despite
his ontological additions of parthood relations and a real distinction of the
whole substance from its essential parts).
v The Causality of Matter and Form
I now return to the worry that Scotus has misunderstood what it is to be

an intrinsic cause. As we have seen, Scotus recognizes that [S] demands a
fairly radical reconstrual of the causality of matter and form, a reconstrual
expressed in
[S3] The effect of an intrinsic cause is something really distinct from the
intrinsic cause.
One might worry, however, that Scotus has tried to make material and formal
causes do the work of efficient causes. This is exactly Ockham’s worry, and one
of his many criticisms of Scotus’s view that a whole substance is something in
addition to matter and form:
I say that intrinsic causes do not cause something absolute or relative in

any way distinct from them, because if they did, then in respect to those
things [which they cause] they would have efficient [causation] and so
they would be extrinsic causes. But the causality of these is nothing other
than to be essential parts of any composite. And this comes about on

pp. 502), Responsio: conclusio istius ultimae rationis posset concedi.
16 Ockham, Quaestiones Variae VI, a. 2, ll. 217–218 (OTh VIII, p. 216), Sed causalitas earum non
est aliud nisi esse partes essentiales alicuius compositi.
account of this, that one is informing and the other is informed, one is
potency and the other is act. This pertains to these from their natures,
having removed any distinction from them.17
Ockham thinks that all it takes for matter and form to be parts of a substance
is for any preventing real relation of distinction to be destroyed; and he thinks
that all it takes for matter and form to be intrinsic causes is to be parts. Scotus
would have been unmoved by Ockham’s objection, for two reasons. First,
Scotus rejects Ockham’s alternative account according to which matter and
form do not need to do anything to compose substance. As we have seen,
Scotus thinks that causing a substance is prior to being a part of a substance,
because he thinks that the production of a substance is the only absolute
change that can explain the relative changes in matter and form when they
acquire parthood relations. Second, he actually considers it a special strength
and not a weakness of his view that it includes that the effect of intrinsic causes
is something really distinct from the intrinsic causes:
If there were no other entity of the whole than the entity of the parts,
there would not be a caused composite, because the parts are not caused
from intrinsic causes. If therefore the composite is nothing other than
the parts, it would follow that the composite would not be caused from
intrinsic causes, because then there would be nothing except those two
entities, which are causes.18
Scotus is quite explicit that the effect of an intrinsic cause is something really
distinct from the intrinsic cause; he embraces [S3] with eyes wide open. If
pressed with Ockham’s objection, Scotus could have distinguished efficient
from material and formal causality in some other way than by saying that the
17 Ockham, Quaestiones Variae VI, a. 2, ll. 214–220 (OTh VIII, p. 216), […D]ico quod causae
intrinsecae non causant aliquid absolutum nec respectivum quocumque modo dis-
tinctum ab eis, quia si sic respectu illius haberent efficientiam et sic essent causae extrin-
secae. Sed causalitas earum non est aliud nisi esse partes essentiales alicuius compositi.
Et hoc fit per hic quod una est informans et alia informata, una est potentia et alia actus.
Hoc convenit eis ex natura earum, circumscripto quocumque distincto ab eis.
18 Lectura III, d. 2, q. 2, n. 83 (Vatican XX, p. 103), Item, si non esset alia entitas totius quam
est entitas partium, non esset compositum causatum, quia partes non sunt causatae ex
causis intrinsecis; si igitur compositum nihil aliud esset quam partes, sequeretur quod
compositum non esset causatum ex causis intrinsecis, quia tunc essent nonnisi illa duo
entia quae sunt causae. Also, Ordinatio III, d. 2, q. 2, n. 76 (Vatican IX, p. 150); Reportatio
Parisiensis III-A, d. 2, q. 1, n. 4 (Wadding XI, p. 428).
72 Chapter 4
effect of the efficient cause is really distinct from the efficient cause whereas
the effect of the intrinsic causes is not really distinct from the intrinsic causes.19
For this reason, Scotus has no reason to abandon [S3]; and this yields [S], and
thus [SO1] → [S].
vi Dispensing with Total Separability?
Suppose someone did not buy Scotus’s total separability thesis about matter
and form—he would therefore reject
not compose a substance.
But [SO3] was used in QMet VIII, q. 4, n. 14 to argue that something must hap-
pen to matter and form in order for them to compose a substance, which led
eventually to
[S] A substance is a whole which is something in addition to its essen-

tial parts.
Can [S] be held, then, if [SO3] is denied? I think Scotus’s answer would be
affirmative. If [SO3] is false, then whenever matter and form exist they com-
pose a substance. But this is not to say that all it is for matter and form to
compose a substance is for both to exist. It could be that whenever matter and
form exist they concurrently cause a substance which is a whole really dis-
tinct from matter and form. Their existing and their causing would be simul-
taneous but nevertheless naturally ordered—the existence would be prior to
the causing, which is exactly what happens when supernatural causes do not
intervene. In the language of QMet IX, the being of form and matter would still
be naturally prior to their principiating or causing. And Scotus offers several
additional arguments for the claim that a whole substance is really distinct
from its matter and form, which do not depend on total separability.20
19 What Scotus cannot say, however, is that intrinsic causes are distinguished from extrinsic
causes in that the former but not the latter become parts of their effect. He cannot say this
because he has argued that all it is to be an essential part is to have an intrinsic causal
relation to the whole.
20 These arguments are well analyzed by Richard Cross, “Duns Scotus’s Anti-Reductionistic
Account of Material Substance,” Vivarium 33:2 (1995), pp. 137–170. See Ordinatio III,
vii Conclusions: Composing, and Composing
Here is a summary of the long argument for [S]. Scotus holds
not compose a substance.
So when matter and form do compose a substance, some change must occur,
and
the destruction of some entity.
But
when matter and form begin to compose a substance is the pro-
duction of some entity (other than matter and form).
is false; therefore
[S1] The sort of change that occurs when matter and form begin to
compose a substance is the production of some entity.
But
a substance cannot be the production of an absolute form,
and the relevant change cannot be the production of God, an angel, or matter, so
pose a substance is the production of a whole substance.
d. 2, q. 2, nn. 74–77 (Vatican IX, pp. 149–150); Lectura III, d. 2, q. 2, nn. 81–83 (Vatican
XX, p. 103).
74 Chapter 4
But this is almost surely not the production of a substance which is composed
of some other matter and form, so
[SO7] Matter and form compose a substance if and only if matter and
form are the intrinsic causes of the whole substance of which
they are the essential parts.
And since
[S3] The effect of two intrinsic causes is something really distinct from
those causes, taken individually or together;
therefore
[S] A substance is a whole which is something in addition to its essen-

tial parts.
Scotus begins from a general claim about unity and motivates his claim that a
substance is more unified than an aggregate by arguing that the essential parts
of a substance can meet the existence requirements of an aggregate and not be
a substance. This is a negative characterization of the unity of substance, how-
ever, in that it tells us what this unity cannot be. In the course of arguing for
[S], Scotus limns a positive account of this unity.
The heart of the positive account lies in Scotus’s idea that matter and form
compose a substance by causing a substance. They are united as co-causes
of a joint effect, and they are parts of their effect by being causally related to
their effect.
This chapter has explored part of Scotus’s answer to the familiar question
about the conditions under which some things compose one thing. The sense
of compose that is usually meant has to do with the special relation things
are said to have when they are parts of a thing, and this is the sense of compose
that has been assumed in this chapter. Another legitimate sense of com
pose has to do with the activity of bringing something into existence, as in
composing a sonnet or symphony. According to this second sense, one way in
which some things compose one thing is through collaboration, for example
when musicians get together to compose a song, or when poets jointly com-
pose a poem. Of course, when metaphysicians ask the familiar question about
composition they are asking about the parts of things. Thus, in the intended
sense of the word, notes and not the musicians compose a song, words and
not the poets compose a poem. The difference between these two senses is
probably pretty clear, but the similarity is harder to get hold of. Given certain
ideas about composition that were popular through most of the twentieth cen-
tury, the connection between these two senses of compose is tenuous and to
wonder whether parts compose by making is about as quaint and misleading
as asking whether composers compose by becoming parts of their artworks.21
For Scotus, however, the connection is neither quaint nor misleading. While he
would deny that composers become parts of their artworks, he does think that
parts compose by making—composing in a very strong sense—the wholes of
which they are parts.
21 I am thinking here of formal mereology, according to which anything is a part of some-

thing just in case it exists. See Peter Simons, Parts: A Study in Ontology (Oxford: Clarendon
Press, 1987).
Chapter 5
Scotistic Pluralism about Substantial Form—Part I
The last two chapters focused on essential parts—matter and substantial

form—and considered Scotus’s account of how matter and form compose a
substance. This chapter and the next focus on integral parts, specifically the
integral parts of organisms, and develops a view about Scotus’s answer to the
question, “How do organic parts—things like bones, flesh, hearts, livers, eyes,
teeth, hands, and so on—compose one substance, an animal?” Scotus thinks
that a substance is a composite of matter and substantial form, and he thinks
that the substances include organisms like plants and animals, inorganic com-
pounds such as bronze, and what he recognizes as the basic elements of such
compounds: earth, water, air, and fire.1 Scotus is also a pluralist about substan-
tial forms: he thinks that living composite substances are composed of matter
and more than one substantial form.2 According to the common scholarly
1 Scotus also thought that there were immaterial substances, but for present purposes we can
ignore these. In this chapter, therefore, by “substance” I just mean “material substance.”
2 Every medieval hylomorphist thought that there were a plurality of forms in a substance, viz.,
at least one substantial form and many accidental forms, where accidental forms modify a
substance and a substantial form (together with matter) composes a substance. In this chap-
ter I am not concerned with accidental forms except in an incidental way, so unless other-
wise specified by “form” I mean “substantial form.” For some of the history of the medieval
debate over the number of substantial forms in composite substances, see Roberto Zavalloni,
“Étude Critique” in Richard de Mediavilla et la Controverse sur la Pluralité des Formes (Louvain:
Éditions de L’institut Supérieur de Philosophie, 1951), pp. 213–504; Daniel A. Callus, “The
Origins of the Problem of the Unity of Form,” in The Dignity of Science: Studies in the
Philosophy of Science, ed. James A. Weisheipl (Washington, dc: Thomist, 1961), pp. 121–149;
James A. Weisheipl, “Albertus Magnus and Universal Hylomorphism: Avicebron (A Note on
Thirteenth-Century Augustinianism),” in Albert the Great: Commemorative Essays, ed. F.J.
Kovach and R.W. Shahan (Norman, ok: University of Oklahoma Press, 1980), pp. 239–260. For
some views about Scotus’s place in this history, see Étienne Gilson, Jean Duns Scot:
Introduction a ses Positions Fondamentales (Paris: Librairie Philosophique J. Vrin, 1952),
pp. 490–497; Prospero Stella, L’ilemorfismo di G. Duns Scoto (Torino: Societa Editrice
Internazionale, 1955), pp. 187–229; Efrem Bettoni, Duns Scotus: The Basic Principles of his
Philosophy, trans. Bernardino Bonansea (Washington, dc: The Catholic University of
America Press, 1961), p. 69; Marilyn McCord Adams, William Ockham, v.2 (Notre Dame, in:
University of Notre Dame Press), pp. 633–670; R. Cross, The Physics of Duns Scotus: The
Scientific Context of a Theological Vision (Oxford: Clarendon Press, 1998), pp. 47–76; Robert
Pasnau, Metaphysical Themes 1274–1671 (Oxford: Oxford University Press, 2011) pp. 581–582.

Scotistic Pluralism about Substantial Form—Part I 77
view these substantial forms are a soul and a form of corporeity, where the
form of corporeity is supposed to be the substantial form by which an animal
is corporeal, and the soul is supposed to be the substantial form by which an
animal is living. Fewer commentators have recognized that Scotus also thinks
that what he and others call the integral parts of animals, things like livers and
hearts, are themselves composites of matter and distinct kinds of substantial
forms, a form of the liver, a form of the heart, and so on.3
It is not clear how this analysis of integral parts is supposed to cohere with
the common view of Scotus’s analysis of essential parts, however. On these two
analyses it appears that there is one form by which an animal is a body, and
many other forms by which an animal has a heart, liver, bones, etc. But, intui-
tively, if we have all of these integral parts, what more do we need in order to
have a body? It seems redundant to suppose that the integral parts of an ani-
mal have their own substantial forms, and that there is in addition to these
another form by which an animal is a body. I argue that, for Scotus, it is redun-
dant to suppose this. Scotus thinks that in a process of embryological develop-
ment many substances are generated—a heart, blood, a brain, and all the rest
of the organs—and under natural conditions these substances can be informed
by a soul. The union of these substances with the soul is the last stage in the
generation of a complete organism, whereby these substances become integral
parts of one animal. For Scotus there is no substantial form of corporeity
whose job it is to make a substance merely corporeal. In the following I develop
and defend this interpretation of Scotus and offer an account of how Scotus
thinks that many substances can be parts (part-substances, as I will call them4)
of one substance. I start by providing some of the medieval background against
which Scotus formed his position.
Avicebron’s Fons Vitae is generally recognized, both by modern scholars and by scholastics
such as Albert the Great and Aquinas, as the source of the pluralist view, but the view is actu-
ally much older than this, having been articulated by Alexander of Aphrodisias ca. 200 ad.
See Alexander’s On the Soul 1.13, trans. A.P. Fotinis, in The De Anima of Alexander of
Aphrodisias: A Translation and Commentary (Washington, dc: University Press of America,
1979), p. 9 (thanks go to Calvin Normore for pointing me towards Alexander). For Avicebron’s
pluralism, see Fons Vitae II.8 and III.3.22, trans. John of Spain and Dominic Gundisalvi, ed.
C. Baeumker (Munster: Aschendorff, 1895), pp. 37–39, 81.
3 Dorothea E. Sharp, Franciscan Philosophy at Oxford in the Thirteenth Century (London: Oxford
University Press, 1930), pp. 311–313; Richard Cross, The Physics of Duns Scotus, pp. 68–71.
4 A part-substance is anything that is both a part of a substance and a substance. It is an admit-
tedly awkward expression, but does a better job of conveying my meaning than more elegant
expressions like partial substance or substantial part.
78 Chapter 5
i Unitarianism and Pluralism about Substantial Form
For most hylomorphists, including Scotus, it is standard to say that a living

substance such as Mole is a composite of body and soul, where body is the mat-
ter and soul is the substantial form, and where soul informs the body. But some
medieval hylomorphists would quibble with this characterization of body and
soul as the matter and form of an organism. Although Aquinas sometimes says
that the soul unites with the body to compose an organism, he really does not
think this.5 Instead, Aquinas thinks that the soul informs prime matter, an
uncharacterized substratum of substantial change. He therefore thinks that
soul and prime matter, rather than soul and body, are strictly speaking the
essential parts of a composite substance. All of the essential properties of a
substance like Mole, not just being animate but also being corporeal—are due
to Mole’s sensitive soul.6 This entails that Mole’s body cannot exist indepen-
dent from his soul, so when Mole dies his corpse is a substance (or substance-
like object) both specifically and numerically distinct from Mole (and from any
former part or parts of Mole). Like Aristotle, Aquinas affirms that Mole’s corpse
is only homonymously Mole’s body, Mole’s severed paw only homonymously a
paw.7 Aquinas also held that Mole’s soul does not begin to inform (prime mat-
ter, the body) until his fetus has developed to a sufficiently advanced stage. He
thinks therefore that there is some instant at which the substantial form of
Mole’s fetus is replaced by Mole’s sensitive soul, resulting in the generation
of a new substance.8 As for Mole so for any composite substance: Aquinas
analyses it as a composite of prime matter and exactly one substantial form.9
We can call Aquinas’s view unitarianism about substantial form.
Other medieval hylomorphists disagree with Aquinas’s unitarianism.
Ockham holds, for example, that the body is itself a composite substance,
5 Aquinas, Summa theologiae Ia, q.76, a.1.

6 Aquinas, Summa theologiae Ia, q.76, a.4.
7 Aristotle, On the Soul II, c.1, 412b 10–24, (Barnes I, p. 657; Meteorology IV, c.12, 389b 30 – 390a 1
(Barnes I, p. 624); Generation of Animals II, c.1, 734b 24–35 (Barnes I, p. 1140); Aquinas, Summa
theologiae III, q.50, a.5, corp.
8 Aquinas, De potentia Dei q.3, a.9, ad 9; Summa contra gentiles II, c.89, 11; Summa theologiae Ia
q.119, a.2.
9 Strictly speaking a unitarian could deny that Mole has only one substantial form, for example
if she denied that Mole is one substance (perhaps he is instead many substances).
Traditionally, however, unitarianism as a metaphysical position was yoked with assumptions
about what counted as substances, things like plants, animals, elements, and compounds of
elements.
composed of prime matter and a form of corporeity. Soul informs the body,
making it living, but its identity as a body is independent from the soul. On this
analysis, therefore, the body persists when the organism dies; the corpse is
both specifically and numerically the same body as the body of the living
organism. In the case of rational animals, humans, Ockham argues that there
is a total of three substantial forms: the form of corporeity, the sensitive soul,
and the rational soul.10 And at least one medieval hylomorphist, Richard of
Middleton, counted four substantial forms in a human: a form of corporeity
together with vegetative, sensitive, and rational souls.11 For present purposes,
we can consider Ockham’s and Richard’s views, along with others like them, a
single view, standard pluralism about substantial form.
Scotus rejects both unitarianism and standard pluralism. On his view the
body is indeed an essential part of an organism, and in this sense he agrees
with the standard pluralists against the unitarians. But Scotus denies that the
body is a substance, and in this sense he agrees with the unitarians against the
standard pluralists. Scotus thinks that the body is in fact composed of many
different kinds of composite substances, corresponding to different integral
parts, and thinks that some integral part-substances are themselves composed
of integral part-substances (along with a substantial form). Following Aristotle,
he thinks that heterogeneous parts—parts like faces, hands, hearts, and eyes—
are partially composed of homogeneous parts—parts like bone, flesh, and
blood.12 These substances compose one complete organism when they are
together informed by the soul, and Scotus thinks that any organism, plant,
brute, or human, has just one soul.13 We can call this version of pluralism about
substantial form, Scotistic pluralism.14
10 Ockham is agnostic on the issue of whether the organic parts of animals have substantial
forms of distinct kinds, but he is definitely committed to a plurality of forms. See Marilyn
McCord Adams, William Ockham, v.2, pp. 633–670, and references to Ockham therein.
For his agnosticism about the forms of organic parts, see Quodlibet III, q.6, (OTh IX,
pp. 225–227).
11 Richard of Middleton, De Gradu Formarum, ed. R. Zavalloni, in Richard de Mediavilla et la
Controverse sur la Pluralité des Formes, pp. 154–157. See also the discussion of Richard in
Dorothea E. Sharp, Franciscan Philosophy at Oxford in the Thirteenth Century, pp. 235–239,
and references therein.
12 Reportatio Parisiensis IV-A d.44, q.1, n.2, in (Wadding XI, p. 854); Aristotle, Parts of Animals
II, c.1 646b 11–27 (Barnes I, p. 1006).
13 Ordinatio IV, d.44, q.1, n.4 (Wadding X, p. 98).
14 Scotus is not the first Scholastic to have held that individual integral parts of an organism
have their own substantial forms. The editors of the Opera Philosophica point us to Peter
John Olivi and Peter de Trabibus as early proponents of the view (Bonaventure IV, p. 382,
80 Chapter 5
Scotus considers several arguments for the view that integral parts are dis-
tinct substances, not all of which he considers cogent. He presents several
arguments that reason from distinction of functions and modal properties to
distinction of substance, but finds reasonable rejoinders to each.15 He finds
surer grounds for his claim in two additional arguments. First, Scotus argues
that the fact that in embryological development some integral part are gener-
ated temporally prior to other integral parts makes it probable that the coming
to be of an integral part of an organism is a distinct substantial generation.16
Second, he argues that where two properties, F and G, cannot inhere in the
same subject in exactly the same way, then the form by which a substance is
F is numerically distinct from the form by which it is G. I discuss each of these
arguments in greater detail below.
Scotus recognizes that a major theoretical weakness of his view that the
integral parts of a substance are distinct substances is that it is not clear how
several substances should be able to compose one substance. For Aristotle and
most Aristotelians, distinct objects can form a substantial unity only if one is
potency and the other is act. As the Philosopher said, “A substance cannot con-
sist of substances present in it actually (for things that are thus actually two are
fn.5). For Peter de Trabibus, see the texts printed in Hildebert Alois Huning, “The Plurality
of Forms according to Petrus de Trabibus,” Franciscan Studies 28 (1968), pp. 137–196. For
Olivi, see Peter John Olivi, Quaestiones in secundum librum Sententiarum, q.51, ed.
Bernardus Jansen (Quarracchi, 1922). And after Scotus Henry of Harclay and Albert of
Saxony held something similar to Scotus’s position. For Harclay, see Ordinary Questions
VIII, ed. Mark Henninger (Oxford: Oxford University Press, 2008), pp. 348–397. Regarding
Albert, in his commentary on Aristotle’s On Generation and Corruption he holds that in
the same substance there are “plures forme substantiales partiales distincte specie exis-
tentes partes integrales unius forme totalis [sic],” and then refers the reader to his com-
mentary on On the Soul for his view of the soul. See Duos libros de generatione et
corruptione I, q.5, in Questiones et decisiones physicales insignium virorum [etc.], ed. Lokert
(Paris, 1516), f.132v. Complete versions of the On the Soul commentary only exist in manu-
script form; for information about the manuscripts, see Angel Muñoz Garcia, “Albert of
Saxony, Bibliography,” Bulletin de Philosophie Médiévale 32 (1990), pp. 161–190. The portion
of Albert’s On the Soul commentary published by Marshall does not deliver a determinate
account of Albert’s pluralism (and Marshall himself is dubious about its attribution to
Albert), but see Peter Marshall, “Parisian Psychology in the Mid-Fourteenth Century,”
Archives d’histoire doctrinale et littéraire du moyen age 50 (1983), pp. 101–193. For other
authors, see Pasnau, Metaphysical Themes: 1274–1671, pp. 630–632.
15 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, nn.11–18, in (Bonaventure
IV, pp. 382–383).
16 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.38 (Bonaventure IV,
pp. 389–390).
never actually one…).”17 Bronze and the form of a sphere can compose one bra-
zen sphere, for example, because the bronze is in potency to that form.18 For the
unitarian, prime matter and substantial form can compose one substance
because prime matter is in potency to substantial form. And for the standard
pluralist, body and soul can compose one substance because body is in potency
to soul. On Scotistic pluralism, however, the potency-act analysis of composition
is inapplicable to the sort of unity that, e.g., Mole’s heart and liver have when
they (partially) compose Mole, even though Scotus wants to say that a heart and
a liver can (partially) compose one substance. Whatever potency and act
amount to, neither Scotus nor any Aristotelian of whom I am aware thought that
one composite substance could be in potency to another composite substance.19
Scotus does think that the heart, liver, and other integral parts are together
informed by Mole’s sensitive soul, and therefore thinks that the integral parts are
together in potency to the soul. But this account of the substantial union of the
soul with the integral parts leaves unresolved what it is for the integral parts to
be together in potency. Why, for example, are this heart, this liver, these bones,
and so on, in potency to Mole’s sensitive soul, and some other substances are
not? Scotus thus presents himself with a theoretical challenge that unitarians
and standard pluralists need not face. His response is that some substances are
able to be informed by the soul if they have a special kind of unity, what Scotus
calls a unity of order, which is the sort of unity that things have when one depends
on another (in a technical sense of depends, which I elaborate below).
ii Scotus against Unitarianism
Aquinas and Scotus shared a commitment to parsimony in theoretical

matters, and both take parsimony to be a reason in favor of unitarianism.20
17 Aristotle, Metaphysics VII, c.13, 1039a 3–5 (Barnes II, p. 1640); this passage is quoted as the
first objection to Scotus’s view that the integral parts of animals have distinct substantial
forms. Scotus, Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.1
(Bonaventure IV, p. 381).
18 Aristotle, Metaphysics VIII, c.6, 1045a 21–36 (Barnes II, p. 1650). As for Aristotle so for
Scotus, the brazen sphere is merely a heuristic tool for explaining hylomorphism: as an
artifact, it is excluded from the class of genuine substances.
19 I qualify substance with composite in this sentence because Aristotle sometimes calls
form and matter substances, and under this description it would be correct to say that a
substance (matter) is in potency to another substance (form). Aristotle, Metaphysics VII,
c.3, 1028b 33 – 1029a 7 (Barnes II, pp. 1624–1625).
20 Aquinas, Summa theologiae Ia, q.76, a.3,4, corp.; Scotus, Ordinatio IV, d.11, p.1, a.2, q.1, n.186,
in (Vatican XII, p. 234). In Ordinatio IV, d.11, p.1, a.2, q.1, Scotus argues against the views of
82 Chapter 5
The unitarian thinks that a material substance, s, of some kind, K, has and can
have just one substantial form, K-form, such that K-form united with the mat-
ter of s gives s all of its essential perfections. But Scotus argues against Aquinas
that a “contradiction in being” arises on the unitarian assumption. When Mole
dies, says Scotus, the soul does not remain but the body does. He concludes
that the form of the body is numerically distinct from the soul.21 The inference
is unwarranted however, since, as Marilyn McCord Adams has pointed out,
Scotus presumes what he is trying to prove: that the body remains.22 On the
unitarian thesis a corpse is not identical with the body or any part of the body
of the organism that precedes it. Mole, body and soul, has been corrupted, and
a new substance with qualitative and quantitative features very similar to
Mole’s has been generated—Mole’s corpse. According to Aquinas, the corrup-
tion of an organism naturally tends toward dissolution to the elements,23 but
there is a plurality of middle stages along the way to the elements—Aquinas
identifies dead body and putrefied body as two separate stages—and each of
these stages is itself a composite of matter and some imperfect or merely tran-
sitional substantial form.24
The unitarian then is not committed to holding a blatant “contradiction in
being,” as Scotus accuses him of holding. But in identifying Mole’s corpse as a
newly generated substance, even an imperfect substance, the unitarian must
confront two significant challenges: explaining how the corpse comes to be, and
explaining how the corpse comes to be so very similar to Mole. If the corpse is a
substance and is not identical with Mole’s body, it is something newly generated
and therefore has an efficient cause. But it is not obvious what this efficient
cause could be. Initially plausible is the suggestion that whatever is responsible
for killing Mole is also responsible for generating the corpse, since in normal
cases of generation an efficient cause brings about a new substance from some
preexisting substance. But, according to Scotus, we would expect different kinds
Henry of Ghent as much as those of Aquinas. Henry of Ghent has an interesting view
according to which only humans have a plurality of substantial forms. But as Aquinas’s
views are better known, and as he is the most famous unitarian, I have found it conve-
nient to focus on him.
21 Ordinatio IV, d.11, p.1, a.2, q.1, n.280 (Vatican XII, p. 265). Sic in proposito forma animae non
manente, corpus manet; et ideo universaliter in quolibet animato, necesse est ponere
illam formam, qua corpus est corpus, aliam ab illa, qua est animatum.
22 Marilyn McCord Adams, William Ockham v.2, p. 648; also in Richard Cross, The Physics of
Duns Scotus, p. 56.
23 Aquinas, In de generatione et corruptione I, l.8, n.3; Summa theologiae III, q.50, a.5, corp.
24 Aquinas, In de generatione et corruptione I, l.8, n.3. Aquinas claims to derive his idea
of imperfect forms from Avicenna’s Sufficienta. See Avicenna, The Metaphysics of
of killers to produce different kinds of corpses, and this is often not the case, for
example, whether Mole drowns, is stabbed, or dies of illness, his corpse would
appear to be of the same kind, or at least of a very similar kind.25
Francisco Suárez—a unitarian late in time—attempted to fill this explana-
tory gap in the unitarian position by claiming that “heaven or the author of
nature” is responsible for providing the substantial form of Mole’s corpse at the
moment Mole passes away. In this Suárez was echoing a traditional way of
resolving the problem of spontaneous generation, where, e.g., maggots are
generated from putrid flesh without any apparent efficient cause. The princi-
ple at work seems to be this: where no sublunary efficient cause can explain a
generation, a celestial efficient cause must be posited. Given the principle,
Suárez’s extension of it to cases of corpse-production seems reasonable.26
Aquinas himself does not seem to have explicitly endorsed this view as a way
to account for the generation of corpses, and I have not found any texts in
which Aquinas posits an alternative theory. As far as I can tell, then, Aquinas
tells us what corpses are (composites of matter and imperfect form) without
telling us how they come about.27
Second, the unitarian must explain not only how the new substance comes
to be, but also how it comes to be similar to Mole—it is black, furry, and the
The Healing VIII, c.2, n.15, trans. M.E. Marmura (Provo, ut: Brigham Young University
Press, 2005), pp. 265–266.
25 Ordinatio IV, p.1, a.2, q.1, n.226 (Vatican XII, p. 246).
26 Francisco Suárez, Disputationes metaphysicae 18, 2, n.28, in Opera omnia, v.25, ed. Carlos
Berton (Paris: Louis Vivès, 1861, pp. 608–609). The view that a heavenly body or God sup-
plies what is lacking in the sublunary causal order has a long history. Aquinas offers a
similar solution to the problem of spontaneous generation in Sententia libri Metaphysicae
VII, l.8, nn.25–26. For some of the ancient context of the view, see Devin Henry,
“Themistius and Spontaneous Generation in Aristotle’s Metaphysics,” Oxford Studies in
Ancient Philosophy 24 (2003), pp. 183–208.
27 While Aquinas did not consider Suárez’s solution, Scotus certainly did (not, of course,
under that description), since he invokes the same principle for the generation of all
organisms whose sublunar efficient causes cause by means of seed on the grounds that
the the form of seed is inferior to soul and therefore incapable of generating a living mate-
rial substance. Cf. Scotus, Lectura II d.18, qq.1–2, n.37, in (Vatican XIX, p. 163); Quaestiones
super Libros Metaphysicorum Aristotelis VII, q.12, nn.32–40 (Bonaventure IV, pp. 204–206);
Reportatio Parisiensis II-A, d.18, q.1, nn.11–12 (Wadding XI, pp. 354–355). Moreover, when
Scotus argues for the existence of substantial forms of organs, he presents but then rejects
an argument which uses exactly the same reasoning as the argument from contradiction,
on the grounds that a “universal generator” could induce a new form at the moment the
organ was removed from the body. Cf. Quaestiones super Libros Metaphysicorum Aristotelis
VII, q.20, nn.11–12 (Bonaventure IV, p. 382).
84 Chapter 5
same size and shape as Mole. According to Aquinas accidents go down with
the ship; if the substance is corrupted, so are the accidents, since accidents by
nature inhere in, are individuated by, and (in non-miraculous cases) depend
for their existence on not matter but substance.28 So in addition to holding
that Mole’s corpse has a substantial form which was never a form of Mole,
Aquinas must also hold that all of its accidents were never accidents of Mole.
Suárez argued instead that quantity and some qualities do not depend on sub
stance for their existence but on matter. Wherever the matter goes, therefore,
so go those accidents. Since Mole’s matter survives Mole’s death (albeit not
under the description “Mole’s matter”) and becomes the matter of a new sub-
stance, Mole’s corpse, the accidents inhering in Mole’s matter survive Mole’s
death, too.29 This move allows Suárez to explain the similarity of the corpse’s
accidents to Mole’s—they are numerically identical—at the cost of making
quantified matter play the role of substance without the title.
The intuitive implausibility of the claim that Mole’s corpse is an entirely
new substance with entirely new accidents, coupled with the difficulty of giv-
ing an efficient causal account of the generation of this new substance with its
new accidents, pushes strongly in favor of the pluralist position, Scotus’s hasty
argument from “contradiction in being” notwithstanding.
iii Scotus against Standard Pluralism
In his dispute with Aquinas, Scotus was concerned merely to show that living
substances have at least one substantial form in addition to the soul. If this
were all he was committed to, then it would be natural to think of the body as
one substance informed by one substantial form of corporeity, where the form
of corporeity is responsible not just for an organism’s being a body, but also for
its being a body with a certain structure, with the sorts of integral parts requi-
site for the life of an organism of some kind. Scotus’s view is more complicated
than this, however, since he thinks that an organism’s integral parts are distinct
kinds of substances, each with its own substantial form. This in itself does not
28 Aquinas, In de generatione et corruptione I, l.10, n.6.

29 Suárez, De generatione et corruptione d.1, q.4 (Padua ms. 133), f.37–39v, also in Francisco
Suárez: “Der ist der Mann,” ed. S. Castellote (Valencia: Facultidad de Teología “San Vincente
Ferrer,” 2004), pp. 490–492, hereafter cited using the convention: (Castellote, p.n); see the
parallel discussion in Disputationes metaphysicae 15, 10, Opera Omnia, v.25, ed. C. Berton
(Paris: Louis Vivès, 1861), pp. 536–557. Suárez credits the view to Diego Astudillo, De gen
eratione et corruptione I, q.2, ed. N. Tyerri (Valladolid, 1532), f. v–ix.
entail that there is not a form of corporeity distinct from these several forms of
organs, however, since it could be that the form of corporeity has some other
role in composing an organism. Nor does it entail that the body is not one sub-
stance. For example, Scotus considers whether the role of the form of corpore-
ity might be to unify the integral parts, making them part-substances of one
substance, the body, which is then ready to be informed by the soul. For rea-
sons described later, however, Scotus rejects this thought. These considerations
indicate that Scotus simply has no need for a form of corporeity in addition to
the substantial forms of organs. It neither endows the integral parts with their
respective functions, nor does it unify them.
In this section I consider Scotus’s primary argument for denying that the
form of corporeity endows the integral parts with their respective functions,
an argument that I call the incompossibility argument.
In QMet VII, q.20, Scotus asks whether the organic parts of animals have
distinct substantial forms that are of different kinds.30 He argues that they do.
In the course of arguing for his position Scotus presents and then criticizes an
argument for the opposing positions [i] that there is just one substantial form
in an animal, and [ii] that there are only two substantial forms in an animal. He
identifies a principle shared by both views and attacks this principle en route to
denying both [i] and [ii]. His presentation of these positions is not pellucid,
but it is intelligible. Here is the text:
Another opinion, [i] that the form of corporeity31 precedes the soul (if
there be another [substantial form]): it would be one for the whole, virtu-
ally containing in itself many perfections on account of which [perfec-
tions] it would perfect diverse parts of matter, and would constitute
30 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20 (Bonaventure IV, pp. 381–
394). Although the explicit topic of this question is whether the integral parts of animals
have distinct substantial forms, it is clear that Scotus thinks that an affirmative answer to
this question entails an affirmative answer to the question about whether the integral
parts of animals are distinct substances. See for example Quaestiones super Libros
Metaphysicorum Aristotelis VII, q.20, n.38 (Bonaventure IV, pp. 389–390), and Lectura III,
d.2, q.3, n.117 (Vatican XX, p. 113).
31 What I translate here as “form of corporeity” is actually “form of the mixed” (forma mixti)
and “form of the mixture” (forma mixtionis). These latter two expressions are used fre-
quently in Ordinatio IV, d.11, p.1, a.2, q.1, where they are interchangeable with “form of
corporeity” (forma corporeitatis). Their interchangeability is evident throughout the
whole question but especially in nn.285–286, (Vatican XII, pp. 267–268). Scotus does not
use the expression forma corporeitatis in Quaestiones super Libros Metaphysicorum
Aristotelis VII, q.20.
86 Chapter 5
diverse incomplete organs, namely as imperfect and quasi-remote prin-

ciples of diverse operations. Thus [ii] if the sensitive soul of the brute
animal from its perfection includes the perfection of such a form of cor-
poreity, and in addition to this its own [perfection], it will be able to be
one really and many virtually. And according to diverse perfections virtu-
ally contained, both properties of it and of the form of corporeity (if it
includes it), it will be able to perfect diverse parts of matter and perfectly
constitute diverse organs.32
[i] says that an animal has two substantial forms, one form of corporeity and
one soul; it is a version of standard pluralism. [ii] is the unitarian view that an
animal has just one substantial form, a soul. According to [i] a form of corpore-
ity virtually contains the perfections of the different organic parts. By “virtually
contains” Scotus means that while the form of corporeity itself does not have
organic parts of different functions, it has the power or (in an archaic sense of
the word) the virtue to compose a substance that does have such parts. By
informing matter the form of corporeity makes a substance with incomplete
organs, suitably disposed to be perfected by a soul. The soul then completes the
organs, that is, it makes them fully functioning organs in the life of an animal.
According to [ii] the soul does all the work that the form of corporeity is sup-
posed to do in [i], plus its normal vivifying activity. The common assumption
between these two views is:
[DP] One form virtually contains many perfections, and can give differ-
ent perfections to different parts of matter.
Scotus has two objections to [dp]. First, it cannot explain how a substantial
form gives one perfection to this part of matter and another perfection to
that part.33 In other words it cannot explain how the parts have the particular
structure they have. Why, for instance, does a substantial form give head
p. 385), Opinio alia: Quod forma mixti praecedens animam, si esset alia, ipsa esset una
totius, virtualiter in se continens perfectiones multas, secundum quas perficeret diversas
partes materiae, et constitueret diversa organa incompleta, scilicet principia imperfecta
et quasi remota operationum diversarum. Quare si sensitiva animalis bruti ex perfectione
sui includat perfectionem talis formae mixti, et praeter hoc propriam, poterit esse una
realiter et multiplex virtualiter. Et secundum diversas perfectiones virtualiter contentas,
tam proprias sibi quam fomae mixti—si eam includat—poterit diversas partes materiae
perficere et diversa organa perfecte constitutere.
33 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.31 (Bonaventure IV, p. 387).
structure and neck structure to adjacent parts of matter rather than head
structure and ankle structure? Second, [dp] cannot explain why a substantial
form does not give all of the perfections that it can give to every part of mat-
ter.34 Scotus thinks that the second objection is the more worrisome of the
two, because he takes it to be impossible for the very same part of matter to
have repugnant perfections: for example, a hand and an eye are both partially
composed of matter, but plausibly the very same matter cannot have both eye
structure and hand structure, since a hand needs to be stiff for grabbing and an
eye needs to be soft for receiving sensory images.
In light of this objection Scotus offers a refinement of [dp] on his oppo-
nent’s behalf that is intended to explain how one form virtually containing
many perfections can distribute its perfections to different parts of matter and
can arrange the parts in the right way. Scotus suggests on his opponent’s behalf
that a substantial form has both a virtual totality and a quantitative totality. As
used here, virtual totality is the sum of all the perfections that a substantial
form gives to matter, and quantitative totality is the structure of the distribu-
tion of a form’s perfections. For example, on the unitarian view a sensitive soul
makes a substance that has all the parts needed for sensing, for taking in nutri-
ents and expelling waste, and for procreating, and additionally ensures that
these parts are distributed and arranged in the appropriate way.35
Scotus finds two reasonable ways to interpret this modified version of [dp],
and objects to both. First, it could be that in introducing the notion of quanti-
tative totality the defender of [dp] is claiming that a substantial form has really
distinct parts of different kinds, one part responsible for giving one perfection
to one part of matter, and another part responsible for giving another perfec-
tion to another part of matter. But this, says Scotus, simply amounts to the view
that Scotus himself eventually defends, which is that the form of corporeity is
just reducible to really distinct substantial forms that are parts of a unity of
order: “What is the difference between this subtle opinion which seems to fol-
low reason,” he rhetorically asks, “and the first [Scotus’s own opinion] which
seems gross but consonant with sense?”36 Second, if the defender of [dp] is
p. 388).
p. 388). Scotus says in this passage that an intellective soul has only virtual totality, and
therefore gives all of its perfections to every part of the matter that it informs.
36 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, nn.36–37 (Bonaventure IV,
p. 389), Item, quae est differentia huius opinionis subtilis, quae videtur sequi rationem, et
primae, quae videtur grossa et est sensui consona?
88 Chapter 5
not claiming that a substantial form has really distinct parts of different kinds,
then a contradiction is supposed to ensue.
Here is how the contradiction is supposed to ensue: Scotus holds that if two
perfections are incompossible in the very same matter, then one and the same
form cannot virtually contain those two perfections. The defender of [dp]
agrees with Scotus that some pairs of perfections are incompossible in the
same matter, but insists that the same form can virtually contain such a pair.
Scotus’s task, then, is to show why the consequent is supposed to follow from
the antecedent. Scotus asserts that two perfections are incompossible in some
third thing if and only if they are repugnant to each other. He stipulates as a
second premise that two or more things are in a third thing (insunt tertio) in
exactly two ways: either when they are virtually contained by a form, or when
they inform the same part of matter. (Logically speaking there could be other
ways in which two or more things could be in a third, but Scotus discusses
none of them here.) From these premises it follows that the reason that some
pair of perfections is incompossible in the same matter—that they are repug-
nant to each other—is equally the reason that those perfections cannot be
virtually contained in a single form. But the defender of [dp] does hold that
some pairs of perfections are incompossible in matter, and therefore is logi-
cally committed to the view that one form cannot contain perfections that are
incompossible in matter.37
The argument’s success depends on the truth of Scotus’s implied claim that
there is at least one sense of being in that applies both to the way that perfec-
tions are in matter and to the way that perfections are in a form, such that if
two perfections cannot be in matter then they cannot be in form (and vice
versa). An argument for this claim would have helped his cause; in its absence,
pp. 388–389), Contra ista: si perfectiones aliquae virtualiter contentae sunt incompossi-
biles in materia, igitur et in forma continente. Probatio consequentiae: numquam est
incompossibilitas aliquorum ut insunt tertio, nisi quia inter se sunt incompossibilia.
Unde si album et nigrum non essent inter se repugnantia quin secundum proprias ratio-
nes simul essent in aliqua essentia continente, ut in rubore, secundum proprias rationes
possent idem perficere. “If the perfections of something virtually contained are incom-
possible in matter, therefore also in the containing form. Proof of the consequence: there
is never incompossibility of things as they belong to a third, unless because they are
incompossible among themselves. Hence, if white and black were not repugnant to each
other so that with their own distinctive natures they might be at the same time in some
essence that contains them, for example, redness, they would with their own distinctive
natures be able to perfect the same thing.” (Thanks go to Martin Tweedale for suggesting
an improvement to the translation of the last sentence of this quotation.)
we are left to speculate. Unfortunately for Scotus there seems to be an impor-

tant and relevant disanalogy between the way that perfections are in form and
the way that forms are in matter. If matter is informed by some substantial
form that virtually contains several perfections, say F and G, the result is a sub-
stance that is F and G. But the form containing those perfections is not itself
F and G, and in general forms are not characterized by the perfections that they
virtually contain. Consider two repugnant perfections, being square and being
circular. Obviously these cannot inhere in the very same matter, since the
resulting object would simultaneously be circular and square. But if these are
virtually contained in one form, it does not follow that that form is square and
circular; it only follows that if that form informs matter then it makes some-
thing that is square and circular. But it is exactly here that the objector’s notion
of quantitative totality is supposed to do metaphysical work. The form con-
taining the perfections of being square and being circular is also endowed with
an abstract structure for how these perfections are arranged in matter, or so the
objector claims. The form, let us say, makes one part of matter square and
another part circular. Thus, on the supposition that forms have quantitative
totality, the defender of [dp] can explain why a form does not give all of its
perfections to every part of matter, contrary to Scotus’s second objection.
Quantitative totality also furnishes the [dp]-defender with a way of explaining
how a form gives structure to the parts of matter. Scotus’s incompossibility
argument appears, therefore, to fail.
Fortunately for Scotus, however, there is an additional reason in favor of
Scotus’s rejection of [dp]. It is the thought that, if repugnant perfections can be
virtually contained in one (non-complex) form, then every perfection can be vir-
tually contained in one (non-complex) form. We might as well hold that the
whole world is informed by one substantial form. But this spells doom for a hylo-
morphic analysis of change, according to which form and matter (along with
privation) are invoked as those principles of nature that explain qualitative and
substantial change.38 For example, Socrates’s change from white to tan is par-
tially explained as the loss of a form of whiteness and the acquisition of a form of
tanness. Thus, on the view that there is simply one form of the world, or more
modestly that whiteness and tanness can be virtually contained by one and the
same form, we lose whatever original motivation we had to posit forms. Better to
hold, then, as Scotus apparently does, that we distinguish between forms at least
partially on the basis of opposed perfections. When two perfections cannot be
actualized in the very same matter, we attribute them to two different forms.39
38 Aristotle, Physics I, 6–8 (Barnes I, pp. 323–328).

39 I discuss this argument in greater detail in Chapter 9.
90 Chapter 5
Fortunately for me, however, my aim here is neither to defend nor refute
Scotus but merely to point out that his incompossibility argument signals his
rejection of the idea that one form can account for the varied and incompati-
ble perfections of an animal, whether this one form is a soul or a form of
corporeity.40
iv An Inconsistent Position about the Form of Corporeity?
Despite his apparent rejection of the form of corporeity in favor of distinct

substantial forms of organic parts in QMet VII, q.20, Scotus sometimes uses the
expression form of corporeity and its synonyms to express his own views, so it
is important to say something about how his apparent rejection of the form of
corporeity in QMet VII, q.20 is consistent with passages in which he does not
reject the form of corporeity. On my view, Scotus’s use of “form of corporeity”
and similar expressions is shorthand for all of the substantial forms that com-
pose the body, so I do not think that Scotus is inconsistent. Richard Cross
40 Scotus’s rejection of the form of corporeity in favor of many forms of integral parts pro-
vides him with a nice response to one of Aquinas’s objections to pluralism—an objection
that has some force against standard pluralism, and that Suárez later directed against
Scotus himself. Aquinas and Suárez argued that pluralism entails that there could be a
body that was not of any determinate kind, that was simply corporeal. From a pre-
Cartesian, Aristotelian perspective, a body that is not of any kind is metaphysically unwel-
come not just because it is difficult or impossible to conceive, but more importantly
because it would tread a middle ground where a middle is excluded. On the Porphyrian
Tree animate and inaminate exhaustively divide the genus body, so every body must be
one or the other; moreover sensitive and nonsensitive exhaustively divide the genus ani-
mate body, so every animate body must be one or the other. Against standard pluralism
this seems like a good objection. Against Scotistic pluralism, however, this criticism misses
the mark: according to Scotus there is no such thing as body as such. As deep as one
wishes to analyze a substance’s compositional structure, one never arrives at a part that is
simply a body. There is always a body of a certain kind: the heterogeneous parts (such as a
heart, liver, or hand) composed of a substantial form together with homogeneous parts
(such as flesh and bones), and the homogenous parts composed of prime matter and sub-
stantial form. For Aquinas and Suárez’s objection, see Aquinas, Sententia libri Metaphysicae
VII, l.12, n.9; Suárez, De generationes et corruptione d.1, q.2 (Castellote, p. 465), […] nam si
forma constituens genus et speciem sunt distinctæ, posset separari inferior, servata supe-
riori, ut posset separari anima, servata forma corporeitatis, et manere hoc corpus sub
nulla specie corporis. “[…F]or if the forms constituting genus and species are distinct, the
inferior can be separated, preserving the superior, as the soul can be separated, preserving
the form of corporeity, and can remain this body under no species of body.”
thinks that Scotus is inconsistent, claiming that whereas QMet VII, q.20 holds
that the forms of organs are really distinct parts of the form of corporeity,
according to Ordinatio IV, d.11, q.3 the forms of organs are in potency to the form
of corporeity.41 On my reading, while Ordinatio IV, d.11, q.3 is less explicit than
QMet VII, q.20 about Scotus’s commitment to a plurality of forms of integral
parts, and his rejection of a form of corporeity distinct from these, it is never-
theless consistent with it. To show this, I consider three passages from Ordinatio
IV, d.11, q.3, including the passage that Cross cites as proof of the inconsistency
between this text and QMet VII, q.20.
The first passage is Ordinatio IV, d.11, q.3, n.56. On Cross’s reading, Scotus
claims that the presence of the organs is not necessary for the inherence of the
form of corporeity.42 If correct, this would show that the form of corporeity is
really distinct from the forms of organic parts, since the former could exist
without the latter. I do not think that this is the correct reading, however. In
context, Scotus is discussing some general features of what it is for a body to be
properly disposed for receiving the intellective soul. A body needs to be dis-
posed in a certain way if it is to be informed by the soul, but the particular
grades of qualities necessary for the soul to inform the body are not necessary
for the form of corporeity to inform matter. The body could have some other
grades of qualities, suitable for it to go on existing, even if these make the body
unsuitable for being informed by the intellective soul. Scotus writes,
For the animation by the intellective soul there is needed a heart and
liver of determinate heat, a brain of determinate frigidity, and so on for
individual organic parts. But such a disposition ceasing, there can still
remain some species or quality which stays with the form of corporeity,
although not that which is required for the existence and operation of
the intellective soul in matter.43
Scotus does not say here that the organs can cease to exist while the form of
corporeity remains united with matter. He says instead that the grades of the
qualities of the organic parts can cease to be suitable for the intellective soul
41 Richard Cross, The Physics of Duns Scotus, p. 69.

42 Richard Cross, The Physics of Duns Scotus, pp. 69–70.
43 Ordinatio IV, d.11, p.1, a.2, q.1, n.284 (Vatican XII, p. 267), Istud patet exemplo, quia ad ani-
mationem ab intellectiva requiruntur cor et hepar determinate calida, cerebrum cetermi-
nate frigidum, et sic de singulis partibus organicis; tali autem dispositione cessante,
potest adhuc manere aliquas species vel qualitas, quae stat cum forma mixtionis, licet
non illa quae requiritur ad esse et operationem intellectivae in materia.
92 Chapter 5
even if they are suitable for the body. Presumably this is how Scotus would
describe the death of an organism where the body remains intact but is no
longer alive.
In the second passage, Scotus argues against what he takes to be Henry of
Ghent’s view that one potency in matter entails that only one substantial form
can actualize that potency. Scotus agrees with Henry that the natural genera-
tion of a substance is completed by exactly one substantial form, but insists
that this final generation may be preceded by several generations of parts each
of which has its own substantial form. Then he gives an example:
For example: if organic parts are posited as differing according to sub-

stantial forms, then the generation of one precedes, not only by nature
but also temporally, both the generation of other parts and that genera-
tion which is simply the generation of the whole, namely [that genera-
tion] by which the form of the whole is induced, the forms of all the parts
already presupposed.44
Here the view is that some organic parts are generated prior to others and prior
to the generation of the whole substance. Scotus’s use of “form of the whole”
here could be read either as the final perfecting substantial form, or as the
whole substance itself which is produced by the concurrent causality of the
essential parts. On either reading, however, Scotus has in mind that change by
which the whole complete substance is brought about, and not that by which
a form of corporeity is brought about.
In the third and final, Scotus, replying to Aquinas’s worry that a plurality of
substantial forms compromises the unity of a substance, writes,
I concede that the formal existence of the whole composite is principally

through one form, and that form is that by which the whole composite is
this being. But that form is the last, coming to all the preceding. And in
this way the whole composite is divided into two essential parts in proper
act, namely the last form by which it is that which it is, and the proper
potency of that act which includes prime matter with all the preceding
forms […] An example of this is in a composite of integral parts; for, the
44 Ordinatio IV, d.11, p.1, a.2, q.1, n.238 (Vatican XII, p. 250), Exemplum huius est si ponantur
partes organicae differre secundum formas substantiales: tunc enim generatio unius
praecedit non solum natura sed etiam tempore generationem alterius partis, et etiam
generationem illam quae est simpliciter generatio totius, qua scilicet inducitur forma
totalis praesuppositis iam formis omnium partium.
more perfect the animate thing, the more it requires a plurality of organs,
and it is probable that [they are] distinct kinds through substantial
forms.45
Clearly this text is at least consistent with Cross’s view, since it could be that “all
the preceding forms” refers to all of the forms of organic parts as well as a sepa-
rate form of corporeity. It is suggestive however that Scotus’s example of his
view makes no reference to a separate form of corporeity but only to integral
parts, which “probably” have distinct substantial forms of distinct kinds. The
“last form” clearly refers to the final perfecting substantial form or soul, and not
a form of corporeity.
Despite these textual considerations, I do not think that Ordinatio IV, d.11,
q.3, taken on its own, presents a determinate view about the relation between
the form of corporeity and the forms of organic parts. Scotus’s explicit conclu-
sion is that in a living substance such as Christ, there is in addition to matter
and soul, at least one form of corporeity.46 And the conclusion from the argu-
ment from contradiction is simply, “It is necessary to posit that form, by which
the body is a body, distinct from that by which it is animate.”47 Both of these
texts are consistent with Cross’s view, and the use of the singular (that form) in
the second reference could be taken to lean in favor of his view. Equally plau-
sible, however, is that the unity of the form of corporeity—the unity by which
it is one—is just the special sort of unity that the substantial forms of organic
parts have, and this is the natural way to read the text in light of QMet VII, q.20.
As I read the texts, what is merely adumbrated in Ordinatio IV, d.11, q.3, is crys-
tallized in QMet VII, q.20. In the following chapter I offer a close reading of this
question which is intended to establish that Scotus abandons the forma corpo
reitas as it is traditionally conceived, in favor of substantial forms of many
organic parts.
45 Ordinatio IV, d.11, p.1, a.2, q.1, n.252 (Vatican XII, pp. 255–256), Concedo quod totale “esse”
totius compositi est principaliter per formam unam, et illa est forma, qua totum composi-
tum est “hoc ens”; illa autem est ultima, adveniens omnibus praecedentibus; et hoc modo
totum compositum dividitur in duas partes essentiales: in actum proprium, scilicet in
ultimam formam, qua est illud quod est, et in propriam potentiam illius actus, quae
includit materiam primam cum omnibus formis praecedentibus […] Exemplum huius est
in composito ex partibus integralibus: quanto enim animatum est perfectius tanto
requirit plura organa, (et probabile est quod distincta specie per formas substantiales.
[My emphases.]
46 Ordinatio IV, d.11, p.1, a.2, q.1, n.285 (Vatican XII, pp. 267–268).
47 Ordinatio IV, d.11, p.1, a.2, q.1, n.280 (Vatican XII, p. 265).
Chapter 6
Scotistic Pluralism about Substantial Form—Part II
i The Special Potency Question
Having divided up the bodies of animals into many different substances,

Scotus tries to find a way to maintain that one animal is still one substance.
Scotus thinks that some substances are part-substances of Mole just in case
they are informed by Mole’s sensitive soul, and in general thinks that some
substances become part-substances of another substance when there is some
substantial form informing those substances. Substantial form is therefore
Scotus’s answer to (a suitably paraphrased version of) van Inwagen’s Special
Composition Question: under what conditions do some substances compose a
substance?1 But Scotus does not think that a substantial form can inform just
any assortment of substances; he thinks instead that there are at least two
restrictions on the composition of substances. First, if some substances p1…pn
can be informed by a substantial form of a kind, K, they themselves are of the
appropriate kinds. A Mole-soul cannot inform my briefcase and the Eiffel
Tower, for example, since these cannot support the activities of a mole. Second,
p1…pn must be unified in some way. Suppose we have two mole fetuses, one
right here and another over there; the bottom half of one and the top half
of the other are not able to be informed by one and the same Mole-soul,
since they do not have the right sort of unity. Scotus therefore asks a second
question about composition, which presupposes his answer to the Special
Composition Question, and which we might call the Special Potency Question:
under what conditions are some substances able to be informed by a unifying
substantial form?
Scotus’s answer is that some substances can be informed by a substantial
form just in case they are of the appropriate kinds and are unified in what he
calls a unity of order. Being of the appropriate kinds, in this context, means
being the sorts of things that a complete substance requires in order to per-
form its specific functions; Mole, for example, needs certain organs for his sen-
sitive and vegetative functions. Unity of order (unitas ordinis) is a technical
term. I will have much more to say about this kind of unity below but for now
it is worth noting that Scotus thinks that if two or more things are so united
then they have more unity than the parts of an aggregate (such as a pile of
1 Peter van Inwagen, Material Beings (Ithaca, ny: Cornell University Press, 1990), p. 21.

Scotistic Pluralism about Substantial Form—part ii 95
bricks), but less unity than the parts of a unity of inherence (the unity that
holds between a subject and its accident).2 In this section I motivate Scotus’s
answer to the Special Potency Question. I do this, first, by analyzing an argu-
ment—what I call the unity argument—that Scotus presents on his opponent’s
behalf, which aims to show that an animal cannot be a substantial unity on the
supposition that its parts are distinct substances; second, I present Scotus’s
response to the argument.
An assumed premise of the unity argument is that an animal is more unified
than a mere aggregate, such that any account of the parts of animals that does
not preserve this difference in degree of unity is ipso facto a bad account. This
is intended to be uncontroversial: there is some intuitive appeal to the idea
that, say, the heart, lungs, bone, liver, and so on of Mole compose something
more unified than the assortment of bricks in a pile of bricks, or an assortment
of organs in an organ bank, even if it is hard to pin down exactly what this
greater unity is supposed to amount to. Scotus himself accepts the premise.
The first part of the argument reads as follows:
That opinion [of Scotus’s, viz., that the organic parts of animals have
distinct substantial forms of different kinds], if it posits some unity in
the animal beyond aggregation (the way in which a heap of stones is
one), it is necessary that one posit one of three modes: either [a] that its
unity is from a final form, namely the sensitive soul, which is simply one
kind in the whole, although extended per accidens through the exten-
sion of the whole; or, [b] that before that there is posited one form of
corporeity, which is a disposition for the sensitive soul; or, [c] the third
mode, that there is not a form that is of one kind, but only forms of
many kinds, from which one form is integrated, by which there is a
unity of composite.3
2 For Scotus’s list of kinds of unity, see Ordinatio I, d.2, p. 2, qq.1–4, n.403 (Vatican II,
pp. 356–357).
3 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.19 (Bonaventure IV, pp. 383–
384), Ista opinio si ponat aliquam unitatem in animali praeter aggregationem—quo modo
unus est acervus lapidum—oportet quod uno trium modorum ponat—Vel quod eius unitas
sit ab ultima forma, scilicet anima sensitiva, quae sit simpliciter una specie in toto, licet
extensa per accidens ad extensionem totius—Vel quod ante istam ponatur una forma mixti,
quae sit dispositio ad sensitivam—Vel tertio modo, quod nulla est ibi una forma specie, sed
tantum multae specie, ex quibus integratur una forma a qua est unitas compositi. (I have
profited from Martin Tweedale’s suggested translation of the part of the quotation following
tertio modo.).
96 Chapter 6
[a] expresses the view that various organic parts are unified when they are
informed by one soul. [b] expresses the view that there is an intermediary
form—the form of corporeity—between the forms of organic parts and the
soul, such that the form of corporeity unifies the organic parts and makes them
capable of receiving the soul. [c] holds that the many different kinds of forms of
organic parts are “integrated” such that they make up a single form, which unites
with matter to compose one substance. [c] does not collapse into a version of
[a] or [b] because in [c] the forms of integral parts are unified, not the integral
parts themselves. The opponent continues the unity argument by showing that
[a], [b], and [c] do not adequately explain the unity of the organic parts.
Against [c] the opponent invokes the closing lines of Metaphysics VII, in
which Aristotle argues for something in addition to the letters, A and B, that
explains their jointly composing the syllable, AB.4 Here Aristotle calls this
additional something substance (οὐσία) but the opponent refers to it as that
which is act with respect to the parts, suggesting that he took Aristotle’s use of
substance (in this passage) as synonymous with substantial form. Taking
Aristotle’s cue the objector reasons by analogy that the forms of organic parts
would become one form only if there could be some higher-order form that
unifies them, just as the letters become one syllable only if there is a form that
unifies them. But, the objector continues, a form cannot inform forms.
Therefore distinct forms cannot compose one form (contrary to [c]). The sec-
ond premise holds given the general scholastic thesis that a form is act and not
in potency to receive the act of an additional form. Therefore [c] cannot
explain how organic parts compose a substantial unity.
One natural modification of [c] is that an additional substantial form uni-
fies the integral parts themselves, and not their forms. [a] and [b] are two such
modifications of [c]: each posits a substantial form that unifies the organic
parts, making them part-substances of an animal. But according to the oppo-
nent both [a] and [b] face difficulties as well:
Both are disproved though this argument: For a unity of perfection there
is presupposed one proper perfectible, such that the proper unity of the
perfectible is presupposed by the unity of perfection and is not from it.
How will you give such a unity in that which is perfectible by the sensitive
[soul], according to the first way, or in that which is perfectible by the
form of the mixture, according to the second? It does not seem possible.5
4 Aristotle, Metaphysics VII, c.17, 1041b 11–33 (Barnes II, p. 1644).

5 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.21 (Bonaventure IV, p. 384),
Uterque improbatur per hanc rationem: unius perfectionis est unum perfectibile proprium
The opponent reasons that something is in potency to receive a substantial

form only if it is already unified in some way. “Unity of perfection” in general
refers to the sort of unity that a form, whether substantial or accidental, makes
with its subject. In the text it refers specifically to the unity that a substantial
form and its subject have. The “perfectible” in general refers to that which is in
potency to a form, and in the text it refers specifically to that which is in
potency to a substantial form. “Proper unity” is used here in a vague way. At
minimum it refers to whatever sort of unity several substances must have if a
substantial form can perfect them. From the argument against [c] we can infer
that this proper unity is something less than substantial unity, because there
the opponent claims that substantial unity only comes about through substan-
tial form. We can also assume that proper unity is greater than aggregative
unity, since aggregates are indifferent to the precise arrangement of their parts,
and also to the kinds to which their parts belong, whereas the proper unity of
a perfectible is indifferent to neither. The arguments against [a] and [b] close
the unity argument.
Since Scotus accepts the opponent’s view that substances must be unified in
some way if they are in potency to a substantial form, the unity argument
leaves him with the formidable challenge of providing some sort of unity that
is less than substantial but more than aggregative. Scotus’s response to the
challenge is difficult and (I think) ultimately impossible to understand without
recourse to texts other than QMet, but here it is:
To [the unity argument] I respond: material parts, which are called ele-
ments at the end of [Metaphysics] VII [1041b 11–16], do not have as much
unity before they receive form as they have from form; just as A and B in
themselves [do not have as much unity] as they have from the form of the
syllable. Therefore a unity of order among the parts of matter in regard to
the unbounded form suffices in some way, namely that the whole matter
from those [parts] may have an order to such a form as to an adequate
act, in regard to which no part of matter would be adequate potency.6
praesuppositum, ita quod illa propria unitas unius perfectibilis praesupponitur unitati per-
fectionis, nec est ab ea. Quam talem unitatem dabis in perfectibili sensitivae, secundum pri-
mam viam; vel in perfectibili formae mixtionis, secundum secundam? Non videtur possibile.
6 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.48 (Bonaventure IV, pp. 392–
393), [P]artes materiales, quae vocantur elementa in fine VII, non habent tantam unitatem
antequam recipiant formam quantam habent a forma; sicut a et b in se, quantam habent a
forma syllabae. Sufficit igitur respectu formae illimitatae aliquo modo unitas ordinis in par-
tibus materiae, scilicet quod tota materia ex illis ordinem habeat ad talem formam ut ad
actum adaequatum, respectu cuius nulla pars materiae esset potentia adaequata.
98 Chapter 6
Scotus calls this unity, a unity of order, which is supposed to be immune to

the argument against [a] and [b] insofar as the organic parts are more unified
than the parts of an aggregate, and simultaneously immune to the argument
against [c], since it is the integral parts and not their forms that have a unity of
order. It is clear from context that by “material parts” Scotus does not have
prime matter in mind, and that by “elements” he does not have earth, water, air,
and fire in mind. Instead, he means those organic parts which when properly
united are in potency to the soul and in this sense are material, just as A and B
are the matter of the syllable, AB. The “unbounded form” is the soul or perfect-
ing substantial form prior to its reception in the body. Several substances
together are “perfectible” by a substantial form, then, if there is some unity of
order among them.7
ii Essential Orders
So what is this unity of order? The expression unitas ordinis and its inflected
variants do not occur often in Scotus’s corpus. In a brief taxonomy of kinds of
7 A slight twist to my translation of Quaestiones super Libros Metaphysicorum Aristotelis VII,

q.20, n.48 yields a very different position, so it is worth defending my take in a little more
detail. In my translation (and in Wolter and Etzkorn’s) the paragraph employs two notions of
order: the unity of order that obtains among the material parts, and because they are so
ordered, the order that that material parts have as to their adequate act. But the Latin could
permit a reading according to which only one notion of order is employed: the unity of order
of all the material parts (taken together) to the substantial form. An anonymous reader sug-
gests that this is the correct way to read the passage. From context, however, it is clear that
this alternate reading won’t do. The unity argument revealed a weakness in Scotus’s position
that organic parts have their own substantial forms, namely, that these part-substances must
be sufficiently unified before they can together be informed by a substantial form. For his
view to stand up, Scotus must offer an adequate account of the unity of these part-sub-
stances. n.48 presents Scotus’s solution to the problem: the part-substances are more than
just an aggregate and they are less than a substance; they are instead a unity of order, and
because of this order they can together be informed by a substantial form. On the reading
espoused by my anonymous referee, Scotus simply asserts what his opponent denies, namely
that several substances can be informed by a substantial form. On my reading Scotus is offer-
ing the conditions under which several substances can be informed by a substantial form,
conditions which are developed precisely to address and dispel his opponent’s objection.
I therefore see no reason to adopt the alternative reading of the paragraph. For Wolter and
Etzkorn’s translation see Scotus, Questions on the Metaphysics of Aristotle, v.2, translated by
Girard J. Etzkorn and Allan B. Wolter (St. Bonaventure, ny: Franciscan Institute Publications,
1998), p. 339.
unity in Ordinatio I, Scotus says simply that a unity of order is a greater unity
than an aggregate and a lesser unity than a relation of informing.8 But this is
not very helpful to us, since we have already gathered that part-substances
must be more unified than an aggregate and less unified than a unity of perfec-
tion. Intuitively, the relevant sort of unity that some parts must have in order
to be potentially alive is a kind of cohesion and/or organization. We will not
get a mole simply by heaping together some mole parts; we need them to be
properly related or ordered to one another. I suggest that we can get a richer
sense of this proper order by attending to Scotus’s use of “unity of order” in his
late treatise, De Primo Principio.9 When he talks of a unity of order in this
work—and he discusses unity of order far more here than in other works—he
has in mind the unity that obtains among the relata of the relations essentially
prior and essentially posterior. Thus, x and y have a unity of order if and only if
one is essentially prior to the other. For example, Scotus says that the world is
a unity of order because everything in the world is essentially ordered both in
terms of perfection and of dependence.10 Also, the four causes—final, effi-
cient, material, formal—are essentially ordered in the production of their joint
effect, and therefore compose a unity of order.11 Plausibly then when Scotus
says that part-substances are sufficiently unified for receiving substantial form
if they compose a unity of order, he means that these parts are in some sense
essentially ordered to one another.
But Scotus identifies several different kinds of essential orders, and it is not
obvious just how part-substances are supposed to be essentially ordered to one
another, at least in a way that is relevant to their joint functioning as the mate-
rial cause of a material substance. Scotus’s text is unfortunately silent about
precisely what order he had in mind. Nevertheless, through some careful read-
ing of De Primo Principio it is possible to fill out Scotus’s account of the unity
of part-substances in a way that is faithful to Scotus’s deep metaphysical com-
mitments, and presents Scotus in QMet VII, q.20, n.48 simply as alluding to a
8 Ordinatio I, d.2, p. 2, qq.1–4, n.403 (Vatican II, pp. 356–357).

9 The work is now generally assumed to be very late. Likely it is a compilation, done with
the help of a secretary, of writings taken from the first book of Ordinatio, a relatively late
work. For discussion of On the First Principle’s genesis, see Allan B. Wolter, A Treatise on
God as First Principle (Chicago: Franciscan Herald Press, 1966), pp. ix–xiii; hereafter cited
by the convention: (Wolter, p.n). For clear exposition of Scotus’s theory of essential orders,
see Marilyn McCord Adams, “Essential Orders and Sacramental Causality,” in John Duns
Scotus, Philosopher: Proceedings of “The Quadruple Congress” on John Duns Scotus, ed.
Mary Beth Ingham and Oleg Bychkov (Münster: Aschendorff Verlag, 2010), pp. 191–206.
10 De Primo Principio III.49 (Wolter, pp. 25–27, 65).
11 De Primo Principio II, 30 (Wolter, pp. 25–27).
100 Chapter 6
well-developed metaphysical view rather than reaching for an ad hoc solution.

To this end, in this section I elaborate the main features of essential orders, and
in the following section I apply the framework of essential orders to the unity
of order of organic parts, raising and then responding to an objection to this
application.
Scotus divides essential orders into two sorts: an order of eminence and an
order of dependence.12 Only the second is relevant here. In an essential order of
dependence, x’s being prior to y is a matter of y depending on x. Scotus distin-
guishes six types of orders of essential dependence. Each of the four Aristotelian
causes is essentially prior to its correlative: anything that is ordered to an end
is essentially dependent on a final cause; any effect is essentially dependent on
an efficient cause; anything made of matter is essentially dependent on a
material cause; and any formed thing is essentially dependent on a formal
cause.13 The remaining two essential orders of dependence—what we might
call nonevident essential orders14—hold among effects of a common cause. In
the following I briefly discuss, first, some general features of essential orders,
next, the causal essential orders, and finally the nonevident essential orders.
In the most general sense, x and y are essentially ordered if and only if either
x is essentially prior to y and y is essentially posterior to x, or y is essentially
prior to x and x is essentially posterior to y.15 Relations of essential priority and
posteriority are transitive (if x is essentially prior to y and y is essentially prior
to z then x is essentially prior to z), but noncircular (if x is essentially prior to y
then y is not essentially prior to x) and irreflexive (x is not essentially prior or
posterior to itself).16 It can and frequently does turn out that some x is essen-
tially prior to some y in one kind of essential order, but essentially posterior to
y in a different kind of essential order. For example, Scotus thinks that matter
is essentially prior to form in the essential order of dependence, but essentially
posterior to form in the order of eminence.17
Not every causal relation is an essential order of dependence. In the orders
of final and efficient causality, Scotus distinguishes essentially from acciden
tally ordered causal series.18 In an accidentally ordered causal series, an effect
does not depend on its cause in its causing. In Scotus’s example, a son can

14 The label was suggested to me by Marilyn McCord Adams, riffing on Scotus’s own descrip-
tion of these orders as nec in se patet.
16 De Primo Principio II.2–7 (Wolter, p. 15).
17 De Primo Principio II.32 (Wolter, p. 27).
procreate whether or not his parents are alive.19 In an essentially ordered

causal series an effect does depend on its cause in its causing. For example, a
paintbrush and a painter are efficient causes of a painting, but the paintbrush
depends for its causality (i.e., its stroking the canvas with paint) on the causal-
ity of the painter (i.e., her moving the brush in the appropriate way). As an
example from the order of final causality, if Mole and Rat are paddling the boat
up the river in order to find a quiet spot for a picnic, then their paddling is
essentially ordered to finding a quiet spot for a picnic—for the sake of finding
a spot for a picnic they are paddling, and they would not be paddling were it
not for finding a spot for a picnic. I am not sure how to describe an accidentally
ordered final causal series, however, and Scotus does not offer an example of
an accidentally ordered final causal series.20
Essential orders are also distinguished from accidental orders in that all the
causes of an essentially ordered causal series are simultaneous in their caus-
ing.21 To see why Scotus thinks this, imagine three terms in an essentially
ordered causal series, c1, c2, and c3, such that c1 causes c2 and c2 causes c3. If c2 and
c3 were not simultaneous, then c2 would be either before or after its effect.
Obviously it would not be after its effect. If it were before its effect, then c1
would be causing c2 to cause c3 for some period of time during which c2 was not
causing c3—a contradiction.
At least some essentially ordered items are not just things but things insofar
as they are causing.22 For example, an end is a (final) cause not because it
causes the efficient cause, but because it causes the causality of the efficient
cause, insofar as the end is desired and is therefore that for the sake of which
the efficient cause (efficiently) causes. Similarly, the efficient cause produces a
material substance by causing the causality of the matter and the form, the
product of which is the material substance itself.
In nonevident essential orders, essential dependence relations obtain
among two or more effects of some common cause. According to Scotus,

20 The problem is that I cannot conceive how the mere existence of x could be for the sake
of y (that is, how y could be the final cause of x and not some activity of x). Final causality
appears to me to be inextricable from action in a way that efficient causality is not. I do
not have any conceptual hang-ups, for example, with the idea that Mole’s parents are
simply the efficient causes of Mole. Even in the theological context in which Scotus was
at home, according to which everything exists for God’s sake, is it not that what is really
meant is that everything (ultimately) acts for God’s sake? But the difficulties raised here
will have to wait their turn.
22 De Primo Principio II.33 (Wolter, pp. 27–29).
102 Chapter 6
If one and the same cause [z] produces a dual effect [x and y], one of
which [x] is such that by its nature it could be caused before the other [y]
and therefore more immediately, whereas the second [y] can be caused
only if the first [x] is already caused, then I say that the second effect
[y] is posterior in the order of essential dependence whereas the more
immediate effect [x] of the same cause [z] is prior.23
Notice that Scotus here leaves it open as to whether x is itself a cause of y; he

says only that y essentially depends on x in such a way that z must cause x
before it can cause y.
Among the nonevident essential orders of dependence, Scotus distinguishes
proximate from remote nonevident essential orders of dependence (I will refer
to these as [pneo] and [rneo], respectively). x and y are in a [pneo] if and
only if x and y are effects which are essentially ordered to their cause, z, and z
could not cause y unless it first caused x. r and t are in a [rneo] if and only if r
and s are effects which are essentially ordered to their cause, q, and s could not
cause its effect, t, unless q first caused r.24
By distinguishing two types of nonevident essential dependence ([pneo]
and [rneo]), Scotus does not provide much by way of guidance for identifying
what the relata of such orders might be. He is not completely silent on this
issue, however. If x essentially depends on y (for its existence, for its causing)
but x is not a final, efficient, formal, or material cause of y, then x and y stand
either in a [pneo] or a [rneo]. Using Scotus’s own example, suppose that a
qualitative form (such as a color form) essentially depends on a form of quan-
tity in the sense that something must be extended if it is to be colored. They are
effects of a common cause in the sense that an efficient cause causes some-
thing that is both quantified and colored. Scotus says that while quantity is an
effect more immediate than quality, “It is not the cause of quality, as is evident
if we go through the causes.”25 We are still inclined to say that a thing must be
extended if it is to be colored, however, so we say that color non-causally (non-
evidently) depends on quantity in a [pneo].26

24 De Primo Principio I.13 (Wolter pp. 7–9); in these formulations of [pneo] and [rneo] I am
assuming noncircularity and irreflexivity since these hold for all essential orders.
25 De Primo Principio II.36 (Wolter, p. 29).
26 Scotus’s denial in De Primo Principio II.36 that quantity is a cause of quality, even a mate-
rial cause of quantity, seems inconsistent with aspects of his eucharistic metaphysics. For
example, in Ordinatio IV, d.12, p. 1, q.2 (Vatican XII, pp. 330–348), Scotus argues that in
transubstantiation, where the substance of bread and wine are converted to the body and
blood of Christ but the accidents of bread and wine remain numerically and specifically
the same and do not become accidents of the body and blood of Christ, quantity is the
iii Essentially Ordered Unity
Applying this procedure to the question about the sort of unity of order that
organic parts are supposed to have, it is clear that they could not be related
causally to one another as formal or material causes. One substance—a heart,
for example—can be a (partial) material cause of Mole in the sense that the
heart, when suitably unified with other organic parts, can receive a substantial
form and become a part of Mole. But a heart cannot be a material cause of a
liver or of any other part-substance, since the heart neither becomes a part of,
nor becomes, a liver. Nor can a substance be the formal cause of another sub-
stance, since it is of the wrong ontological kind—only forms are formal causes.
But there are reasons to think that part-substances—or at least some of
them—are essentially ordered to one another in both final and efficient causal
series. Scotus hints that this is what he has in mind in QMet VII, q.20, n.38,
where he invokes Aristotle’s authority for the view that in the generation of
blooded animals the heart is formed prior to other parts:
According to the Philosopher the heart is generated first—even tempo-

rally—before the other parts of an animal. And he would assign in the
generation of an animal—speaking of the whole—many complete
changes to many forms of parts, one before the other in time.27
In at least one important passage in which Aristotle expresses this view, he says
not only that the heart is generated first but also that it is an efficient cause of
the generation of the other parts, and is generated for the sake of other parts.28
Scotus’s allusion to Aristotle’s embryology strongly suggests that he would
endorse Aristotle’s idea that organic parts are causally ordered to one another.
In the following I limn these two ways in which Scotus might have thought that
the parts of an animal are essentially ordered, starting with final causality.
First, there is good reason to think the unity of order that obtains among
several part-substances is the unity of an essentially ordered final causal series.
subject of quality. Within Scotus’s metaphysics and in general scholastic Aristotelian

metaphysics, it is not clear how x could be the subject of y without x being a material
cause of y. Whether this signals a change of mind is a question I leave for others to answer.
pp. 389–390), Secundum Philosophum cor primo generatur—etiam tempore—ante alias
partes animalis. Et esset assignare in generatione animalis—loquendo de tota—multas
mutationes completas ad multas formas partium, unam ante aliam tempore.
28 Aristotle, On the Generation of Animals II, c.5, 739b 32 – 740a 24; c.6, 742b 3–5 (Barnes I,
pp. 1148, 1152).
104 Chapter 6
On this view, two part-substances are unified just if one is, or comes to be, or
acts, for the sake of another. On Scotus’s understanding of final causality, if pro-
ducing blood and blood vessels are final causes of producing the heart—if
(part of) the heart’s function is to produce blood and blood vessels—then
there is a causal series such that an efficient cause brings about the heart for the
sake of blood and blood vessels. Some blood vessels are produced for the sake
of carrying nutriments from the mother’s uterus, and some are produced for
the sake of transporting blood to the region where the brain develops. The
brain is produced for the sake of maintaining an optimum ratio of coldness to
heat, and this ratio is supposed to be the mechanism by which the developing
embryo can transform nutriments into organic parts—and therefore it is
brought about for the sake of these organic parts.29 Each final causal chain
terminates at some activity of the whole organism composed of many part-
substances, and therefore every part-substance comes to be for the sake of the
whole, even if there is some pair of part-substances such that neither is a final
cause of the other. When there is such a pair of substances, say p1 and p2, p1 and
p2 can still be essentially ordered to one another even if neither is a final cause
of the other, for example in a nonevident essential order. Let w be a whole sub-
stance for the sake of which p1 and p2 are produced. If p1 would not be final-
causally ordered to w unless p2 were final-causally ordered to w first, or vice
versa, and neither is final-causally ordered to the other, then p1 and p2 stand in
a [pneo], since each is a proximate effect of a common final cause, w. Now
suppose that three more part-substances of w, p3, p4 and p5, are essentially
ordered such that p3 is a final cause of p5, p1 is a final cause of p3 and p4, but p4 is
not a final cause of p5. If p5 would not be final-causally ordered to p3 unless p4
were first final-causally ordered to p1, then p4 and p5 stand in a [rneo], since
each are remote effects of a common final cause, w.
There is also some reason to think that the sort of unity of order that obtains
among organic part-substances is a unity of an essentially ordered efficient
causal series. Again taking the cue from Scotus’s endorsement of Aristotle’s
embryology, the heart’s activity is supposed to initiate the efficient causal
series that results in a complete organic body. It produces the blood vessels and
the blood, the vessels transport the blood to the brain and the nutriments from
the uterus, these effects go on to produce their effects, and so on. If there is an
essential order of efficient causation among these organic part-substances, it
does not obtain simply in virtue of one or more part-substances efficiently
causing the existence of another. The efficient causal order by which all the
parts are brought into being fails to be an essential order because the causes do
29 Aristotle, On the Generation of Animals II, c.6, 743a 1 -743b 32 (Barnes I, pp. 1153–1154).
not cause simultaneously—the heart is generated before the blood, the blood
before the brain, and so on. But if we think of a more or less complete organic
body at some advanced developmental stage, where roughly all the parts have
been produced, then it does seem plausible to suppose that there is an essen-
tial order of efficient causation among these parts, where the activity of one
part, p1, is dependent on the activity of another part, p2, at the moment of p1’s
activity, and where some first part—the heart according to Aristotle—is the
first cause in the series. For example, according to Aristotle the brain’s cooling
and heart’s heating are partial causes of all that goes on in the body. If the heart
or the brain is removed it is not simply the case that the body loses a part-
substance; additionally, the functioning of all the other parts cease, since these
activities are dependent on the ongoing activity of the heart or brain. In an
essential efficient causal order such as this, the causal line can arrive at a fork,
as in an order of final causality. Two effects on different prongs can neverthe-
less be nonevidently essentially ordered to one another such that every part is
essentially ordered to every other part.
To say that every part is essentially ordered to every other part is not to
say, however, that the parts of the body are interdependent or make up an
organic system in which all the parts (or at least the vital parts) are needed not
only for the life of the organism but for the ongoing functioning of any one of
the parts. Instead, it is just to say that every part is either essentially prior or
essentially posterior to every other part. Scotus would deny that that the parts
could be interdependent because he is committed to the noncircularity of
essentially ordered items. So if the brain is essentially dependent on the heart
(in the order of efficient causality) and the heart is essentially dependent on
the brain (in the order of final causality), then the heart is not efficient-causally
essentially dependent on the brain, and the brain is not final-causally essen-
tially dependent on the heart. For Scotus an organic first cause is always
required.
One might object to my characterization of the unity of order of organic
parts in terms of final and efficient essential orders along the following line.
One of the major motivations for pluralism about substantial form (both stan-
dard and Scotistic pluralism) is that a corpse seems like it is continuous with
the living organism whose corpse it is. Standard pluralism allows us to say that
Mole’s corpse is the same body as Mole’s body, because they are the same sub-
stance, informed by the same substantial form. But if the unity of Mole’s fetal
body consists in its parts being efficient- and final-causally ordered to one
another (which unity we rightly suppose the full-grown animal retains), then
Mole’s corpse cannot be the same body as Mole’s body, because whatever unity
the corpse has is not the unity Mole’s body has.
106 Chapter 6
The objection forces a nice clarification. On the view I am attributing to

Scotus, Mole’s corpse is not a substance, but neither was Mole’s body. The
organism was a substance because it was a composite of soul and an ordered
unity of organic parts. The organs of an organism do not themselves cease to
exist at the moment the organism ceases to exist, but the body they composed
does cease to exist, because that body was constituted not only by organs but
also by the final- and efficient-causal relations that these parts bore one to
another, and these relations cease to exist when the organism ceases to exist.
(Or, they cease to exist when the organism ceases to exist under usual circum
stances. There is nothing in Scotus’s theory which prevents these relations
from persisting after death, since death would be a matter of the soul separat-
ing from the body, and according to the theory these relations can obtain prior
to ensoulment [e.g., during fetal development].) Thus, while it may be true that
Mole’s corpse is not the same body as Mole’s, it is true that the (non-relative)
parts of Mole’s corpse are the same as the (non-relative) parts of Mole’s body.30
To sum up Scotus’s answer to the Special Potency Question: when Scotus
says that several substances must have a unity of order in order to be perfect-
ible by a substantial form or soul he means that the substances must be essen-
tially ordered to one another. Of the six essential orders of dependence I have
focused on final causality and efficient causality as plausible candidates for the
sort of unity of order that Scotus may have had in mind in QMet VII, q.20, n.48.
iv The Role of Soul in Scotus and Two Unitarians
In Buridan’s commentary on Aristotle’s On Generation and Corruption, he

denies that integral parts have distinct substantial forms. According to him,
the soul informs matter prior to the development of organs, so if there were
substantial forms of integral parts, then they would inform the composite of
30 To go a little deeper on this issue: what is supposed to be scandalous about the unitarian
position with respect to corpses is not just that the corpse is not the same body as the
body of the organism whose corpse it is, but also and more startlingly that no part of the
corpse is the same object as an any part of the organism whose corpse it is. On some criti-
cisms of Aquinas, the corpse and the organism do not even share the same prime matter.
Now, while Scotus is committed in the end to the idea that the corpse is not the same
body as the body of the organism whose corpse it is, he is not committed to the idea that
the parts of the corpse were never parts of the organism. Indeed, all of the (non-relative)
parts of the corpse (and many of its relative parts) were parts of the organism (at least
within a sufficiently short period post mortem). I am grateful to Martin Tweedale for rais-
ing the objection that prompted this clarification.
soul and matter. But it is unfitting (inconveniens) that there should be a sub-
stantial form of an organ such as a bone (Buridan’s example) informing this
composite, since the soul is more perfect than bone. Therefore there is not a
substantial form of bone.31 Buridan therefore holds that organ terms such as
“bone,” “flesh,” “head,” and “finger” do not pick out substances but are instead
accidental and connotative terms necessarily defined with reference to specific
functions of a living organism.32 On Buridan’s view the gradual process of dif-
ferentiation into the sorts of parts that can support the life functions of Mole
seems to be directed by Mole’s sensitive soul; some mere sequence of changes is
a process on the way to the generation of a living thing because there is a single
form at work, structuring matter in the appropriate ways.33 The sequence of
changes is therefore a sequence of changes in one substance, Mole.34
Scotus would object to Buridan’s account on the grounds that a substantial
form perfects whatever it informs, such that a composite of matter and a sub-
stantial form of some kind, K, is a full-fledged K, and not in process of becom-
ing a K.35 But Scotus also holds that a K-form can only inform a subject that is
31 John Buridan, Quaestiones super libros De generatione et corruptione Aristotelis I.8, ed.
Michiel Streijger, Paul J.J.M. Bakker, Johannes M.M.H. Thijssen (Leiden and Boston: Brill,
2010), p. 84, ll.15–25, hereafter cited using the convention: (Streijger p.n, ll.m).
32 Ibid. (Streijger, p. 87, ll.4–7).
33 Ibid., (Streijger, p. 84, ll.8–25). According to Buridan, bones, nerves, hands, and other body
parts develop after the soul begins to inform matter (Streijger, p. 84, ll.22–23, anima sub-
stantialiter generatur et vivit embrio, antequam ossa fiant sive antequam sint ossa); the
forms of these body parts are accidental forms (Streijger, p. 84, ll.26–27, forma qua os dici-
tur os est forma accidentalis); and the dispositions for these accidental forms come after
the soul (Streijger, p. 84, ll.30–31−p. 85, ll.1–2, dispositio qua os dicitur os vel qua nervus
dicitur nervus […] adveniunt post adventum animae). It seems natural to interpret
Buridan as holding that the dispositions for these accidental forms of body parts arise not
just after but in some sense because of the soul.
34 These claims about Buridan are controversial. I make them on the basis of Buridan’s com-
mentary on On Generation and Corruption, but the better known commentary on On the
Soul presents what at first blush looks like a contrasting picture of the soul’s relation to
organic parts. In this text, the soul apparently informs a body that is already differentiated
into different kinds of organic parts. Buridan, Quaestiones in Aristotelis De Anima (de
prima lectura) II, qq.4–5, ed. Benoît Patar, in Benoît Patar, Le Traité de L’âme de Jean
Buridan (Louvain-la-Neuve: Éditions de L’institut Supérieur de Philosophie, 1991),
pp. 249–278. I think that the On the Soul account can be read as consistent with my inter-
pretation of the On Generation and Corruption account, but for the purposes of this paper
I am interested in the view I attribute to Buridan only insofar as it serves as a foil to
Scotus’s view, so I reserve a defense of this claim for future work.
35 For those animals and plants that generate offspring by means of seed, Scotus says that
the form of the seed does have some causal role in the generation of the organism, but
108 Chapter 6
ready to support the sorts of activities that a K characteristically performs, and

therefore requires a more or less fully developed body in which to perform its
characteristic functions.36 As far as I can tell he does not argue for this claim; it
appears to be a basic feature of Scotus’s hylomorphism about substance. This
leaves Scotus in need of some explanation of how a sequence of substantial
changes (a heart develops, then blood, then a brain, and so on) is also a process
on the way to the generation of Mole. As we have seen, Scotus deploys a meta-
physics of essential orders, and in particular an essential order of final causes,
to explain how discrete generations of substances are in fact generations of
things destined to be parts of an organism.
Aquinas agrees with Scotus that the soul begins to inform matter not at the
moment of conception but at some more advanced stage of fetal development
and therefore both agree that a mole fetus is not really Mole (or any other
mole) until Mole’s sensitive soul is induced. But unlike Scotus, since Aquinas is
committed to unitarianism, and since the fetus is supposed to be either a sub-
stance or composed of substances, it is strictly speaking false to say that the
soul begins to inform the fetus; instead the fetus and all of its integral parts are
corrupted at the moment at which Mole is generated.37 For Aquinas, then, it is
inaccurate to say that the soul directs the development of the organs, since the
organism and all of its parts begin to exist at the instant the soul begins to
inform matter. For both Buridan and Aquinas, then, the soul does not simply
unify organs but makes them.
A striking consequence of Scotus’s metaphysics of part-substances is how
little work it leaves the soul to do, relative to Buridan’s and Aquinas’s hylomor-
phic accounts. According to Scotus the organs of Mole both exist and function
(in some sense) prior to becoming organs of Mole. It is then the soul’s job to
unify the organs, making them part-substances of one living substance. The
special mark of a substantial unity by Scotus’s lights is that there is some activ-
ity or activities that cannot be attributed to a part or parts of that unity. In the
case of an animal sensing—seeing, tasting, etc.—is the paradigmatic activity
that indicates that an animal is one substance (albeit composed of part-
substances) and not some less-than-substantial unity of part-substances.38
that form is not the soul and moreover is corrupted at the moment when the first part of
the animal is generated. Quaestiones super Libros Metaphysicorum Aristotelis VII, q.12,
nn.40–44 (Bonaventure IV, pp. 206–208).
36 Ordinatio IV, d.11, p. 1, a.2, q.1, n.284 q.3 (Vatican XII, p. 267).
37 Aquinas, De potentia Dei q.3, a.9, ad 9; Summa contra gentiles II, c.89, 11; Summa theologiae
Ia q.119, a.2.
p. 393). I discuss this idea in some detail in Chapter 9.
The soul and the body, then, compose the whole substance that is the proper
subject of sensing and other activities (if there be any others).39 But all of the
organs needed for performing these activities are already there, waiting, as it
were, for the soul.
39 Scotus thinks that a substance is really distinct from its parts, and thus literally means
that the whole substance and not its parts (taken singly or together) is the proper subject
of some activities and properties. This aspect of Scotus’s metaphysics of parts is discussed
in Richard Cross, “Duns Scotus’s Anti-Reductionistic Account of Material Substance,”
Vivarium 33:2 (1995), pp. 137–170.
Chapter 7
Contingent Supposits and Contingent Substances
In the last two chapters I have argued that Scotus thinks that some integral
parts of material substances—paradigmatically organs of organisms—are
composites of matter and at least one substantial form. For example, Scotus
thinks that the heart of Mole is a composite of matter and a substantial form of
the heart. Does it follow from this that a heart is a substance? If it does, it fol-
lows straightaway that a material substance can have material substances as
integral parts—a view alien to a common way of interpreting the theory of
substance offered in Aristotle’s Metaphysics VII.1 I think it does, and I think
Scotus thinks it does, for the very simple reason that a composite of matter and
substantial form just is a substance. But the issue is not quite as simple as it
appears. A substance is supposed to be that which exists per se, where per se
can be understood in a literal way as through oneself, by oneself, or on its own.
Even if a heart can exist per se, for example if it is removed from the body and
stored under the right conditions, it apparently does not so exist while it is a
fully functioning part of a living organism. Not only does the heart, while it is
a part, depend on the organism (or perhaps other parts of the organism) for its
existence and functioning, but in a deeper sense its existence and functioning
seem wholly for the sake of the organism; it is ordered to the existence of
something else in a way in which the organism itself is not so ordered. Could it
be, then, that a heart is contingently a substance—not a substance while a part
of an organism, but a substance when it is not?
The short answer is probably not, but the reason is not what you might
expect. You might expect that a heart is not contingently a substance because
no substance is contingently a substance. Being a substance, you might think,
is a necessary feature of a thing in the sense that for any x, if x is a substance
then if x ceases to be a substance x ceases to be. This is not Scotus’s reason,
however. Scotus thinks—and I will argue for this claim in this chapter—that
being a substance is a contingent feature of at least some substances. One and
the same thing might first not be a substance, and then become a substance, or
might first be a substance and then cease to be a substance. He also thinks, and
I will also argue that he thinks, that any substance can be a part of another
substance, if only by divine power. This implies that being a supposit—that is,
being the sort of substance that is not a part of any other substance and is a
1 Aristotle, Metaphysics VII, c.13, 1039a 3–5 (Barnes II, p. 1640).

Contingent Supposits And Contingent Substances 111
member of a determinate Aristotelian kind, such as an individual human

being—is also a contingent feature of at least some substances; any created
supposit can cease to be a supposit but not cease to exist, and some non-sup-
posits can become supposits and continue to exist. So the full story about
Mole’s heart is rather more complicated than first appears. It is a substance, a
composite of matter and substantial form; under certain (tragic) conditions it
might cease to be a part of Mole while continuing to be the very substance it is.
But because a heart is a substance by nature disposed to be a part of an organ-
ism it is not a member of a determinate Aristotelian kind; it therefore fails to be
a supposit even when removed from Mole’s chest, and probably cannot become
a supposit. On my reading, then, Scotus’s doctrine of the substancehood
of integral parts includes the claim that some substances are not supposits.
In this chapter I expand this reading and defend some additional claims
about substances and supposits. Not only are some substances not supposits,
but being a supposit is a merely contingent feature of some substances, and
being a substance is a merely contingent feature of some things. I begin by
providing some further defense of my claim that Scotus thinks that some sub-
stances can be parts. I then turn to Scotus’s metaphysics of the Incarnation to
defend the claims that Scotus thinks that not every substance is a supposit,
that any created supposit might not be a supposit but still exist, and that some
non-supposits can become supposits. Finally I examine Scotus’s metaphysics
of continua to support my claim that Scotus thinks that being a substance is a
merely contingent feature of at least some substances. Along the way I exam-
ine some relevant texts from Ockham to illuminate Scotus’s views by compari-
son and contrast.
First, however, it is worth pointing out that early in his career Scotus appar-
ently held that a part of a substance is or can be a substance only in an equivo-
cal sense. In Quaestiones super Praedicamenta Aristotelis q.15, Scotus asks
whether the parts of substances are substances. He endorses the first argu-
ment that they are not, according to which,
Substance in the most general sense is a being per se; [but] no part of a
substance is a being per se while it is a part of a substance, because then
it would be a hoc aliquid, and one substance would be from many hoc
aliquid, which does not seem true.2
2 Quaestiones super Praedicamenta Aristotelis q.15, n.1 (Bonaventure I, p. 383), Quod non, vide-
tur: Quia substantia ut est generalissimum est per se ens; nulla pars substantiae est per se ens
dum est pars substantiae, quia tunc esset “hoc aliquid”, et una substantia esset ex multis hoc
aliquid, quod non videtur verum.
112 Chapter 7
Yet Scotus qualifies. Substance is an equivocal term: it can signify either an ens
per se, or a principle of an ens per se. And, of course, it is only in the second
sense that parts of substances can be said to be substances.3 In this second
sense parts are in the genus of substance through reduction to the per se prin-
ciples of the species.4
Quaestiones super Praedicamenta Aristotelis is among Scotus’s earliest
works; it is probably earlier than any except for the Quaestiones in Librum
Porphyrii Isagoge.5 That Scotus offers in other works a considerably different
answer to the question about whether substances can be parts of sub-
stances can therefore be chalked up to doctrinal development rather than
inconsistency.
i Three Modes of per se Being
One of these other works is Quodlibet IX. Here Scotus asks whether an angel
can inform matter. Arguing that an angel cannot, Scotus aims to show that an
angel is a being per se in a special sense according to which it cannot be a part
of something that is itself a per se being. He distinguishes three senses of
beings per se:
First, it can designate something which exists in isolation or apart from a

subject, in the third mode of per se referred to in the Posterior Analytics.6
In this sense, an accident can be a per se being when it does not inhere
in a subject. Second, a per se being is contrasted with one that exists in
another, and in this sense it is a thing which neither actually inheres in
another nor has an aptitude to do so. Every substance, not only one that
is composite, but matter and form as well, are all beings per se in this
sense, for though a substantial form is in the matter it informs, it does not
inhere in it like an accident, for “to inhere” says that it does not inform
its subject per se. What inheres is neither an act simply, but only in a
3 Quaestiones super Praedicamenta Aristotelis q.15, n.10 (Bonaventure I, p. 386).

4 Ibid.
5 “Quaestiones super Praedicamenta Aristotelis: Introduction,” (Bonaventure I, p.xxxvii).
6 Aristotle, Posterior Analytics I.4, 73b6-10. “[…W]hat is not said of some other underlying sub-
ject […] and whatever signifies some ‘this,’ is just what it is without being something else.”
Scotus’s point is that even accidents (under miraculous conditions) can exist without inher-
ing in a subject, and in this sense are like substance. Aristotle seems to have had only sub-
stances in mind in the text quoted here.
qualified sense, nor does it form one thing per se with the subject in
which it inheres. What informs per se has the opposite characteristics.
Third, a per se being may refer to one which has its ultimate actuality, so
that it is simply unable to be ordered per se to some ulterior act beyond
that which it has, where the ulterior actualization would belong to it per
se, either in a primary or in a participated sense. A per se being in this
sense is called a suppositum, and if it is of an intellectual nature, it is
called a person. Only this third is properly said to be subsisting, in the
sense the Philosopher has in mind when he says: “Matter is only poten-
tially ‘a this’ and the form is that in virtue of which a thing is called ‘a this,’
but that third being compounded of both matter and form is called ‘a
this.’”7 In other words, something subsisting per se has its ultimate actual-
ization so that it is unable to be ordered per se to some ulterior act.8
First, anything that can exist on its own (solitarie) is a being per se. This includes
not only substances, substantial forms, and matter, but also qualitative and
quantitative forms, which can exist without inherence in a substance (if only
by divine power). Second, anything that does not inhere, neither actually nor
aptitudinally, is a being per se. The second mode rules out accidents, since even
7 Aristotle, On the Soul II.1, 412a6-10 (Barnes I, p. 656).

8 Scotus, Quodlibet IX, a.2, n.7, trans. Felix Alluntis and Allan B. Wolter (Princeton and London:
Princeton University Press, 1975, p. 220. Alluntis and Wolter’s translation is based on Alluntis’
Latin edition in Obras del Doctor Sutil Juan Duns Escoto (edicion biligue): Cuestiones
Cuodlibetales, trans. (with Latin edition) by Felix Alluntis (Madrid: Biblioteca de autores cris-
tianos, 1968), pp. 344–345: Ens per se potest intelligi tripliciter: uno modo, intelligitur ens per
se solitarie, prout accipitur I Posteriorum in tertio modo per se, et hoc modo accidens potest
esse ens per se quando non est in subiecto. Secundo modo, dicitur ens per se prout distingui-
tur contra ens in alio et sic per se ens est idem quod non inhaerens actualiter nec aptitudi-
naliter; et hoc modo quaecumque substantia, non tantum composita sed etiam materia et
forma, est ens per se, quia forma substantialis, licet insit materiae informando, non tamen
inhaeret; quia inhaerere dicit non per se informare, quia inhaerens nec est actus simpliciter,
sed actus secundum quid, nec cum illo cui inhaeret facit per se unum; opposita conveniunt
ei quod per se informat. Tertio modo, ens per se dicitur illud quod habet actualitatem ulti-
mam, ita quod non est per se ordinabile ad aliquem actum simpliciter, ultra istum quem
habet, qui quidem actus ulterior possit esse actus eius per se, et hoc vel primo vel participa-
tive; quod hoc modo est per se end communiter dicitur suppositum et in natura intellectuali
dicitur persona. Hoc modo intelligitur maior de ente per se. Istud solum dicitur proprie sub-
sistens, sicut Philosophus loquitur II De Anima, dicens: “Quod materia est potentia hoc aliq-
uid, species autem secundum quam aliquid dicitur hoc aliquid; tertium, quod est ex ipsis,
quod simpliciter est hoc aliquid”, scilicet per se subsistens habet actualitatem ultimam non
ordinabilem per se ad aliquem actum ulteriorem.
114 Chapter 7
if they can exist without inhering, by nature they have aptitude to inhere.
It rules in substances, substantial forms, and matter. Finally, anything that has
ultimate actuality (actualitatem ultimam) is a being per se. A being with ulti-
mate actuality is a per se being that is not per se orderable [per se ordinabile] to
another per se being, where each occurrence of “per se being” can refer to any
of the three modes of per se being. A separated accident is per se orderable to a
substance in the sense that its natural aptitude is to inhere in a substance;
prime matter and substantial form are per se orderable to a substance in the
sense that their natural aptitudes are to be essential parts of a substance. But a
per se being of the third mode, called a supposit or (if its nature is rational) a
person, is not per se orderable in the sense that it can neither inhere in nor be
an integral or essential part of a substance. A supposit has “ultimate actuality”
in the sense that while it may have parts (if it is complex), there is no supposit
of which it is or can be a part. It is in this third sense that an angel is a per se
being and therefore an angel cannot inform matter.9
Notice that in the second mode, not just a substantial form or matter but a
composite substance itself is a per se being. So at least some substances belong
in the second mode and at least some substances belong in the third. Now, it is
trivially true that if a composite substance is in the third mode then it is also in
the second mode, since a third-mode per se being is a being that cannot inhere
in something else and a second-mode per se being is a being that need not
inhere in something else—necessity implies possibility. But it is not trivially
true that if a substance is in the second mode then it is also in the third. In fact,
this is almost certainly not true. If Scotus thinks that for all x, x is a substance
if and only if x is a per se being in the third mode, then it simply would be
redundant to place any substance in the second mode. Of course, Scotus does
not explicitly commit himself to holding that there are any substances that are
in the second mode but not in the third. But given that he bothers to list sub-
stances among the kinds of things falling under the second mode, it seems
highly likely that he thinks, at the very least, that it is possible that there is
some y, such that y is a substance in the second mode but not in the third. In
other words, it seems highly likely that Scotus thinks it is possible that there is
a substance that is not a supposit.
On my reading of Scotus’s pluralism about substantial form, organic sub-
stances have any number of per se second mode substances as parts. To show
this, it will be helpful to reconsider briefly a text already discussed. In QMet VII,
9 Quodlibet IX, a.2, n.7, in Obras del Doctor Sutil Juan Duns Escoto (edicion biligue): Cuestiones
Cuodlibetales, trans. (with Latin edition) by Felix Alluntis (Madrid: Biblioteca de autores
cristianos, 1968), pp. 344–345.
q.20, Scotus endorses Aristotle’s hypothesis that the heart is the first organ to
be generated in human embryogenesis and applies the results of his quaestio
on whether organic parts have substantial forms, saying,
[Aristotle] would assign in the generation of an animal (speaking of the

whole) many complete changes to many forms of parts, one before the
other in time.10
We know from context that the forms mentioned here are substantial forms,
the substantial forms of the various organic parts. These parts develop tempo-
rally prior to the generation of the whole animal, and therefore exist in some
sense independently of the animal whose parts they are waiting to become.
Since these parts have substantial forms, it is safe to conclude that they are
substances. It is clear also that when the whole animal comes into existence
these parts remain actual. After all, Scotus deploys a classic argument for plu-
ralism, namely that a corpse is not a new substance but is identical with some
part(s) of the deceased animal. If Scotus claimed that these parts ceased to be
actual when they composed an organism, then he would have the same
explanatory gap as the unitarians—where an efficient cause reducing poten-
tial beings to actual beings should be, there is none. It follows, then, that the
organic parts remain actual when they begin to compose an animal. The ques-
tion, then, is whether they remain substances when they begin to compose an
animal. If they are substances they are obviously not per se beings in the third
mode (since they are parts). But they may be per se beings in the second mode.
Since they were substances before they became parts, since they continue to
exist when they become parts, and since Scotus thinks that some substances
(those existing per se in the second mode) can make a per se unity with another
per se unity, it is highly likely that these organic parts are actual substances (in
the second mode of per se being) actually composing an organism (in the third
mode of per se being). They are not supposits, but they are substances, com-
posing a supposit.
ii Ockham on the Distinction between Substance and Supposit
Ockham holds a similar view. Although he declared agnosticism on the

issue of whether organic parts have specifically different substantial
p. 390).
116 Chapter 7
forms,11 he thought that the integral parts of a composite substance were

all actual, and since he thought that both matter and substantial forms
(with the exception of the rational soul) were equally extended, he held
that the integral parts of one extended composite substance were them-
selves composites of parts of matter and parts of form. Like Scotus, he
denies that these parts are supposits.12 But he thinks that they become sup-
posits if they are separated from their whole, and he thinks they were sup-
posits if they existed before they were parts of their whole:
The same part of air or a man first is not a supposit when it is a part, and
afterward, when it is not a part, it is a supposit. And similarly numerically
the same nature is first a supposit when it is not an integral part, and
afterward, when it is an integral part, it is not a supposit. Example: if
some part of a man or of air were separated from the whole, first it was a
part and consequently not a supposit, and after it is a supposit, because it
is not a part when it is separated from the whole and exists per se.13
Ockham emphasizes that what becomes a supposit when separated from the
whole is the very same thing as that which was a part and not a supposit. He
does not think that the new supposit is a new thing. Likewise, what ceases to
be a supposit when it becomes a part is the very same thing as that which was
a supposit and not a part. The new part is not a new thing. This entails that
being a supposit is a contingent property.14 But of what is it a contingent prop-
erty? Not of form, or of prime matter, but a composite of the two. And the
composite is not a potential being but an actual.15 But an actual composite of
matter and form just is a substance. For Ockham too, then, there are substances
that are not supposits.
11 Ockham, Quodlibet III, q.6 (OTh IX, pp. 225–227).

12 Ockham’s definition of supposit is similar to Scotus’s third mode of per se existence.
Ockham, Quodlibet IV, q.7, ll.10–12 (OTh IX, p. 328), Dico quod suppositum est ens comple-
tum, incommunicabile per identitatem, nulli natum inhaerere, et a nullo sustentatum.
13 Ockham, Quodlibet IV, q.7, ll.28–32 (OTh IX, p. 329), Similiter eadem pars aeris vel hominis
primo non est suppositum quando est pars, et postea quando non est pars, est supposi-
tum. Et similiater eadem natura numero est primo suppositum quando non est pars inte-
gralis, et postea quando est pars integralis, non est suppositum. Exemplum: si aliqua pars
hominis vel aeris separaretur a toto, primo erat pars et per consequens non suppositum,
et post est suppositum, quia non est pars cum sit separata a toto et existat per se.
14 Not everything that is a supposit is contingently a supposit, however. The divine persons
are supposits of the divine essence and cannot not be.
15 Ockham, Quaestiones Variae III, q.1, ll.320–328 (OTh VIII, p. 78).
Ockham gives a different analysis of the ontological status of separated

essential parts, and this analysis reveals that Ockham does not think that all it
takes for a substance to be a supposit is that it not be a part of one. According
to his version of pluralism, a human is composed of at least three kinds of sub-
stantial forms: a form of corporeity, a sensitive soul, and a rational soul. In
Quodlibet II, q.10, he canvasses a natural objection to this view: if the sensitive
and rational souls are really distinct and if (as is apparently the case) it is pos-
sible that the composite of body and sensitive soul ceases to be informed by
the rational soul, then it is possible that there be a sensitive animal that is no
kind of animal.16 Ockham admits that the rational-soulless animal would be
living and would be neither rational nor irrational! But it is an animal only in
an equivocal sense of the term, because it is an animal not truly belonging to a
specific kind. Moreover,
It is not a complete being existing per se in a genus, but is by nature an

essential part of something existing per se in a genus. And no such thing
is in the genus of substance or of animal per se.17
Here Ockham seems to imply that the rational-soulless animal would not be a
supposit, since it is incomplete, i.e., it does not belong to a specific kind and its
nature is to be an essential part of something that does. But it exists as an
actual composite of body and sensitive soul. It is therefore a substance, but not
a supposit.
In recent work, Calvin Normore has denied that, for Ockham, integral
parts of substances are substances. According to Normore, while Ockham
thinks that the integral parts of a continuum (such as a piece of wood) are all
actual parts, and that there are actually infinitely many such parts in any
continuum,
[He] does nonetheless deny that a piece of wood is made up of infinitely

many other substances. His thought here is that an integral part of a substance,
while it is in act in the sense that it exists in the world, is not in act in another
sense which sense requires precisely that it not be part of anything else.18
16 Ockham, Quodlibet II, q.10, ll.83–85 (OTh IX, p. 160).

17 Ockham, Quodlibet II, q.10, ll.111–114 (OTh IX, p. 161), Et tota ratio est, quia non est ens
completum existens per se in genere, sed est natum esse pars essentialis alicuius existen-
tis per se in genere. Et nullum tale est in genere substantiae vel animalis per se.
18 Calvin Normore, “Ockham’s Metaphysics of Parts,” p. 748, fn.29. Normore does not
cite a text for this claim. From his discussion it is clear that he is examining Ockham’s
118 Chapter 7
Implied here is that a necessary condition of a thing’s being a substance is

that it not be a part of anything else; the special way in which substances are in
act is that they exist, but do not exist as parts of other substances. And this
entails that any substance is ipso facto a supposit. But, as we have seen, Ockham
does not think that every substance is a supposit. So the distinction Normore
draws between being in act and being in act and not being a part of anything else
is better seen as a distinction between being in act and being a supposit, rather
than as a distinction between being in act and being a substance. Since being a
supposit is a contingent feature of any (created) substance, the very same
thing can be in act in the first sense and then be in act in the second (or vice
versa). And, more to the point, the very same substance can be in act in the first
sense and then be in act in the second (or vice versa).
Ockham’s account of supposits yields two necessary conditions for being a
supposit: 1) not being a part of something; 2) being of a complete, determinate
kind. Ockham’s macabre example, the rational soulless, human-like organism,
fails to be a supposit because, while it is not a part of anything, it does not
belong to a complete species. Some part of a puddle of water would be a sub-
stance but not a supposit and it would fail to be a supposit just because it is a
part of a substance, the puddle. But remove that part and it would become a
supposit since it is already a homogeneous substance of the elemental kind,
water, and would not be a part of anything once separated from the puddle.
While Scotus does not explicitly include condition (2) in his own account of
what a supposit is, he would likely endorse it. But he parts ways with Ockham
inasmuch as he would deny that a non-detached part of a homogeneous sub-
stance (such as a puddle of water) is a substance—of which more, below.
iii What’s Special about Supposits?
Medieval reflection on the metaphysics of the Incarnation yielded a similar

distinction between supposits and mere substances. Marilyn McCord Adams
asks, “What’s metaphysically special about supposits?” in light of medieval
questions de continuo from Quaestiones in Libros Physicorum Aristotelis. In these ques-

tions Ockham argues, among other things, that the integral parts of a continuum are in
act while they are parts (q.68), that they have their own actuality independent of the
actuality of the whole of which they are parts (q.69), that there are infinitely many actual
parts in any continuum (q.70), and that these parts are all really distinct from each other
(q.71). But Ockham does not claim in these questions that nothing that is a part is not a
substance. Cf. Ockham, Quaestiones in Libros Physicorum Aristotelis, qq.66–71 (OPh
VI, 582–597).
innovations of Aristotle’s theory of substance motivated by Chalcedonian

Christology.19 The Chalcedonian Definition holds that in Christ there are two
natures—divine and human—but only one person and this person is the Son
of God.20 Medieval theologians interested in bringing Aristotle to bear on the-
ology knew that in the Categories Aristotle distinguishes two kinds of sub-
stances, primary and secondary, where secondary substances are substance
kinds, such as humanity, and primary substances are substance individuals,
such as this human.21 Which is the human nature of Christ, primary or second-
ary? It is not secondary, since Christ is truly man and therefore truly is an indi-
vidual man. But identifying Christ’s human nature with primary substance
also has problems. If the Son of God unites himself with a man, that is, some
guy already on the scene, then either the union will be a union of two persons,
which violates Chalcedon, or in taking on human nature the Son of God
destroys the person and takes his human nature, which is either uncouth or
impossible. An unbaptized Categories-based metaphysics is therefore inade-
quate for an orthodox metaphysics of the Incarnation. Summarizing some
medieval approaches, Adams writes,
[The doctrine of the Incarnation requires] us to complicate Aristotle’s

bipartite division between primary and secondary substances. [It forces]
a further distinction between complete concrete individual substance things
that are natures but not persons or supposits, and those that are persons or
supposits.22
What the Son of God assumes, therefore, is not a “person or supposit” but a
“complete concrete individual substance thing.” God—Father, Son, and Holy
Spirit—creates this individual substance thing and therefore it causally
depends on God. Under natural conditions, when God creates an individual
substance nature he creates it as a supposit, a primary substance that neither
inheres in nor is a part of anything. In Scotus’ language a supposit is an indi-
vidual substance nature that lacks both the aptitude to depend on another
19 In this paragraph I closely follow Marilyn McCord Adams’s exceptionally clear account of
the role of the Incarnation in medieval innovations to Aristotle’s theory of substance in
“What’s Metaphysically Special about Supposits?”
20 “The Definition of Chalcedon (451),” in Creeds of the Churches: A Reader in Christian
Doctrine from the Bible to the Present, 3rd edition, ed. John H. Leith (Louisville, KY: John
Knox Press, 1982), pp. 34–36.
21 Aristotle, Categories I.5, 2a11-19 (Barnes I, p. 4).
22 Adams, “What’s Metaphysically Special about Supposits?” p. 26 (italics hers).
120 Chapter 7
(since it is a substance nature), and lacks actual dependence on another. But

Scotus thinks that there is nothing repugnant about God’s creating an indi-
vidual substance nature as a (non-causally) dependent entity in a way analo-
gous to the dependence of an accident on a substance.23 And in the case of the
Incarnation, Scotus thinks, the Son’s individual human nature is created as a
thing depending on the Son of God. In his language the Son’s individual human
nature lacks the aptitude to depend on another (again, since under normal
conditions it would be created as a supposit), but it actually depends on the
Son. Since Scotus and others thought that the individual human nature of
Christ could be (but in fact never is) disunited from the Son of God and, as it
were, become his own man, it follows that for that very individual substance
nature, whether it is a supposit or not is a contingent affair.
We can neatly transpose Scotus’s distinction between the second and third
modes of per se being onto his Incarnational metaphysics. Under normal con-
ditions God creates an individual substance nature as a person, a per se being
in the third mode. But in the special case of the Incarnation he creates an indi-
vidual substance nature as dependent on the Son of God. The substance nature
is a per se being in the second mode, and it depends on a per se being of the
third mode, the second person of the Trinity. In the pro-Scotist conclusion of
his book, The Metaphysics of the Incarnation, Richard Cross writes,
If substances are individualized natures, then it is clear that Christ’s

human nature is a substance. As such it is a subject of properties and
accidents—albeit not an ultimate subject of properties and accidents.24
In other words, Christ’s human nature is a part-substance in the second mode

of per se being, and it is a part of Christ which is a supposit in the third
mode of per se being.
The previous sections of this chapter have shown that conclusions about
the contingency of being a supposit were not limited to the Christological case.
What I would like to emphasize here, however, is that if we grant that Scotus
and Ockham allow for substances that are not supposits, e.g., substances that
are parts of other substances and/or are not of some determinate species, we
should also grant that they allow, or at least are committed to, Christ’s human
nature being a substance that is not a supposit. In fact this is the most natural
23 Cross notes that while Scotus’s preferred “model” for articulating the relationship between
the two natures of Christ is a substance-accident model, he sometimes uses a part-whole
model. Richard Cross, The Metaphysics of the Incarnation, pp. 128–133.
24 Richard Cross, The Metaphysics of the Incarnation, p. 314.
way to interpret Scotus’ application of the distinction between a “supposit or

person” and “a complete concrete individual substance thing” to the distinc-
tion between human nature as found in Socrates or Plato (persons) and human
nature as found in Christ (a complete concrete individual substance thing).
The claim that any created substance that is a supposit is only contingently
a supposit (that is, it could cease to be a supposit without ceasing to be) is not
to say, however, that every created substance can be a supposit. Ockham’s
rational soulless human-like thing is a substance but fails to be a supposit
because it is not of a complete species. It can become a supposit only by reduc-
tion in the sense that if it begins to be informed by a rational soul it becomes
an essential part of a human supposit. And Scotus’s organic part-substances
really are substances, whether attached to or detached from an organism, but
not even detached organic part-substances are supposits because, while they
are not a part of any substance and therefore satisfy one criterion for being a
supposit, they lack “ultimate actuality,” and therefore fail to be supposits.
iv Arbitrary Part-Substances?
While Scotus allows for many different kinds of part-substances, he would

deny that arbitrary parts are part-substances. His reasons reveal a commit-
ment not just to the contingency of a substance’s being a supposit, as we have
already discussed, but to the contingency of a thing’s being a substance, where
this is understood in the radical sense that one and the same thing might first
exist and not be a substance and then, continuing to exist, become a substance.
By arbitrary parts I have in mind a variation of what Van Inwagen calls the
Doctrine of Arbitrary Undetached Parts, according to which
For every material object, M, if R is the region of space occupied by M at

time t, and if sub-R is any occupiable sub-region of R whatever, there
exists a material object that occupies the region sub-R at t.25
For example, imagine a block of marble and then imagine a statue inside the
marble, waiting to emerge. In reality there is no “joint” to carve between the
statue in the block and the rest of the marble surrounding the statue. It is in
this sense arbitrary, and we might well entertain doubts about whether it really
exists.
25 Peter van Inwagen, “The Doctrine of Arbitrary Undetached Parts,” in Ontology, Identity,
and Modality: Essays in Metaphysics (Cambridge: Cambridge University Press, 2001), p. 75.
122 Chapter 7
By arbitrary part-substances I have in mind arbitrary parts that are sub-

stances. I have argued that Scotus thinks that, e.g., Mole’s heart, liver, lungs,
and other organs are substances, part-substances. These certainly aren’t arbi-
trary part-substances, since each of these has a distinct function or role in the
organism. But now consider Mole’s body and imagine a spherical region of
Mole’s body in whose boundaries fall parts of several organs. Like the statue in
the block of marble this is, if it exists, an arbitrary part because there is no
natural joint to carve between it and its surrounding bodies.
What would Scotus say? To see his answer, it’s necessary first to distinguish
between hetero- and homogeneous bodies. The block of marble is marble
throughout (let’s say); every part of the marble block is marble. As we will see
in the following chapter, Scotus denies that the four chemical elements remain
actual in a compound substance (that is, a substance compounded of two or
more kinds of elements), so he holds that there can be non-elemental homo-
geneous substances (that is, homogeneous substances other than earth, water,
air, and fire). But Mole’s body is heterogeneous; bone is a different kind of stuff
from flesh, and a heart is a different kind of thing from a liver. Scotus’s answer
to the question about arbitrary part-substances would begin with a distinction
between homogeneous and heterogeneous substances because he would give
different answers for each.
We will first examine homogeneous substances. A homogeneous substance
can be a substance that is not a part of any substance, for example a block of
marble or a bar of bronze (if marble and bronze are homogeneous). Or a
homogeneous substance can be a part of a substance, for example Scotus
thinks that bone is a homogeneous substance, and bone is part of a man. The
question about whether there are arbitrary part-substances of homogeneous
substances concerns both homogeneous substances that are parts of sub-
stances and those that are not. So, for example, consider a bone, a femur. It is a
discrete object with a distinct function in the life of a human, so it is probably
a substance in its own right, at least on Scotus’s lights. What of its parts, such as
its top and bottom halves? According to Scotus, a homogeneous substance
has infinitely many actual parts but none of these parts (proper parts, in
the language of mereology) is a substance. Each becomes a substance only if
separated.26 Since each part is actual but is not actually a substance, and since
26 Ordinatio I, d.17, p.2, q.1, n.232 (Vatican V, p. 251). Also see Cross’s discussion in The Physics
of Scotus, pp. 139–158 and especially pp. 146–147. By contrast, Ockham thought that the
parts of a homogeneous body were not only actual but actually substances, pre- and post-
detachment. See Ockham, Expositio in Libros Physicorum Aristotelis VI, c.13, §6 (OPh V,
pp. 563–564), and discussion in Pasnau, Metaphysical Themes, pp. 610–613.
each part can become a substance if separated, it follows that Scotus is com-
mitted not just to the contingency of being a supposit—a substance can
become a supposit and continue to be the very substance it is—but also to the
contingency of being a substance—an actual region of a homogeneous sub-
stance can become, if separated, a substance and continue to be the very thing
that it is.27
Heterogeneous arbitrary parts require a different analysis. Homogeneous
arbitrary parts such as a region of bone are, if not discrete things, members
of some natural kind, such as bone, flesh, water, marble, and so on. So
while Scotus would deny that a non-detached region of a femur is a part-
substance, it seems reasonable to suppose that it is the sort of thing that
can become a substance—even if we could only describe it as some bone
rather than a bone. But now consider a heterogeneous region of Mole’s
body, such as a spherical region in whose boundaries fall arbitrary parts of
several different organs. All of the parts of the sphere are actual so the
sphere is actual. But it is not a part-substance. The one criterion Scotus
offers for picking out substances—for identifying some complex material
object as a substance—is that it have some activity that is not reducible to
the activity of its parts taken either singly or together.28 But our arbitrary
sphere has no such activity. So even detached from the Mole we would
have no reason to think that it becomes a substance. At best its homoge-
neous parts become substances once detached—á la the analysis of homo-
geneous arbitrary parts offered above.
In conclusion, neither arbitrary homogeneous parts nor arbitrary heteroge-
neous parts are substances while they are parts. But the former can, while the
27 Wholly mysterious here is whether undetached arbitrary homogeneous parts have their
own haecceities. Scotus assigns an haecceity to any individual, and of course the class of
individuals is far larger than the class of substances, including individual forms and indi-
vidual matter. So the failure of such undetached parts to be substances is not itself reason
to deny that they have their own haecceities. In favor of the view that such parts do have
haecceities is the thought that since it’s one and the same thing that first isn’t a substance
and then is, and since it indisputably has an haecceity when it is detached, it’s hard to see
how it—that very thing—could exist prior to its detachment and not have its own
haecceity. If it could, then why would it ever require one? Against this view is the very
plausible thought that if every arbitrary homogeneous part had its own haecceity, and if
any homogeneous continuum is infinitely divisible, then there should be as many hae-
ceities actually in the continuum as there are potentially detached from the continuum—
infinitely many. I leave this aporia for another time.
p. 393).
124 Chapter 7
latter cannot, become substances when they cease to be parts (e.g., by being
detached from their wholes). And this, together with the assumption that one
and the same thing first is not a substance while a part and then is a substance
while not a part, entails that, for at least some kinds of substances, being a
substance is a contingent property.
Chapter 8
The Mereological Status of the Elements

in a Mixture
The last section of the last chapter examined the metaphysics of continuous
homogeneous substances and teased out some implications of Scotus’s claim
that the integral parts of such substances are not themselves substances. There
I drew a distinction between elemental and non-elemental homogeneous sub-
stances, where a non-elemental homogeneous substance is a homogeneous
substance that is not one of the four elements, earth, water, air, and fire. That
there are such substances is a commonplace of medieval natural philosophy.
And this commonplace of medieval natural philosophy is one of the most
dramatic differences between medieval and contemporary philosophical out-
looks. Contemporary people, folk and philosophers alike, are taught early on
that the material objects we perceive with our technologically unaided senses
can be analyzed not only into incomprehensibly small parts, but incomprehen-
sibly small parts of kinds very different from the kinds of thing we perceive
with our senses. Organisms are composed of organs, organs of cells, cells of
organelles, organelles of molecules, molecules of atoms, and so on, down to the
elementary particles. Any macroscopic object really is composed of these smaller
things; each of these smaller things is an actual part of the whole organism.
Surface reflections on medieval chemistry suggest an analogous analysis
of material objects into some sort of composition of two or more of the four
elements—earth, air, fire, and water—which were commonly understood to
be the basic ingredients of all material objects. Yet Scotus, along with most
medieval philosophers after Aquinas, denied this. They believed instead that
while every material substance other than the four basic elements was in some
sense a mixture or compound of these four elements, the sense in which it is a
mixture or compound excludes these elements being actual parts of the mate-
rial substance compounded from them. Instead, the common assumption was
that when these elements combine to produce a compound substance, they
cease to exist and remain, if at all, only virtually, that is, they remain only in the
sense that the resulting compound has some of the properties characteristic of
the elements of which it is compounded. A homogeneous substance such as
bone (for so they thought of bone) really and truly is bone all the way down;
there is no integral part of bone that is not bone.

126 Chapter 8
Scotus tows this party line, and this by itself is not surprising. But this is
surprising when considered in combination with the view about part-
substances that I have been attributing to Scotus. Since Scotus holds in general
that there is nothing metaphysically gauche with the position that a substance
can be a part of a substance, his denials that the elements remain in a mixture,
and that the quantitative parts of a homogeneous body are substances, are
prima facie surprising. Having toiled to make room for part-substances in his
Aristotelian ontology, one might have thought that he could let substances be
composed of substances all the way down. For all that has been said so far, any
part-substance of a material substance could itself be composed of part-
substances, and any part-substance of that part-substance could be composed
of part-substances, and so on. Since Scotus rejects atomism, he holds that any
quantified object is infinitely divisible, so it could turn out—again, for all that
has been said so far—that any material substance has an infinitely branching
structure of part-substances. In fact, Scotus does hold that every material sub-
stance is infinitely divisible,1 but he denies (i) that there is a material substance
that has an infinitely branching structure of part-substances, and that (ii) there
is actually an infinite number of part-substances in any material substance.
Scotus denies (i) because he holds that an analysis of a material substance into
its part-substances eventually terminates at part-substances which are not
themselves composed of part-substances, what he calls the homogeneous
parts, things (or stuffs) like bone, flesh, blood, and nerve. He denies (ii) because,
although Scotus is committed to the view that a material substance is infinitely
divisible, he denies that the quantitative parts of a whole continuous homoge-
neous substance are themselves substances.
In what follows, I present some of Scotus’ reasons for denying that the
elements are actual part-substances of the material substances compounded
from them, with an eye to pinpointing the difference between the elements of a
compound substance like bone and the sort of substance that is composed of
part-substances, such that the latter but not the former actually exist in the sub-
stances they make up. The difference lies, I argue, in the natural efficient- and
final-causal activity of elemental substances and organic part-substances. Only
substances whose natural functions are for the sake of the functioning of the
whole and of other parts of the whole are the sort of substances that naturally are
part-substances, and elemental substances have no such functions; their natural
functions, if they have functions, are to move to their respective sublunary places.
1 For exceptionally clear exposition of Scotus’ arguments for infinite divisibility, see Cecilia
Trifogli, “Scotus and the Medieval Debate about the Continuum,” Medioevo 29 (2004),
pp. 233–266.
The Mereological Status Of The Elements In A Mixture 127
Anneliese Maier concluded that medieval attempts to combine a metaphys-

ical analysis of substance into matter and form with a chemical analysis of
substance as a mixture of elements were ultimately unsuccessful, and that
no such attempt could be successful. As she saw it, the chemical and meta-
physical analyses were fundamentally irreconcilable—a medieval version of
Eddington’s Two Tables.2
Scholastic philosophy treated material substance in two ways: as a com-

posite (compositum) of matter and form and as a compound (mixtum)
of the four elements—earth, water, air, and fire. The first interpretation
derived from scholastic-Aristotelian metaphysics, while the second, the
theory of the elements, played a fundamental role in numerous medieval
disciplines and was a self-contained, fully developed system. Both theo-
ries deal with the question of the structure of physical substance and
each offers a solution, but the two answers do not coincide. The meta-
physical approach cannot be squared with the viewpoint of natural
philosophy.3
I myself do not share Maier’s pessimism; I think that Scotus’s solution is suc-
cessful in the sense that it is consistent and is rooted in deep-structure theo-
retical commitments about composition and unity. In short, as Scotus thinks
that organs potentially compose an organism because they are essentially
ordered to each and in that sense unified, so he denies that elements poten-
tially compose a compound substance—elements are not and cannot be
essentially ordered to each other.
i Mixed Opinions about Mixtures
According to the Greek chemistry utilized by scholastic philosophers, the four

elements—fire, air, water, and earth—are the basic ingredients of every mate-
rial substance;4 in Scotus’s words, they are “the lowest bodies and the first from
2 Arthur Eddington, The Nature of the Physical World (New York City and Cambridge, uk: The
MacMillan Company and Cambridge University Press, 1929), pp. ix–xvii.
3 Anneliese Maier, “The Theory of the Elements and the Problem of the Participation in the
Compound,” in On the Threshold of Exact Science: Selected Writings of Anneliese Maier on Late
Medieval Natural Philosophy, ed. and trans. Steven D. Sargent (Philadelphia: University of
Pennsylvania Press, 1982), pp. 124–142, especially pp. 125–126.
4 For an introduction to medieval chemistry, see Anneliese Maier, “The Theory of the Elements
and the Problem of the Participation in the Compound.”
128 Chapter 8
which others are generated.”5 Theoretically each can exist unmixed with
any other, and discrete quantities of a pure element are themselves sub-
stances, capable of acting on surrounding material substances, including other
elemental substances.6 Like any natural substance, an elemental substance
produces its characteristic effect, and occupies its characteristic place, when-
ever it can, e.g., fire strives to heat and to move to the outermost region of the
sublunary sphere. According to Aristotle each element has characteristic qual-
ities: fire is hot and dry, air is wet and hot, water is cold and wet, and earth is
dry and cold.7
Elements can be combined to make up substances with natures distinct from
the nature of any individual element. The precise nature of a mixture of the ele-
ments is supposed to be in some sense determined by the proportion of the
elements that make it up. Importantly, a mixture of elements is distinguished
from a mere juxtaposition of elements. In a juxtaposition the elements are sim-
ply next to one another, like a blend of spices. In a mixture no quantitative part,
however small, is a pure elemental body; according to Aristotle, “If combination
has taken place, the compound must be uniform—any part of such a com-
pound being the same as the whole, just as any part of water is water.”8
The dialectical agenda of medieval discussions of the status of the elements
in a mixture was set by Avicenna and Averroës.9 Avicenna held that the
essences of the elements were “fixed and permanent” in a mixture, but their
qualities were “changed and converted.”10 Averroës disagreed, holding that an
5 Lectura II, d.15, q.un, n.47 (Vatican XIX, p. 153), [S]unt infima corpora et sunt prima cor-
pora ex quibus generantur alia.
6 The widespread medieval commitment to the idea that the elements are substances—
composites of prime matter and substantial form—seems to have been the source of all
of the hullabaloo about their status in the mixture. At the end of an era Suàrez was able
to say that if the elements were not substances they would cause no mereological trouble:
they would remain in the mixture. He himself thought they were substances, however,
and adopted the Thomistic position that they remain in the mixture virtually. Francisco
Suàrez, Disputationes Metaphysicae XV, sec.X, ed. Berton (Paris: Vivès, 1866), pp. 536–557.
7 Aristotle, On Generation and Corruption II.3 330b 4–5 (Barnes I, p. 540).
8 Ibid., I.10 328a 10–12 (Barnes I, p. 536).
9 For background, see Abraham Stone, “Avicenna’s Theory of Primary Mixture,” in Arabic
Sciences and Philosophy 18 (2008), pp. 99–119; Rega Wood and Michael Weisberg, “Interpreting
Aristotle on Mixture: Problems about Elemental Composition from Philoponus to Cooper,”
Studies in the History and Philosophy of Science 35 (2004), pp. 681–706; Anneliese Maier, “The
Theory of the Elements and the Problem of the Participation in the Compound.”
10 Avicenna, Sufficienta I. c.10 (Venice 1508, I, f.19rb), Et harum formae essentiales sunt fixae
et permanentes, sed accidentia earum…mutantur et convertuntur. Quoted in Vatican
XIX, p. 139.
element could not exist without its qualities. He held instead that both the ele-
ments and their qualities remained in a mixture, but in a diminished way, such
that the mixture itself is a medium between the elements and their qualities.11
Averroës thought both that one substance could be more or less intense
than another of the same species, and that numerically the same substance
could change in its degree of intensity, and it is specifically the second claim
that is operative in his account of the existence of the elements in a mixture.
Averroës holds that the substantial forms of elemental substances are dimin-
ished and become as it were a medium between a substantial form and a quali-
tative form. The qualities of the elements are also diminished, such that a
mixture of these diminished elements exhibits a more or less uniform com-
plexion—the fire in the mixture becomes cooler and moister, the water in the
mixture becomes warmer and drier, and so on.
Aquinas’s criticisms of Avicenna and Averroës and his own account of the
status of the elements in a mixture were influential on subsequent thinkers,
including non-Dominicans such as Scotus, Ockham, Buridan, and Albert of
Saxony,12 and was a departure from the view of his teacher, Albert the Great,
which was a modified Avicennan theory.13 When Peter Aureole surveyed the
scholarly literature on the status of the elements, Aquinas’s view had become
popular enough that he was able to distinguish three major perspectives:
Avicenna’s, Averroës’s, and the Moderns’, where it is clear from the text that the
Modern way is Aquinas’s.14 Against Avicenna Aquinas reasoned that, since no
substance can have more than one substantial form, and since the elements
were taken to be composites of prime matter and substantial form, at best a
mixture of elements could be an aggregate of very tiny bodies.
For it is impossible for the same matter to sustain diverse forms of ele-
ments. If, therefore, in the mixed body the substantial forms of the ele-
ments are preserved, it will be necessary for them to be in different parts
of the matter. But it is impossible for matter to have different parts unless
quantity is already understood to be in matter, for having taken away
11 Averroës De caelo III com. 67. [D]icemus quod formae istorum elementorum substantia-
les sunt deminutae a formis substantialibus perfectis, et quasi suum esse est medium
inter formas et accidentia. Quoted in Vatican XIX p. 143.
12 Ockham, Quodlibet III, q.5 (OTh IX, pp. 220–224); Buridan, De generatione et corruptione I,
q.22 (Streijger, pp. 232–237); Albert of Saxony, De generatione et corruptione I, q.18, ed.
Lockert (Paris, 1516), f. 141–142.
13 Albert the Great, De generatione et corruptione I, tr.6, c.5 (Cologne ed. V.2, p. 172).
14 Peter Aureole, Commentarium in secundum librum Sententiarum, d.15, a.1 (Rome: ex
Typographia Vaticanae, 1596), p. 206.
130 Chapter 8
quantity the substance remains indivisible, as is clear from Physics I.

But from matter existing under quantity and the arriving substantial
form a physical body is constituted. Therefore the parts of matter subsist-
ing under the forms of the elements take on the character of several bod-
ies. But it is impossible for [one body] to be many bodies at once;15
therefore the four elements will not be in every part of the mixed body.
And thus it will not be a true mixture, but an apparent one, as happens in
an aggregation of bodies that are imperceptible on account of their
smallness.16
Not only would such a mixture fail to be a true mixture, it would fail to be one
body or substance. But every body was taken to be a mixture of elements. So
Avicenna’s theory has the consequence that ordinary bodies like organisms are
not per se unified.
Against Averroës Aquinas asserted the authority of Aristotle: for two
extremes to form a medium, they must be opposites within the same genus.
For example, white and black are the extremes of an intermediate quality
(grey) because they are both colors. Because a substance or substantial form
and an accident do not belong to the same genus, they cannot make something
intermediate between them.17 Moreover substances do not come in degrees
and therefore do not admit of intermediaries.18
According to Aquinas, the elements do not actually exist in a substance. But
the characteristic qualities of the elements interact in such a way that some
intermediary quality is generated, and this intermediary quality disposes the
matter of the elements to receive the form of the mixed body. When the form
of the mixed body is received, the substantial forms of the elements recede.
15 This clause follows Paul Vincent Spade’s translation: http://pvspade.com/Logic/docs/

mixture.pdf.
16 Aquinas, De mixtione elementorum, Impossibile est enim materiam secundum idem
diuersas formas elementorum suscipere; si igitur in corpore mixto formae substantiales
elementorum saluentur, oportebit diuersis partibus materie eas inesse. Materie autem
diuersas partes accipere est impossibile nisi preintellecta quantitate in materia, sublata
enim quantitate substantia indiuisibilis permanet, ut patet in I Phisicorum; ex materia
autem sub quantitate existente et forma substantiali adueniente corpus phisicum consti-
tuitur: diuerse igitur partes materie formis elementorum subsistentes plurium corporum
rationem suscipiunt. Multa autem corpora impossibile est esse simul; non igitur in qual-
ibet parte corporis mixti erunt quatuor elementa: et sic non erit uera mixtio, sed secun-
dum sensum, sicut accidit in aggregatione corporum insensibilium propter paruitatem.
17 Aristotle, Metaphysics X, 1057a18-34 (Barnes II, p. 1670).
18 Aristotle, Categories V, 3b24-4a7 (Barnes I, p. 7).
But the elements continue to exist virtually due to the existence in the mixture
of the intermediary quality.
It is necessary therefore to come up with some other way, by which both

the truth of mixture is preserved, and by which the elements are not
totally corrupted, but remain somehow in the mixture. It must be consid-
ered therefore that the active and passive qualities of the elements are
contrary to each other, and are receptive or more and less. But from con-
trary qualities that receive more and less, there can be constituted a mid-
dle quality that shares in the nature of both extremes, just as pallor
between white and black, and warm between hot and cold. So, therefore,
the excellence of the qualities of the elements remitted, there is consti-
tuted from these a certain middle quality that is the proper quality of the
mixed body, differing however in diverse bodies according to a diverse
proportion of the mixture, and this certain quality is the proper disposi-
tion for the form of the mixed body, as a simple quality is for the form of
a simple body. Therefore, just as extremes are found in the middle that
participates in the nature of both, so the qualities of the simple bodies
are found in the proper quality of the mixed body. But the quality of the
simple body is something other than its substantial form; nevertheless it
acts by the power of the substantial form; otherwise, heat would only
heat, while through its action the substantial form would not be educed
in the action, since nothing acts beyond its species. So, therefore, the
powers of the substantial forms of the simple bodies are preserved in the
mixed bodies. There are then forms of the elements in mixed bodies, not
in act but in power.19
19 Aquinas, De mixtione elementorum, Oportet igitur alium modum inuenire, quo et ueritas
mixtionis saluetur, et tamen elementa non totaliter corrumpantur, sed aliqualiter in
mixto remaneant. Considerandum est igitur quod qualitates actiue et passiue elemento-
rum contrarie sunt ad inuicem, et magis et minus recipiunt. Ex contrariis autem qualitati-
bus que recipiunt magis et minus, constitui potest media qualitas que sapiat utriusque
extremi naturam, sicut pallidum inter album et nigrum, et tepidum inter calidum et frigi-
dum. Sic igitur remissis excellentiis qualitatum elementarium, constituitur ex hiis
quedam qualitas media que est propria qualitas corporis mixti, differens tamen in diuer-
sis secundum diuersam mixtionis proportionem; et hec quidem qualitas est propria dis-
positio ad formam corporis mixti, sicut qualitas simplex ad formam corporis simplicis.
Sicut igitur extrema inueniuntur in medio quod participat naturam utriusque, sic quali-
tates simplicium corporum inueniuntur in propria qualitate corporis mixti. Qualitas
autem simplicis corporis est quidem aliud a forma substantiali ipsius, agit tamen in uir-
tute forme substantialis; alioquin calor calefaceret tantum, non autem per eius actionem
132 Chapter 8
Scotus adopted the basic Thomistic solution, denying that the elements actu-
ally in the mixture but trying to carve out some way in which they exist in the
mixture virtualiter. Although he apparently held the Averroistic view according
to which one substance could be more or less intense than another of the same
species, he denied that numerically the same substance could diminish or
intensify according to its substantial form.20 Thus he could not adopt the
Averroistic theory of the elements, according to which one and the same ele-
mental substance moves from a high degree of intensity outside of the mixture,
to a low degree of intensity within the mixture. His rejection of the Avicennan
theory is, as I have already indicated, somewhat surprising, since Scotus’s plu-
ralism about substantial form enables him in general to hold that one sub-
stance can have many substances as parts. All the more surprising is that Scotus
seems to think that the principle of parsimony alone should establish that the
elements do not remain in a mixture. He writes, “If, against [Avicenna’s] way, I
had nothing other than ‘plurality is not to be posited without necessity,’ I would
not proclaim this way.”21 Yet Scotus offered additional arguments for his view,
and in the following sections I analyze several of these, mostly tracking the
arguments given in Lectura II, d.15, q.un, but at times looking abroad.
ii The Argument from Quantitative Forms
In the argument from quantitative forms Scotus reasons from the premise that
every material substance has its own quantitative form.
forma substantialis educeretur in actum, cum nichil agat ultra suam speciem. Sic igitur
uirtutes formarum substantialium simplicium corporum in corporibus mixtis saluantur.
Sunt igitur forme elementorum in corporibus mixtis, non quidem actu sed uirtute.
20 Quaestiones super Libros Metaphysicorum Aristotelis VIII, qq.2-3, nn.220, 233 (Bonaventure
IV, pp. 477, 484). Scotus held the first apparently only on the basis of Proposition 124 of
Bishop Tempier’s list of condemned propositions in 1277: Quod inconveniens est ponere
aliquos intellectus nobiliores aliis, quia, cum ista diversitas non posset esse a parte corpo-
rum, oportet quod sit a parte intelligentarium: et sic animae nobiles et ignobiles essent
necessario diversarum specierum, sicut intelligentiae—Error, quia sic anima Christi non
esset nobilior anima Iudae. “That it is unfitting to posit that some minds are more noble
than others because, since that diversity could not be on the part of the body, it would be
necessary that it be on the part of the minds: and so souls noble and ignoble would neces-
sarily be of diverse species, just like minds—Error, because then the soul of Christ would
not be nobler than the soul of Judas.” Cf. Chartularium Universitatis Parisiensis, prop. 124,
ed. H. Denifle and A Chatelain (Paris 1889, p. 1550).
21 Lectura II, d.15, q.un, n.21 (Vatican XIX, p. 143), Sed si contra hanc viam non haberem aliud
nisi “pluralitas non est ponenda sine necessitate,” hanc viam non dicerem.
[E]very positive accident has for its proper subject a composite sub-
stance, and consequently no quantity is in prime matter as in its proper
subject; similarly, there is no quantity which follows a form that is prior
to the form of an element (because one posits that as the first), but flesh
[a kind of mixture] is quantified and has its proper quantity. Either there-
fore the forms of the elements have their proper quantities, or not. If not,
then there would be some generable and corruptible form without its
proper quantity, which I hold to be unfitting. But if each one has its
proper quantity, then there will be several quantities at once, because
“every part of the mixed is mixed.”22
A mixture is a material substance, so it has its own quantitative form. If the

elements remain in a mixture, then they have their own quantitative forms
since they too are substances in their own right. But it is a feature of any true
mixture that “every part of the mixed is mixed.” So if the elements remain in a
mixture, then for any quantitative part of a mixture, each element present in
that mixture exists in that part—every part is fire and every part is earth and so
on. Therefore the quantities of these elements exist in every part as well. In the
text that Wadding printed as Ordinatio II, d.15, q.un,23 Scotus or someone writing
22 Lectura II, d.15, q.un, n.22 (Vatican XIX, pp. 144–145), Praeterea, omne accidens positivum
habet pro subiecto proprio substantiam compositam, et per consequens nulla quantitas
est in materia prima ut in proprio subiecto; similiter, nulla quantitas est quae consequitur
formam priorem forma elementari (quia illam ponit primam), sed caro est quanta et
habet propriam quantitatem. Aut igitur formae elementares habent suas proprias quanti-
tates, aut non. Si non, tunc esset aliqua forma generabilis et corruptibilis sine propria
quantitate, quod habeo pro inconvenienti. Si autem quaelibet habeat propriam quantita-
tem, tunc erunt plures quantitates simul, quia “quaelibet pars mixti est mixta.”
23 The Vatican editors left out d.12 (along with all of Book II, Distinctions 15–25) of their criti-
cal edition (Vatican VIII). According to the editors these distinctions were originally parts
of other works of Scotus or of William of Alnwick’s Additiones Magnae (Vatican VIII,
p. 224), a compilation of Scotist material which is generally taken to be faithful to Scotus’s
own mind. According to Thomas Williams, “Three manuscripts of Additiones 2 contain an
explicit attributing the Additiones to Scotus and identifying Alnwick not as their author
but as their compiler: ‘Here conclude the Additions to the second book of Master John
Duns, extracted by Master William of Alnwick of the Order of Friars Minor from the Paris
and Oxford lectures of the aforesaid Master John.’ In their earliest appearances, the
Additiones were identified as an appendix to Scotus’s Ordinatio, but they gradually came
to be inserted into the Ordinatio itself to supply material where Scotus had left the
Ordinatio incomplete—a process that attests to the belief of Scotus’s contemporaries
and immediate successors in the authenticity of the Additiones.” Thomas Williams,
“Introduction,” The Cambridge Companion to Duns Scotus, ed. Thomas Williams. The
134 Chapter 8
ad mentem Scoti explains that the problem of several quantities existing in
every part of the mixture is equivalent to the problem of co-located bodies.
Scotus does not draw an explicit conclusion—the reader is left to wonder
whether Scotus thinks that this result is absurd or merely embarrassing—but
in either case the conclusion is supposed to be unwelcome. In the Reportatio
II-A, d.15, q.un version of the argument Scotus writes,
The rationale [for denying that the elements remain in a mixture] is that
quantity follows the composite, as a property of substance follows body.
But the same property does not follow several supposits at once. Therefore
it would be necessary that there would be as many quantities in act there
as there are elements, and then there will be a juxtaposition of elements
and not a mixture.24
Reportatio II-A takes for granted that elements never exist without their proper
quantities, but modifies what is supposed to be problematic about a genuine
mixture of several quantified bodies. Scotus assumes that if several bodies
could occupy the same extended place, then they would all share numerically
the same quantitative form. But several bodies cannot share numerically the
same quantitative form (“the same property does not follow several supposits
at once”). Therefore the elements cannot be truly mixed. At best there is just a
juxtaposition of elements—many tiny bodies existing in close proximity to
each other, but not composing one bodily substance.
The arguments taken together generate the following dilemma: if the ele-
ments remained in a mixture, either they would be co-located, or they would
be merely juxtaposed. Exactly what is supposed to be problematic with these
options is not totally clear, given some of Scotus’s other commitments.
Co-location may be counter-intuitive, but Scotus argues that it is possible, if
only for God.25 Scotus’s worry must be, then, that since no natural agent can
manuscript from which Williams quotes is Oxford, Balliol College, ms 208, f.40v. For addi-
tional confirmation of the reliability of the Additiones Magnae, see Stephen Dumont,
“Did Duns Scotus Change His Mind on the Will?” Miscellanea Mediaevilla 28 (2001),
pp. 719–794.
24 Reportatio Parisiensis II-A, d.15, q.un, n.5 (Wadding-Vivès XXIII, pp. 64–65), Item rationale
est quod quantitas consequatur compositum, sicut passio substantiae corpus; sed non
sequitur eadem passio plura supposita immediata; igitur oportet quod sint tot quanti-
tates in actu ibi, quot elementa, et tunc erit juxtapositio elementorum et non mixtum.
25 See Marilyn McCord Adams’s discussion in Some Later Medieval Theories of the Eucharist
(Oxford: Oxford University Press, 2010), pp. 111–138, and especially pp. 120–127. Also,
Ordinatio IV, d.10, p. 1, q.2 (Vatican XII, pp. 77–109), especially n.159 (p. 101).
bring it about that several bodies exist in the same place at once, any mixture
of the elements (a paradigm of natural change) would be miraculous.
Mere juxtaposition would be problematic, since nothing made from the ele-
ments would be genuinely one substance. But why think that juxtaposed
elemental bodies are not in potency to a perfecting substantial form (such as
the form of the mixture), analogous to the way in which several organic part-
substances are in potency to the soul? In Wadding’s Ordinatio II, d.15, q.un,
Scotus reasons,
If there are several forms of the elements in a mixed body, each one con-
stitutes a supposit. Thus in every mixed body there would be several sup-
posits, because there will be a supposit of water, and a supposit of fire
[etc.] of which each one subsists per se by nature—which is unfitting.
Likewise it is unfitting that one subsistent thing could have two specifi-
cally different forms, of which one is not naturally perfected by the other—
but this would be the case, if the elements were posited in the mixture
according to their forms.26
The reason why the elements cannot exist in a mixed body is not, according to
this text, because no substance (or supposit) can be composed of substances
(or supposits). Instead, it is because elemental substances are not naturally
perfected by the substantial forms either of the other elements or of the mixed
body. But organic parts are so perfected. A mole heart and mole brain and so
on come into existence for the sake of contributing to the overall functioning
of Mole; they are tailor-made for him (or at least his kind). In arguing that
some substances have a plurality of substantial forms, Scotus reasoned that
the single esse of a composite substance can be constituted by several substan-
tial forms, provided there is some final form whose special role it is to complete
the substance:
[…T]he whole composite is one esse, and nevertheless includes many

partial esse, just as a whole is one being and nevertheless has and includes
26 Ordinatio II, d.15, q.un, n.5 (Wadding VI, p. 753), Item, forma elementaris nata est cum
materia constitutere suppositum per se subsistens in genere Substantiae: ergo si sint
plures formae elementares in mixto, quaelibet constituet suppositum: & sic in omni
mixto essent plura supposita, quia ibi erit suppositum aquae, & suppositum ignis, quo-
rum quodlibet natum est per se subsistere: quod est inconveniens, inconveniens etiam
est, quod subsistens possit habere duas formas specificas, quarum una non est nata per-
fici ab alia: hoc autem poneretur, si elementa ponerentur in mixto secundum formas suas.
(Italics mine.)
136 Chapter 8
many partial entities […Nevertheless] I concede that the total esse of the
whole composite is principally through one form, and that is the form by
which the whole composite is this being; but that [form] is the last, com-
ing to all the preceding […And] I concede that that total esse is completed
by one form […] An example of this is a composite of integral parts: for
the more perfect is the animal, the more it requires many organs (and it
is probable that these are of distinct species through substantial forms);
and nevertheless it is more truly one…27
Scotus therefore thinks that no conjunction of elemental bodies is ever in

potency to a further substantial form, such that they cannot both exist and be
part-substances of a more complex substance. (And this must be why he says
that if elements were actual in a mixture, they would be supposits, rather than
substances, where a supposit is a substance that is not a part of another sub-
stance and is of a determinate kind.) But why think that the elements are never
thus in potency? Perhaps Scotus is assuming that the elements are never suit-
ably ordered to one another to be together in potency to a perfecting substan-
tial form. Recall from earlier chapters that Scotus argues that only when
composing a unity of order are organic part-substances together in potency to
a soul. I argued that this unity of order is at least one of two kinds of essential
orders: final and efficient causality. One organic part comes to be for the sake
of another as well as for the whole organism, and one organic part’s efficient
causal activity enables the efficient causal activity of another organic part.
If Scotus thinks that composing a unity of order in either or both of these
senses is a necessary condition on n substances potentially composing one
substance (by being together informed by one substantial form), then in the
absence of such order(s) among elemental bodies, they are not potentially one
substance. It is not clear to me how someone might go about thoroughly show-
ing whether the elements can have such an order among themselves, but it is
possible to take a couple steps in the right direction. First, it seems clear that
27 Ordinatio IV, d.11, p. 1, a.2 q.1, nn.251–254 (Vatican XII, pp. 255–256), […T]otius compositi
est unum “esse,” et tamen includit multa “esse” partialia, sicut “totum” est unum ens et
tamen multas partiales entitates habet et includit […Tamen] concedo totale “esse” totius
compositi est principaliter per formam unam, et illa est forma, qua totum compositum
est “hoc ens”; illa autem est ultima, adveniens omnibus praecedentibus […Et] conedo
quod “esse” istud totale est completive ab una forma […] Exemplum huius est in com-
posito ex partibus integralibus: quanto enim animatum est perfectius, tanto requirit plura
organa (et probabile est quod distincta specie per formas substantiales); et tamen ipsum
est verius unum…
no two elemental bodies of different kinds are efficient-causally ordered to one

another, since the elements are naturally contrary to one another and, when
two elemental bodies exist in proximity and one is more powerful than the
other, the more powerful corrupts the less. Second, insofar as the elements can
be described as final-causally ordered at all, it seems that the final cause of
elemental motion is simply for each element to arrive at its proper location.28
To sum up, Scotus seems to argue that the elements are not co-located in
a mixture because this would unsuitably involve miraculous intervention in
a paradigmatically natural change. Elements are not merely juxtaposed in a
mixture because, if they were, no mixed body would be a genuine substance.
Given the natural efficient- and final-causal activity of elemental bodies,
there is some reason to think that elemental bodies cannot be ordered to
one another so as to compose something that is potentially one bodily
substance.
iii The Generation and Corruption Argument
In Scotus’s second argument he reasons from two common assumptions, first,

that a mixture is generated from the elements, and second, that a mixture is
corrupted into the elements.
Next, a mixture is generated from elements, and also a mixture is cor-

rupted into elements: therefore it can be a term “from which” and a term
“to which” of generation, just as the elements among themselves, of
which one is generated from another and is corrupted into it. Therefore a
mixture and the elements have incompossibility between them, as the
elements have. Therefore just as one element does not remain in another,
so neither will an element remain in a mixture.29
28 In On the Heavens, Aristotle says that the natural movement of the element, earth, is to
the center of the cosmos, and that to reach the center is the element’s goal (II, 296a6-14
(Barnes I, p. 487)). He also says that the elements have functions (III, 298a24-34 (Bares I,
p. 489)). And, in Parts of Animals he says that when there is a final end of action, the
action is always for the sake of the final end (I, 642b24-26 (Barnes I, p. 998)). These in
conjunction suggest that elements have final causes, and that these final causes are the
goals of elemental natural motion. See the discussion in Caleb Kinlaw, “Elemental
Teleology and an Interpretation of the Rainfall Example in Physics 2.8,” unpublished mas-
ters thesis (University of Edinburgh, 2008), pp. 25–27.
29 Lectura II, d.15, q.un, n.23 (Vatican XIX, pp. 145), Praeterea, mixtum generatur ex elemen-
tis, et etiam corrumpitur mixtum in elementa: potest igitur esse terminus “a quo” et
138 Chapter 8
The generation of one substance, s1 from another, s2, involves the corruption of
s2. So if a mixture is generated from the elements, the elements are corrupted.
But if they are corrupted when the mixture is generated, they cannot exist in
the mixture.
The generation and corruption argument is bad. Of course, if a mixture is
generated from the elements in the normal Aristotelian fashion, such that
the generation of one substance is always the corruption of another,
then the elements are corrupted when a mixture is generated. But this is
exactly what is at issue—whether or not the elements remain when a mix-
ture is generated from them. What might be going on in the argument is that
Scotus recognizes that any natural generation does involve the corruption of
something. If the elements are not corrupted when a mixture is generated
(supposing that they remain in a mixture) then what is? No other candidates
are on offer. Even this, however, will not work given other of Scotus’s commit-
ments. Scotus holds that a process of generation can be broken up into
several stages. For example, he holds that in the generation of a blooded
animal the heart is generated first, then the other organs, and finally the ani-
mal itself when the soul begins to inform a sufficiently prepared organic
body. The final stage—the union of the soul with the body—is strictly speak-
ing the generation, according to Scotus.30 But this stage in the process does
not involve the corruption of anything. The corruption requisite for the gen-
eration of the animal occurs earlier in the process, at the generation of the
first organ, and at the subsequent generations of other organs, as the devel-
oping organic body changes nutriments into organic parts. Given this model,
it could be open for Scotus to hold that the corruption requisite for the gen-
eration of a mixture occurs at some early stage, say, when elements corrupt
other elements and take on an arrangement and proportion that disposes
them to receive a form of a mixed body. As it turns out, and as we will see in
more detail below, Scotus does not think it is possible for a mixture to be
generated by the activity of the elements on one another, but he thinks this
for reasons altogether different from those advanced in the generation and
corruption argument.
terminus “ad quem” generationis sicut elementa inter se, quorum unum generatur ex alio
et corrumpitur in illud; igitur mixtum et elementum habent incompossibilitatem inter se
qualem habent elementa: sicut igitur unum elementum non manet in alio, ita nec ele-
mentum manebit in mixto.
30 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.52–53 (Bonaventure IV,
pp. 393–394).
iv The Violence Argument
The violence argument reasons from the natural opposition of the elements. If
the elements actually composed a mixed substance, at least one of the ele-
ments would be there unnaturally since it would be out of its proper sublunary
place and would strive to return there. But this violence is inconsistent with
the stability and harmony of natural substances.
Next, if an element remained in a generated mixture, so that one element

remained there and another [remained there], because there is there the
quality of one element as well as another, therefore since the elements do
not have the same place, only one element would be there naturally, and
the other [would be there] violently, and consequently the parts of the
mixed would be violently in the mixture—which is false, because a part
has natural existence in the whole.31
Suppose two elements remained in a mixture, and that the mixture occupied
the natural place of one the elements. For example, suppose a mixture of earth
and fire were very close to the center of the earth. Then the fire in that mixture
would be there violently rather than naturally—its natural tendency would be
to move closer to the moon, whereas the earth in the mixture would stay put.
In this case the mixture itself would be unnatural since its parts are together
unnaturally. The problem however is that every material substance other than
the pure elements are mixtures or are composed of mixtures. So if the ele-
ments remained every material substance would be unnatural. This is false,
Scotus says, “because a part has natural existence in the whole.”32 Presumably
Scotus means something like, “A part has natural existence in the whole, if the
whole is a natural substance, that is, a substance with a specific nature.”
An opponent whom the Vatican editors identify as Richard of Middleton
argued that the existence of the elements in a natural mixture explained the
intrinsic corruption of substance.33 As Aristotle observed, material substances
31 Lectura II, d.15, q.un, n.7 (Vatican XIX, p. 139), Praeterea, si elementum maneret in mixto
generato, qua ratione unum elementum ibi maneret, et aliud, quia ita est ibi qualitas
unius elementi sicut alterius; igitur cum non habeant elementa eundem locum, tantum
unum elementum esset ibi naturaliter et alia violenter, et per consequens partes mixti
violenter essent in mixto—quod falsum est, quia pars habet naturale esse in toto.
32 Ibid.
33 Richard of Middleton, Quaestio de gradu formarum in corp. c.3, n.3 (PhM II, p. 125), quoted
in Vatican XIX, p. 138, Nisi enim essent elementa in mixto…in ipsis mixtis non esset
140 Chapter 8
break down through both internal and external causes.34 In reply, Scotus
asserted that while organic substances do undergo intrinsic corruption, inor-
ganic substances do not, unacquainted as he was with radioactive decay: “if an
extrinsic contrary agent were not corrupting the stone, supposing the general
conservation of God, the stone would be conserved eternally just like the heav-
ens.”35 If the elements remained in the stone, however, we would not expect it
to be so stable. The fact that organic substances do undergo intrinsic corrup-
tion is not, for Scotus, a good reason to suppose that in them the elements do
remain. Scotus argues that their internal corruption can be explained by the
fact that the parts of such substances are heterogeneous, with contrary domi-
nating qualities. I will explore his reasons for thinking this in some detail later
in this chapter.
v Generation from the Elements
Based on the above arguments, among others, Scotus denies that the elements
remain in a mixture.36 In what way, then, is a mixture a mixture of the ele-
ments? Scotus recognizes the grammatical awkwardness and bites the bullet:
the elements are not elements strictly speaking; in reality, prime matter and
the kinds of substances that can be actual parts of other substances are truly
elemental. Mixtures are, strictly speaking, mixtures of prime matter and sub-
stantial form.37
Scotus argues that if the elements do not remain in a mixture, they cannot
be the efficient causes of a mixture, since at the moment the mixture comes to
be, ex hypothesi no element is there bringing it into being.38 Nor can a mixture
principium corruptionis intrinsecum…for unless the elements be in a mixture, there

would not be in mixture itself a principle of intrinsic corruption.
34 Aristotle, On Youth, Old Age, Life and Death, and Respiration 23(17) 478b 24–26 (Barnes I,
p. 760), There is violent death and again natural death, and the former occurs when the
cause of death is external, the latter when it is internal…
35 Lectura II, d.15, q.un, n.43 (Vatican XIX, pp. 151).
36 Ibid., n.26 (Vatican XIX, p. 146).
37 Ibid., n.47 (Vatican XIX, p. 153), [D]ico quod secundum rei veritatem nomen “elementi” et
eius definitio convenit materiae primae, et similiter materiae si habens formam priorem
possit manere sub forma posteriore. Sed vere et proprie non convenit igni et aquae etc.
“I say that according to the truth of the matter the name ‘element’ and its definition per-
tain to prime matter, and similarly to the matter that, having a prior form, is able to remain
under a posterior form. But truly and properly it does not pertain to fire and water, etc.”
be produced through any action of the elements on one another. Take two ele-
ments, fire and water, where it is supposed that a mixture is generated from
their mutual interaction (mutuo agentibus):
I ask: either fire remains when it corrupts water, or not; if not, therefore
that which is nothing, corrupts another; if it remains, therefore after the
corruption of water fire will remain.39
Scotus is therefore led to give an analysis of the generation of a mixture from

the elements roughly in terms of ordinary cases of generation and corruption:
an agent operates on a patient (one or more elemental bodies), educing a new
substantial form (the form of the mixture) from the potency of the matter.40
He notes, however, this peculiar feature of the generation of mixtures: the mix-
ture is more like the elements from which it is generated, than it is like its effi-
cient cause.41 This entitles the elements to some sort of presence in a mixture,
so Scotus says that they remain in the mixture sicut in effecti communi42 and
also (though less frequently) virtualiter.43 As far as I can tell, all that Scotus
means by the virtual presence of the elements in a mixture is that the dominat-
ing quality of a mixture specifically resembles a mean quality of the elements
from which the mixture is generated.
Scotus does not attempt to explain why generation from the elements has
this peculiar feature. Since he holds that both the substantial forms and quali-
ties of the elements perish when the mixture is generated, he cannot explain
the similarity of the mixture and the elements as a sharing of numerically the
same qualities—with a new substance comes new accidents. Perhaps surpris-
ingly, however, he thinks that the new qualities of the new mixture must also
be specifically different from the qualities of the elements—the heat or cool-
ness or moistness or dryness of a mixture is both numerically and specifically
different from the four qualities of the four elements.44 Since Scotus does not
in general hold that qualitative accidents of a certain kind can inhere in only
one kind of substance, his denial that elemental qualities can inhere in mix-
tures must be rooted in the idea that the elemental qualities are properties of
39 Ibid., n.30 (Vatican XIX, p. 147), [Q]uaero: aut ignis manet quando corrumpit aquam, aut
non; si non igitur illud quod nihil est, corrumpit aliud—si maneat, igitur post corruptio-
nem aquae remanebit ignis.
42 Ibid., nn.27, 38 (Vatican XIX, pp. 146, 149–150).
142 Chapter 8
the elements in the strict sense that they are necessary accidents belonging to
one and only one kind of substance, in just the way that risibility is a property
of a human. Just why it is that a mixture has qualities that resemble one or
more of the elemental qualities is left unexplained, but the similarity of these
qualities is the reason that it is correct to describe a mixture as having a domi-
nating element, whose characteristic activities are similar to those of the char-
acteristic activities of the element.
vi Mixtures and Organic Parts
Richard of Middleton had argued that if the elements did not remain in a mix-
ture, there would be no explanation of the intrinsic corruption that substances
undergo. Scotus seems to have thought that an inanimate mixture (such as a
stone) was totally impervious to internal corruption, and took this as evidence
that the elements did not remain in mixtures. What then of substances, such
as animals, that do suffer internal corruption?
In Parts of Animals Aristotle theorized that organic parts are both homoge-
neous and heterogeneous, where the heterogeneous parts are in some sense
composed of homogeneous parts and the homogeneous exist for the sake of
the heterogeneous. So for example bones, sinews, and flesh—homogeneous
parts by Aristotle’s lights—compose and exist for the sake of parts like a hand
or a face.45 Scotus adopts this basic division:
[A man] is composed of many and diverse parts, certain homogeneous

and certain heterogeneous, which are suited to the diverse operations
performed by a man, for which there come to be many organs, because
one organ does not suffice for all of those performed operations […] But
these diverse parts are composed of homogeneous parts […]46
Scotus reasons that while one mixture does not contain contrary elements
as parts and therefore will not suffer internal corruption, one organism is
composed of parts of different kinds of mixtures, all of which have their own
45 Aristotle, Parts of Animals II, 1 646b 11–27 (Barnes I, p. 1006).

46 Reportatio Parisiensis IV-A d.44, q.1, n.2 (Wadding XI, p. 854), Deinde componitur ex mul-
tis et diversis partibus, quibusdam homogeneis et quibusdam heterogeneis, quae congru-
unt diversis operationibus ab homine excercendis, quarum nata sunt esse plura organa,
quia unum organum non sufficit ad omnes illas operationes excercendas […] Hae autem
partes diversae componuntur ex partibus homogeneis […]
dominating element and associated dominating quality, what Scotus calls

their complexions (complexiones):
Just as the organic parts in an animal have different substantial forms (as
it will be said in Book III47) so also they have different predominating
qualities […] I say then that the parts of animals have various and diverse
complexions, just as the brain is cold and the heart is hot, and when in
the brain coldness excessively dominates, then the other parts are made
colder, against their natural complexions; and thus corruption is caused
in the animal when the principal parts are excessively altered by contrary
qualities.48
Scotus’s account here leaves it an open possibility that some organic parts are
themselves composed of mixtures with contrary complexions, such that an
organ itself can be corrupted both from within and from without, i.e., from
other organs of the same organism. Presumably the only internally stable
organic parts are those which are not composed of mixtures of contrary com-
plexions, and I take it that such parts would be the homogeneous parts, since
any heterogeneous part is by definition composed of more than one kind of
mixture, and for any two distinct kinds of mixtures they will have distinct and
therefore (to some extent) contrary qualities. For example, if one part of a het-
erogeneous part is slightly hotter than another part of the same heterogeneous
part, the two will be opposed with respect to hot and cold (even if it is only a
light skirmish), and therefore the heterogeneous part will be susceptible to
internal corruption. Presumably, the homogeneous parts, like stones, will only
be susceptible to external corruption, since each part of a homogeneous part
will have the same complexion. This suggests that, in a Scotistic analysis of
part-substances, once you get to the homogeneous substances—e.g., “fleshes
and bones”49—you get to the bottom of things. The four elements then are not
47 Lectura III d.2, q.3 (Vatican XX, pp. 110–117).

48 Lectura II, d.15, q.un, nn.41–42 (Vatican XIX, pp. 150–151), Unde sicut partes organicae in
animali habent alias formas substantiales (ut in III dicetur) ita etiam habent alias quali-
tates praedominantes […] Dico tunc quod partes animalis habent varias et diversas com-
plexiones, sicut cerebrum est frigidum et cor calidum; et quando in cerebro nimis
dominatur frigiditas, tunc magis frigidat alias partes, contra naturalem complexionem
earum—et sic corruptio causatur in animali quando partes principales nimis alterantur
secundum qualitates contrarias.
49 Reportatio Parisiensis IV-A d.44, q.1, n.2 (Wadding XI, p. 854), carnibus et ossibus.
144 Chapter 8
for him the fundamental building blocks of pluriformed material substances.

This role belongs instead to the homogeneous mixtures, those mixtures com-
posed of prime matter and exactly one substantial form.50
50 Of course, homogeneous mixtures can be divided into smaller portions, each of which is
of the same kind as the whole mixture and of the same kind as every other portion. But
Scotus does not seem to think that such portions are actual individual substances while
they compose a homogeneous mixture. Any region of a homogeneous mixture is merely
a potential homogeneous substance. For preliminary studies of Scotus’s metaphysics of
continua, see Chapter 8 of Richard Cross, The Physics of Duns Scotus, pp. 139–158, and Neil
Lewis, “Space and Tme,” in The Cambridge Companion to Duns Scotus, pp. 69–99.
Chapter 9
Why the World is Not a Substance
That the world is in some sense a whole is undeniable. It is therefore undeni-

able that the world is in some sense unified. Since the world is evidently com-
plex, and undeniably unified, it makes sense to speak of the things in the world
as parts of the world, and therefore as composing the world. But as we have
seen throughout this book, Scotus recognizes several kinds of composition,
and the kind of composition that the things in the world make when they
make up the world is, Scotus claims, a unity of order, which is the sort of unity
things have when they are essentially ordered. Scotus thinks that everything in
the world is essentially ordered to a first efficient and final cause, God, and
through this common cause also essentially ordered to everything else in the
world. In fact, Scotus thinks that being a world simply consists in the totality of
things being essentially ordered.1 The essential order of all things to each other
and to God is precisely the difference between these things composing a world
and these things just being lots and lots of things. As a consequence of this
understanding of the world’s unity, where there are completely distinct unities
of order, ipso facto there are distinct worlds. If there were two first causes, each
with their own effects, then there would be two worlds, “because these beings
and those will not be ordered to each other nor to the same thing.”2
1 De Primo Principio III.49 (Wolter, pp. 64–65), Sine unitate ordinis non est unitas universi. One
consequence of Scotus’s view that the unity of the world is a unity of order, a consequence
not discussed in this chapter except in this footnote, is that since a whole unity of order is
nothing over and above its parts (the terms of the essential order and the relations these
terms have toward each other) the whole world is identical with its parts, including its rela-
tive parts—unlike a substance, which Scotus emphatically insists is a thing over and above
its essential and integral parts.
2 De Primo Principio III.49 (Wolter, p. 65), [S]i sit aliud primum et aliorum, erit illorum aliud
universum, quia entia illa et ista nec ordinabuntur inter se nec ad idem. Sine unitate ordinis
non est unitas universi. Scotus recognizes the possibility that there might be more than one
such unity of order, and that any additional unities of order would constitute different uni-
verses from our own. Any such additional universe, if it contained more than one thing,
would be composed of things that are essentially ordered to each other but not essentially
ordered to anything in our universe. Given Scotus’s understanding of what makes things into
a universe of things, it is logically impossible for God to create more than one universe, since
everything God creates is essentially ordered to God and therefore constitutes one universe.
So for there to be two or more universes, there must be two or more beings necessary
of themselves, that is, two or more Gods. But by the principle of parsimony we should not

146 Chapter 9
The concept of unity of order has already made some appearances in this
book. Recall from Chapter 6 that Scotus describes the sort of unity that a com-
plex subject has when it is in potency to a substantial form as a unity of order;
for example, in embryological development organic parts are generated prior
to ensoulment and are in potency to soul only when they are sufficiently devel-
oped and together compose a unity of order. This sort of unity simply repre-
sents on a local scale exactly the sort of unity that obtains on a cosmic scale.
Unity of order also made an appearance in Chapter 8, in which I examined
whether Scotus’s denial that elemental substances—the four chemical
elements: earth, water, air, and fire—actually compose mixed substances is
consistent with his affirmation that in general and in the particular case of
organic substances, substances can and do actually compose more complex
substances. The conclusion was that he is consistent. There are principled rea-
sons why he denies for elemental substances what he affirms for other sorts of
substances, one of which is that the four chemical elements are not capable of
composing a unity of order and therefore are not capable of co-instantiating
one of the necessary conditions for being part-substances.
If the last chapter was about chemistry, this one is about cosmology. The fact
that Scotus countenances part-substances motivates an investigation not only
into the reasons why he denies that elemental substances are actual part-
substances, but also into the reasons why he denies that the world as a whole
is a single substance composed of all other substances. If a “medium-sized dry
good” like an organism can be and is composed of substances, of what
substance(s) might an organism itself be a part-substance? Could it be that a
substance like Mr. Mole is not after all a substance that is composed of sub-
stances but composes no other substances—a natural terminus of substantial
composition—but is himself just a part-substance of some yet more complex
substance? And, could it be that the composition of part-substances continues
all the way up, to the composition of one world-substance?
In some sense of could, the answer to both questions is yes, according to
Scotus. But, according to Scotus, Mole is not in fact a part-substance of any-
thing else and the world is not in fact a single substance. Figuring out just why
is tricky. At first glance, if an organic body is in potency to a substantial form or
soul in virtue of its parts exhibiting a unity of order, why not suppose that the
world itself is in potency to a substantial form or soul? Why not be, to coin a
phrase, a hylomorphic monist [HM]?
needlessly posit entities, including Gods and universes. So we should not hold that there
are any others but this universe and its God. De Primo Principio III.26 (Wolter, pp. 57–59).
Why The World Is Not A Substance 147
[HM] The world is one substance, composed of matter and form. Its
form is the forma mundi and its matter is everything in the world
except the forma mundi.
That [HM] is at least possible is supported by Scotus’s contention that many

substances can and do have part-substances, parts that are themselves sub-
stances. Unsurprisingly, though, Scotus rejects [HM]. From the fact that both
the world and an organic body such as Mole’s fetus are unities of order we can-
not infer that the world is in potency to a substantial form or soul, since there
could be, as Scotus thinks there are, other reasons why Mole’s fetus is in
potency to a substantial form, which do not apply to the unity of order that is
the world itself.3
The examination of these reasons yields a counter-intuitive conclusion:
there is no significant metaphysical difference between the sort of substance
that naturally composes a substance, such as a heart, and the sort of substance
that does not, such as Mole. Or, to be more precise, there is no difference
between these sorts of substances that is relevant to questions about composi-
tion.4 This chapter argues for this conclusion and also offers Scotus’s reasons
for denying that the world as a whole is a substance, and specifically for deny-
ing [HM]. The plan is, first, to motivate interest in the admittedly abstruse
question of whether the world is a single substance by saying some things
about what is at stake in answers to the question. Then I will present three
arguments for denying that the world is a substance. The second and third of
these are Scotus’s own arguments and the first is inspired by him.
3 Scotus’s opposition to the idea that the world as a whole is a hylomorphic compound or any
kind of substantial unity runs against a venerable tradition in philosophy. Timaeus teaches
that the world as a whole is a living thing (30b), as does Philebus (30a-d), and A.E. Taylor
traces this idea back to Anaximines in A Commentary on Plato’s Timaeus (Oxford: Oxford
University Press, 1962), p. 81. Some Stoics conceived of God as something like a form or soul
of the whole world; see texts and discussion in The Hellenistic Philosophers, Volume 1:
Translations of the Principal Sources, with Philosophical Commentary, ed. A.A. Long and
D.N. Sedley (Cambridge: Cambridge University Press, 1987), pp. 274−279; and F.H. Sandbach,
The Stoics, 2nd ed. (Indianapolis, in: Hackett, 1989), pp. 69−75. As late as the twelfth century
Honorius Augustodunensis likened the world to an egg and its elements to the parts of an
egg. See De Imagine Mundi, Liber I, c.1 (PL 172, col.0121A).
4 The one significant difference between such substances is that those that are not naturally
apt to be part-substances belong to some determinate Aristotelian natural kind. But this fact
is not relevant to the question about whether these substances can compose another sub-
stance, as we saw in Chapter 7.
148 Chapter 9
i Motivating Monism
Scotus thinks that there are part-substances, substances that are themselves
parts of other substances, but it is clear that he thinks that many substances
are not naturally parts of other substances. Mole, for example, is not naturally
a part of any other substance, even though he is a part of many other kinds of
things: the Riverbank community, the story called The Wind in the Willows, an
ecosystem, the mereological sum of Mole and all the stoats, and so on. It is also
clear that he thinks that many things which are naturally part-substances, such
as organs, are not necessarily part-substances. (A kidney in an organ bank is
not a part-substance while it is in the organ bank, but it is a substance none-
theless.) So Scotus thinks that some substances naturally are parts while some
naturally are not.
Scotus does not discuss, as far as I have been able to determine, what it is
about some substances that makes them unapt for being part-substances.
However, he indirectly discusses what it is about the non-substantial wholes
that some substances do in fact compose such that these non-substantial
wholes are not substances. In an argument discussed below, Scotus argues that
the world is not a substance because it has no special operation that cannot be
reduced to the operation(s) of one or more of its parts.5 The unity of the world
is merely a unity of order and not a substantial unity. On the assumption, then,
that having some non-reducible operation of a whole is both necessary and
sufficient for that whole’s being a substance, we can say that, of the wholes of
which substances like Mole or Socrates are in fact parts, such as the world, an
ecosystem, a city, and so on, these are not substances, and Mole or Socrates
are therefore not part-substances, because these wholes do not have some
one operation that cannot be reduced to the operation(s) of one or more of
their parts.6
Still, what we would like is something stronger, something that tells us not
just why the non-substantial wholes that substances like Mole or Socrates
make up are not substances, but also why substances like Mole or Socrates are
not apt to make up other substances. Could there be, even if there is not in fact,
a substance which had Socrates as a part, analogous to the way in which
Socrates’ heart is a part-substance of Socrates?
I think that Scotus’s answer should be yes, for reasons having to do with his
metaphysics of the incarnation and his theory of essential orders. I think that
5 Quaestiones super Libros Metaphysicorum Aristotelis VII, q.20, n.51 (Bonaventure IV, p. 393).
6 In Ordinatio IV, d.24, q.un (Wadding IX, p. 515), Scotus says that, like the unity of the world,
the unity of the Church and the unity of the State are unities of order.
Scotus thinks that there is no meaningful distinction between these two sorts
of substance. To show this, it will be helpful first to describe what I take to be a
natural intuition or commonsense view about the differences between natural
part-substances like organs and natural non-part-substances like organisms, a
view that Scotus will ultimately deny.
The special mark of a part-substance, we might suppose, is that it has no
essential activity that is not for the sake of some other substance. The para-
digm here is an organ of an organism. To say what an organ is for, or even to say
what it does, is to specify its role in the life of something else—its organism.
The eyes are for seeing, the ears are for hearing, the liver is for filtering blood,
and the heart is for pumping blood; but eyes don’t see nor do the ears hear, and
the heart pumps and the liver filters not their own blood but something else’s.
These things, we might say, are teleologically instrumental.
Contrast teleologically instrumental substances with what we can call teleo
logically autonomous substances. Complete organisms, for example, do have
some essential activity that is not for the sake of some other substance. A man
might be a part of many things—a family, a state, an orchestra—and some if
not all of what it means to be a part of such things is to do certain things for
their sakes: he supports his spouse, raises his children, pays taxes, votes, prac-
tices his instrument, performs with his colleagues, etc. But by referencing the
man’s actions to other things we don’t (and don’t intend to) mean that these
things are what the man is for, as we do mean that the heart is for pumping
blood when we say that it pumps blood. We might say that qua member of an
orchestra a man’s activities are ordered to the end for which the orchestra
exists; even his eating and sleeping, again qua member of an orchestra, are
ordered to the orchestra, since these activities are necessary for the man to
keep on playing. But we would not say that the man simpliciter plays for the
orchestra, or eats and sleeps so he can keep on playing. His playing for the
orchestra is one slice of a wide range of ends that his playing an instrument
serves. For example, he might play for camaraderie, for health, for pleasure, or
for money, and none of these things involves activity that is ordered to the
good of something other than the man himself. And in addition to his playing
he does many other things, some of which are ordered to things other than the
man himself but not to his orchestra, and others of which are ordered to
himself.
So one way to cash out the difference between substances that naturally are
part-substances and substances that naturally are not, is to say that the former
are teleologically instrumental and the latter are teleologically autonomous.
But Scotus would resist this distinction. He would deny that being teleologi-
cally instrumental is the special mark of part-substances because he holds that
150 Chapter 9
everything and every activity is final-causally ordered to God, which is to say

that the ultimate final cause of everything and every activity is God.7 Part of
what this entails is that apparently teleologically autonomous activity (such as
playing an instrument for pleasure) is itself ordered to some end that ulti-
mately extends beyond the musician. Or, put more generally, there is nothing
in the world of created nature and no activity in the world of created nature,
that is truly for its own sake.8
To the extent, therefore, that being essentially ordered is a mark of being the
sort of unity that can be a substance, the whole world passes the test, and there
is no relevant distinction between the final-causal relations of organs to an
organism, on the one hand, and the final-causal relations of Socrates and Mole
to other kinds of wholes, including the whole universe, on the other.
Another cadre of reasons also clears the ground for the possible substance-
hood of the whole world. Scotus’s metaphysics of the incarnation includes the
claims (1) that in the hypostatic union with the Son of God a concrete individ-
ual human nature is a part of the God-Man, Jesus Christ;9 (2) that the Son of
God could have been hypostatically united with any created nature;10 (3) that
the Son of God could have been hypostatically united to several created natures
simultaneously,11 and for that matter, (4) that he could have been hypostati-
cally united to the whole world.12 So any substance can be a part-substance,
7 De Primo Principio III.27–34 (Wolter, pp. 58–61).

8 Contra Aristotle, who did not anticipate Scotus’s Anselmian distinction between affectio
commodi and affectio iustitiae. An unregulated affectio commodi would lead us to be true
Aristotelians: we would think that our own happiness represents a genuine terminus of a
final causal chain in nature. Scotus urges us to see eudaimonia as itself subordinated to
higher ends.
9 Ordinatio III, d.11, q.3 (Vatican IX, pp. 373–378). Scotus’s official position is that the rela-
tion of Christ’s human nature to the Son of God should be conceived along the lines of an
accident’s dependence on its subject and not of the relation of part to whole. For analysis
see Marilyn McCord Adams, Christ and Horrors: The Coherence of Christology (Cambridge:
Cambride University Press, 2006), pp. 123–143; Richard Cross, The Metaphysics of the
Incarnation: Thomas Aquinas to Duns Scotus (Oxford: Oxford University Press, 2002).
But Scotus does imply that Christ, the God-Man, is a whole composed of the Son of God
and an individual human nature. For discussion, see Richard Cross, The Metaphysics of the
Incarnation, pp. 128–133.
10 Ordinatio III, d.1, p. 1, q.3, nn.126–127 (Vatican IX, pp. 57–58).
11 Ibid.
12 Ordinatio III, d.1, p. 1, q.1, nn.36–50 (Vatican IX, pp. 16–23). For the way that Scotus’s meta-
physics of the incarnation inspires innovations in the theory of substance, see Marilyn
McCord Adams, “What’s Metaphysically Special about Supposits? Some Medieval
and the world can be a part of one substance, even if it cannot be a

part-substance.
So is it merely incidental, just a contingent feature of the way the world was
set up, that substances like hearts and lungs are naturally parts of other
substances, while substances like Mole and Socrates are naturally not parts of
other substances? Scotus does not give us a final answer, but it seems to me
that his answer should be yes. There is nothing about simply being a substance,
of any kind, that precludes being able to be a part of a substance.
ii The Argument from the Distinguishing of Forms
The previous section argued that there is no metaphysically significant

way to characterize the difference between substances naturally apt to be part-
substances and substances naturally apt not to be. One consequence of
this conclusion is that there is no reason not to think that any material
substance, regardless of its natural conditions, could be a part-substance.
And since this consequence clears conceptual space for what I have called
hylomorphic monism [HM], the idea that the world as a whole is one great
hylomorphic compound having as its matter all the things in the world, in
this section and the following sections I would like to present Scotus’s rea-
sons for rejecting [HM].
There are really two distinct kinds of hylomorphic monism: unitarian hylo-
morphic monism [UHM] and pluralistic hylomorphic monism [PHM]. They are
distinguished in this way:
[UHM] There is just one substantial form in the world, which is the sub-
stantial form of the world, the forma mundi.
[PHM] There is one substantial form of the world, the forma mundi, and
many substantial forms of things in the world.13
Variations on Aristotelian Substance,” Aristotelian Society Supplemental Volume 79:1

(2005), pp. 15–52, along with Chapter 7, above.
13 We can distinguish these in another way, following Jonathan Schaffer’s helpful target/
count distinction. Both count by tokens but [UHM] targets (hylomorphic) substances and
[PHM] targets substances that are not parts of other substances. [UHM] says that there is
just one (hylomorphic) substance, whereas [PHM] says that there is just one substance
that is not a part of another substance. Jonathan Schaffer, “Monism,” The Stanford
Encyclopedia of Philosophy (Fall 2008 Edition), Edward N. Zalta (ed.), url = <http://plato
.stanford.edu/archives/fall2008/entries/monism/>.
152 Chapter 9
According to [UHM], all of the things in the world, Mole, the Eifel Tower, my
briefcase, the Rocky Mountains, and so on, are all parts of one substance and
are not themselves substances. And according to [PHM] all of these things
might turn out to be substances, and all the substances in the world are
together the matter of the whole world, informed by forma mundi. Scotus’s two
arguments against [HM] are specifically against [PHM], and I discuss these
arguments in the following sections. In this section I present an argument
against [UHM] which draws on some of the material of Chapter 5.
Recall, from Chapter 5, that Scotus rejects
[DP] One form virtually contains many perfections, and can give differ-
ent perfections to different parts of matter.
Scotus argued against [DP] by claiming that it entailed an impossibility, namely

that one and the same thing, in this case a form, can be the subject of repug-
nant perfections simultaneously. I criticized this argument on the grounds that
the sense in which a form is the subject of perfections, whatever it amounts to,
does not amount to a form being characterized by its perfections. A form, let’s
say, whose quantitative totality is a map or blueprint for making one part of
matter square and another part circular, is not itself square and circular.
Scotus might have argued against [DP] on the grounds that it is incompati-
ble with a hylomorphic analysis of change. In Chapter 5 I argued on Scotus’s
behalf that, if repugnant perfections can be virtually contained in one form,
then every perfection can be virtually contained in one form, a forma mundi,
but that on the supposition that there is just one form of the whole world,
we lose whatever motivation we have for being hylomorphists; therefore a
hylomorphist should reject the forma mundi. Here I would like to develop this
argument a bit more.
The Aristotelian argument from change, discussed in Chapter 1, distin-
guishes matter from form in order to explain how change occurs, both substan-
tial and accidental change. A subject is able to undergo change not on the basis
of what it is (form), but on the basis of what it can be (matter). So matter is the
potentiality of a subject to undergo change. Pale Socrates can become tan
because the pallor of Socrates is a thing really distinct from Socrates himself,
who can be the subject of pallor, or tanness, or several other colors.
A second way to motivate the distinction between form and matter, namely
by reflecting on the distinction between a thing and what a thing is made of, is
ultimately parasitic on the argument from change. We know that the statue
is not altogether the same thing as the clay it is made of, because we know that
the statue can be destroyed without the clay being destroyed. Even if the clay
and the statue began to exist at the same time (perhaps the ingredients of the
clay were mixed together in a statue mold) and ceased to exist at the same time
(perhaps the clay statue was incinerated in an extremely hot furnace), we
would still know that they were not identical because we know that they differ
in their modal properties: the clay could have gone on existing while the statue
ceased to exist, if something else had happened.14 A hylomorphic analysis of
the relationship between the clay and the statue yields that the clay is a part of
the statue (its matter), and that the statue itself is a composite of its matter and
a form, statue-form. The fact of change or the possibility of change are the data
meant to be explained by the hylomorphist’s recourse to a division between
form and matter.
It might be argued that there is a third motivation for hylomorphism,
namely through reflection on the source or cause of the unity of a complex
object like an organism. The thought would be that something besides mate-
rial parts is needed to explain the special kind of unity that some material
objects have. But ask, why? Why is something besides these needed to explain
this unity? If you really want to get an argument for hylomorphism from unity
off the ground, you need to supply additional motivation, and the relevant
kind of additional motivation is going to invoke something about change. The
organism, you might argue, cannot be identical with its material parts, because
organisms are just the sort of objects that survive the gain and loss of material
parts. In order to explain how this is possible we need some sort of unifying
force or principle, such that wherever that unifier is unifying some material
parts, you have the very same organism. But this argument for hylomorphism
hearkens back to the argument from change; the source or cause of the unity
of an object, for most complex, structured material objects, only becomes a
metaphysical problem when we start wondering how such an object can stay
unified through change.
Still, not every structured material object can survive the gain and loss of
parts; or, at least, for many objects it is not obvious whether they can survive
the gain and loss of parts. And yet, for objects that cannot, or for many of them,
it is meaningful to ask, “In virtue of what are they unified?” A ham sandwich, to
14 For a selective history of modern discussions of the philosophical issues at stake in

the statue’s relation to the clay, see Alan Gibbard, “Contingent Identity,” Journal of
Philosophical Logic 4 (1975), pp. 187–221; David Wiggins, Sameness and Substance Renewed
(Cambridge: Cambridge University Press, 2001); Michael Rea, “The Problem of Material
Constitution,” The Philosophical Review 104:4 (1995), pp. 525–552; Ryan Wasserman,
“Material Constitution,” Stanford Encyclopedia of Philosophy, 2009 (http://plato.stanford
.edu/entries/material-constitution/#2).
154 Chapter 9
borrow an example from Kit Fine, might not be the sort of thing that can sur-
vive the gain or loss of its parts—or at least its most important parts, bread and
ham.15 Suppose it is not. Yet we know that the cause of its unity, its being a ham
sandwich, is something in addition to its bread and ham. We know this because
we could have this bread and ham and not have the sandwich, by rearranging
bread and ham and thus destroying the sandwich, or by neglecting to arrange
the bread and ham and so never producing a sandwich. So when the bread and
ham do compose a sandwich there is something besides the bread and ham—
and hylomorphists traditionally call this something else a form.
The ham sandwich might, for some, motivate the idea that we ought to
invoke hylomorphism not only as an account of what unifies an object dia-
chronically, through time and change, but also as an account of what unifies an
object, such as a ham sandwich, synchronically, at a time. Of course, if hylo-
morphism is a good account of diachronic unity then it’s a good account of
synchronic unity, too, since if an organism is unified through time by its form
then it’s unified by its form at every time in the duration of time through which
it is unified. So in this sense we can invoke hylomorphism as an account of
synchronic unity. But would we ever be interested in an account of the unity of
a complex object at a time and not really be interested in the unity of such an
object through time, as a subject (and agent) of change? The ham sandwich
might interest us in such an account, because the sandwich, at a time, has a
certain structure that cannot be reduced to its bread and ham. So to account
for its being a sandwich right now we have to appeal (says the hylomorphist) to
form. But even here, we are led to postulate the existence of a sandwich form
because we know that a sandwich is the sort of thing the ingredients of which
preexist the sandwich and can survive the sandwich. Sandwiches, in other
words, begin to exist after their parts begin to exist and can cease to exist before
their parts cease to exist (by taking apart the sandwich). Were it not for these
facts about sandwiches, we would never come to know or ever have good rea-
son to postulate the existence of sandwich forms. So I maintain that all roads
to hylomorphism merge into the boulevard of the need to account for change
or the possibility of change, and I assert the following thesis, which I will call
Motivation [M]:
[M] Any version of hylomorphism that does not or cannot explain

change or the possibility of change, should be rejected as theoreti-
cally wasteful; it’s multiplying entities without necessity.
15 Kit Fine, “Things and Their Parts,” Midwest Studies in Philosophy 23 (1999), pp. 61–74.
My argument against [UHM] in brief is that, since a forma mundi cannot

explain change, according to [M] a hylomorphist should deny that there is a
forma mundi and hence deny that the world is a substance, or more precisely
deny [UHM]. And in general, if we show that a candidate version of hylomor-
phism does not or cannot explain change, then we show (not, strictly speaking,
that it’s false, but) that there is no good reason to accept it. If a version of hylo-
morphism entails either that there is no change in the world (either substan-
tial or accidental change), or that there is change but every change is something
other than a change of substantial forms in matter (hylomorphic substantial
change) or a change of accidental forms in a substance (hylomorphic acciden-
tal change), then that version of hylomorphism should be rejected. This is
what the following argument against [UHM] shows.
[U1] If [UHM], then (a) either the world does not have any part-
substances, or (b) it does have part-substances but these part-
substances are not composites of matter and their own substantial
forms.
For [U1b] I have in mind non-hylomorphic theories of substance, such as a

Lockean “bare particular” or Humean “bundle” view of substance.16 [U1a] is the
view that all the things in the world are either (non-substantial) parts of,
or properties of, the world. First I will argue against the second disjunct of
[U1], [U1b].
[U2] If [U1b], then (a) it is not the case that a true account of substan-
tial change requires a division between substantial form and mat-
ter, or (b) there is no substantial change in the world.
[U2a] violates [M]. It violates [M] because it holds that hylomorphism is true
but that hylomorphism is irrelevant to a theory of substantial change. So,
16 It seems entirely possible to me that, of several theories of the metaphysical structure of

material substances on offer, T1…Tn, T1 might accurately describe some substances in the
world and not others, T2 might accurately describe substances T1 does not, and so on.
A hylomorphist has no reason to think that all substances are hylomorphic; only sub-
stances that undergo or can undergo change come into the hylomorphist’s consideration.
If there are things that (a) do not and cannot change and (b) for whatever reason deserve
the title “substance,” we would use or develop some non-hylomorphic theory to describe
their metaphysical structure.
156 Chapter 9
[U3] not-[U2a].
[U4] If [U2b], then (a) there is no change in the world, or (b) there is only
accidental (that is, non-substantial) change in the world.
[U4a] simply rejects the data that hylomorphism is designed to explain, so
[U5] not-[U4a].
[U6] If [U4b], then (a) At least some accidental change in the world is hylo-
morphic (i.e., the advent of a new accidental form in the world and the
recession of a prior accidental form from the world), or (b) none is.
[U6b] violates [M] because it holds that hylomorphism is true but that
hylomorphism is irrelevant to a theory of accidental change. So,
[U7] not-[U6b].
This leaves [U6a]. It asserts that some accidental change in the world is
hylomorphic. Is there good reason to think that according to [UHM] at least
some accidental change in the world is hylomorphic? I do not think so. [UHM]
is itself motivated by [DP], which says that it is possible that repugnant perfec-
tions can be virtually contained in one and the same form. If repugnant perfec-
tions can be virtually contained in one form, then every perfection can be
virtually contained in one form, a forma mundi, and this is exactly what [UHM]
claims. But [DP] would apply as well to accidental forms as substantial forms.
Given the theoretical innovations that led to positing a forma mundi, then, we
can equally well posit a forma accidentalis mundi; so a forma accidentalis mundi
makes a hylomorphic account of accidental change nugatory as much as a
forma mundi makes a hylomorphic account of substantial change nugatory.
So just as [DP] inspires [UHM], it equally inspires
[U8] There is just one accidental form in the world, a forma accidentalis
mundi.
But
[U9] If [U8], then no accidental change in the world is hylomorphic, i.e.,

no accidental change involves the advent of a new accidental form in
the world and the recession of a prior accidental form in the world.
[U9] holds because, given [U8], if there is any accidental change in the world,
this change does not involve the advent and recession of accidental forms
because there is only one accidental form. Therefore,
[U10] No accidental change in the world is hylomorphic, i.e., no acci-

dental change involves the advent of a new accidental form in
the world and the recession of a prior accidental form in the
world.
[U11] not-[U6a]. ([U10], contradiction).
[U12] not-[U4b]. ([U6], [U7], [U11], and modus tollens).
Now I turn to [U1a], which says that, given [UHM], the world does not have
any part-substances. It is the view that all the things in the world are either
(non-substantial) parts of, or properties of, the world. Against [U1a] we can
simply reprise parts of the argument against [U1b].
[U15] If [U1a], then (a) there is accidental (i.e., non-substantial) change

in the world, or (b) there is not.
[U16] not-[U15b] ([M]).
As [U6] says about [U4b] so we can say about [U15a]:
[U17] If [U15a], then [U6a] or [U6b].
[U18] not-[U15a]. ([U7], [U11], [U17], and modus tollens).
[U19] not-[U1a]. ([U15], [U16], [U18], and modus tollens).
[U20] not-[UHM]. ([U14], [U19], [U1], and modus tollens).
The argument shows pretty convincingly that [UHM] is hopeless. [UHM] is

nominally hylomorphic in the sense that it posits a form/matter split in the
world as a whole, but its hylomorphism does no theoretical work and should
158 Chapter 9
be rejected. But the argument says nothing about the plausibility of [PHM],
and it is [PHM] rather than [UHM] that was the target of Scotus’ own arguments
against [HM]. To his two arguments I now turn.
iii The World/Organism Analogy
Scotus explicitly questions whether the world is a substance in QMet VII,

q.20, n.22, in the form of an objection to his own view that it is not. According
to the objection, if several substances can be part-substances of one sub-
stance, then “how will it be disproved that there is one form of the whole
universe […] or one form for any number of disparate things?”17 Scotus’s plu-
ralism involves the claim that some collection of substances which is not
itself a substance can be in potency to a substantial form. But intuitively not
just any collection of substances which is not itself a substance is in potency
to substantial form—my briefcase and the Eiffel tower, for example, do not
seem apt to be part-substances of the same substance. Scotus’s claim that
several substances are in potency to a substantial form when they compose
a unity of order seemed like the sort of move that would allow one to distin-
guish the substances that can be part-substances from those that cannot. But
if everything in the world is essentially ordered to everything else, then this
move is obviously insufficient. If a unity of order is all that it takes for several
substances to be able to be part-substances of the same substance, then any
arbitrary collection of substances is able to be a collection of part-substances
of the same substance, and indeed everything in the world is able to be a
part-substance of the world-substance.
Scotus offers two attempts to avoid the conclusion that there is one world-
substance. First, he says,
To that objection about the soul of the world, it is argued that it is not one
form, because then the universe would be imperfect wherever an indi-
vidual is corrupted.18
p. 384), Confirmatur prima propositio: quomodo enim improbabitur una forma totius
universi […] vel quorumcumque disparatorum?
p. 393), Ad illud de anima mundi arguitur quod non est una forma, quia tunc universum
esset imperfectum, quocumque individuo corrupto.
The idea here is that, on the assumption that the universe as a whole is one
substance, each individual in the universe—e.g., Mole, Rat, Badger, etc.—
would be analogous to an organ of organism, as displayed in Table 1:
Table 1 Word/organism analogy, first version
Part Whole
Heart, eyes, hands, liver, etc. The animal

Individual substances The world
Remove an eye or hand and the organism is injured or impaired; remove a heart
or brain and the organism is corrupted. If the analogy holds then the universe
as a whole is injured or impaired when an individual such as Mole corrupted,
just as Mole is injured or impaired when he loses an eye or hand. But (with all
due respect to Mr. Mole), the universe is not thus impaired or injured when
Mole is corrupted, so the universe is not really an organism. In short,
[W1] That the universe is one substance implies that it is injured when
an individual substance is corrupted.
[W2] But the universe is not injured when an individual substance is

corrupted.
[W3] So the universe is not one substance.
But Scotus finds an objection to his own argument, the basic claim of which is
that the organ/organism analogy was not apt. He writes:
Against this [[W1]-[W3]]: is an animal really imperfect wherever a part of

its flesh is removed? Is not an individual in the universe held to be like
homoœmeric parts and a species like non-homoœmeric?19
The objection expressed in the first rhetorical question is that an animal is not
necessarily made imperfect if it loses a part. For example, it just doesn’t sound
p. 393), Contra: numquid animal imperfectum quacumque parte carnis amota? Nonne indi-
vidua in universo ponerentur quasi partes homoœmereae, et species quasi anomoœmereae?
160 Chapter 9
right to say that I become less perfect if I lose some skin cells (indulging
the anachronism), clip my fingernails, or trim my hair. The second rhetorical
question advances the claim that, if the whole universe is a substance, then its
analogue of organs or parts the removal of which would make the body less
perfect are not individuals but whole species. Aristotle himself said that com-
plex organs like hearts, eyes, and hands are in some sense composed out of
several different kinds of homoœmeric parts, such as flesh, bone and nerve.20
Scotus’s idea is that individuals would be like these homoœmeric parts, some-
how composing whole species. These relationships are displayed in Table 2:
Table 2 World/organism analogy, second version
Homoœmeric parts Heterogeneous parts The whole
Bone, flesh, blood, nerve etc. Heart, eyes, hands, liver, etc. The animal
Individual substances Species The world
It would thus take the corruption of a whole species to impair the universe as a
whole.
From Aristotle’s point of view, this last claim would entail that the universe
could never be impaired, since he held that it was necessary that the species
are eternal.21 He could make the following argument:
[W4] If the universe is one substance, then it is possible that it be

injured.
[W5] It is possible that the universe be injured if and only if it is pos-

sible that a species be corrupted.
[W6] It is not possible that a species be corrupted.
[W7] So it is not possible that the universe be injured.
[W8] So the universe is not one substance.
20 Aristotle, Parts of Animals II, 1 646b 11–27 (Barnes I, p. 1006).

21 Aristotle, Physics III.6, 206a25-27 (Barnes I, p. 351).
Scotus would not quite agree with Aristotle about the necessary eternity of the
species, since he doubtless would hold that God could simply annihilate every
individual of a species. But since the formal contents or notae of any creatable
nature are repugnant or non-repugnant of themselves and not by divine voli-
tion, God could not make it the case that the non-repugnant notae composing
any given nature become repugnant,22 and thus he could not altogether eradi-
cate a species at least in the sense that he could not eradicate his own idea of
the species. So for reasons other than Aristotle’s Scotus would accept the argu-
ment [W4]-[W8].
iv Properties of the Whole That Are Not Properties of the Parts
The second argument sets forth a kind of procedure for picking out the sub-
stances, and argues that according to this procedure the world does not make
the cut:
Otherwise it is argued that operation discloses form. Therefore where

beyond the proper operations corresponding to the parts according to
their proper forms, we see some one operation common to them—such
as sensing in an animal—there we conclude that there is a form of the
whole actuating all the parts together. In the parts of the universe, beyond
the proper operations belonging to them according to their proper forms,
we do not see another common [operation], as in fire and water, beyond
heating and cooling. Hence, etc.23
Consider my briefcase and the Eiffel Tower. Maybe they are informed or can be
informed by one substantial form and so compose or can compose a substance;
maybe they are not and so do not. To find out, Scotus would seek to discover
whether there is some operation that my briefcase and the Eiffel Tower have,
22 On this see Calvin Normore, “Scotus, Modality, Instants of Nature and the Contingency
of the Present,” in John Duns Scotus: Metaphysics and Ethics, ed. Ludger Honnefelder, Rega
Wood, and Mechthild Dreyer (Leiden: E.J. Brill, 1996), pp. 161–175, especially pp. 162–164.
Scotus, Ordinatio I, d.36.
p. 393), Aliter arguitur: quod operatio arguit formam. Igitur ubi ultra proprias operationes
correspondentes partibus secundum proprias formas videmus aliquam unam operatio-
nem communem eis—sicut in animali “sentire”—ibi concludimus formam totius actu-
antem omnes partes communiter. In partibus universi, ultra proprias operationes
convenientes eis secundum proprias formas, nullam videmus aliam communem, ut in
igne et aqua, ultra calefacere et frigescere; quare etc.
162 Chapter 9
that is not an operation of the briefcase, the Eiffel Tower, or the two together,
an operation attributable only to the whole substance of which they are (puta-
tive) material parts. But we do not see (to use Scotus’s word) any such opera-
tion, and therefore do not infer that they compose a substance.
In another text Scotus offer risibility as an example of a property that cannot
be reduced to a part or parts of a substance.24 The ability to laugh requires both
the ability to find things humorous—which is limited to rational beings—as
well as the ability to produce the motions and sounds associated with
laughter—which is limited to corporeal beings. Thus only a human—a ratio-
nal animal—is risible. Neither matter nor form is a rational animal, so neither
can be the proper subject of risibility. Scotus also denies that “both together”
can be the proper subject of risibility. “Both together” is ambiguous, however;
it might mean that risibility inheres in matter and in form, such that matter is
risible and form is risible, or it might mean that the feature which is in “both
together” is divided between the two, for example, the rational “part” of risibil-
ity inheres in the soul while the bodily “part” inheres in the matter. “Socrates is
risible” would turn out to be an abbreviation of a complex attribution such as,
“The soul of Socrates can find things funny and the body of Socrates can pro-
duce motions associated with laughter, and these can occur at the same time.”
Someone who was committed to Ockham’s claim that a substance is identical
with its essential parts (such that any property of a substance is ipso facto a
property of one or more parts of a substance), and also to the claim that risibil-
ity is an essentially rational ability, would be forced to offer an analysis of risi-
bility along these lines. Ockham himself seems to deny that risibility is a
rational ability, claiming that it pertains [conveniat] only to the body.25 This
seems false, and does not do justice to the traditional idea that risibility is a
proper accident of the human species, since on Ockham’s analysis anything
bodily should turn out to be risible.
There is a deeper problem lurking in the way in which Ockham states his
opposition to Scotus’s view, however. Ockham says that properties like risibil-
ity pertain to the body as opposed to the soul, but body itself is, for both Ockham
and Scotus, a composite of matter and form. Therefore, if Ockham really means
that risibility pertains to the body, then he is committing himself unwittingly
to Scotus’s position, that a property like risibility is a property of a composite
substance and not of the parts of the substance. If by body he really means
24 Reportatio III-A, d.2, q.1, n.5 (Wadding XI, p. 428), Item, aliquod ens habet per se passio-
nem, et primo, ut ponitur de homine respectu risibilis, et tamen ista passio nec est mate-
riae primo, nec formae, nec amborum.
25 Ockham, Quaestiones Variae VI, a.2, ll.234–237 (OTh VIII, p. 217).
matter, then his position is obviously false, for it would imply that something
could be risible without being either rational or a body.
Ockham gives another and even odder example of a property that pertains
only to the body and not to the composite of body and soul: he says that
descending [descendere] pertains only to the body.26 Again, if Ockham really
means that a body descends, then he is committed to Scotus’s view. But if he
really means that matter descends then he is wrong, for matter does not do
anything unless it be informed.
Ockham also seems to be unable to respond to the following worry. If a sub-
stance, s, has some proper accident, passion, or operation, F, if and only if one
or more of its essential parts have F, then we can ask whether the parts were F
before they composed s. If they were, then F is not really a proper accident,
passion, or operation of s. If they were not, then Ockham should provide some
account of what changes when the parts compose s such that one or more of
them becomes F. If, for example, matter becomes risible when and only when
it is informed by the substantial forms that compose a human, then what is it
about receiving those forms that makes the matter risible? If it is because mat-
ter is now a part of a rational animal, then why go on attributing the risibility
only to the matter and not to the rational animal? Ockham did not address
these concerns, and I am not sure how to address them for him.
In contrast to substances like human beings, which have operations like
sensing and proper accidents like risibility, the common operations and acci-
dents of two or more substances that do not form a substantial unity can be
reduced, Scotus thinks, to the operations or accidents of one or more sub-
stances. Consider Scotus’s own example of substances that do not compose a
substance, some fire and some water. If these had a common operation, it
would probably be warming. So why not think that fire and water compose a
substance when they warm? The answer is supposed to be that warming is
simply reducible to the respective operations of fire and water, the way that the
warm water from your tap is simply the confluence of the cold water and the
hot. Finally, consider the world itself. Suppose (in submission to the Parisian
Condemnations of 1277) that the world as a whole can move rectilinearly.27
This would not support the view that the world is a substance, since the world’s
rectilinear motion could be accounted for by the rectilinear motion of all the
parts. The point is that Scotus finds no reason for supposing that the world has
some operation that cannot be reduced to one or more of its parts.
26 Ibid.
27 Cf. Edward Grant, A Sourcebook of Medieval Science (Cambridge, ma: Harvard University
Press, 1974), p. 48.
164 Chapter 9
It could turn out that there are no substances composed of other substances
if, for example, there is nothing whose activity cannot be reduced to its
part-substances. In such a world there could still be substances, but there
would be no substance which has substances as parts. Aquinas, for example,
should not be troubled by a forma mundi, or at least not by [PHM], since
Aquinas denies that one substance can have more than one substantial form
(and therefore that one substance can be composed of substances).
Scotus’s response to the worry about the forma mundi shows that compos-
ing a unity of order is merely a necessary and not a sufficient condition for
being in potency to a substantial form. If we ask why this particular unity of
order is in potency to a substantial form the answer would be that this unity of
order is itself essentially ordered in the order of final causality to a whole sub-
stance of a certain kind. The whole substance is the final cause of the final
causality of all the part-substances, and the union of substantial form with
these part-substances is one step in a final causal chain. When a unity of order
is not in potency to substantial form, the causality of each of its causes is not
ordered to a substantial union.
Chapter 10
Scotistic Hylomorphism and the Problem

of Homonymy
In this final chapter I will argue for a thesis which I believe to be Scotus’s but
which I also believe to be true. I will argue that a hylomorphist should hold
that a material substance can be composed of material substances, that he
should be what I call a Scotistic pluralist. The plan for the chapter is as follows:
I will review some of the features that make Scotistic hylomorphism Scotistic,
and contrast these with certain features of different versions of hylomorphism.
I will move on to pose a serious problem with hylomorphism, first raised, as far
as I know, by J.L. Ackrill.1 Then I will sketch the ways in which different versions
of hylomorphism can respond to Ackrill’s problem, and argue that Scotistic
hylomorphism has the best response.
i Scotistic Hylomorphism Among Other Varieties
Take it for granted, that for any material substance, there is something or
some things that it is made out of. Let us say, following an ordinary way of
speaking (at least in metaphysics), that what a substance is made out of con
stitutes that substance, and (following an older way of speaking) let’s call
that which constitutes a substance the matter of that substance. All hylo-
morphists agree that a substance is not identical with that which constitutes
it, is not identical with its matter. They argue that the matter of any material
substance is formed, structured, organized, in some way and that an expla-
nation of the form, structure, or organization of the matter of a substance
must involve some ontological commitment to something in addition to the
matter. Hylomorphists call this additional something, form. A material sub-
stance turns out to be, on nearly every hylomorphic theory, some sort of
composite of matter and form.2
1 J.L. Ackrill, “Aristotle’s Definitions of ‘Psuche’,” Proceedings of the Aristotelian Society 73 (1972–
1973), pp. 119–133. Ackrill’s problem is concisely expressed by Christopher Shields, “The
Fundamental Problem about Hylomorphism,” Stanford Encyclopedia of Philosophy 2010,
http://plato.stanford.edu/entries/aristotle-psychology/suppl1.html.
2 I know of one exception: Mark Johnston, “Hylomorphism,” Journal of Philosophy 103:12 (2006,
pp. 692–698. Johnston denies that forms are parts of the objects whose forms they are.

166 Chapter 10
On older versions of hylomorphism, such as Scotus’s, a division of forms

into substantial and accidental is put to heavy use. A substantial form is sup-
posed to be the sort of form that structures matter and makes a substance,
while an accidental form structures a substance in a certain way, characteriz-
ing it but not changing its essence or nature. For the purposes of this chapter I
am concerned exclusively with substantial form.
Perhaps the most significant disagreement between hylomorphists is the
one I will be focusing on here. It concerns whether or not the matter of a sub-
stance, s, can itself be one or more substances while it is the matter of s, which
by transitivity of parthood (on the assumption that matter is a part of the sub-
stance whose matter it is) amounts to a disagreement about whether or not a
substance can have one or more substances as parts.
Aquinas held that for every substance you count you must count no more or
less than one substantial form. (Following my earlier usage, I will call him and
others who agree with him unitarians about substantial form.) Unitarians
therefore deny that a substance can be composed of substances. Scotus and
Ockham held that one substance can have more than one substantial form.
(We will continue to call them pluralists about substantial form.) Pluralists
therefore affirm that a substance can be composed of substances. But there are
two very different kinds of pluralism, what I have called (a little unimagina-
tively) standard pluralism and Scotistic pluralism. In order to understand the
differences between these kinds of pluralisms, recall another term of art from
earlier chapters: a part-substance is a substance that is a part of a substance.
Pluralists think there are part-substances; unitarians do not. For both standard
and Scotistic pluralists, a whole substance can be composed of several part-
substances, and one part-substance can be composed of several part-sub-
stances. But, to speak loosely, whereas the standard pluralist will deny that
integral parts like hands, eyes, and hearts can be individual substances while
they compose a substance, the Scotistic pluralist affirms this. More strictly, for
Standard Pluralists,
[Standard Pluralism] For any substance, w, part-substance of w, x, and

part-substance of w, y, if x ≠ y then either x is a part-substance of y or y is
a part-substance of x.
But for Scotistic pluralists,
[Scotistic Pluralism] For some substance, w, part-substance of w, x, and

part-substance of w, y, such that x ≠ y, x is not a part-substance of y and y
is not a part-substance of x.
Scotistic Hylomorphism and the Problem of Homonymy 167
p3
p2
p1
Figure 1 Standard pluralism
p1 p2
p3 p4 p5 p6
Figure 2 Scotistic pluralism
Given part-substances, p1 … pn, and substance, w, standard and Scotistic

pluralism can be represented graphically in Figures 1 and 2, above.
In Figure 2, each node represents either a part-substance or a whole
substance, and the lines connecting the nodes represent part of relations, such
that for any two nodes connected by a line, the lower is a part of the higher
(but not vice versa). Thus, p3 is a part of p1, which itself is a part of w. But p1 and
p3 are not parts of p2, even though p2 is a part of w. In Figure 1, each circle rep-
resents either a part-substance or a whole substance, and being contained in a
larger circle represents the part of relation. p1, p2, and p3 therefore are parts of w,
p1 and p2 are parts of p3, and p1 is a part of p2.
On Scotistic pluralism, by way of example, suppose that a heart and a femur
are part-substances of a human substance. Suppose also that marrow is a part-
substance of a femur. Then, whereas marrow is a part-substance of a femur
and is therefore a part-substance of a human, it is not a part-substance of a
heart, even though a heart is a part-substance of a human. Standard pluralism
does not really deny this rather intuitive point; instead it restricts the exten-
sion of substance in such a way that while the body might count as a sub-
stance, its organic parts do not. Thus, if a body is a part-substance of a vegetative
168 Chapter 10
substance, and if a vegetative substance is a part-substance of a sensitive

substance, then a body is a part-substance of a sensitive substance. Notice
that while Scotistic pluralism does not rule out that part-substances can be
“layered,” as Figure 1 represents, Standard pluralism does rule out that part-
substances can be “branched,” as Figure 2 represents. Scotistic pluralism is
therefore a more inclusive theory.
Crucial to my argument is that any version of hylomorphism will be unitarian,
standard pluralist, or Scotistic pluralist. Obviously any version of hylomorphism
will be either unitarian or pluralist, since a theory that posits zero substantial
forms is not a theory of hylomorphism. For the claim that any theory of plural-
ism is either standard or Scotistic, consider the following. If we semi-formalize
the definitions of standard and Scotistic pluralism in the following ways,
[Standard Pluralism] ∀w∀x∀y ((Sw and xPw and yPw) implies ((x ≠ y)
implies (xPy or yPx))), where S can be read as “is a substance,” and P can
be read as “is a part-substance of;”
[Scotistic Pluralism] ∃w∃x∃y (Sw and xPw and yPw and x ≠ y and not-
xPy and not-yPx), where S can be read as “is a substance,” and P can be
read as “is a part-substance of;”
we can show that one is simply the negation of the other through a simple
argument, argument [N]:
[N1] ∀w∀x∀y ((Sw and xPw and yPw) then (x ≠ y then (xPy
or yPx))) and ∃w∃x∃y (Sw and xPw and yPw and x ≠ y and not-
xPy and not-yPx) (assume for reductio)
[N2] ∀w∀x∀y ((Sw and xPw and yPw) then (x ≠ y then (xPy or yPx)))
([N1], simplification)
[N3] not-∃w∃x∃y not-((Sw and xPw and yPw) then (x ≠ y then (xPy or
yPx))) ([N2], quantifier negation x3)
[N4] not-∃w∃x∃y ((Sw and xPw and yPw) and (x ≠ y and not-(xPy or
yPx))) ([N3], negated conditional decomposition x2)
[N5] not-∃w∃x∃y (Sw and xPw and yPw and x ≠ y and not-xPy and
not-yPx) ([N4], DeMorgan, removed parentheses)
[N6] ∃w∃x∃y (Sw and xPw and yPw and x ≠ y and not-xPy and not-
yPx) ([N1], simplification)
[N7] Therefore, not-[N1] ([N5], [N6], contradiction)
Scotistic and standard pluralism offer different accounts of how part-

substances are structured when a substance has more than one part-substance,
and the argument above shows that these accounts are mutually exclusive.
This does not mean that there cannot be more than three theories of hylomor-
phism—unitarianism, standard pluralism, Scotistic pluralism; it just means
that any version can be sorted under one of these.
ii Ackrill’s Problem
As a slow approach to Ackrill’s problem, consider a familiar heuristic case used

to motivate hylomorphism. Suppose that a cube of bronze is shaped into a
sphere. Is the cube of bronze identical with all of the bronze that constitutes
it? Is the sphere of bronze? If they are, then we seem committed to the identity
of the cube of bronze with the sphere of bronze, which seems false, since one
and not the other rolls well. On a hylomorphist analysis, the sphere and
the cube share the same matter, but they differ in form. They are two objects
sharing (at different times) the same matter.
Relevant in this case for setting up Ackrill’s problem are the facts that the
matter of the cube persists through the change of the cube into a sphere, and
(what immediately follows from this) that the identity of the bronze does not
depend on its being informed by the form of the cube or the form of the sphere.
Ackrill puts it this way:
It seems then that [the matter-form distinction depends] upon the

idea that something that is actually the case might not have been: this
stuff might not have been so arranged, the capacity being now displayed
might have remained undisplayed. ‘It is the nature of matter to be
capable both of being and of not being < such and such > (Met. Z.15
1039b29).’3
Generalized to describe any hylomorphic compound substance, s, we can say

that the matter of s preexists s and goes on existing after s ceases to be and
therefore the matter of s can exist even if s does not exist. Call this generaliza-
tion the Contingently Informed Principle [cip]:
3 Ackrill, “Aristotle’s Definitions of ‘Psuche’,” p. 125. Cf. Aristotle, Metaphysics VII 1039b29
(Barnes II, p. 1641).
170 Chapter 10
[CIP] For any hylomorphic compound substance, s, we can say that the
matter of s preexists s and goes on existing after s ceases to be and there-
fore the matter of s can exist even if s does not exist.
Recall, however, that for Aristotle and all medieval Aristotelians, a bronze cube
and a bronze sphere are not actually substances, but artifacts. Another way of
saying this is that the form of the cube and the form of the sphere are not sub-
stantial forms; instead they are accidental forms, modifying a substance, the
homogeneous mixture, bronze. So the example of the cube and the bronze is
at best merely heuristic.
But its usefulness even as a heuristic falls under suspicion when we attempt
to make the transition to genuine Aristotelian substances, especially living
things. For it is not clear what would be the analogue of bronze in the
metaphysical makeup of a living thing. The standard thing to say, and what
Aristotle himself does say in De Anima II, is that the soul is the substantial form
of the body, where it is implied that the body functions as the matter of the com-
posite. But for Aristotle this cannot be quite right, since according to him the
corpse is not a human body; the corpse is only homonymously a human body,
the severed hand is only homonymously a hand.4 But why think that a corpse is
not really a human body, or that a severed hand is not really a hand, or (naturally
extending the range of examples) that the removed kidney (waiting in the organ
bank to be implanted in a patient) is not really a kidney? The answer lies in a
certain picture of the relationship between being a certain kind of thing and
being able to perform a certain function, namely this: that something is a mem-
ber of a kind, K, if and only if it is able to perform the characteristic function or
functions of the Ks. Call this relationship the Homonymy Principle [hp]:
[HP] A thing, x, is a member of a kind, K, if and only if x is able to perform

the characteristic function or functions of the Ks.
According to [hp], since the severed hand cannot do what hands character-
istically do (they enable a human to manipulate its environment in certain
ways conducive to its flourishing)—it is only homonymously a hand. And
since the kidney in the organ bank does not filter blood (to narrow for sim-
plicity’s sake the range of functions that kidneys do), it is only homony-
mously a kidney.
4 Aristotle, On the Soul II, c.1, 412b 10–24 (Barnes I, p. 657); Meteorology IV, c.12, 389b 30 – 390a
1 (Barnes I, p. 624); Generation of Animals II, c.1, 734b 24–35 (Barnes I, p. 1140); Aquinas,
Summa theologiae III, q.50, a.5, corp.
Reflecting on the implications of [hp] for the hylomorphic analysis of sub-

stances, Ackrill concluded that the body cannot be the matter of a living thing
in any straightforward sense, since its identity is dependent on its being
informed by the soul. A better way to put this is to say that the body of a living
thing only exists when the living thing exists, and therefore fails to be the
matter of a living thing according to [cip].
The problem with Aristotle’s application of the matter-form distinction to

living things is that the body that is here the matter is itself ‘already’ neces-
sarily living. For the body is this head, these arms, etc. (or this flesh, these
bones, etc.), but there was no such things as this head before birth and
there will not be a head, properly speaking, after death. In short—and I am
of course only summarizing Aristotle—the material in this case is not
capable of existing except as the material of an animal, as matter so in
formed. The body we are told to pick out as the material ‘constituent’ of the
animal depends for its very identity on its being alive, in-formed by psuche.5
So here is Ackrill’s Problem with hylomorphism: Given [hp], it is necessary

that for any substance, s, the matter of s is informed by the form of s. But,
given [cip], it is not necessary that for any substance, s, the matter of s is
informed by the form of s. Aristotle’s hylomorphism seems to involve com-
mitment to both [cip] and [hp]. But they cannot both be true.
There are different kinds of hylomorphism, but all are alike, as far as I know,
in being committed to [cip]. In fact, it is impossible for me to think of a way in
which some theory can count as hylomorphism at all if it does not include
such a commitment. So the possible responses to Ackrill’s Problem involve
either rejecting [hp], or modifying [hp] and [cip] in such a way that their
conjunction does not after all generate a contradiction. Unitarians about sub-
stantial form—Aquinas their chief representative—opt for the latter route,
while pluralists about substantial form—Scotus among them—opt for the
former.
iii Aquinas’s Response
As discussed in Chapter 5, even though Aquinas sometimes says that the soul
or substantial form informs the body, for example in st I, q.76, a.1, consistent
5 Ackrill, “Aristotle’s Definitions of ‘Psuche’,” p. 126.

172 Chapter 10
with Aristotle’s language in De Anima II, it is clear that he does not really mean
this. Instead he thinks that the soul informs prime matter, an uncharacterized
and in itself purely potential substratum of change. All of the essential proper-
ties of a living thing, not just being sensitive or being alive but also being
sighted or being quadrupedal, are due to the living thing’s one substantial
form, its soul. It follows from this that the corpse of Mole is numerically dis-
tinct from the body of Mole, and is therefore only homonymously a talpid
body. Aquinas therefore holds on to [hp] while rejecting the understanding of
the body as the matter of a living thing.
Does Aquinas also hold on to [cip]? This I think is a hard question to answer.
On the one hand, Aquinas’s Aristotelian analysis of change, presented for
example in Principles of Nature 2, demands that prime matter persist through
substantial change:
Since in generation the matter or subject persists (permanet) but not the
privation or the composite from matter and privation, thus the matter
which is not considered with privation is persistent (permanens); but
that which is considered with privation passes away (transiens).6
The privation is simply that capacity or disposition of a subject to receive some

form; thus, before some bronze is shaped into a statue it has a privation of
statue-form. When the bronze is shaped into a statue it loses this privation,
but the bronze itself persists through the change. Aquinas is clear that both
proximate matter, such as bronze, as well as prime matter, persist through sub-
stantial change.7
So the short answer is that he does support [cip]. On the other hand, there
are reasons to question whether Aquinas can consistently hold [cip]. For
example, it is not clear to me how Aquinas is entitled to hold that prime matter
persists, given his well-known commitment to the idea that prime matter
is pure potency, having in itself no actuality. For Aquinas matter is actual
insofar as it is informed by some form; it is not in itself a hoc aliquid. It is not
obvious to me whether it follows from this that matter cannot persist through
substantial change; however neither is it obvious to me whether it is consistent
with this that matter can persist through substantial change. We might be
inclined to think that if matter depends for its actuality on some particular
6 Aquinas, De principiis naturae c.2, Et quia in generatione materia sive subiectum permanet,
privatio vero non, neque compositum ex materia et privatione, ideo materia quae non
importat privationem, est permanens: quae autem importat, est transiens.
7 Aquinas, De principiis naturae c.2.
form, f1, then when the composite of matter and f1 is corrupted and f1 ceases to
exist, the matter ceases to exist as well. Then again we might think that what
Aquinas is committed to is not that this particular matter depends for its actu-
ality on this particular form, but rather that it depends on some form or other,
such that so long as it is informed by some form or other, it remains actual and
the very same matter. Thus when f1 recedes and f2 is introduced in a process of
generation, perhaps the matter is passed off, as in a game of Hot Potato, from f1
to f2 in an instantaneous change, remaining actual across the generation. This
picture of prime matter’s relationship to form sees any particular form as a
merely contingent actualizer of any particular prime matter, such that one and
the same parcel of prime matter can be actualized successively by indefinitely
many forms, just as one and the same hot potato can be held aloft by indefi-
nitely many hands in succession.
I myself do not know how to decide between these competing views.
Fortunately a commitment one way or another is not necessary to my argu-
ment, since it is open to a unitarian simply to deny that prime matter depends
for its actuality on being informed. Suffice it to say that Aquinas himself was
committed to [cip], questions about the consistency of this view with other of
his views notwithstanding.
iv The Standard Pluralistic Response
Standard pluralists take issue with Aquinas’s support of [hp]; according to

them there are good reasons for rejecting it. I will focus here on just one argu-
ment; it is inspired by a very common medieval argument for pluralism.
Suppose that Mole is a substance and that Mole’s corpse is a substance. It’s
clear that Mole’s corpse is not the same substance as Mole. Either Mole’s corpse
is identical with a part of Mole (such as his body), or it is not. By the indiscern-
ibility of identicals, for any x and any y, if x is identical to y then, if Fx then Fy.
From this it would follow that if Mole’s corpse is identical with a part of Mole,
then Mole has at least one substance as a part, letting “F” represent “is a sub-
stance.” This follows given that Mole’s corpse is a substance and that Mole’s
corpse is not the same substance as Mole. Now the unitarian denies that Mole
has a substance as a part. So the unitarian takes the other disjunct: he holds
that it is not the case that Mole’s corpse is identical with a part of Mole.
But if the corpse is not identical with Mole or with a part of Mole, then it has
an efficient cause whose efficient causing of the corpse is distinct from the
efficient causing by which Mole comes to be. The corpse must be generated at
the moment Mole is corrupted. This assumption is rooted in the ideas that
174 Chapter 10
every created substance is efficiently caused, and that one efficient causing
produces at most one substance. But going through the list of causes, we find
no plausible candidates for this efficient cause of the corpse. God or an angel
or an invisible corpse-generator could do it, but the pluralist insists that there
is no reason to think that anything does do it. So by modus tollens the pluralist
denies that the corpse is not identical with Mole or with a part of Mole, and
therefore accepts the first disjunct, from which he derives that Mole has at
least one part-substance. In outline, the argument against unitarianism, argu-
ment [R], runs as follows:
[R1] Mole’s corpse is a substance.

[R2] Mole is a substance.
[R3] Mole’s corpse is not the same substance as Mole.
[R4] For any x and for any y, x = y implies that if x is a substance then
y is a substance. (indiscernibility of identicals)
[R5] Either [R5a] Mole’s corpse is identical with a part of Mole (such
as his body), or [R5b] it is not. ([R3], excluded middle)
[R6] If [R5a], then [R6a] Mole has at least one substance as a part—
his body. ([R1], [R4], [R5], universal elimination)
[R7] If [R5b], then [R7a] Mole’s corpse has an efficient cause that is
not the efficient cause of Mole. (assumption)
[R8] not-[R7a]. (“induction”)
[R9] not-[R5b]. ([R7], [R8], modus tollens)
[R10] Therefore [R5a]. ([R5], [R9], disjunctive syllogism)
[R11] Therefore [R6a]. ([R6], [R10], modus ponens)
[R12] Therefore [Pluralism] a substance has at least one substance as a
part. ([R2], [R11])
[R8] is disputable. It might seem weak because the unitarian might well be
willing to tolerate a metaphysics in which God or an angel has to generate
corpses constantly, if that is the price to pay for preserving unitarianism. So to
a certain extent the pluralist argument simply relies on an intuition that it is
more likely that there are part-substances than that there are non-apparent
efficient causes of corpses.
This intuition is strengthened, I think, by the consideration, discussed in

Chapter 5, that the non-apparent efficient cause would not only be responsible
for generating a new substance, but would also be responsible for generating a
substance that in many cases has specifically the same accidents as the living
substance that precedes it. The corpse of Mole, like Mole, is black, small, and
furry. According to Aquinas accidents do not survive the substances they
inhere in except in miraculous circumstances; if the substance is corrupted, so
are the accidents, since accidents by nature inhere in, are individuated by, and
(in non-miraculous cases) depend for their existence on not matter but sub-
stance.8 So in addition to holding that Mole’s corpse has a substantial form
which was never a form of Mole, Aquinas must also hold that all of its acci-
dents were never accidents of Mole. Again, the lack of a causal explanation of
these accidents casts doubt on the cogency of the unitarian position.
[R4] and [R10] require some heavy qualification. There are two problems
here. The first is that a strict application of the law of the indiscernibility of
identicals leads straightaway to the false conclusion that one and the same
thing cannot undergo change. We want to be able to say, for example, that the
man, Socrates, is the very same person or human being as the boy, Socrates. But
man-Socrates is taller than boy-Socrates; let us say that boy-Socrates is four-
feet tall and man-Socrates is six-feet tall. The indiscernibility of identicals tells
us that from the identity claim, “Boy-Socrates is identical to man-Socrates,” we
can infer that man-Socrates is four-feet tall (and that boy-Socrates is six-feet
tall). But these are false. Rather than deny the indiscernibility of identicals,
Aristotelians say that boy-Socrates and man-Socrates are not altogether identi-
cal. Aristotelians account for identity through changes of a substance like
Socrates by distinguishing the substance, Socrates, from various accidental
modifications of Socrates, such as his height. Time carries on and Socrates
grows up, but he is the same substance he always was. Relevant to the present
discussion is that since being a substance is just the sort of feature of a thing
that is supposed to remain constant over a thing’s many accidental changes,
the application of the indiscernibility of identicals expressed in [R4] and the
identity claim made in [R10] would not fall prey to any worries about the sort
of changes Mole’s body undergoes when it becomes Mole’s corpse. We could
say, cumbersomely, “The substance designated ‘Mole’s body’ while Mole is
alive is identical to the substance designated ‘Mole’s corpse’ when Mole dies,”
from which we could infer that if one is a substance then the other is, too.
The second problem is more complicated. In Chapter 7 I argued that Scotus
thinks that being a substance is, at least for some substances, contingent, such
8 Aquinas, In de generatione et corruptione I, l.10, n.6.

176 Chapter 10
that one and the same thing might first not be a substance and then be a sub-
stance, remaining the very same thing through the change. If Scotus is right,
then being a substance, for some substance, might turn out to be like any old
accident of a thing such as height or skin color. Just as we cannot infer from the
identity of boy-Socrates and man-Socrates that man-Socrates is four-feet tall,
so we might be wrong to infer from the identity of Mole’s body and Mole’s
corpse that Mole’s body is a substance. Maybe Mole’s body and Mole’s corpse
are the very same thing and that thing gains a new property—being a
substance—when Mole dies. Now, this chapter is not primarily a defense of
any claim of Scotus, but it is written in a Scotistic spirit, so this worry ought to
be worrisome. In reply, first, as far as I can tell Scotus has no reason not to
admit that any substance could cease to be a substance (and go on existing).
But also, second, as far as I can tell the only context in which he affirms that
being a substance is a contingent feature of a thing is in his discussion of the
metaphysics of continua, in which he claims that an arbitrary homogeneous
chunk of a continuous homogeneous substance itself becomes a substance
only when separated from the continuous homogeneous substance.9 The issue
here seems to be that since any continuous substance is infinitely divisible,
and any homogeneous continuous substance is infinitely divisible into parts of
the same specific nature, the claim that any arbitrary chunk of such a homoge-
neous continuous substance is itself a substance (a part-substance) would
entail that any homogeneous continuous substance has actually infinitely
many substances as parts—a conclusion Scotus would wish to deny. But in the
case of organic bodies this issue does not arise. And since Scotus explicitly
affirms that organisms have at least one substantial form in addition to a soul,
as discussed in Chapters 5 and 6, we have no reason to suppose that he himself
would deny that if Mole’s corpse is one substance then Mole’s body was one
substance, too. In fact, Scotus thinks that Mole’s corpse is composed of more
than one substance and therefore would think that his body was composed of
more than substance, but this is not directly relevant to the discussion of stan-
dard pluralism’s argument against unitarianism. The other premises of the
argument are not controversial, and the argument is valid.
For the standard pluralist, then, in a living thing there will be one substan-
tial form in virtue of which it is a body, and one or more souls in virtue of which
it is animate, sensitive, or rational. Standard pluralists therefore unequivocally
hold that the soul informs the body. The body of Mole is a substance all by
itself, a composite of prime matter and a substantial form of the body,
9 Ordinatio I, d.17, p. 2, q.1, n.232 (Vatican V, p. 251). Also see Cross’s discussion in The Physics of
Scotus, pp. 139−158 and especially pp. 146−147.
potentially informed by a soul and therefore potentially a living body, but not
dependent for its being a body on being informed by the soul. The standard
pluralist therefore embraces [cip]—Mole’s body can exist without Mole’s
soul—and rejects [hp]—Mole’s corpse really is a mole body.
v Scotus’s Response
From the perspective of Scotistic pluralism, however, the standard pluralist

does not take his rejection of [hp] far enough. Since on the standard pluralist
view the whole body is a composite of prime matter and one substantial form,
any integral part of the body, such as the organs, hands, and head, is dependent
for its identity on being a part of the body. Remove a kidney, therefore, and it
does not merely cease to be a part of the body but ceases to be altogether. That
thing in the organ bank is not really a kidney but, let’s say, a homonymous kid-
ney. It follows, therefore, that the standard pluralist rejects [hp] when it comes
to bodies, yet accepts [hp] when it comes to body parts.
For the Scotistic pluralist it seems that parity of reasoning should lead one
to reject [hp] for body parts as well as bodies. The argument is very similar to
the standard pluralist argument against unitarianism, but complicates it with
three substances—Mole, Mole’s removed kidney, and Mole’s removed paw—
whereas the earlier argument deals with just two. It also relies on an additional
but indubitable premise, namely that a kidney is not part of a paw or vice versa.
Consider the main reasons for adopting standard pluralism, that on the uni-
tarian thesis the corpse is a new substance, that if it is a new substance then it
must have had an efficient cause, but that there is no reasonable candidate in
the offing for what this efficient cause might be. Now, consider a case of organ
transplantation. On both the unitarian and standard pluralist views, once the
kidney is removed from the donor it ceases to exist, and something else comes
to be from it—what I have called the homonymous kidney. Now suppose the
homonymous kidney is installed in a recipient; at some instant it ceases to
exist and a kidney comes into existence. Note, however, that we have no reason
to think that the kidney that comes into existence is the very same kidney
that once existed as a part of the donor. So on both the unitarian and standard
pluralist views, an organ transplantation turns out to involve three distinct
objects—two kidneys and a homonymous kidney. But the homonymous kid-
ney appears to be continuous with the donor’s kidney and the recipient’s kid-
ney, so this is a reason in favor of thinking that it is the same kidney. And
supposing the removed kidney to be a homonymous kidney requires there to
be an efficient cause of the homonymous kidney, of which, again, there is none
178 Chapter 10
in the offing. So by parity of reason the Scotistic pluralist concludes that at

least some integral parts of the body have their own substantial forms and are
therefore substances. Repeat this process for some other integral part of the
body that is not a part of the kidney and of which the kidney is not a part, and
you arrive at Scotistic pluralism. In outline, the argument against standard plu-
ralism, argument [P], runs as follows:
[P1] Mole is a substance.

[P2] Mole’s removed kidney is a substance.
[P3] Mole’s removed paw is a substance.
[P4] Mole’s removed kidney is not the same substance as Mole.
[P5] Mole’s removed paw is not the same substance as Mole.
[P6] Mole’s removed kidney is not the same substance as Mole’s
removed paw.
[P7] For any x and for any y, x = y implies that if x is a substance then
y is a substance. (indiscernibility of identicals)
[P8] Either [P8a] Mole’s removed kidney and removed paw are identi-
cal with parts of Mole (such as one of his kidneys and one of his
hands), or [P8b] they are not. ([P4], [P5], excluded middle)
[P9] If [P8a], then [P9a] Mole has at least two substances as parts—
one of his kidneys and one of his paws. ([P2], [P3], [P6], [P7],
[P8], universal elimination)
[P10] If [P8b], then [P10a] Mole’s removed kidney and removed paw
have efficient causes that are not the efficient cause of Mole.
(assumption)
[P11] not-[P10a]. (“induction”)
[P12] not-[P8b]. ([P10], [P11], modus tollens)
[P13] Therefore [P8a]. ([P8], [P12], disjunctive syllogism)
[P14] Therefore [P9a]. ([P9], [P13], modus ponens)
[P15] A hand is not a part of a kidney nor is a kidney a part of hand.
[P16] Therefore, [Scotistic Pluralism]: ∃w∃x∃y (Sw and xPw and yPw
and x ≠ y and not-xPy and not-yPx). ([P14], [P15], existential
introduction x3)
[P17] Therefore not-[Standard Pluralism]: not-(∀w∀x∀y ((Sw and xPw

and yPw) then (x ≠ y then (xPy or yPx)))). ([N], above)
vi Scotus, Aquinas, and the Ultimate Subject of Substantial Change
Scotus’s understanding of part-substances allows for a finer-grained analysis of

the composition of substances than my focus on organs has suggested. Scotus
himself thought that many part-substances are themselves composed of part-
substances. Informed by the science of his day he thought that complex organs
such as the heart were composed of several different kinds of homogeneous mix-
tures, what he describes loosely as bone, flesh, and nerves. For Scotus an analysis
of part-substances ends at these homogeneous mixtures; each is composed not
of additional substances but of prime matter and one substantial form.
Scotus’s analysis of the bottom layer of part-substances should sound remi-
niscent of Aquinas’s solution to Ackrill’s problem. Recall that Aquinas attempts
to hold on to both [cip] and [hp] by identifying not the body but prime matter
as the matter of a substance. It might seem then that the Scotistic pluralist
simply delays recourse to prime matter. However many part-substances the
Scotistic pluralist allows to compose one substance, eventually he gets down to
a substance or substances whose matter is not itself a substance but is instead
simply the bare potency to become all things.
If like me you are worried about the consistency of Aquinas’s commitment
to [cip] given his doctrine of matter as pure potency actualized only through
form, then the Scotistic pluralist’s recourse to prime matter might seem like a
strike against him. But Scotus himself has a quick fix to any problems that
might lurk in Aquinas’s doctrine of prime matter. Scotus simply denies that
matter is not actual of itself. For Scotus it is all by itself a hoc aliquid, whose
nature it is to be the ultimate substrate of any substantial form.
vii Faulty Metaphysics or Faulty Chemistry? Scotistic Hylomorphism

and the Four Elements
Anyone familiar with ancient and medieval chemistry should immediately

wonder why Scotus’s analysis of part-substances stops at homogeneous mix-
tures. After all, such mixtures are supposed to be mixtures of at least two of the
four elements: earth, water, air, and fire. If Scotus has countenanced part-
substances at all, why not countenance elemental part-substances?
As discussed in Chapter 8, Scotus directly tackles this problem in Lectura II,
d.15. Scotus’s two strongest arguments against the claim that the elements
180 Chapter 10
actually exist in a mixed body represent two very different strategies. On the one
hand is Scotus’s argument from quantitative forms, which tries to show that if
the elements actually remained in a mixed body, they would either be co-located
and therefore mixable only by divine power, or they would juxtaposed and
therefore not compose a genuine substance. On the other hand is Scotus’s vio-
lence argument, which tries to show that any mixture of actually existing ele-
ments would subject at least one element of the mixture to non-natural motion,
since under natural conditions any mixed body is located within the sphere of
the moon and therefore located within one of the four elemental regions. This
violent motion is doubly problematic for the thesis that the elements remain in
a mixture, since it entails both that any mixed body is partially the result of non-
natural motions, and that the misplaced elements in any mixed body will strive
to return to their proper places, rendering the mixed body inherently unstable.
The two strategies are relevantly similar in at least one respect: each hangs
for its soundness on crucial features of medieval chemical theory, and not just
in the obvious sense that each argues that the four chemical elements cannot
exist in mixture. The argument from quantitative forms blocks what would
seem to be the natural Scotistic move—allowing non-mixed elemental bodies
to be bottom-level part-substances of complex bodily substances—by denying
that elemental bodies are perfectible by additional substantial forms. And this
denial, as I speculated, is rooted in the intra-theoretically plausible thesis that
elemental bodies cannot compose a unity of order: they are neither efficient
nor final-causally related to one another, since elements act for the sake of
reaching their proper sublunary region, and they naturally corrupt elemental
bodies of other kinds if they are able. The argument from violence, too, depends
on claims about natural elemental motion, specifically that since every ele-
mental body strives to reach its proper place, no mixed body composed of
them would exhibit the stability that natural substances like lions, lambs, and
Mole exhibit.
Contrast this with Aquinas’s denial in De mixtione elementorum that the
elements remain in a mixture. For Aquinas, given that the elements are sub-
stances, any composition of elements is bound to have no more than merely
aggregative unity.10 This is expected given Aquinas’s general endorsement of
10 Aquinas, De mixtione elementorum, Impossibile est enim materiam secundum idem

diuersas formas elementorum suscipere; si igitur in corpore mixto formae substantiales
elementorum saluentur, oportebit diuersis partibus materie eas inesse. Materie autem
diuersas partes accipere est impossibile nisi preintellecta quantitate in materia, sublata
enim quantitate substantia indiuisibilis permanet, ut patet in I Phisicorum; ex mate-
ria autem sub quantitate existente et forma substantiali adueniente corpus phisicum
unitarianism. For Scotus, however, it seems fairly clear that his denial that the
elements exist as part-substances of a substance is due not to metaphysical
considerations about the nature of composition, but is instead wedded to an
antiquated conception of elemental motion. This invites the speculation that
a different chemistry would persuade Scotus to allow an even finer-grained
account of part-substances, one, for example, that could include cells and their
organelles, molecules and their atoms, perhaps all the way down to elementary
particles.11
constituitur: diuerse igitur partes materie formis elementorum subsistentes plurium cor-
porum rationem suscipiunt. Multa autem corpora impossibile est esse simul; non igitur in
qualibet parte corporis mixti erunt quatuor elementa: et sic non erit uera mixtio, sed
secundum sensum, sicut accidit in aggregatione corporum insensibilium propter paruita-
tem. “For it is impossible for the same matter to sustain diverse forms of elements. If,
therefore, in the mixed body the substantial forms of the elements are preserved, it will
necessary for them to be in different parts of the matter. But it impossible for matter to
have different parts unless quantity is already understood to be in matter, for having taken
away quantity the substance remains indivisible, as is clear from Physics I. But from mat-
ter existing under quantity and the arriving substantial form a physical body is consti-
tuted. Therefore the parts of matter subsisting under the forms of the elements take on
the character of several bodies. But it is impossible for [one body] to be many bodies at
once; therefore the four elements will not be in every part of the mixed body. And thus it
will not be a true mixture, but an apparent one, as happens in an aggregation of bodies
that are imperceptible on account of their smallness.”
11 Kathrin Koslicki’s recent work on hylomorphism bears a striking affinity with Scotistic
hylomorphism, and especially the scientifically updated Scotistic hylomorphism I am
interested in here. See The Structure of Objects (Oxford: Oxford University Press, 2008),
especially pp. 186–187, from which it is worth quoting at length: “Consider once again the
table which, we said, is composed of some material components (the legs, top, hardware),
arranged in the manner dictated by the table’s formal components; it is the job of these
latter components to specify the variety and configuration which must be exhibited by
the material components out of which a whole of this kind may be composed. Consider
now a proper part of (a proper part of…) one of the table’s material components, e.g., a
single molecule which might be, say, a proper part of (a proper part of…) one of the table’s
legs. By the transitivity of parthood, the single molecule in question is a proper part of the
table as well. ¶ If tables are hybrid objects, consisting of formal and material components,
then so are molecules, since the same considerations apply in both cases. For the relation
between a molecule and the particles which constitute it is exactly the same as that which
holds between a table’s material components and the table itself: the molecule and the
particles that constitute it occupy the same region of space-time, but they do not share all
of their properties (e.g., the particles might exist before or after the molecule exists; they
need not constitute the molecule in questions; etc.) […] The same considerations which
motivated us to recognize within the table a certain amount of structural complexity,
182 Chapter 10
The Scotistic solution to Ackrill’s problem, then, is to reject the Homonymy

Principle. The severed paw or surgically removed kidney cannot, to be
sure, last very long as a paw and a kidney without being in auspicious
surroundings—a living organism or something simulating an organic environ-
ment. But being removed from, no longer a part of, an organism is for Scotus no
reason to conclude that they have ceased to exist. More importantly, Scotus
also would deny that being unable (in some sense) to perform the functions of
a paw or kidney excludes something from the class of paws and kidneys. The
kidney in the organ bank, so long as it is there on the shelf, cannot (in one
sense) perform the functions of a kidney. But it is in the organ bank—and not
in the biohazardous waste bin—because it can (in some other sense) perform
the functions of a kidney. It is open to Scotus to exploit this weaker potentiality
of the removed kidney in making his case to someone committed to the
Homonymy Principle. But Scotus, as far as I know, does not directly pursue this
route; I won’t either—for now. The difficulty here is that, for Scotus as for any
Aristotelian, any material thing can (eventually) become any other kind of
material thing, and therefore any material thing can (perhaps only in some
extremely remote sense) perform the function of any kind of material thing.
The trick is to determine a minimum level of potentiality such that anything
that can (at least at this minimum level) perform the functions of a K, is a K,
and if it cannot (at that minimum level) then it is not a K. But this seems like a
very hard task.
which we traced to the presence of additional components within the table over and
above its material components, therefore apply with the same force to molecules as well.
More generally, the material components of mereologically complex objects, as well as
their material components’…material components, can themselves be expected to exhibit
the same dichotomous nature as the wholes of which they are part. ¶ Only objects (if
there are any) which lie at the very bottom of the compositional hierarchy, i.e., objects
which are not themselves constituted by anything, would present us with an exception to
this generalization: if there are such things, they would be non-hybrid; or, at least, the
considerations which led us to ascribe a hybrid nature to such objects as tables would not
apply to this special case. For the job of an object’s formal components is to specify the
variety and configuration that must be exhibited by an object’s material components in
order for a whole of this kind to exist; but an object that is not constituted by anything has
no material components, and hence has no proper parts that must be of a certain variety
and configuration.”
Bibliography
Ackrill, J.L. “Aristotle’s Definitions of ‘Psuche’,” Proceedings of the Aristotelian Society 73

(1972–1973), pp. 119–133.
Adams, Marilyn McCord. Christ and Horrors: The Coherence of Christology. Cambridge:
Cambridge University Press, 2006.
——. “Essential Orders and Sacramental Causality,” in John Duns Scotus, Philosopher:
Proceedings of “The Quadruple Congress” on John Duns Scotus, ed. Mary Beth Ingham
and Oleg Bychkov. Münster: Aschendorff Verlag, 2010, pp. 191–206.
——. Some Later Medieval Theories of the Eucharist: Thomas Aquinas, Giles of Rome,
Duns Scotus, and William Ockham. Oxford: Oxford University Press, 2010.
——. “What’s Metaphysically Special about Supposits? Some Medieval Variations
on Aristotelian Substance,” Aristotelian Society Supplemental Volume 79:1 (2005),
pp. 15–52.
——. William Ockham, 2 vols. South Bend, in: University of Notre Dame Press, 1987.
Albertus, Magnus. Opera Omnia, 41 vols. Cologne ed. Monasterii Westfalorum:
Aschendorff, 1960.
Aphrodisias, Alexander of. On the Soul, trans. A.P. Fotinis, in The De Anima of Alexander
of Aphrodisias: A Translation and Commentary. Washington, dc: University Press of
America, 1979.
Aquinas, Thomas. Opera Omnia, ed. Robert Busa, S.J. Milan: Editoria Elettronica
Editel, 1992. Revised edition on the internet, by Enrique Alarcón, http://www
.corpusthomisticum.org/iopera.html, 2005.
Aristotle, The Complete Works of Aristotle, 2 vols., ed. J. Barnes. Princeton, nj: Princeton
University Press, 1984.
Astudillo, Diego. De generatione et corruptione, ed. N. Tyerri. Valladolid, 1532.
Augustodunensis Honorius. De Imagine Mundi. Patrologia Latina 172, ed. Migne, 1854.
Aureole, Peter. Commentarium in libros Sententiarum, 2 vols. Rome: ex Typographia
Vaticanae, 1596.
Avicebron. Fons Vitae, trans. John of Spain and Dominic Gundisalvi, ed. C. Baeumker.
Munster: Aschendorff, 1895.
Avicenna. The Metaphysics of the Healing, trans. M.E. Marmura. Provo, ut: Brigham
Young University Press, 2005.
Bettoni, Efrem. Duns Scotus: The Basic Principles of His Philosophy, trans. Bernardino
Bonansea. Washington, dc: The Catholic University of America Press, 1961.
Buridan, Jean. Le Traité de L’âme de Jean Buridan, ed. Benoît Patar. Louvain-la-Neuve:
Éditions de L’institut Supérieur de Philosophie, 1991.
184 Bibliography
——. John Buridan, Quaestiones super libros De generatione et corruptione Aristotelis,

ed. Michiel Streijger, Paul J.J.M. Bakker, Johannes M.M.H. Thijssen. Leiden and
Boston: Brill, 2010.
Callus, Daniel A. “The Origins of the Problem of the Unity of Form,” in The Dignity of
Science: Studies in the Philosophy of Science, ed. James A. Weisheipl. Washington, dc:
Thomist, 1961, pp. 121–149.
Cross, Richard. “Duns Scotus’s Anti-Reductionistic Account of Material Substance,”
Vivarium 33:2 (1995), pp. 137–170.
——. Duns Scotus on God. Burlington, vt: Ashgate, 2005.
——. “Identity, Origin, and Persistence in Duns Scotus’s Physics,” History of Philosophy
Quarterly 16 (1999), pp. 1–18.
——. The Metaphysics of the Incarnation: Thomas Aquinas to Duns Scotus. Oxford:
Oxford University Press, 2002.
——. “Ockham on Part and Whole,” Vivarium 37:2 (1999), pp. 143–167.
——. “Philosophy of Mind,” in The Cambridge Companion to Duns Scotus, ed. Thomas
Williams. Cambridge: Cambridge University Press, 2003. pp. 263–284.
——. The Physics of Duns Scotus: The Scientific Context of a Theological Vision. Oxford:
——. Richard Cross, “Relations, Universals, and the Abuse of Tropes,” Aristotelian
Society Supplementary Volume 79:1 (2005), pp. 53–72.
Dumont, Stephen. “Did Duns Scotus Change His Mind on the Will?” Miscellanea
Mediaevilla 28 (2001), pp. 719–794.
——. “The Origin of Scotus’s Theory of Synchronic Contingency,” The Modern
Schoolman 72 (1995), pp. 149–167.
——. “The Question on Individuation in Scotus’ Quaestiones super Metaphysicam,” in
Via Scoti: Methodologia ad mentem Ioannis Duns Scoti, vol. I, ed. Leonardo Sileo.
Rome: Edizioni Antonianum, 1995, pp. 193–227.
Ebrey, David Buckley. “Aristotle’s Motivation for Matter,” Doctoral Dissertation. ucla,
2007.
Eddington, Arthur. The Nature of the Physical World. New York, ny and Cambridge, uk:
The MacMillan Company and Cambridge University Press, 1929.
Fine, Kit. “Things and Their Parts,” Midwest Studies in Philosophy 23 (1999), pp. 61–74.
Ghent, Henry of. Summae Quaestionum Ordinarium, 2 vols., ed. Jodicus Badius. Paris,
1520. Reprinted by St. Bonaventure, ny: The Franciscan Institute, 1953.
Gibbard, Allan. “Contingent Identity,” Journal of Philosophical Logic 4 (1975),
pp. 187–221.
Gilson, Étienne. Jean Duns Scot: Introduction a ses Positions Fondamentales. Paris:
J. Vrin, 1952.
Grant, Edward. A Sourcebook of Medieval Science. Cambridge, ma: Harvard University
Press, 1974.
Bibliography 185
Harclay, Henry of. Ordinary Questions, 2 vols, ed. Mark Henninger, trans. Mark
Henninger and Raymond Edwards. Oxford: Oxford University Press, for the British
Academy, 2008.
Henninger, Mark. Relations: Medieval Theories 1250–1325. Oxford: Oxford University
Press, 1989.
Henry, Devin. “Themistius and Spontaneous Generation in Aristotle’s Metaphysics,”
Oxford Studies in Ancient Philosophy 24 (2003), pp. 183–208.
Hume, David. An Enquiry Concerning Human Understanding, with A Letter from a
Gentleman to His Friend in Edinburgh, and Hume’s Abstract of A Treatise of Human
Nature, 2nd edn, ed. Eric Steinberg. Indianapolis, in: Hackett, 1993.
Huning, Hildebert Alois. “The Plurality of Forms According to Petrus de Trabibus,”
Franciscan Studies 28 (1968), pp. 137–196.
Johnston, Mark. “Hylomorphism,” Journal of Philosophy 103:12 (2006).
King, Peter. “Duns Scotus on Possibilities, Powers, and the Possible,” in Potentialität und
Possibilität: Modalaussagen in Der Geschichte Der Metaphysik, ed. T. Buchheim,
K. Lorenz, and C.H. Kneepkens. Stuttgart: Frommann-Holzboog, 2001, pp. 175–199.
Kinlaw, Caleb. “Elemental Teleology and an Interpretation of the Rainfall Example in
Physics 2.8,” Masters Thesis. University of Edinburgh, 2008.
Knuuttila, Simo. Modalities in Medieval Philosophy. London and New York: Routledge,
1993.
——. “Interpreting Scotus’s Theory of Modality: Three Critical Remarks,” in Via Scoti:
Methodologia ad mentem Joannis Duns Scoti, vol. 1, ed. Leonardo Sileo. Rome:
Edizioni Antonianum, 1995, pp. 295–303.
Koslicki, Kathryn. The Structure of Objects. Oxford: Oxford University Press, 2008.
Langston, Douglas. “Scotus and Possible Worlds,” in Knowledge and the Sciences in
Medieval Philosophy, vol. 2, ed. Simo Knuuttila, Reijo Työrinoja, and Stan Ebbesen,
pp. 240–247. Helsinki: Yliopistopaino, 1990.
Leith, John H. (ed.). Creeds of the Churches: A Reader in Christian Doctrine from the Bible
to the Present, 3rd edn. Louisville, ky: John Knox Press, 1982.
Lewis, David. “Causation,” Journal of Philosophy 70 (1973), pp. 556–567.
——. “Causation as Influence,” Journal of Philosophy 97 (2000), pp. 182–197.
Long, A.A. and Sedley D.N. The Hellenistic Philosophers, Volume 1: Translations of
the Principal Sources, with Philosophical Commentary. Cambridge: Cambridge
Lowe, E.J. “Non-Cartesian Dualism,” Philosophy of Mind: A Guide and Anthology, ed.
J. Heil. Oxford: Oxford University Press, 2004.
Maier, Anneliese. On the Threshold of Exact Science: Selected Writings of Anneliese
Maier on Late Medieval Natural Philosophy, ed. and trans. Steven D. Sargent.
Philadelphia, pa: University of Pennsylvania Press, 1982.
186 Bibliography
Marshall, Peter. “Parisian Psychology in the Mid-Fourteenth Century,” Archives

d’histoire doctrinale et littéraire du moyen age 50 (1983), pp. 101–193.
Massobrio, Simona. “Aristotelian Matter as Understood by St. Thomas Aquinas and
John Duns Scotus.” Doctoral Dissertation. McGill University, 1991.
Middleton, Richard of. De Gradu Formarum, ed. Roberto Zavalloni, in Richard de
Mediavilla et la Controverse sur la Pluralité des Formes. Louvain: Éditions de L’institut
Supérieur de Philosophie, 1951.
Muñoz, Garcia, Angel. “Albert of Saxony, Bibliography,” Bulletin de Philosophie
Médiévale 32 (1990), pp. 161–190.
Noone, Timothy. “Scotus’ Critique of the Thomistic Theory of Individuation and the
Dating of the Quaestiones in libros Metaphysicorum VII, q. 13,” in Via Scoti:
Methodologia ad mentem Ioannis Duns Scoti, vol. I, ed. Leonardo Sileo. Rome:
Edizioni Antonianum, 1995, pp. 391–406.
Normore, Calvin. “Duns Scotus’s Modal Theory,” in The Cambridge Companion to John
Duns Scotus, ed. Thomas Williams. Cambridge: Cambridge University Press, 2002,
pp. 129–160.
——. “Ockham’s Metaphysics of Parts,” The Journal of Philosophy 103:12 (2006), pp.
737–754.
——. “Scotus, Modality, Instants of Nature and the Contingency of the Present,” in
John Duns Scotus: Metaphysics and Ethics, ed. Ludger Honnefelder, Rega Wood, and
Mechthild Dreyer. Leiden: E. J. Brill, 1996, pp. 161–174.
Ockham, William of. Opera Philosophica et Theologica, 17 vols. St. Bonaventure, ny: The
Franciscan Institute, 1967–1988.
Olivi, Peter John. Quaestiones in secundum librum Sententiarum, 2 vols, ed. Bernardus
Jansen. Quarracchi, 1922.
Paasch, J.T.. Divine Production in Late Medieval Trinitarian Theology: Henry of Ghent,
Duns Scotus, and William Ockham. Oxford: Oxford University Press, 2012.
Pasnau, Robert. Metaphysical Themes 1274–1671. Oxford: Oxford University Press, 2011.
Plato. Complete Works, ed. John M. Cooper. Indianapolis and Cambridge: Hackett
Publishing Company, 1997.
Rea, Michael. “The Problem of Material Constitution,” The Philosophical Review 104:4
(1995), pp. 525–552.
Sandbach, F.H. The Stoics, 2nd edn. Indianapolis, in: Hackett, 1989.
Saxony, Albert of. De generatione et corruptione. ed. Lockert. Paris, 1516.
Schaffer, Jonathan. “Monism”, The Stanford Encyclopedia of Philosophy, Fall 2008
Edition, ed. Edward N. Zalta. url: <http://plato.stanford.edu/archives/fall2008/
entries/monism/>.
Scotus, John Duns. A Treatise on God as First Principle, trans. Allan B. Wolter. Chicago:
Franciscan Herald Press, 1966.
Bibliography 187
——. Obras del Doctor Sutil Juan Duns Escoto (edicion biligue): Cuestiones Cuodlibetales.
Trans. (with Latin edition) by Felix Alluntis. Madrid: Biblioteca de autores cristia
nos, 1968.
——. Opera omnia, 12 vols, ed. Luke Wadding. Lyon: Durand, 1639.
——. Opera Omnia, 18 vols, ed. Scotistic Commission. Citta del Vaticano: Typis
Vaticanis, 1950.
——. Opera Philosophica, 5 vols. St. Bonaventure, ny: The Franciscan Institute,
1991–2006.
——. Questions on the Metaphysics of Aristotle, 2 vols., trans. Girard J. Etzkorn and
Allan B. Wolter. St. Bonaventure, ny: The Franciscan Institute, 1997.
Sharp, Dorothea E. Franciscan Philosophy at Oxford in the Thirteenth Century. London:
Shields, Christopher. “The Fundamental Problem about Hylomorphism,” Spring 2011
Edition, ed. Edward N. Zalta. url: http://plato.stanford.edu/entries/aristotle
-psychology/suppl1.html.
Simons, Peter. Parts: A Study in Ontology. Oxford: Clarendon Press, 1987.
Stella, Prospero. L’ilemorfismo di G. Duns Scoto. Torino: Societa Editrice Internazionale,
1955.
Stone, Abraham. “Avicenna’s Theory of Primary Mixture,” Arabic Sciences and
Philosophy 18 (2008), pp. 99–119.
Suárez, Francisco. De generatione et corruptione, in Francisco Suárez: “Der ist der Mann,”
ed. S. Castellote. Valencia: Facultidad de Teología “San Vincente Ferrer,” 2004,
pp. 435–682.
——. Opera omnia, 28 vols, ed. Carlos Berton. Paris: Louis Vivès, 1856–1861.
Taylor, A.E. A Commentary on Plato’s Timaeus. Oxford: Oxford University Press, 1962.
Trifogli, Cecilia. “Scotus and the Medieval Debate about the Continuum,” Medioevo 29
(2004), pp. 233–266.
van Inwagen, Peter. Material Beings. Ithaca: Cornell University Press, 1990.
——. “The Doctrine of Arbitrary Undetached Parts,” in Ontology, Identity, and Modality:
Essays in Metaphysics. Cambridge: Cambridge University Press, 2001, pp. 75–94.
Ward, Thomas M. “Relations without Forms: Some Consequences of Aquinas’s
Metaphysics of Relation,” Vivarium 48:3–4 (2010), pp. 279–301.
Wasserman, Ryan. “Material Constitution,” Stanford Encyclopedia of Philosophy,
Summer 2013 Edition, ed. Edward N. Zalta, forthcoming. url: <http://plato.stanford
.edu/archives/sum2013/entries/material-constitution/>
Weisheipl, James A. “Albertus Magnus and Universal Hylomorphism: Avicebron
(A Note on Thirteenth-Century Augustinianism),” in Albert the Great: Commemora
tive Essays, ed. F.J. Kovach and R.W. Shahan. Norman, ok: University of Oklahoma
Press, 1980. pp. 239–260.
188 Bibliography
Wiggins, David. Sameness and Substance Renewed. Cambridge: Cambridge University

Press, 2001.
Williams, Scott M. “Henry of Ghent on Real Relations and the Trinity: The Case for
Numerical Sameness without Identity,” Recherches de Théologie et Philosophie
Médiévales 79:1 (2012), pp. 109–148.
Williams, Thomas. “Introduction: The Life and Works of John Duns the Scot,” in The
Cambridge Companion to Duns Scotus, ed. Thomas Williams. Cambridge: Cambridge
Wood, Rega and Weisberg Michael. “Interpreting Aristotle on Mixture: Problems about
Elemental Composition from Philoponus to Cooper,” Studies in the History and
Philosophy of Science 35 (2004), pp. 681–706.
Wyatt, Nicole. “Did Duns Scotus Invent Possible Worlds Semantics?” Australasian
Journal of Philosophy 83 (2000), pp. 196–212.
Zavalloni, Roberto. Richard de Mediavilla et la Controverse sur la Pluralité des Formes.
Louvain: Éditions de L’institut Supérieur de Philosophie, 1951.
Index
Absolute Aggregate 41, 45, 48–52, 74, 94–5,

Absolute accident 58 97–9, 129
Absolute change 42, 55–56, 58, Albertus Magnus 76 fn. 2, 129,
62–8, 71 129 fn. 13
Absolute effect 19, 23 Albert of Saxony 80 fn. 14, 129,
Absolute entity, see Absolute Thing 129 fn. 12
Absolute essence 19, 23, 63, 69 al-Ghāzāli 9
Absolute form 54, 58, 61, 73 Alnwick, William 31, 133 fn. 23
Absolute genus 55 Analysis 1, 22, 65 fn. 7, 77, 79, 117, 123,
Absolute nature, see Absolute Essence 125–7, 141, 143, 150 fn. 9, 162, 172, 179
Absolute power 20–23 Hylomorphic analysis 43 fn. 7, 89,
Absolute property 41, 64 152–3, 169, 171
Absolute ratio 20 Potency-act analysis 81
Absolute term (of a relation) 64 Angel 10, 10 fn. 7, 29, 49, 61–2, 73, 112,
Absolute thing 35, 37 fn. 35, 38 fn. 36, 114, 174
41 fn. 3, 55–56, 59, 66, 68–70 Animal 1, 25, 32, 48, 76–7, 79 fn. 10, 81 fn.
Accident 11, 41 fn. 3, 51, 54, 84, 95, 102 fn. 17, 85, 85 fn. 30, 86, 90, 94–6, 103, 105,
26, 112, 120, 130, 133, 141, 175–6 107 fn. 35, 108, 115, 117, 136, 138, 142–3,
Accidental essence 28 159–62, 171
Accidental form 12, 30, 43 fn. 7, 49, 54, Blooded animal 103, 138
76 fn. 2, 97, 107 fn. 33, 155–6, 166, 170 Animate 48, 56 fn. 46, 78, 90 fn. 40,
Accident Dependent on substance 93, 176
120, 150 fn. 9 Animation 26, 91
Necessary accident 142 Antichrist 19, 28, 30
Proper accident 162–3 Aphrodisias, Alexander of 77 fn. 2
Separated accident 114 Appetite 11–2, 18
Ackrill, J.L. 165, 169, 171, 179, 182 A priori 1
Act, see Actuality Aptitude 11, 52, 112, 114, 119–20
Action 15, 101 fn. 20, 149 Aquinas, Thomas 5, 27, 28 fn. 5, 30 fn. 12,
Action of an efficient cause 12, 37, 63 39–40, 42 fn. 3, 45–6, 77 fn. 2, 78, 81–82, 82
Action of an element 131, 141 fn. 20 & 24, 83, 83 fn. 26 & 27, 84, 90 fn. 40,
Divine action 8 92, 106 fn. 30, 108, 125, 129–30, 164, 166, 170
Final end of action 137 fn. 4, 171–3, 175, 179–80
Actuality 27, 39–40, 47, 113–4, 118 fn. 18, Aristotle 2, 5, 9, 13 fn. 10, 32, 43 fn. 7, 45–8,
121, 172–3 53, 58, 62, 64, 66–7, 78–80, 81 fn. 18, 96,
Adams, Marilyn McCord 6 fn. 1, 30 fn. 16, 103–6, 110, 112 fn. 6, 115, 119, 119 fn. 19, 128,
55 fn. 42, 59 fn. 50, 76 fn. 2, 79 fn. 10, 82, 82 130, 137 fn. 28, 139, 142, 150 fn. 8, 160–1,
fn. 22, 99 fn. 9, 100 fn. 14, 118, 119, 119 fn. 19 170–2
& 22, 134 fn. 25, 150 fn. 9 & 12 Arrangement 49, 97, 138
Affectio commodi 150 fn. 8 Astudillo, Diego 84 fn. 29
Affectio iustitiae 150 fn. 8 Augustodunensis, Honorius 147 fn. 3
Agent 9, 12–8, 37, 65, 140–1, 154 Aureole, Peter 129
Created agent, see Natural agent Averroës 46, 128–30
Divine agent 8, 10 Avicebron 76 fn. 2
Natural agent 9–10, 25, 38–9, 134 Avicenna 45–7, 82 fn. 24, 128–30, 132
190 Index
Bare Particular 155 Change as motivation for

Bettoni, Efrem 76 fn. 2 hylomorphism 8–12, 152–4
Body 4–5, 28, 41, 43–5, 48, 48 fn. 27, 53, Change in degree 129
58, 64, 77–9, 81–2, 83 fn. 26, 84–5, 90 fn. 40, Relational change, see Relative change
91–3, 98, 102, 104–6, 106 fn. 30, 107 fn. 33 & Relative change 41, 54, 56, 60–1, 63–4,
34, 108–110, 117, 122–3, 126, 128–32, 132 fn. 66–8
20, 134–5, 137–8, 146–7, 160, 162–3, 167–8, Substantial change 7–8, 11, 17, 27, 30,
170–80, 181 fn. 10 40, 78, 89, 108, 155–6, 172, 179
Bradley’s Regress 55 fn. 42 Chemistry 2, 4, 125, 127, 146, 179, 181
Bronze 6–7, 25, 27, 43, 43 fn. 7, 76, 81, 122, Christology 119
169–70, 172 Co-location 134, 137, 180
Buridan, Jean 106–8, 107 fn. 33 & 34, 129 Complex 1, 4–5, 44, 49–50, 59, 89, 114, 123,
Bundle Theory 155 136, 145–6, 153–4, 160, 162, 179, 180
Complexion 129, 143
Callus, Daniel A. 176 fn. 2 Complexity 181 fn. 11
Capacity 169, 172 Composition 5, 34, 44, 54, 58–60, 62,
Category 41 fn. 3, 119 74–5, 81, 90 fn. 40, 94, 125, 127, 143, 147,
Cause 179–81, 182 fn. 11
Accidentally ordered causal series Compound 4, 76, 78, 113, 122, 125–8, 147
100–1 fn. 3, 151, 169–70
Co-cause 65, 67, 74 Composite 1–3, 7, 13, 20, 23–5, 30–3, 38–9,
Concurrent cause 63, 67, 72, 92 45–9, 54–6, 66, 68–71, 76 fn. 2, 77–9, 81–3,
Efficient cause 4, 8, 12–5, 18–9, 37–9, 81 fn. 19, 92, 95, 106–7, 110–2, 114, 116–7,
63, 65, 70–2, 82–4, 83 fn. 27, 99–106, 101 127–9, 128 fn. 6, 133–6, 153, 155, 162, 165, 170,
fn. 20, 115, 126, 136–7, 140–1, 145, 173–5, 172–3, 176–7
177–8, 180 Condemnations of 1277 163
Essentially ordered causal series 4, Connotative 107
99, 101–6, 127, 145, 164 Conservation 38, 140
Extrinsic cause 70, 72 fn. 19, 140, Contingency 120–1, 123
Final cause 4, 99–106, 101 fn. 20, 108, Contingent 4–5, 35, 64, 110–1, 116, 116 fn.
126, 136–7, 137 fn. 28, 145, 150, 150 fn. 8, 14, 118, 120–1, 124, 151, 169, 173, 175–6
164, 180 Continuum 111, 117, 118 fn. 18, 123 fn. 27, 144
Formal cause 62, 68, 70–1, 99–100, fn. 50, 176
102–3 Contradiction 22, 35–6, 88, 93, 101,
Intrinsic cause 61–2, 64 fn. 2, 64, 157, 168
70–2, 74 Contradiction between Homonymy
Material cause 99–100, 102–3, 102 Principle and Contingently Informed
fn. 26 Principle 5, 171
Secondary cause 37, 39 Contradiction in being 82–4
Cell 125, 160, 181 Corpse 5, 43, 78–9, 82–4, 105–6, 106 fn.
Chalcedon 119 30, 115, 170, 172–7
Change 1, 3, 6, 10, 16 12–4, 18, 25, 28, 43, Corporeity, see Form of corporeity
53–5, 57–62, 73, 89, 92, 103, 107, 115, 135, Corruption 10, 13–5, 17, 108, 133, 137, 158–9
137–8, 152–5, 163, 169, 172–3, 175–6 Corruption of elements 4, 131, 137–8,
Absolute change 41–2, 54, 56, 58, 60–1, 141, 180
63, 65–6, 71 Corruption of relations 68–9
Accidental change 8, 9, 11, 27, 89, 128, Extrinsic corruption 143
155–7
Index 191
Form not inclined to corruption Real distinction

11–2, 16 Ockham’s two kinds of real distinc
Intrinsic corruption 139–40, 142, tion 57–59, 71
Substantial corruption 11, 13–4, 17–8, Real distinction between matter and
52, 66, 69, 82, 84, 138, 141, 173, 175 form 30, 43, 152, 169, 171
Cosmos 3, 137 fn. 28 Real distinction of the persons of the
Creation 38–9 Trinity 30, fn. 15
Cross, Richard 24 fn. 33, 29 fn. 12, 34 fn. Real distinction of substance from its
28, 37 fn. 34, 38–9, 42, 90–1, 93, 109 fn. 39, parts 3, 70
120, 120 fn. 23, 122 fn. 26 Specific distinction 115, 135, 141
Target/count distinction 151 fn. 13
Definition 47–8, 48 fn. 24, 55, 140 fn. 37, Divine
143 Divine agent, see Agent, Divine
Real definition 47–8, 47 fn. 23 Divine nature, see Nature, Divine
Nominal definition 47 fn. 23 Divine person, see Person, Divine
Democritus 8 Divine power, see Power, Divine
Denominative Predication 33 Divine volition 161
Dependence Dumont, Stephen 22 fn. 27, 33 fn. 25, 134
Dependence of an agent on a fn. 23
patient 14, 18
Causal dependence 23 fn. 30 Ebrey, David 13 fn. 10
Essential dependence relations Eddington, Arthur 127 fn. 2
45, 101 Effect
Essential order of dependence, see Effect of accidentally ordered
Essential order, of dependence causes 100
Non-causal dependence 102, 120 Effect of a first cause 145
Destruction Effect of an angel 10
Destruction of an accident 58 Effect of an efficient cause 9, 13–8, 72,
Destruction of an entity 53, 57, 59–61, 99–100
73 Effect of an element 128, 141
Destruction of an essential part 52 Effect of an intrinsic cause 58, 62–3,
Destruction of a relation 58, 67–8 65, 70–2, 74
Diachronic 154 Effect of a power 19–23
Disposition 26, 91, 95, 131, 172 Effect of a rational power 10 fn. 6
Distinction Effect of essentially ordered causes
Distinction between denominative and 67, 99–102, 104
essential predication 32–3 Effect of God 10
Distinction between essential and integral Elements
parts 43 Body parts described as elements
Distinction between homogeneous and 97–8
heterogeneous substances 122 The Four elements (earth, air, fire,
Distinction between natural and rational water) 2–5, 7, 76, 78 fn. 9, 82, 118,
powers 10 fn. 6 122, 125–43, 146, 179–80
Distinction between subjective and Matter and form as elements strictly
objective potency 29, 40 speaking 140
Distinction between supposits and Qualities of elements 128–9, 143
substances 118–21 Embryo 6, 8, 26, 104
Numerical distinction 78, 141 Embryology 77, 80, 103–4, 115, 146
192 Index
Esse 135–6 Dependence of part on whole for

Essence 37, 48, 166 existence 110
Essence cannot contain repugnant Essential dependence of x on y for
perfections 88 fn. 37 existence 102
Essence of a power 19, 23 Existence conditions 41, 49, 74
Essence of composite substance Existence conditions of soul in
46–7 matter 91
Essence of matter 37, 39 Existence from something else 7
Essence understood as form 46 Existence of elements in a
Divine essence 116 fn. 14 mixture 129, 139
Individual essence 48 Existence of form 39, 154
Objective potency of an essence Existence of informative relation 57
28–30 Existence of intellective soul in
Part of an essence 46–8 matter 26
Essential Existence of matter and form prior to their
Essential activity 149 causing 72
Essential dependence, see Essential order, Existence of one part efficiently caused by
of Dependence another 104
Essential nature 52 Existence of part for the sake of the
Essential notion of power 19, 20, 23 whole 110, 135
Essential order, see Essential Order Existence of part prior to whole 115
Essential parts, see Parts, Essential Existence of prime matter 7, 9, 29, 33,
Essential perfections 82 37 fn. 35, 38–9
Essential posteriority 100 Existence of relations of union 66
Essential predication, see Predication, Existence of substantial forms of organic
Essential parts 83
Essential priority 100 Form strives to conserve existence 8
Essential properties 6, 33, 51 fn. 33, Independence of part on whole for
78, 172 existence 5
Essential unity, see Unity, Essential Natural existence of part in
Essential order 3, 45, 100, 102 fn. 24, whole 139
104–5, 108, 145, 148 Objective potency of an essence to its own
Essential order of dependence 35, 45, existence 28
47, 100, 106 Parthood relations posterior to existence
Essential order of efficient of whole 65
causation 104–5, 136 Per se existence 116 fn. 12
Essential order of eminence 35 Possible exisence 19, 28
Essential order of final causation Requirement of the prior for the existence
105, 108, 136 of the posterior 36
Nonevident essential order Subjective potency of a subject to receive
Proximate nonevident essential existence from another 28
order 102, 104 Virtual existence 131
Remote nonevident essential Zapping into existence 9
order 102, 104 Ex nihilo 10 fn. 6, 13–4, 17
Eudaimonia 150 fn. 8 Extended
Existence 6, 21, 92, 101 fn. 20, 177 (for Ockham) matter as such is
Dependence of accidents on matter for extended 27
existence 84 (for Scotus) matter as such is not
Dependence of accidents on substance for extended 27
existence 84, 175 Extended parts, see Parts, Extended
Index 193
Extension Generation 9, 15, 52 fn. 35, 82, 172–3

Of a whole 95 Active generation 12
Of substance 167 Generation ex nihilo 13
Extreme (of a relation), see Term (of a Generation from elements 137–8,
relation) 140–2
Generation of corpses 83
Fine, Kit 154 Generation of parts 103
Form Generation of parts prior to generation of
Accidental form 30, 49, 54, 76 fn. 2, whole 92, 108, 115
107 fn. 33, 155–7, 166, 170 Generation of an organism 77, 103,
Angelic form 49 107, 138
Form of a part (see Forma partis) Generation of a relation 55
Form of corporeity 4, 26, 77, 79, 84–7, Generation of a substance 11–2, 14,
90–3, 117 63–4, 66, 69, 78, 80, 84, 92
Form of the whole (see Forma totius) Passive generation 12
Forma accidentalis mundi 156 Persistence through generation
Forma mundi 147, 151–2, 155–6, 164 12–8
Forma partis 45–7 Spontaneous generation 83
Forma totius 45–7 Genus 32, 51, 55–6, 90 fn. 40, 112, 117, 130
Immaterial form 29 Gibbard, Allan 153 fn. 14
Substantial form 2–5, 11–2, 24, 27, 29, Gilson, Étienne 30 fn. 16, 76 fn. 2,
33, 36, 41–3, 45–7, 58, 76–9, 81–99, 101, Giver of Forms 83
103, 105–7, 110–7, 128–32, 135–6, 140–1, Grade (of qualities) 91
143–4, 146–7, 151, 155–6, 158, 161, 163–4, Grahame, Kenneth vi
166, 168, 170–2, 176–81 Grant, Edward 163 fn. 27
Unbounded form 97–8 God
Foundation (of a relation) 19, 58, 66, 68 Argument against plurality of gods
Frog Prince 25 145 fn. 2
Function Everything exists and acts for God’s
Argument for distinct part-substances sake 101, 150
from distinction of functions 80, God alone can produce ex nihilo 13
122 God alone can produce matter 7
Dependence for functioning of some parts God and occasionalism 8
on others 105 God as first efficient and final
Function of a substance of a specific cause 145
kind 108 God as giver of forms 83 fn.
Function of elements 126, 137 fn. 28 26, 174
Function of integral parts 85–6, 107–8, God can conserve matter without
122 form 31
Function of prime matter 3, 7 God can create withour secondary
Function of the heart 109–10 causes 37, 39
Joint-functioning of parts 99 God: Father, Son, and Holy Spirit
Parts function for the sake of the 119
whole 126, 135 God’s power 25–6, 39, 40, 134, 145 fn.
Parts required for a substance to perform 2, 161, 174
its specific function 94 Matter in obediential potency toward
Powers exist prior to their God 8
functioning 21 Son of God 119–20, 150
Relationship between kind and Stoic views of God 147 fn. 3
function 5, 170, 182 Haecceity 123 fn. 27
194 Index
Ham Sandwich 153–4 Johnston, Mark 44–5, 165 fn. 2

Henninger, Mark 42 fn. 3, 80 fn. 14 Juxtaposition 128, 134–5
Henry, Devin 83 fn. 26
Henry of Ghent 20, 20 fn. 23, 82 fn. 20, 92 Kind
Henry of Harclay 80 fn. 14 Four kinds of elements 4, 118, 122
Hoc aliquid 111, 172, 179 Kinds of causality 42
Homonymy 5, 165, 170, 182 Kinds of change 7
Hot Potato 173 Kinds of composition 145
Hume, David 23, 23 fn. 29, 155 Kinds of essential order 35, 99, 136
Huning, Hildebert Alois 80 fn. 14 Kinds hylomorphism 171
Hypostatic Union 150 Kinds of mixture 142, 179
Hyle (see Matter) Kinds of power 19, 27
Hylomorphic Kinds of substantial form 94,
Hylomorphic analysis of change 89, 107, 117
152–7, 165, 171 Kinds of unity 1, 45, 94, 95 fn. 2
Hylomorphic compound 1, 147, 151, Kinds of whole 2, 41, 150
169–70, Membership in a kind 5, 170, 182
Hylomorphic Monism, see Monism, Natural kind 4, 123
Hylomorphic Organic parts as different kinds of
Hylomorphic theory 108, 165 substance 5, 48, 93
Hylomorphism 1–2, 4–5, 7, 9, 43–4, 81, King, Peter 24 fn. 33
108 153–7, 165–9, 171, 179, 181 Kinlaw, Caleb 137 fn. 28
Knuuttila, Simo 22 fn. 27
Identity Koslicki, Kathryn 181 fn. 11
Identity of matter independent of
form 79, 169 Langston, Douglas 22 fn. 27
Real identity between individual and Leith, John H. 119 fn. 20
essence 48 Lewis, David 23, 23 fn. 30
Inanimate 142 Long, A.A. 147 fn. 3
Incarnation 111, 118–20, 148, 150,
Inclination 24 Maier, Anneliese 127–8
Incompossible 10, 88 Malebranche 9
Indiscernibility of Identicals 173–5, 178 Marshall, Peter 80 fn. 14
Individual 1, 12, 26, 34–5, 39, 48, 54, 57, Massobrio, Simona 29 fn. 12
62, 74, 79, 91, 111, 119–21, 123, 128, 144, 150, Matter
158–61, 166 Matter incapable of destruction by natural
Infinite powers 58
Actual infinite 117, 122, 126, 176 Prime matter 2–3, 6–9, 11–2, 21, 23–9,
Infinitely divisible 123 fn. 27, 126, 176 45, 61–2, 78–9, 81, 92, 98, 114, 116, 129,
Infinite regress 7, 44, 45, 54 133, 140, 144, 172–3, 176–7, 179
Inherence 12, 42 fn. 3, 80, 84, 89, 91, 95, Proximate matter 6–8, 24, 26, 30, 172
112–4, 119, 141, 162, 175 Mereological sum 49, 148
Instant Metaphysics 1, 2, 58, 102 fn. 26, 108, 111,
Naturally ordered instants 65 118–9, 125, 127, 148, 150, 165, 174, 176, 179
Instant of corruption 13–5, 78, 177 Miraculous 58, 84, 112, 135, 137, 175
Instant of generation 78, 108, 177 Mixture 2, 3, 4, 125–44 180
Instant of nature 65, 65 fn. 7 Form of the mixture 85 fn. 31, 96
Temporal instant 22, 65, 67 Homogeneous mixture 144, 170, 179
Jesus Christ 150 Mole, see Mr. Mole,
Index 195
Monism 148, 151 Opposite 9, 9 fn. 4, 10, 130

Morphé, see Form Power for opposites 15 fn. 15
Motion Opposition
Elemental motion 137, 180 Logical opposition 10
Local motion 53–4, 58 Metaphysical opposition 10
Non-natural motion 180 Opposition of elements 139
Rectilinear motion 163 Opposition of forms 12, 16
Mr. Mole 1, 5, 41, 43–4, 50, 56, 78, 81–4, Organ Donation 177
94–5, 99, 101, 103, 105–8, 110–1, 122–3, 125, Organism 2, 4–5, 43, 45, 76–9, 80, 82–5,
135, 146–8, 150–2, 159, 172–8, 180–2 92, 104–8, 110–1, 115, 118, 121–2
Muñoz Garcia, Angel 80 fn. 14 Organization 1, 99, 165
Natural order, also see Instant, of Paasch, JT 20 fn. 23, 21 fn. 24, 22 fn. 28
nature 65, 72, Parmenides 8
Nature Parsimony 59, 81, 132, 145 fn. 2
Author of nature 83 Part(s)
Divine nature 119 Actual parts 7, fn. 2, 117, 122, 140
Human nature 119, 121, 150 Arbitrary parts 121–24
Individual substance nature 119–20 Essential parts 2–3, 41–2, 47–8,
Nature (in general) 6 52, 61, 68–70, 72, 74, 76–8, 92, 114, 117,
Nature of accidents 84, 175 121, 163
Nature of causal power 19–21 Extended parts 4, 41
Nature of change 9 Form as part 1, 20, 31, 35 fn. 29, 41,
Nature of elements 128 43–8, 66, 71, 74, 153
Nature of form 16, 71 Heterogeneous parts 79, 90 fn. 40, 123,
Nature of prime matter 7, 12–3, 18, 140, 142–43, 160
33–4, 39–40, 71, 169 Homogenous parts 79, 90 fn. 40, 118,
Nature of substance 32 123, 123 fn. 27, 142–43, 160, 176
Necessary 3, 7–9, 11–2, 24–5, 39, 47, 54–5, Integral parts 2, 41, 54, 76–7, 80–1, 84,
59, 61, 64–5, 91, 93, 95, 110, 118, 122, 129, 131, 90 fn. 40; 92, 96, 98, 106, 108, 110–11, 114,
134, 136, 142, 146, 148–9, 160–1, 164, 171, 173 116, 125, 177
Necessity 114, 132, 154 Logical parts 48
Noone, Timothy 33 fn. 25 Material parts 1, 4, 41, 97–8,
Normore, Calvin 22 fn. 27, 41 fn. 1, 65 fn. 153, 160
7, 77 fn. 2, 117, 117 fn. 18, 118, 161 fn. 22 Matter as part 11, 13–14, 16, 18, 24, 26,
28, 31, 33, 35 fn. 29, 39, 47–8, 66, 71, 74,
Objective potency, see Potency, Objective 153, 166
Ockham’s Razor, see Parsimony Parts of a continuum 117, 118 fn. 18,
Ockham, William 27, 42–3, 46, 51–62, Parts Of an aggregate 48–9, 52, 94,
70–1, 78–9, 111, 115–8, 120–1, 129, 162–3, 166 97–8
Olivi, Peter John 79 fn. 14 Parts Of an essence 46–7
Ontology 3, 27, 126 Parts Of a unity of inherence 95
Operation 5, 86, 161–3 Parts Of form 116
Diversity of operations in human Parts Of matter 86–8, 116, 129–30, 152,
beings 142 180 fn. 10
Operation discloses form 161 Parts Of substance 111–12, 151
Operation of intellective soul 26, 91 Parts Organic 5–6, 26, 76, 85–6, 90–3,
Operation of whole not reducible to 96–8, 98 fn. 7, 100, 103, 106, 115, 135–36,
operations of parts 148, 163 138, 142–43, 146, 167
196 Index
Part(s) (cont.) Aquinas’s objection to pluralism 90

Part outside of part 27 fn. 40
Part-substance 4, 77, 79, 85, 94, 96, 98 Ockham’s pluralism 117
fn. 7, 99, 103–5, 108, 120–21, 126, 135–36, Scotistic pluralism 5, 42 fn. 5, 79, 81,
146–51, 155, 157–58, 164, 166, 168–69, 94, 105, 114–5, 132, 158, 166–9, 177–8
176, 178–81 Standard pluralism 79, 84, 86, 105,
Passive power as part 20, 25 166–9, 176–7
Predication of a property of the whole to a Porphyrian Tree 56 fn. 46, 90 fn. 40
part, see Denominative predication Possible
Parts prior to whole 2 Co-location possible by divine
Proper parts 181 fn. 11 power 134
Quantitative parts 43, fn. 8; 126, 128 Logically possible for essentially prior exist
Relative parts 145 fn. 1, without essentially posterior 36
Substantial parts 43, fn. 8 Logically possible for matter to exist
Parthood relations, see Relation, of part to without form 36
whole Not possible for species to be
Particle 125, 181 fn. 11 corrupted 160
Pasnau, Robert 76 fn. 2, 80 fn. 14, Not possible for substance to exist without
122 fn. 26 causality of matter and form 47
Passion 12, 163 Possible being 27
Patient Possible existence 19, 28
Elements as patient 141 Possible for matter and form to exist and
Patient of change 13–4 not compose a substance 51
Patient on which an agent acts 9–18 Possible substancehood of the whole
Perfection world 150
Essential order of Possibility
eminence/perfection 35, 99 Logical possibility 36
Essential perfection 82 Necessity implies possibility 114
Many perfections virtually contained in Possibility of change as motivation for
one form 85–7, 89, 152 hylomorphism 153–4
Repugnant perfections 87–90, Synchronic possibility 22 fn. 27
152, 156 Posterior
Unity of perfection 96–7, 99 Essentially posterior 36, 99–100
Persistence Naturally posterior 19, 23
Persistence of body through death 79 Relation of whole to part naturally
Persistence of matter through posterior to relation of part to
change 3, 7, 8, 12–3, 16–8, 28, 30, whole 65
36, 40, 43, 169, 172 Relations in matter and form
Persistence of relations through naturally posterior to
death 106 composite 56, 64–5
Person 113–4, 119–21, 175 Potency
Divine person 30 fn. 15, 116 fn. 14, Adequate potency 97
119–20 Immediate potency 25
Physics 1–2 Juxtaposed elements not in potency to
Place 53–4, 58, 134–5 substantial form 135–6
Places of elements 126, 128, 139, 180 Minimum level of potentiality required for
Plato 64, 121 membership in a kind 182
Pluralism (about substantial form) Obediential potency 8, 23–26
105, 173 Objective potency 28–30, 40
Index 197
Passive potency 24–5 Individuals are prior to their relations to

Potency as a mode of being 19 other individuals 57
Potency as a principle 19 Matter is actual prior to
Potency of matter 8, 26, 51–2 form 57
Pure potency 40, 51 fn. 33, 172, 179 Matter prior to form in essential order of
Several substances together in potency to dependence 35–6
substantial form 3, 81, 97–8, Naturally prior 20, 64–5, 72
146, 158 Parts prior to whole 2
Special potency question, see Special Powers prior to their effects 21
Potency Question Temporally prior 80, 115
Subjective potency 27–30, 40 The essentially prior cannot exist without
The world in potency to a soul? 147 the essentially posterior 36
Potential Principiation 72
Potential effect of a power 21 Principle
Potential of a plurality of substances to Aristotelian principle about relational
compose one substance 4, 127 change 62–4, 66–8
Potential relation 21 Contingently informed principle
The sense in which prime matter is a 169
potential entity 29 Passive power as a principle 20
Unity of order required for several things Homonymy principle 5, 170, 182
potentially to compose one Principle of nature 89, 140 fn. 33
substance 99 Potency as a principle 19
Potentiality, see Potency Principle of charity 36
Power Principle of parsimony, see
Active power 11, 13 Parsimony
Causal power 15–9, 25 Principle of unity 2–3, 44–5, 153
Creation of prime matter not within the Substance in the sense of a principle of an
power of creatures 38, 58 ens per se 112
Divine power 2–3, 25, 27–8, 39, 41, 49, Process 178
110, 113, 180 Infinite process, see Infinite, Regress
Matter as a power 27 Process of generation 10, 77,
Natural power 10 fn. 6 107–8, 138
Passive power 3, 7–8, 11, 13, Production, 153
18–23, 40 Four causes essentially ordered in
Power of substantial form of an production of their joint effect 99
element 131 Production ex nihilo 13
Powers as absolute entities 23 Production of a corpse 83
Rational power 10 fn. 6 Production of a material substance
Will power 10 fn. 7 13, 16, 58, 71
Presence Production of a power’s effect 21
Presence of a patient weakens the power Production of an absolute
of an agent 16 form 54, 58
Presence of form in matter 6 Production of an entity explains changes
Presence sicut in effecti communi 141 in relations 3, 42, 71
Virtual presence 141 Production of an entity is a kind of
Prior change 53
Essentially prior 66–7, 99–100, 105 Production of an essential part necessary
Form prior to matter in essential order of for generation 52 fn. 35
eminence 35–6 Production of matter 39
198 Index
Property 24 fn. 33, 41, 80, 86, 109 fn. 39, Mutual relation 19, 23, 42 fn. 3, 66,
120, 134, 155, 161, 176 68–9
Contingent property 116, 124 Parthood, see Relation of part to
Essential property 78, 172 whole
Modal property 80, 153 Relation of part to whole 41–2, 63–71,
Opposition between properties 10 150 fn. 9, 167
Property in the technical sense of Transitivity of relation of part to
proprium 3, 125, 141–2, 162 whole 166, 181 fn. 11
Relative property 41 fn. 3 Relation of whole to part 46, 65
Totality relation 65–6, 68–9
Qualitative Union relation 64–5, 67–9
Qualitative accident 141 Relative 20, 41, 41 fn. 3, 45, 49, 50, 54–8,
Qualitative change 10 fn. 7, 89 61–2, 66, 69–71, 73, 100, 106, 106 fn. 30, 145
Qualitative form 27, 102, 113, 129 fn. 1
Qualitative feature 82 Reproduction 30
Quality 26, 91 Repugnant 52, 87–9, 120, 152, 156, 161
Category of quality 41 fn. 3 Richard of Middleton 79, 139, 142
Intermediary quality 130 Risibility 142, 162–3
Quality depends on quantity 102–3
Quality of an element 139, 141, 143 Sameness
Quantitative Numerical sameness 79
Quantitative form(s) 27, 64, 113, 179–80 Specific sameness 79
Argument from quantitative Sandbach, F.H. 147 fn. 3
forms 132–137 Schaffer, Jonathan 151 fn. 13
Quantitative part(s), see Parts, quantitative Sedley, D.N. 147 fn. 3,
Quantitative totality 87–89, 152 Sensitive, see Soul, sensitive
Quantity Separability
Category of quantity 41 fn. 3 Particular separability 34–39
Form of quantity 27, 102, 129–30, Total separability 3, 34–40, 53, 72
133–34, 180 fn. 10 Sequence 107–8
Quantity as foundation of a relation 56 Sharp, Dorothea 77 fn. 3, 79 fn. 11
Quantity cause of quality 102, fn. 26 Shields, Christopher 165 fn. 1
Quantity dependent on matter for Simons, Peter 49 fn. 29, 75 fn. 21
existence 84 Simple 1, 49, 110, 131, 168
Quality dependent on quantity 102 Simplicity 8, 59, 120
Quantity subject of quality 102–3, Soul 1, 4, 8, 25–6, 28, 31, 41, 43–6, 48, 53,
fn. 26 58, 64, 77–9, 80 fn. 14, 81–2, 84–7, 90 fn. 40,
91, 93–6, 98, 106–9, 116–8, 121
Rational, see Soul, rational Appetitive soul, see Vegetative
Rea, Michael 153 fn. 14, Intellective soul, see Rational
Relation Rational soul 26, 48, 53, 58, 79, 82 fn.
Background on medieval views about 35, 91, 116–8, 121
relation 41 fn. 3 Sensitive soul 41, 43, 48, 56 fn. 46,
Causal relation 21, 41–2, 62–3, 65, 78–9, 81, 86–7, 94–6, 107–8, 117, 168, 172,
68–9, 72 fn. 19, 100, 106, 150 176
Co-causal relation 62–3, 66–7, 69 Vegetative soul 79, 94, 167–8
Extrinsic relation 64, 66 Special Composition Question 44, 94
Intrinsic intrinsic 64 Special Potency Question 94–5, 106
Index 199
Species 1, 26, 32, 46, 55, 91, 112, 118, 120–1, The Philosopher, see Aristotle
129, 131–2, 136, 159, 160–2 Thisness, see Haeceeity
Stella, Prospero 76 fn. 2, Tree of Porhphyry, see Porphyrian Tree
Stone, Abraham 128 fn. 9 Trifogli, Cecilia 126 fn. 1
Structure 1, 6, 42, 49, 67, 84, 86–7, 89, 90 Trinity 30 fn. 15, 120
fn. 40, 126 Two Tables 127
Suárez, Francisco 83, 128 fn. 7
Subject 3–4, 11–2, 16, 21, 23, 25–30, 33, 34 Unitarianism (about substantial form)
fn. 24, 37, 40, 41 fn. 3, 64, 80, 95, 97, 103 fn. 78–9, 81–3, 86–7, 106, 108, 115, 151, 161, 166,
26, 107, 109, 112–3, 120, 133, 146, 150 fn. 9, 168–9, 171, 173–7, 181
152, 154, 162, 172, 179–80 Unity 74, 94, 127, 153
Subjective potency, see Potency, Subjective Accidental unity 51, 64
Substance Degrees of unity 48–51, 94–5
Composite substance 3, 38–9, 48, 55, Diachronic unity 154
68, 76, 78–9, 81, 114, 116, 133, 135, 162 Different kinds of unity 45
Immaterial substance 76 fn. 1 Elements lack the unity needed to
Material substance 2, 4, 7, 11, 16, 24, 33, compose a substance 5
42, 44, 46–9, 51, 56, 58, 61, 72, 76, 82–3, Essential unity 33 fn. 24
99, 101, 109–10, 125–8, 132–3, 139, 144, Principle of unity 2, 44
151, 155, 165 Proper unity 96–7
Part-substance, see Part, part-substance Substantial unity 1, 41, 50–1, 74, 80, 92,
Primary substance 119 95–6, 108, 163
Secondary substance 119 Synchronic unity 154
Substance really distinct from parts Unity of aggregate 51, 180
58, 72, 74, 109 fn. 39 Unity of inherence 95
World-substance 5, 146, 158 Unity of matter and form 41 fn. 2, 45
Substratum 27, 78, 172 fn. 14, 52
Subtle Doctor, see Scotus, John Duns Unity of order 4, 45, 81, 87, 93–4, 97–8,
Supposit 4, 111, 113–4, 116–8, 121 100, 103, 105–6, 136, 145–6, 158, 164, 180
Christ’s human nature not a Unity of perfection 96–7
supposit 120 Unity of the world 145 fn. 1, 148
Created substance only contingently a Unity per se 48–9, 115
supposit 4, 110–1, 116, 120 Unity per accidens 49
Distinction between substance and Universe 145 fn. 2, 150, 158–61
supposit 4, 115, 118–9
Elemental substance is a supposit van Inwagen Peter 44, 94, 121
135–6 Violence 139–40, 140 fn. 34, 180
Parts can become supposits 116 Virtual 86
Virtual containment 85–9, 152, 156
Taylor, A.E. 147 fn. 3 Virtual existence 125, 128 fn. 6, 131–2,
Teleological 141
Teleological autonomy 149–50 Virtual totality 87
Teleological instrumentality 149 Virtue 3, 11, 18, 24, 32, 44, 56, 63, 65, 86,
Term 104, 113, 128, 146, 153, 176
Term of a change 10, 12, 18, 30, 39, 137,
146, 150 fn. 8, Wadding, Luke 31, 31 fn. 20, 34, 133 135
Term of a relation 20, 23, 42 fn. 3, 55, Wasserman, Ryan 153 fn. 14
64, 101, 105 145 fn. 1 Weisheipl, James A. 76 fn. 2
200 Index
Weisberg, Michael 128 fn. 9 Relation of part to whole 65, 72

Wiggins, David 153 fn. 14 fn. 19
Williams, Scott M. 20 fn. 23 Relation of whole to part 46, 69
Williams, Thomas 31 fn. 20, 133 fn. 23 Structured whole 1
Whole 47, 85, 92 Whole body 177
Composite whole 2, 44, 46, 92, 135–6 Whole continuous substance 126
Different kinds of wholes 41, 49, 51, Whole identical with parts 41, 49
148, 150 Whole matter 97
Form of the whole 45–7 Whole substance 33, 47, 50,
Generation of whole 103, 115 54–5, 57, 61, 73, 92, 104, 109, 136, 166–7
Non-reducible operation of whole World as a whole 5, 89, 145–7, 151–2,
148, 161 159–60
Parts for the sake of the whole 104 Wood, Rega 22 fn. 27, 128 fn. 9, 161 fn. 22
Parts have natural existence in the World 1–3, 5, 23, 89, 99, 117, 145 fn. 1,
whole 139 145–8, 150–2, 155–64
Parts prior to whole 2 Wyatt, Nicole 22 fn. 27
Real distinction of whole from parts
3, 70–2 Zavalloni, Roberto 76 fn. 2, 79 fn. 11

Duns Scotto

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Duns Scotto

Uploaded by

Copyright:

Available Formats

John Duns Scotus on Parts, Wholes, and Hylomorphism

Investigating Medieval Philosophy

The titles published in this series are listed at brill.com/imp

Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands.

This book is printed on acid-free paper.

1 The Purpose of Prime Matter 6

2 The Ontology of Prime Matter 27

3 How Matter and Form Compose a Substance—Part I 41

4 How Matter and Form Compose a Substance—Part II 60

5 Scotistic Pluralism about Substantial Form—Part I 76

6 Scotistic Pluralism about Substantial Form—Part II 94

i The Special Potency Question 97

iii Essentially Ordered Unity 103

iv The Role of Soul in Scotus and Two Unitarians 106

7 Contingent Supposits and Contingent Substances 110

i Three Modes of per se Being 112

ii Ockham on the Distinction between Substance and Supposit 115

iii What’s Special about Supposits? 118

iv Arbitrary Part-Substances? 121

8 The Mereological Status of the Elements in a Mixture 125

i Mixed Opinions about Mixtures 127

ii The Argument from Quantitative Forms 132

iii The Generation and Corruption Argument 137

iv The Violence Argument 139

v Generation from the Elements 140

vi Mixtures and Organic Parts 142

9 Why the World is not a Substance 145

i Motivating Monism 148

ii The Argument from the Distinguishing of Forms 151

iii The World/Organism Analogy 158

10 Scotistic Hylomorphism and the Problem of Homonymy 165

i Scotistic Hylomorphism Among Other Varieties 165

ii Ackrill’s Problem 169

iii Aquinas’s Response 171

iv The Standard Pluralistic Response 173

v Scotus’s Response 177

vi Scotus, Aquinas, and the Ultimate Subject of Substantial Change 178

vii Faulty Metaphysics or Faulty Chemistry? Scotistic Hylomorphism

Reportatio Duns Scotus, Reportatio Parisiensis

Let me introduce you to Mr. Mole, or Mole, for short. He is an animal, an

© koninklijke brill nv, leiden, 2014 | doi 10.1163/9789004278974_002

broader appeal. Hylomorphism’s history and Duns Scotus’s place in it should

The Purpose of Prime Matter

i Distinguishing Matter from Form

For hylomorphists, that in reality which corresponds to our experience of the

© koninklijke brill nv, leiden, 2014 | doi 10.1163/9789004278974_003

In the second sense, matter, or prime matter, is an unobservable theoretical

independent of assumptions about matter’s persistence, and his argument for

ii Motivating Matter: The Argument from Change

What is prime matter supposed to do? What philosophical problem does it

existence at the moment it looks as if they come from another (al-Ghāzāli,

[Aristotle’s] argument is formed in this way:

Scotus takes [A1] to be obvious on empirical grounds. Natural agents produce

iii Why Must Matter Persist through Change?

persists through change.10 Instead, it merely entails that there must be

[B2] In the instant of corruption the agent generates ex nihilo,12

[B2*] In the instant of corruption the agent generates ex nihilo2.

[B3] No agent can generate ex nihilo2 without also generating ex nihilo1.

Since no one would be willing to attribute generation ex nihilo1 to a merely

[B5] A natural15 agent necessarily produces its effect unless there is an

absence of some contrary patient (but rather it is weakened through its

It is clear that the argument depends on an understanding of “impediment”

[B2] In the instant of corruption the agent generates ex nihilo.

But there are two senses of generation ex nihilo, namely,